From 08ec8c3878cea0bf91f2ba3c0badf44b383752d0 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 16 Jan 2009 11:57:00 -0500 Subject: jbd2: On a __journal_expect() assertion failure printk "JBD2", not "EXT3-fs" Otherwise it can be very confusing to find a "EXT3-fs: " failure in the middle of EXT4-fs failures, and it makes it harder to track the source of the failure. Signed-off-by: "Theodore Ts'o" --- include/linux/jbd2.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index b45109c61fba..b28b37eb11c6 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -308,7 +308,8 @@ void buffer_assertion_failure(struct buffer_head *bh); int val = (expr); \ if (!val) { \ printk(KERN_ERR \ - "EXT3-fs unexpected failure: %s;\n",# expr); \ + "JBD2 unexpected failure: %s: %s;\n", \ + __func__, #expr); \ printk(KERN_ERR why "\n"); \ } \ val; \ -- cgit v1.2.3-70-g09d2 From 57064d213d2e44654d4f13c66df135b5e7389a26 Mon Sep 17 00:00:00 2001 From: Seth Heasley Date: Fri, 23 Jan 2009 12:43:38 -0800 Subject: PCI: irq and pci_ids patch for Intel Tigerpoint DeviceIDs This patch adds the Intel Tigerpoint LPC Controller DeviceIDs. Signed-off-by: Seth Heasley Signed-off-by: Jesse Barnes --- arch/x86/pci/irq.c | 1 + include/linux/pci_ids.h | 1 + 2 files changed, 2 insertions(+) (limited to 'include') diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c index 4064345cf144..fecbce6e7d7c 100644 --- a/arch/x86/pci/irq.c +++ b/arch/x86/pci/irq.c @@ -572,6 +572,7 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route case PCI_DEVICE_ID_INTEL_ICH7_1: case PCI_DEVICE_ID_INTEL_ICH7_30: case PCI_DEVICE_ID_INTEL_ICH7_31: + case PCI_DEVICE_ID_INTEL_TGP_LPC: case PCI_DEVICE_ID_INTEL_ESB2_0: case PCI_DEVICE_ID_INTEL_ICH8_0: case PCI_DEVICE_ID_INTEL_ICH8_1: diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index d56ad9c21c09..6f260b50253b 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2421,6 +2421,7 @@ #define PCI_DEVICE_ID_INTEL_ICH7_0 0x27b8 #define PCI_DEVICE_ID_INTEL_ICH7_1 0x27b9 #define PCI_DEVICE_ID_INTEL_ICH7_30 0x27b0 +#define PCI_DEVICE_ID_INTEL_TGP_LPC 0x27bc #define PCI_DEVICE_ID_INTEL_ICH7_31 0x27bd #define PCI_DEVICE_ID_INTEL_ICH7_17 0x27da #define PCI_DEVICE_ID_INTEL_ICH7_19 0x27dd -- cgit v1.2.3-70-g09d2 From 1cf3eb2ff6b0844c678f2f48d0053b9d12b7da67 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Tue, 27 Jan 2009 23:48:59 +0200 Subject: kmalloc: return NULL instead of link failure The SLAB kmalloc with a constant value isn't consistent with the other implementations because it bails out with __you_cannot_kmalloc_that_much rather than returning NULL and properly allowing the caller to fall back to vmalloc or take other action. This doesn't happen with a non-constant value or with SLOB or SLUB. Starting with 2.6.28, I've been seeing build failures on s390x. This is due to init_section_page_cgroup trying to allocate 2.5MB when the max size for a kmalloc on s390x is 2MB. It's failing because the value is constant. The workarounds at the call size are ugly and the caller shouldn't have to change behavior depending on what the backend of the API is. So, this patch eliminates the link failure and returns NULL like the other implementations. Signed-off-by: Jeff Mahoney Cc: Martin Schwidefsky Cc: Heiko Carstens Cc: Christoph Lameter Cc: Pekka Enberg Cc: Matt Mackall Cc: Nick Piggin Cc: [2.6.28.x] Signed-off-by: Andrew Morton Signed-off-by: Pekka Enberg --- include/linux/slab_def.h | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/linux/slab_def.h b/include/linux/slab_def.h index 39c3a5eb8ebe..6ca6a7b66d75 100644 --- a/include/linux/slab_def.h +++ b/include/linux/slab_def.h @@ -43,10 +43,7 @@ static inline void *kmalloc(size_t size, gfp_t flags) i++; #include #undef CACHE - { - extern void __you_cannot_kmalloc_that_much(void); - __you_cannot_kmalloc_that_much(); - } + return NULL; found: #ifdef CONFIG_ZONE_DMA if (flags & GFP_DMA) @@ -77,10 +74,7 @@ static inline void *kmalloc_node(size_t size, gfp_t flags, int node) i++; #include #undef CACHE - { - extern void __you_cannot_kmalloc_that_much(void); - __you_cannot_kmalloc_that_much(); - } + return NULL; found: #ifdef CONFIG_ZONE_DMA if (flags & GFP_DMA) -- cgit v1.2.3-70-g09d2 From 94cd3e6cbebf85903b4d53ed2147bdb4c6e08625 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 27 Jan 2009 17:45:10 -0800 Subject: net: wrong test in inet_ehash_locks_alloc() In commit 9db66bdcc83749affe61c61eb8ff3cf08f42afec (net: convert TCP/DCCP ehash rwlocks to spinlocks), I forgot to change one occurrence of rwlock_t to spinlock_t I believe sizeof(raw_spinlock_t) might be > 0 on !CONFIG_SMP if CONFIG_DEBUG_SPINLOCK while sizeof(raw_rwlock_t) should be 0 in this case. Fortunatly, CONFIG_DEBUG_SPINLOCK adds fields to both spinlock_t and rwlock_t, but at this might change in the future (being able to debug spinlocks but not rwlocks for example), better to be safe. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/inet_hashtables.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index f44bb5c77a70..d0a043153cc6 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -182,7 +182,7 @@ static inline int inet_ehash_locks_alloc(struct inet_hashinfo *hashinfo) size = 2048; if (nr_pcpus >= 32) size = 4096; - if (sizeof(rwlock_t) != 0) { + if (sizeof(spinlock_t) != 0) { #ifdef CONFIG_NUMA if (size * sizeof(spinlock_t) > PAGE_SIZE) hashinfo->ehash_locks = vmalloc(size * sizeof(spinlock_t)); -- cgit v1.2.3-70-g09d2 From f9129a2e537a4849c5194f98577bc274cda0d726 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Mon, 12 Jan 2009 18:12:13 -0300 Subject: V4L/DVB (10230): v4l2-device: fix buggy macro Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab --- include/media/v4l2-device.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/media/v4l2-device.h b/include/media/v4l2-device.h index 9bf4ccc93dbf..55e41afd95ef 100644 --- a/include/media/v4l2-device.h +++ b/include/media/v4l2-device.h @@ -94,16 +94,16 @@ void v4l2_device_unregister_subdev(struct v4l2_subdev *sd); /* Call the specified callback for all subdevs matching grp_id (if 0, then match them all). Ignore any errors. Note that you cannot add or delete a subdev while walking the subdevs list. */ -#define v4l2_device_call_all(dev, grp_id, o, f, args...) \ +#define v4l2_device_call_all(dev, grpid, o, f, args...) \ __v4l2_device_call_subdevs(dev, \ - !(grp_id) || sd->grp_id == (grp_id), o, f , ##args) + !(grpid) || sd->grp_id == (grpid), o, f , ##args) /* Call the specified callback for all subdevs matching grp_id (if 0, then match them all). If the callback returns an error other than 0 or -ENOIOCTLCMD, then return with that error code. Note that you cannot add or delete a subdev while walking the subdevs list. */ -#define v4l2_device_call_until_err(dev, grp_id, o, f, args...) \ +#define v4l2_device_call_until_err(dev, grpid, o, f, args...) \ __v4l2_device_call_subdevs_until_err(dev, \ - !(grp_id) || sd->grp_id == (grp_id), o, f , ##args) + !(grpid) || sd->grp_id == (grpid), o, f , ##args) #endif -- cgit v1.2.3-70-g09d2 From 7b24fc4d7eb611da367dea3aad45473050aacd6c Mon Sep 17 00:00:00 2001 From: "Martin K. Petersen" Date: Sun, 4 Jan 2009 02:43:38 -0500 Subject: block: Don't verify integrity metadata on read error If we get an I/O error on a read request there is no point in doing a verify pass on the integrity buffer. Adjust the completion path accordingly. Signed-off-by: Martin K. Petersen Signed-off-by: Jens Axboe --- fs/bio-integrity.c | 25 +++++++++++++++---------- include/linux/bio.h | 1 - 2 files changed, 15 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c index 77ebc3c263d6..8396d741f804 100644 --- a/fs/bio-integrity.c +++ b/fs/bio-integrity.c @@ -465,7 +465,7 @@ static int bio_integrity_verify(struct bio *bio) if (ret) { kunmap_atomic(kaddr, KM_USER0); - break; + return ret; } sectors = bv->bv_len / bi->sector_size; @@ -493,18 +493,13 @@ static void bio_integrity_verify_fn(struct work_struct *work) struct bio_integrity_payload *bip = container_of(work, struct bio_integrity_payload, bip_work); struct bio *bio = bip->bip_bio; - int error = bip->bip_error; + int error; - if (bio_integrity_verify(bio)) { - clear_bit(BIO_UPTODATE, &bio->bi_flags); - error = -EIO; - } + error = bio_integrity_verify(bio); /* Restore original bio completion handler */ bio->bi_end_io = bip->bip_end_io; - - if (bio->bi_end_io) - bio->bi_end_io(bio, error); + bio_endio(bio, error); } /** @@ -525,7 +520,17 @@ void bio_integrity_endio(struct bio *bio, int error) BUG_ON(bip->bip_bio != bio); - bip->bip_error = error; + /* In case of an I/O error there is no point in verifying the + * integrity metadata. Restore original bio end_io handler + * and run it. + */ + if (error) { + bio->bi_end_io = bip->bip_end_io; + bio_endio(bio, error); + + return; + } + INIT_WORK(&bip->bip_work, bio_integrity_verify_fn); queue_work(kintegrityd_wq, &bip->bip_work); } diff --git a/include/linux/bio.h b/include/linux/bio.h index 18462c5b8fff..18fc4a281a7b 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -312,7 +312,6 @@ struct bio_integrity_payload { void *bip_buf; /* generated integrity data */ bio_end_io_t *bip_end_io; /* saved I/O completion fn */ - int bip_error; /* saved I/O error */ unsigned int bip_size; unsigned short bip_pool; /* pool the ivec came from */ -- cgit v1.2.3-70-g09d2 From 16642eb68216d8e0e136a99e514e9166e7125838 Mon Sep 17 00:00:00 2001 From: Alberto Bertogli Date: Mon, 5 Jan 2009 10:18:53 +0100 Subject: Fix small typo in bio.h's documentation Signed-off-by: Alberto Bertogli Signed-off-by: Jens Axboe --- include/linux/bio.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/bio.h b/include/linux/bio.h index 18fc4a281a7b..5175aa3103c6 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -144,7 +144,7 @@ struct bio { * bit 1 -- rw-ahead when set * bit 2 -- barrier * Insert a serialization point in the IO queue, forcing previously - * submitted IO to be completed before this oen is issued. + * submitted IO to be completed before this one is issued. * bit 3 -- synchronous I/O hint: the block layer will unplug immediately * Note that this does NOT indicate that the IO itself is sync, just * that the block layer will not postpone issue of this IO by plugging. -- cgit v1.2.3-70-g09d2 From 213d9417fec62ef4c3675621b9364a667954d4dd Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 6 Jan 2009 09:16:05 +0100 Subject: block: seperate bio/request unplug and sync bits Signed-off-by: Jens Axboe --- block/blk-core.c | 5 ++++- include/linux/bio.h | 18 +++++++++++------- include/linux/blkdev.h | 2 ++ 3 files changed, 17 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/block/blk-core.c b/block/blk-core.c index a824e49c0d0a..9e2e86fb78b8 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -1125,6 +1125,8 @@ void init_request_from_bio(struct request *req, struct bio *bio) if (bio_sync(bio)) req->cmd_flags |= REQ_RW_SYNC; + if (bio_unplug(bio)) + req->cmd_flags |= REQ_UNPLUG; if (bio_rw_meta(bio)) req->cmd_flags |= REQ_RW_META; @@ -1141,6 +1143,7 @@ static int __make_request(struct request_queue *q, struct bio *bio) int el_ret, nr_sectors; const unsigned short prio = bio_prio(bio); const int sync = bio_sync(bio); + const int unplug = bio_unplug(bio); int rw_flags; nr_sectors = bio_sectors(bio); @@ -1244,7 +1247,7 @@ get_rq: blk_plug_device(q); add_request(q, req); out: - if (sync || blk_queue_nonrot(q)) + if (unplug || blk_queue_nonrot(q)) __generic_unplug_device(q); spin_unlock_irq(q->queue_lock); return 0; diff --git a/include/linux/bio.h b/include/linux/bio.h index 5175aa3103c6..f53568c5852e 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -163,12 +163,15 @@ struct bio { #define BIO_RW 0 /* Must match RW in req flags (blkdev.h) */ #define BIO_RW_AHEAD 1 /* Must match FAILFAST in req flags */ #define BIO_RW_BARRIER 2 -#define BIO_RW_SYNC 3 -#define BIO_RW_META 4 -#define BIO_RW_DISCARD 5 -#define BIO_RW_FAILFAST_DEV 6 -#define BIO_RW_FAILFAST_TRANSPORT 7 -#define BIO_RW_FAILFAST_DRIVER 8 +#define BIO_RW_SYNCIO 3 +#define BIO_RW_UNPLUG 4 +#define BIO_RW_META 5 +#define BIO_RW_DISCARD 6 +#define BIO_RW_FAILFAST_DEV 7 +#define BIO_RW_FAILFAST_TRANSPORT 8 +#define BIO_RW_FAILFAST_DRIVER 9 + +#define BIO_RW_SYNC (BIO_RW_SYNCIO | BIO_RW_UNPLUG) /* * upper 16 bits of bi_rw define the io priority of this bio @@ -194,7 +197,8 @@ struct bio { #define bio_segments(bio) ((bio)->bi_vcnt - (bio)->bi_idx) #define bio_sectors(bio) ((bio)->bi_size >> 9) #define bio_barrier(bio) ((bio)->bi_rw & (1 << BIO_RW_BARRIER)) -#define bio_sync(bio) ((bio)->bi_rw & (1 << BIO_RW_SYNC)) +#define bio_sync(bio) ((bio)->bi_rw & (1 << BIO_RW_SYNCIO)) +#define bio_unplug(bio) ((bio)->bi_rw & (1 << BIO_RW_UNPLUG)) #define bio_failfast_dev(bio) ((bio)->bi_rw & (1 << BIO_RW_FAILFAST_DEV)) #define bio_failfast_transport(bio) \ ((bio)->bi_rw & (1 << BIO_RW_FAILFAST_TRANSPORT)) diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 044467ef7b11..75426e4b8cdf 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -108,6 +108,7 @@ enum rq_flag_bits { __REQ_RW_META, /* metadata io request */ __REQ_COPY_USER, /* contains copies of user pages */ __REQ_INTEGRITY, /* integrity metadata has been remapped */ + __REQ_UNPLUG, /* unplug queue on submission */ __REQ_NR_BITS, /* stops here */ }; @@ -134,6 +135,7 @@ enum rq_flag_bits { #define REQ_RW_META (1 << __REQ_RW_META) #define REQ_COPY_USER (1 << __REQ_COPY_USER) #define REQ_INTEGRITY (1 << __REQ_INTEGRITY) +#define REQ_UNPLUG (1 << __REQ_UNPLUG) #define BLK_MAX_CDB 16 -- cgit v1.2.3-70-g09d2 From 1dfa17f4ab8543d82caf4d36636b93916a18f456 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 6 Jan 2009 09:21:49 +0100 Subject: block: add bio_rw_flagged() for testing bio->bi_rw The existing functions for checking bio->bi_rw are badly named. So lets mirror what we do for bio->bi_flags testing, use a properly named function so that it's immediately obvious what is being tested. Maintain compatability names for the old macros, eventually we'll get rid of these. Signed-off-by: Jens Axboe --- include/linux/bio.h | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/linux/bio.h b/include/linux/bio.h index f53568c5852e..0942765cf8c0 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -173,6 +173,24 @@ struct bio { #define BIO_RW_SYNC (BIO_RW_SYNCIO | BIO_RW_UNPLUG) +#define bio_rw_flagged(bio, flag) ((bio)->bi_rw & (1 << (flag))) + +/* + * Old defines, these should eventually be replaced by direct usage of + * bio_rw_flagged() + */ +#define bio_barrier(bio) bio_rw_flagged(bio, BIO_RW_BARRIER) +#define bio_sync(bio) bio_rw_flagged(bio, BIO_RW_SYNCIO) +#define bio_unplug(bio) bio_rw_flagged(bio, BIO_RW_UNPLUG) +#define bio_failfast_dev(bio) bio_rw_flagged(bio, BIO_RW_FAILFAST_DEV) +#define bio_failfast_transport(bio) \ + bio_rw_flagged(bio, BIO_RW_FAILFAST_TRANSPORT) +#define bio_failfast_driver(bio) \ + bio_rw_flagged(bio, BIO_RW_FAILFAST_DRIVER) +#define bio_rw_ahead(bio) bio_rw_flagged(bio, BIO_RW_AHEAD) +#define bio_rw_meta(bio) bio_rw_flagged(bio, BIO_RW_META) +#define bio_discard(bio) bio_rw_flagged(bio, BIO_RW_DISCARD) + /* * upper 16 bits of bi_rw define the io priority of this bio */ @@ -196,16 +214,6 @@ struct bio { #define bio_offset(bio) bio_iovec((bio))->bv_offset #define bio_segments(bio) ((bio)->bi_vcnt - (bio)->bi_idx) #define bio_sectors(bio) ((bio)->bi_size >> 9) -#define bio_barrier(bio) ((bio)->bi_rw & (1 << BIO_RW_BARRIER)) -#define bio_sync(bio) ((bio)->bi_rw & (1 << BIO_RW_SYNCIO)) -#define bio_unplug(bio) ((bio)->bi_rw & (1 << BIO_RW_UNPLUG)) -#define bio_failfast_dev(bio) ((bio)->bi_rw & (1 << BIO_RW_FAILFAST_DEV)) -#define bio_failfast_transport(bio) \ - ((bio)->bi_rw & (1 << BIO_RW_FAILFAST_TRANSPORT)) -#define bio_failfast_driver(bio) ((bio)->bi_rw & (1 << BIO_RW_FAILFAST_DRIVER)) -#define bio_rw_ahead(bio) ((bio)->bi_rw & (1 << BIO_RW_AHEAD)) -#define bio_rw_meta(bio) ((bio)->bi_rw & (1 << BIO_RW_META)) -#define bio_discard(bio) ((bio)->bi_rw & (1 << BIO_RW_DISCARD)) #define bio_empty_barrier(bio) (bio_barrier(bio) && !bio_has_data(bio) && !bio_discard(bio)) static inline unsigned int bio_cur_sectors(struct bio *bio) -- cgit v1.2.3-70-g09d2 From a229fc61ef0ee3c30fd193beee0eeb87410227f1 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Mon, 19 Jan 2009 10:37:38 +0100 Subject: include/linux: Add bsg.h to the Kernel exported headers bsg.h in current form is perfectly suitable for user-mode consumption. It is needed together with scsi/sg.h for applications that want to interface with the bsg driver. Currently the few projects that use it would copy it over into the projects. But that is not acceptable for projects that need to provide source and devel packages for distros. This should also be submitted to stable 2.6.28 and 2.6.27 since bsg had a stable API since these Kernels and distro users will need the header for these kernels a swell Signed-off-by: Boaz Harrosh Acked-by: FUJITA Tomonori CC: stable@kernel.org Signed-off-by: Jens Axboe --- include/linux/Kbuild | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 12e9a2957caf..2124c063a7ef 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -41,6 +41,7 @@ header-y += baycom.h header-y += bfs_fs.h header-y += blkpg.h header-y += bpqether.h +header-y += bsg.h header-y += can.h header-y += cdk.h header-y += chio.h -- cgit v1.2.3-70-g09d2 From bc58ba9468d94d62c56ab9b47173583ec140b165 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 23 Jan 2009 10:54:44 +0100 Subject: block: add sysfs file for controlling io stats accounting This allows us to turn off disk stat accounting completely, for the cases where the 0.5-1% reduction in system time is important. Signed-off-by: Jens Axboe --- block/blk-core.c | 90 ++++++++++++++++++++++++++++++-------------------- block/blk-sysfs.c | 28 ++++++++++++++++ include/linux/blkdev.h | 6 ++++ 3 files changed, 88 insertions(+), 36 deletions(-) (limited to 'include') diff --git a/block/blk-core.c b/block/blk-core.c index ae75c047f45d..ca69f3d94100 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -64,11 +64,12 @@ static struct workqueue_struct *kblockd_workqueue; static void drive_stat_acct(struct request *rq, int new_io) { + struct gendisk *disk = rq->rq_disk; struct hd_struct *part; int rw = rq_data_dir(rq); int cpu; - if (!blk_fs_request(rq) || !rq->rq_disk) + if (!blk_fs_request(rq) || !disk || !blk_queue_io_stat(disk->queue)) return; cpu = part_stat_lock(); @@ -599,8 +600,7 @@ blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id) q->request_fn = rfn; q->prep_rq_fn = NULL; q->unplug_fn = generic_unplug_device; - q->queue_flags = (1 << QUEUE_FLAG_CLUSTER | - 1 << QUEUE_FLAG_STACKABLE); + q->queue_flags = QUEUE_FLAG_DEFAULT; q->queue_lock = lock; blk_queue_segment_boundary(q, BLK_SEG_BOUNDARY_MASK); @@ -1663,6 +1663,55 @@ void blkdev_dequeue_request(struct request *req) } EXPORT_SYMBOL(blkdev_dequeue_request); +static void blk_account_io_completion(struct request *req, unsigned int bytes) +{ + struct gendisk *disk = req->rq_disk; + + if (!disk || !blk_queue_io_stat(disk->queue)) + return; + + if (blk_fs_request(req)) { + const int rw = rq_data_dir(req); + struct hd_struct *part; + int cpu; + + cpu = part_stat_lock(); + part = disk_map_sector_rcu(req->rq_disk, req->sector); + part_stat_add(cpu, part, sectors[rw], bytes >> 9); + part_stat_unlock(); + } +} + +static void blk_account_io_done(struct request *req) +{ + struct gendisk *disk = req->rq_disk; + + if (!disk || !blk_queue_io_stat(disk->queue)) + return; + + /* + * Account IO completion. bar_rq isn't accounted as a normal + * IO on queueing nor completion. Accounting the containing + * request is enough. + */ + if (blk_fs_request(req) && req != &req->q->bar_rq) { + unsigned long duration = jiffies - req->start_time; + const int rw = rq_data_dir(req); + struct hd_struct *part; + int cpu; + + cpu = part_stat_lock(); + part = disk_map_sector_rcu(disk, req->sector); + + part_stat_inc(cpu, part, ios[rw]); + part_stat_add(cpu, part, ticks[rw], duration); + part_round_stats(cpu, part); + part_dec_in_flight(part); + + part_stat_unlock(); + } +} + /** * __end_that_request_first - end I/O on a request * @req: the request being processed @@ -1698,16 +1747,7 @@ static int __end_that_request_first(struct request *req, int error, (unsigned long long)req->sector); } - if (blk_fs_request(req) && req->rq_disk) { - const int rw = rq_data_dir(req); - struct hd_struct *part; - int cpu; - - cpu = part_stat_lock(); - part = disk_map_sector_rcu(req->rq_disk, req->sector); - part_stat_add(cpu, part, sectors[rw], nr_bytes >> 9); - part_stat_unlock(); - } + blk_account_io_completion(req, nr_bytes); total_bytes = bio_nbytes = 0; while ((bio = req->bio) != NULL) { @@ -1787,8 +1827,6 @@ static int __end_that_request_first(struct request *req, int error, */ static void end_that_request_last(struct request *req, int error) { - struct gendisk *disk = req->rq_disk; - if (blk_rq_tagged(req)) blk_queue_end_tag(req->q, req); @@ -1800,27 +1838,7 @@ static void end_that_request_last(struct request *req, int error) blk_delete_timer(req); - /* - * Account IO completion. bar_rq isn't accounted as a normal - * IO on queueing nor completion. Accounting the containing - * request is enough. - */ - if (disk && blk_fs_request(req) && req != &req->q->bar_rq) { - unsigned long duration = jiffies - req->start_time; - const int rw = rq_data_dir(req); - struct hd_struct *part; - int cpu; - - cpu = part_stat_lock(); - part = disk_map_sector_rcu(disk, req->sector); - - part_stat_inc(cpu, part, ios[rw]); - part_stat_add(cpu, part, ticks[rw], duration); - part_round_stats(cpu, part); - part_dec_in_flight(part); - - part_stat_unlock(); - } + blk_account_io_done(req); if (req->end_io) req->end_io(req, error); diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index b538ab8dbbd1..e29ddfc73cf4 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -197,6 +197,27 @@ queue_rq_affinity_store(struct request_queue *q, const char *page, size_t count) return ret; } +static ssize_t queue_iostats_show(struct request_queue *q, char *page) +{ + return queue_var_show(blk_queue_io_stat(q), page); +} + +static ssize_t queue_iostats_store(struct request_queue *q, const char *page, + size_t count) +{ + unsigned long stats; + ssize_t ret = queue_var_store(&stats, page, count); + + spin_lock_irq(q->queue_lock); + if (stats) + queue_flag_set(QUEUE_FLAG_IO_STAT, q); + else + queue_flag_clear(QUEUE_FLAG_IO_STAT, q); + spin_unlock_irq(q->queue_lock); + + return ret; +} + static struct queue_sysfs_entry queue_requests_entry = { .attr = {.name = "nr_requests", .mode = S_IRUGO | S_IWUSR }, .show = queue_requests_show, @@ -249,6 +270,12 @@ static struct queue_sysfs_entry queue_rq_affinity_entry = { .store = queue_rq_affinity_store, }; +static struct queue_sysfs_entry queue_iostats_entry = { + .attr = {.name = "iostats", .mode = S_IRUGO | S_IWUSR }, + .show = queue_iostats_show, + .store = queue_iostats_store, +}; + static struct attribute *default_attrs[] = { &queue_requests_entry.attr, &queue_ra_entry.attr, @@ -259,6 +286,7 @@ static struct attribute *default_attrs[] = { &queue_nonrot_entry.attr, &queue_nomerges_entry.attr, &queue_rq_affinity_entry.attr, + &queue_iostats_entry.attr, NULL, }; diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 75426e4b8cdf..d08c4b8219a6 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -451,6 +451,11 @@ struct request_queue #define QUEUE_FLAG_STACKABLE 13 /* supports request stacking */ #define QUEUE_FLAG_NONROT 14 /* non-rotational device (SSD) */ #define QUEUE_FLAG_VIRT QUEUE_FLAG_NONROT /* paravirt device */ +#define QUEUE_FLAG_IO_STAT 15 /* do IO stats */ + +#define QUEUE_FLAG_DEFAULT ((1 << QUEUE_FLAG_IO_STAT) | \ + (1 << QUEUE_FLAG_CLUSTER) | \ + 1 << QUEUE_FLAG_STACKABLE) static inline int queue_is_locked(struct request_queue *q) { @@ -567,6 +572,7 @@ enum { #define blk_queue_stopped(q) test_bit(QUEUE_FLAG_STOPPED, &(q)->queue_flags) #define blk_queue_nomerges(q) test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags) #define blk_queue_nonrot(q) test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags) +#define blk_queue_io_stat(q) test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags) #define blk_queue_flushing(q) ((q)->ordseq) #define blk_queue_stackable(q) \ test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags) -- cgit v1.2.3-70-g09d2 From 94df7de0289bc2df3d6e85cd2ece52bf42682f45 Mon Sep 17 00:00:00 2001 From: Sebastien Dugue Date: Mon, 1 Dec 2008 14:09:07 +0100 Subject: hrtimers: allow the hot-unplugging of all cpus Impact: fix CPU hotplug hang on Power6 testbox On architectures that support offlining all cpus (at least powerpc/pseries), hot-unpluging the tick_do_timer_cpu can result in a system hang. This comes from the fact that if the cpu going down happens to be the cpu doing the tick, then as the tick_do_timer_cpu handover happens after the cpu is dead (via the CPU_DEAD notification), we're left without ticks, jiffies are frozen and any task relying on timers (msleep, ...) is stuck. That's particularly the case for the cpu looping in __cpu_die() waiting for the dying cpu to be dead. This patch addresses this by having the tick_do_timer_cpu handover happen earlier during the CPU_DYING notification. For this, a new clockevent notification type is introduced (CLOCK_EVT_NOTIFY_CPU_DYING) which is triggered in hrtimer_cpu_notify(). Signed-off-by: Sebastien Dugue Cc: Signed-off-by: Ingo Molnar --- include/linux/clockchips.h | 1 + kernel/hrtimer.c | 4 ++++ kernel/time/tick-common.c | 26 +++++++++++++++++++------- 3 files changed, 24 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h index cea153697ec7..3a1dbba4d3ae 100644 --- a/include/linux/clockchips.h +++ b/include/linux/clockchips.h @@ -36,6 +36,7 @@ enum clock_event_nofitiers { CLOCK_EVT_NOTIFY_BROADCAST_EXIT, CLOCK_EVT_NOTIFY_SUSPEND, CLOCK_EVT_NOTIFY_RESUME, + CLOCK_EVT_NOTIFY_CPU_DYING, CLOCK_EVT_NOTIFY_CPU_DEAD, }; diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index 8fea312ca36c..647a40e2fea1 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -1608,6 +1608,10 @@ static int __cpuinit hrtimer_cpu_notify(struct notifier_block *self, break; #ifdef CONFIG_HOTPLUG_CPU + case CPU_DYING: + case CPU_DYING_FROZEN: + clockevents_notify(CLOCK_EVT_NOTIFY_CPU_DYING, &scpu); + break; case CPU_DEAD: case CPU_DEAD_FROZEN: { diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index 63e05d423a09..21a5ca849514 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -273,6 +273,21 @@ out_bc: return ret; } +/* + * Transfer the do_timer job away from a dying cpu. + * + * Called with interrupts disabled. + */ +static void tick_handover_do_timer(int *cpup) +{ + if (*cpup == tick_do_timer_cpu) { + int cpu = cpumask_first(cpu_online_mask); + + tick_do_timer_cpu = (cpu < nr_cpu_ids) ? cpu : + TICK_DO_TIMER_NONE; + } +} + /* * Shutdown an event device on a given cpu: * @@ -297,13 +312,6 @@ static void tick_shutdown(unsigned int *cpup) clockevents_exchange_device(dev, NULL); td->evtdev = NULL; } - /* Transfer the do_timer job away from this cpu */ - if (*cpup == tick_do_timer_cpu) { - int cpu = cpumask_first(cpu_online_mask); - - tick_do_timer_cpu = (cpu < nr_cpu_ids) ? cpu : - TICK_DO_TIMER_NONE; - } spin_unlock_irqrestore(&tick_device_lock, flags); } @@ -357,6 +365,10 @@ static int tick_notify(struct notifier_block *nb, unsigned long reason, tick_broadcast_oneshot_control(reason); break; + case CLOCK_EVT_NOTIFY_CPU_DYING: + tick_handover_do_timer(dev); + break; + case CLOCK_EVT_NOTIFY_CPU_DEAD: tick_shutdown_broadcast_oneshot(dev); tick_shutdown_broadcast(dev); -- cgit v1.2.3-70-g09d2 From 0648e10d71c8e510d80772c4cb4220c97e9c34c7 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 2 Feb 2009 08:43:48 +0100 Subject: block: fix inconsistent parenthesisation of QUEUE_FLAG_DEFAULT Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index d08c4b8219a6..dcaa0fd84b02 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -455,7 +455,7 @@ struct request_queue #define QUEUE_FLAG_DEFAULT ((1 << QUEUE_FLAG_IO_STAT) | \ (1 << QUEUE_FLAG_CLUSTER) | \ - 1 << QUEUE_FLAG_STACKABLE) + (1 << QUEUE_FLAG_STACKABLE)) static inline int queue_is_locked(struct request_queue *q) { -- cgit v1.2.3-70-g09d2 From 20b636bf7c946da260391cd4570b16506f140a2c Mon Sep 17 00:00:00 2001 From: Alberto Bertogli Date: Mon, 2 Feb 2009 12:41:07 +0100 Subject: Fix misleading comment in bio.h The comment says "remember to add offset!", but the function already adds it. Signed-off-by: Alberto Bertogli Signed-off-by: Jens Axboe --- include/linux/bio.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/bio.h b/include/linux/bio.h index 0942765cf8c0..99728320cc05 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -451,8 +451,8 @@ extern struct biovec_slab bvec_slabs[BIOVEC_NR_POOLS] __read_mostly; #ifdef CONFIG_HIGHMEM /* - * remember to add offset! and never ever reenable interrupts between a - * bvec_kmap_irq and bvec_kunmap_irq!! + * remember never ever reenable interrupts between a bvec_kmap_irq and + * bvec_kunmap_irq! * * This function MUST be inlined - it plays with the CPU interrupt flags. */ -- cgit v1.2.3-70-g09d2 From c52440a69df22dca69794936a91e2fb529a707fb Mon Sep 17 00:00:00 2001 From: Alberto Bertogli Date: Mon, 2 Feb 2009 12:41:07 +0100 Subject: bio.h: If they MUST be inlined, then use __always_inline bvec_kmap_irq() and bvec_kunmap_irq() comments say they MUST be inlined, so mark them as __always_inline. Signed-off-by: Alberto Bertogli Signed-off-by: Jens Axboe --- include/linux/bio.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/bio.h b/include/linux/bio.h index 99728320cc05..2aa283ab062b 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -456,7 +456,8 @@ extern struct biovec_slab bvec_slabs[BIOVEC_NR_POOLS] __read_mostly; * * This function MUST be inlined - it plays with the CPU interrupt flags. */ -static inline char *bvec_kmap_irq(struct bio_vec *bvec, unsigned long *flags) +static __always_inline char *bvec_kmap_irq(struct bio_vec *bvec, + unsigned long *flags) { unsigned long addr; @@ -472,7 +473,8 @@ static inline char *bvec_kmap_irq(struct bio_vec *bvec, unsigned long *flags) return (char *) addr + bvec->bv_offset; } -static inline void bvec_kunmap_irq(char *buffer, unsigned long *flags) +static __always_inline void bvec_kunmap_irq(char *buffer, + unsigned long *flags) { unsigned long ptr = (unsigned long) buffer & PAGE_MASK; -- cgit v1.2.3-70-g09d2 From 0afd4a21ba7d75e93fa79cf05d7a21774e149c0f Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 2 Feb 2009 13:27:44 -0800 Subject: net: Fix userland breakage wrt. linux/if_tunnel.h Reported by Andrew Walrond Changeset c19e654ddbe3831252f61e76a74d661e1a755530 ("gre: Add netlink interface") added an include of linux/ip.h to linux/if_tunnel.h We can't really let that get exposed to userspace because this conflicts with types defined in netinet/ip.h which userland is almost certainly going to have included either explicitly or implicitly. So guard this include with a __KERNEL__ ifdef. Signed-off-by: David S. Miller --- include/linux/Kbuild | 2 +- include/linux/if_tunnel.h | 3 +++ 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 12e9a2957caf..6a9bb9753235 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -89,7 +89,6 @@ header-y += if_ppp.h header-y += if_slip.h header-y += if_strip.h header-y += if_tun.h -header-y += if_tunnel.h header-y += in_route.h header-y += ioctl.h header-y += ip6_tunnel.h @@ -235,6 +234,7 @@ unifdef-y += if_phonet.h unifdef-y += if_pppol2tp.h unifdef-y += if_pppox.h unifdef-y += if_tr.h +unifdef-y += if_tunnel.h unifdef-y += if_vlan.h unifdef-y += igmp.h unifdef-y += inet_diag.h diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h index aeab2cb32a9c..82c43624c067 100644 --- a/include/linux/if_tunnel.h +++ b/include/linux/if_tunnel.h @@ -2,7 +2,10 @@ #define _IF_TUNNEL_H_ #include + +#ifdef __KERNEL__ #include +#endif #define SIOCGETTUNNEL (SIOCDEVPRIVATE + 0) #define SIOCADDTUNNEL (SIOCDEVPRIVATE + 1) -- cgit v1.2.3-70-g09d2 From 720eba31f47aeade8ec130ca7f4353223c49170f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 3 Feb 2009 13:31:36 +1030 Subject: modules: Use a better scheme for refcounting Current refcounting for modules (done if CONFIG_MODULE_UNLOAD=y) is using a lot of memory. Each 'struct module' contains an [NR_CPUS] array of full cache lines. This patch uses existing infrastructure (percpu_modalloc() & percpu_modfree()) to allocate percpu space for the refcount storage. Instead of wasting NR_CPUS*128 bytes (on i386), we now use nr_cpu_ids*sizeof(local_t) bytes. On a typical distro, where NR_CPUS=8, shiping 2000 modules, we reduce size of module files by about 2 Mbytes. (1Kb per module) Instead of having all refcounters in the same memory node - with TLB misses because of vmalloc() - this new implementation permits to have better NUMA properties, since each CPU will use storage on its preferred node, thanks to percpu storage. Signed-off-by: Eric Dumazet Signed-off-by: Rusty Russell Signed-off-by: Linus Torvalds --- include/linux/module.h | 25 ++++++++++++++++--------- kernel/module.c | 35 +++++++++++++++++++++++++---------- 2 files changed, 41 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/linux/module.h b/include/linux/module.h index 4f7ea12463d3..f3b8329eb5b8 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -219,11 +219,6 @@ void *__symbol_get_gpl(const char *symbol); #endif -struct module_ref -{ - local_t count; -} ____cacheline_aligned; - enum module_state { MODULE_STATE_LIVE, @@ -344,8 +339,11 @@ struct module /* Destruction function. */ void (*exit)(void); - /* Reference counts */ - struct module_ref ref[NR_CPUS]; +#ifdef CONFIG_SMP + char *refptr; +#else + local_t ref; +#endif #endif }; #ifndef MODULE_ARCH_INIT @@ -395,13 +393,22 @@ void __symbol_put(const char *symbol); #define symbol_put(x) __symbol_put(MODULE_SYMBOL_PREFIX #x) void symbol_put_addr(void *addr); +static inline local_t *__module_ref_addr(struct module *mod, int cpu) +{ +#ifdef CONFIG_SMP + return (local_t *) (mod->refptr + per_cpu_offset(cpu)); +#else + return &mod->ref; +#endif +} + /* Sometimes we know we already have a refcount, and it's easier not to handle the error case (which only happens with rmmod --wait). */ static inline void __module_get(struct module *module) { if (module) { BUG_ON(module_refcount(module) == 0); - local_inc(&module->ref[get_cpu()].count); + local_inc(__module_ref_addr(module, get_cpu())); put_cpu(); } } @@ -413,7 +420,7 @@ static inline int try_module_get(struct module *module) if (module) { unsigned int cpu = get_cpu(); if (likely(module_is_live(module))) - local_inc(&module->ref[cpu].count); + local_inc(__module_ref_addr(module, cpu)); else ret = 0; put_cpu(); diff --git a/kernel/module.c b/kernel/module.c index e8b51d41dd72..ba22484a987e 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -573,13 +573,13 @@ static char last_unloaded_module[MODULE_NAME_LEN+1]; /* Init the unload section of the module. */ static void module_unload_init(struct module *mod) { - unsigned int i; + int cpu; INIT_LIST_HEAD(&mod->modules_which_use_me); - for (i = 0; i < NR_CPUS; i++) - local_set(&mod->ref[i].count, 0); + for_each_possible_cpu(cpu) + local_set(__module_ref_addr(mod, cpu), 0); /* Hold reference count during initialization. */ - local_set(&mod->ref[raw_smp_processor_id()].count, 1); + local_set(__module_ref_addr(mod, raw_smp_processor_id()), 1); /* Backwards compatibility macros put refcount during init. */ mod->waiter = current; } @@ -717,10 +717,11 @@ static int try_stop_module(struct module *mod, int flags, int *forced) unsigned int module_refcount(struct module *mod) { - unsigned int i, total = 0; + unsigned int total = 0; + int cpu; - for (i = 0; i < NR_CPUS; i++) - total += local_read(&mod->ref[i].count); + for_each_possible_cpu(cpu) + total += local_read(__module_ref_addr(mod, cpu)); return total; } EXPORT_SYMBOL(module_refcount); @@ -894,7 +895,7 @@ void module_put(struct module *module) { if (module) { unsigned int cpu = get_cpu(); - local_dec(&module->ref[cpu].count); + local_dec(__module_ref_addr(module, cpu)); /* Maybe they're waiting for us to drop reference? */ if (unlikely(!module_is_live(module))) wake_up_process(module->waiter); @@ -1464,7 +1465,10 @@ static void free_module(struct module *mod) kfree(mod->args); if (mod->percpu) percpu_modfree(mod->percpu); - +#if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP) + if (mod->refptr) + percpu_modfree(mod->refptr); +#endif /* Free lock-classes: */ lockdep_free_key_range(mod->module_core, mod->core_size); @@ -2011,6 +2015,14 @@ static noinline struct module *load_module(void __user *umod, if (err < 0) goto free_mod; +#if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP) + mod->refptr = percpu_modalloc(sizeof(local_t), __alignof__(local_t), + mod->name); + if (!mod->refptr) { + err = -ENOMEM; + goto free_mod; + } +#endif if (pcpuindex) { /* We have a special allocation for this section. */ percpu = percpu_modalloc(sechdrs[pcpuindex].sh_size, @@ -2018,7 +2030,7 @@ static noinline struct module *load_module(void __user *umod, mod->name); if (!percpu) { err = -ENOMEM; - goto free_mod; + goto free_percpu; } sechdrs[pcpuindex].sh_flags &= ~(unsigned long)SHF_ALLOC; mod->percpu = percpu; @@ -2282,6 +2294,9 @@ static noinline struct module *load_module(void __user *umod, free_percpu: if (percpu) percpu_modfree(percpu); +#if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP) + percpu_modfree(mod->refptr); +#endif free_mod: kfree(args); free_hdr: -- cgit v1.2.3-70-g09d2 From 2999b58b795ad81f10e34bdbbfd2742172f247e4 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Sun, 1 Feb 2009 20:46:39 +0400 Subject: ide/libata: fix ata_id_is_cfa() (take 4) When checking for the CFA feature set support, ata_id_is_cfa() tests bit 2 in word 82 of the identify data instead the word 83; it also checks the ATA/PI version support in the word 80 (which the CompactFlash specifications have as reserved), this having no slightest chance to work on the modern CF cards that don't have 0x848A in the word 0... Signed-off-by: Sergei Shtylyov Signed-off-by: Jeff Garzik --- include/linux/ata.h | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/ata.h b/include/linux/ata.h index a53318b8cbd0..08a86d5cdf1b 100644 --- a/include/linux/ata.h +++ b/include/linux/ata.h @@ -731,12 +731,17 @@ static inline int ata_id_current_chs_valid(const u16 *id) static inline int ata_id_is_cfa(const u16 *id) { - if (id[ATA_ID_CONFIG] == 0x848A) /* Standard CF */ + if (id[ATA_ID_CONFIG] == 0x848A) /* Traditional CF */ return 1; - /* Could be CF hiding as standard ATA */ - if (ata_id_major_version(id) >= 3 && - id[ATA_ID_COMMAND_SET_1] != 0xFFFF && - (id[ATA_ID_COMMAND_SET_1] & (1 << 2))) + /* + * CF specs don't require specific value in the word 0 anymore and yet + * they forbid to report the ATA version in the word 80 and require the + * CFA feature set support to be indicated in the word 83 in this case. + * Unfortunately, some cards only follow either of this requirements, + * and while those that don't indicate CFA feature support need some + * sort of quirk list, it seems impractical for the ones that do... + */ + if ((id[ATA_ID_COMMAND_SET_2] & 0xC004) == 0x4004) return 1; return 0; } -- cgit v1.2.3-70-g09d2 From 99cf610aa4840d822cdc67d194b23b55010ca9bd Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 29 Jan 2009 20:31:32 +0900 Subject: libata: clear dev->ering in smarter way dev->ering used to be cleared together with the rest of ata_device in ata_dev_init() which is called whenever a probing event occurs. dev->ering is about to be used to track probing failures so it needs to remain persistent over multiple porbing events. This patch achieves this by doing the following. * Instead of CLEAR_OFFSET, define CLEAR_BEGIN and CLEAR_END and only clear between BEGIN and END. ering is moved after END. The split of persistent area is to allow hotter items remain at the head. * ering is explicitly cleared on ata_dev_disable() and when device attach succeeds. So, ering is persistent throug a device's life time (unless explicitly cleared of course) and also through periods inbetween disablement of an attached device and successful detection of the next one. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 4 ++-- drivers/ata/libata-eh.c | 7 +++++++ include/linux/libata.h | 18 ++++++++++-------- 3 files changed, 19 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index d006e5c4768c..564c03c4ebb3 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -5404,8 +5404,8 @@ void ata_dev_init(struct ata_device *dev) dev->horkage = 0; spin_unlock_irqrestore(ap->lock, flags); - memset((void *)dev + ATA_DEVICE_CLEAR_OFFSET, 0, - sizeof(*dev) - ATA_DEVICE_CLEAR_OFFSET); + memset((void *)dev + ATA_DEVICE_CLEAR_BEGIN, 0, + ATA_DEVICE_CLEAR_END - ATA_DEVICE_CLEAR_BEGIN); dev->pio_mask = UINT_MAX; dev->mwdma_mask = UINT_MAX; dev->udma_mask = UINT_MAX; diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index aafe82bf5e2e..90092c1aae53 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -1194,6 +1194,11 @@ void ata_dev_disable(struct ata_device *dev) ata_acpi_on_disable(dev); ata_down_xfermask_limit(dev, ATA_DNXFER_FORCE_PIO0 | ATA_DNXFER_QUIET); dev->class++; + + /* From now till the next successful probe, ering is used to + * track probe failures. Clear accumulated device error info. + */ + ata_ering_clear(&dev->ering); } /** @@ -2765,6 +2770,8 @@ static int ata_eh_revalidate_and_attach(struct ata_link *link, readid_flags, dev->id); switch (rc) { case 0: + /* clear error info accumulated during probe */ + ata_ering_clear(&dev->ering); new_mask |= 1 << dev->devno; break; case -ENOENT: diff --git a/include/linux/libata.h b/include/linux/libata.h index bca3ba25f52a..9dcefee5843c 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -580,7 +580,7 @@ struct ata_device { acpi_handle acpi_handle; union acpi_object *gtf_cache; #endif - /* n_sector is used as CLEAR_OFFSET, read comment above CLEAR_OFFSET */ + /* n_sector is CLEAR_BEGIN, read comment above CLEAR_BEGIN */ u64 n_sectors; /* size of device, if ATA */ unsigned int class; /* ATA_DEV_xxx */ unsigned long unpark_deadline; @@ -605,20 +605,22 @@ struct ata_device { u16 heads; /* Number of heads */ u16 sectors; /* Number of sectors per track */ - /* error history */ - int spdn_cnt; - struct ata_ering ering; - union { u16 id[ATA_ID_WORDS]; /* IDENTIFY xxx DEVICE data */ u32 gscr[SATA_PMP_GSCR_DWORDS]; /* PMP GSCR block */ }; + + /* error history */ + int spdn_cnt; + /* ering is CLEAR_END, read comment above CLEAR_END */ + struct ata_ering ering; }; -/* Offset into struct ata_device. Fields above it are maintained - * acress device init. Fields below are zeroed. +/* Fields between ATA_DEVICE_CLEAR_BEGIN and ATA_DEVICE_CLEAR_END are + * cleared to zero on ata_dev_init(). */ -#define ATA_DEVICE_CLEAR_OFFSET offsetof(struct ata_device, n_sectors) +#define ATA_DEVICE_CLEAR_BEGIN offsetof(struct ata_device, n_sectors) +#define ATA_DEVICE_CLEAR_END offsetof(struct ata_device, ering) struct ata_eh_info { struct ata_device *dev; /* offending device */ -- cgit v1.2.3-70-g09d2 From 9062712fa9ed13b531dfc2228086650b8bd6a255 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 29 Jan 2009 20:31:36 +0900 Subject: libata: implement HORKAGE_1_5_GBPS and apply it to WD My Book 3Gbps is often much more prone to transmission failures. It's usually okay to let EH handle speed down after transmission failures but some WD My Book drives completely shutdown after certain transmission failures and after it only power cycling can revive them. Combined with the fact that external drives often end up with cable assembly which is longer than usual and more likely to have intervening gender, this makes these drives very likely to shutdown under certain configurations virtually rendering them unusable. This patch implements HOARKGE_1_5_GBPS and applies it to WD My Book such that 1.5Gbps is forced once the device is identified. Please take a look at the following bz for related reports. http://bugzilla.kernel.org/show_bug.cgi?id=9913 Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 41 +++++++++++++++++++++++++++++++++++++++++ include/linux/libata.h | 1 + 2 files changed, 42 insertions(+) (limited to 'include') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 3c5965d56c47..9fbf0595f3d4 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -2232,6 +2232,40 @@ retry: return rc; } +static int ata_do_link_spd_horkage(struct ata_device *dev) +{ + struct ata_link *plink = ata_dev_phys_link(dev); + u32 target, target_limit; + + if (!sata_scr_valid(plink)) + return 0; + + if (dev->horkage & ATA_HORKAGE_1_5_GBPS) + target = 1; + else + return 0; + + target_limit = (1 << target) - 1; + + /* if already on stricter limit, no need to push further */ + if (plink->sata_spd_limit <= target_limit) + return 0; + + plink->sata_spd_limit = target_limit; + + /* Request another EH round by returning -EAGAIN if link is + * going faster than the target speed. Forward progress is + * guaranteed by setting sata_spd_limit to target_limit above. + */ + if (plink->sata_spd > target) { + ata_dev_printk(dev, KERN_INFO, + "applying link speed limit horkage to %s\n", + sata_spd_string(target)); + return -EAGAIN; + } + return 0; +} + static inline u8 ata_dev_knobble(struct ata_device *dev) { struct ata_port *ap = dev->link->ap; @@ -2322,6 +2356,10 @@ int ata_dev_configure(struct ata_device *dev) return 0; } + rc = ata_do_link_spd_horkage(dev); + if (rc) + return rc; + /* let ACPI work its magic */ rc = ata_acpi_on_devcfg(dev); if (rc) @@ -4223,6 +4261,9 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { /* Devices that do not need bridging limits applied */ { "MTRON MSP-SATA*", NULL, ATA_HORKAGE_BRIDGE_OK, }, + /* Devices which aren't very happy with higher link speeds */ + { "WD My Book", NULL, ATA_HORKAGE_1_5_GBPS, }, + /* End Marker */ { } }; diff --git a/include/linux/libata.h b/include/linux/libata.h index 9dcefee5843c..5d87bc09a1f5 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -380,6 +380,7 @@ enum { ATA_HORKAGE_ATAPI_MOD16_DMA = (1 << 11), /* use ATAPI DMA for commands not multiple of 16 bytes */ ATA_HORKAGE_FIRMWARE_WARN = (1 << 12), /* firwmare update warning */ + ATA_HORKAGE_1_5_GBPS = (1 << 13), /* force 1.5 Gbps */ /* DMA mask for user DMA control: User visible values; DO NOT renumber */ -- cgit v1.2.3-70-g09d2 From 35626129abcd6a7547e84c817ef5b6eff7a8758b Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 2 Feb 2009 16:00:29 -0800 Subject: sched: add missing kernel-doc in sched.h Add kernel-doc notation for @lock: include/linux/sched.h:457: No description found for parameter 'lock' Signed-off-by: Randy Dunlap Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 5a7c76388731..2127e959e0f4 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -443,6 +443,7 @@ struct pacct_struct { * @utime: time spent in user mode, in &cputime_t units * @stime: time spent in kernel mode, in &cputime_t units * @sum_exec_runtime: total time spent on the CPU, in nanoseconds + * @lock: lock for fields in this struct * * This structure groups together three kinds of CPU time that are * tracked for threads and thread groups. Most things considering -- cgit v1.2.3-70-g09d2 From 97c44836cdec1ea713a15d84098a1a908157e68f Mon Sep 17 00:00:00 2001 From: "Timothy S. Nelson" Date: Fri, 30 Jan 2009 06:12:47 +1100 Subject: PCI: return error on failure to read PCI ROMs This patch makes the ROM reading code return an error to user space if the size of the ROM read is equal to 0. The patch also emits a warnings if the contents of the ROM are invalid, and documents the effects of the "enable" file on ROM reading. Signed-off-by: Timothy S. Nelson Acked-by: Alex Villacis-Lasso Signed-off-by: Jesse Barnes --- Documentation/filesystems/sysfs-pci.txt | 13 ++++++++++++- arch/ia64/sn/kernel/io_acpi_init.c | 2 +- arch/ia64/sn/kernel/io_init.c | 2 +- drivers/pci/pci-sysfs.c | 4 ++-- drivers/pci/rom.c | 8 +++++--- include/linux/pci.h | 2 +- 6 files changed, 22 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/Documentation/filesystems/sysfs-pci.txt b/Documentation/filesystems/sysfs-pci.txt index 68ef48839c04..9f8740ca3f3b 100644 --- a/Documentation/filesystems/sysfs-pci.txt +++ b/Documentation/filesystems/sysfs-pci.txt @@ -9,6 +9,7 @@ that support it. For example, a given bus might look like this: | |-- class | |-- config | |-- device + | |-- enable | |-- irq | |-- local_cpus | |-- resource @@ -32,6 +33,7 @@ files, each with their own function. class PCI class (ascii, ro) config PCI config space (binary, rw) device PCI device (ascii, ro) + enable Whether the device is enabled (ascii, rw) irq IRQ number (ascii, ro) local_cpus nearby CPU mask (cpumask, ro) resource PCI resource host addresses (ascii, ro) @@ -57,10 +59,19 @@ used to do actual device programming from userspace. Note that some platforms don't support mmapping of certain resources, so be sure to check the return value from any attempted mmap. +The 'enable' file provides a counter that indicates how many times the device +has been enabled. If the 'enable' file currently returns '4', and a '1' is +echoed into it, it will then return '5'. Echoing a '0' into it will decrease +the count. Even when it returns to 0, though, some of the initialisation +may not be reversed. + The 'rom' file is special in that it provides read-only access to the device's ROM file, if available. It's disabled by default, however, so applications should write the string "1" to the file to enable it before attempting a read -call, and disable it following the access by writing "0" to the file. +call, and disable it following the access by writing "0" to the file. Note +that the device must be enabled for a rom read to return data succesfully. +In the event a driver is not bound to the device, it can be enabled using the +'enable' file, documented above. Accessing legacy resources through sysfs ---------------------------------------- diff --git a/arch/ia64/sn/kernel/io_acpi_init.c b/arch/ia64/sn/kernel/io_acpi_init.c index c5a214026a77..d0223abbbbd4 100644 --- a/arch/ia64/sn/kernel/io_acpi_init.c +++ b/arch/ia64/sn/kernel/io_acpi_init.c @@ -443,7 +443,7 @@ sn_acpi_slot_fixup(struct pci_dev *dev) size = pci_resource_len(dev, PCI_ROM_RESOURCE); addr = ioremap(pcidev_info->pdi_pio_mapped_addr[PCI_ROM_RESOURCE], size); - image_size = pci_get_rom_size(addr, size); + image_size = pci_get_rom_size(dev, addr, size); dev->resource[PCI_ROM_RESOURCE].start = (unsigned long) addr; dev->resource[PCI_ROM_RESOURCE].end = (unsigned long) addr + image_size - 1; diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c index 4e1801bad83a..e2eb2da60f96 100644 --- a/arch/ia64/sn/kernel/io_init.c +++ b/arch/ia64/sn/kernel/io_init.c @@ -269,7 +269,7 @@ sn_io_slot_fixup(struct pci_dev *dev) rom = ioremap(pci_resource_start(dev, PCI_ROM_RESOURCE), size + 1); - image_size = pci_get_rom_size(rom, size + 1); + image_size = pci_get_rom_size(dev, rom, size + 1); dev->resource[PCI_ROM_RESOURCE].end = dev->resource[PCI_ROM_RESOURCE].start + image_size - 1; diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index db7ec14fa719..dfc4e0ddf241 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -768,8 +768,8 @@ pci_read_rom(struct kobject *kobj, struct bin_attribute *bin_attr, return -EINVAL; rom = pci_map_rom(pdev, &size); /* size starts out as PCI window size */ - if (!rom) - return 0; + if (!rom || !size) + return -EIO; if (off >= size) count = 0; diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c index 132a78159b60..29cbe47f219f 100644 --- a/drivers/pci/rom.c +++ b/drivers/pci/rom.c @@ -63,7 +63,7 @@ void pci_disable_rom(struct pci_dev *pdev) * The PCI window size could be much larger than the * actual image size. */ -size_t pci_get_rom_size(void __iomem *rom, size_t size) +size_t pci_get_rom_size(struct pci_dev *pdev, void __iomem *rom, size_t size) { void __iomem *image; int last_image; @@ -72,8 +72,10 @@ size_t pci_get_rom_size(void __iomem *rom, size_t size) do { void __iomem *pds; /* Standard PCI ROMs start out with these bytes 55 AA */ - if (readb(image) != 0x55) + if (readb(image) != 0x55) { + dev_err(&pdev->dev, "Invalid ROM contents\n"); break; + } if (readb(image + 1) != 0xAA) break; /* get the PCI data structure and check its signature */ @@ -159,7 +161,7 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size) * size is much larger than the actual size of the ROM. * True size is important if the ROM is going to be copied. */ - *size = pci_get_rom_size(rom, *size); + *size = pci_get_rom_size(pdev, rom, *size); return rom; } diff --git a/include/linux/pci.h b/include/linux/pci.h index 48890cf3f96e..7bd624bfdcfd 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -684,7 +684,7 @@ int pci_enable_rom(struct pci_dev *pdev); void pci_disable_rom(struct pci_dev *pdev); void __iomem __must_check *pci_map_rom(struct pci_dev *pdev, size_t *size); void pci_unmap_rom(struct pci_dev *pdev, void __iomem *rom); -size_t pci_get_rom_size(void __iomem *rom, size_t size); +size_t pci_get_rom_size(struct pci_dev *pdev, void __iomem *rom, size_t size); /* Power management related routines */ int pci_save_state(struct pci_dev *dev); -- cgit v1.2.3-70-g09d2 From 7b2cd92adc5430b0c1adeb120971852b4ea1ab08 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Thu, 5 Feb 2009 16:48:24 +1100 Subject: crypto: api - Fix zeroing on free Geert Uytterhoeven pointed out that we're not zeroing all the memory when freeing a transform. This patch fixes it by calling ksize to ensure that we zero everything in sight. Reported-by: Geert Uytterhoeven Signed-off-by: Herbert Xu --- crypto/api.c | 20 ++++++++++---------- include/linux/crypto.h | 7 ++++++- 2 files changed, 16 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/crypto/api.c b/crypto/api.c index 9975a7bd246c..efe77df6863f 100644 --- a/crypto/api.c +++ b/crypto/api.c @@ -557,34 +557,34 @@ err: return ERR_PTR(err); } EXPORT_SYMBOL_GPL(crypto_alloc_tfm); - + /* - * crypto_free_tfm - Free crypto transform + * crypto_destroy_tfm - Free crypto transform + * @mem: Start of tfm slab * @tfm: Transform to free * - * crypto_free_tfm() frees up the transform and any associated resources, + * This function frees up the transform and any associated resources, * then drops the refcount on the associated algorithm. */ -void crypto_free_tfm(struct crypto_tfm *tfm) +void crypto_destroy_tfm(void *mem, struct crypto_tfm *tfm) { struct crypto_alg *alg; int size; - if (unlikely(!tfm)) + if (unlikely(!mem)) return; alg = tfm->__crt_alg; - size = sizeof(*tfm) + alg->cra_ctxsize; + size = ksize(mem); if (!tfm->exit && alg->cra_exit) alg->cra_exit(tfm); crypto_exit_ops(tfm); crypto_mod_put(alg); - memset(tfm, 0, size); - kfree(tfm); + memset(mem, 0, size); + kfree(mem); } - -EXPORT_SYMBOL_GPL(crypto_free_tfm); +EXPORT_SYMBOL_GPL(crypto_destroy_tfm); int crypto_has_alg(const char *name, u32 type, u32 mask) { diff --git a/include/linux/crypto.h b/include/linux/crypto.h index 3bacd71509fb..1f2e9020acc6 100644 --- a/include/linux/crypto.h +++ b/include/linux/crypto.h @@ -552,7 +552,12 @@ struct crypto_tfm *crypto_alloc_tfm(const char *alg_name, const struct crypto_type *frontend, u32 type, u32 mask); struct crypto_tfm *crypto_alloc_base(const char *alg_name, u32 type, u32 mask); -void crypto_free_tfm(struct crypto_tfm *tfm); +void crypto_destroy_tfm(void *mem, struct crypto_tfm *tfm); + +static inline void crypto_free_tfm(struct crypto_tfm *tfm) +{ + return crypto_destroy_tfm(tfm, tfm); +} int alg_test(const char *driver, const char *alg, u32 type, u32 mask); -- cgit v1.2.3-70-g09d2 From 412e87ae5d852bc3d836f475c19d954b3324363d Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Thu, 5 Feb 2009 16:51:25 +1100 Subject: crypto: shash - Fix tfm destruction We were freeing an offset into the slab object instead of the start. This patch fixes it by calling crypto_destroy_tfm which allows the correct address to be given. Signed-off-by: Herbert Xu --- include/crypto/hash.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/crypto/hash.h b/include/crypto/hash.h index cd16d6e668ce..d797e119e3d5 100644 --- a/include/crypto/hash.h +++ b/include/crypto/hash.h @@ -222,7 +222,7 @@ static inline struct crypto_tfm *crypto_shash_tfm(struct crypto_shash *tfm) static inline void crypto_free_shash(struct crypto_shash *tfm) { - crypto_free_tfm(crypto_shash_tfm(tfm)); + crypto_destroy_tfm(tfm, crypto_shash_tfm(tfm)); } static inline unsigned int crypto_shash_alignmask( -- cgit v1.2.3-70-g09d2 From 32bd671d6cbeda60dc73be77fa2b9037d9a9bfa0 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 5 Feb 2009 12:24:15 +0100 Subject: signal: re-add dead task accumulation stats. We're going to split the process wide cpu accounting into two parts: - clocks; which can take all the time they want since they run from user context. - timers; which need constant time tracing but can affort the overhead because they're default off -- and rare. The clock readout will go back to a full sum of the thread group, for this we need to re-add the exit stats that were removed in the initial itimer rework (f06febc9: timers: fix itimer/many thread hang). Furthermore, since that full sum can be rather slow for large thread groups and we have the complete dead task stats, revert the do_notify_parent time computation. Signed-off-by: Peter Zijlstra Reviewed-by: Ingo Molnar Signed-off-by: Ingo Molnar --- include/linux/sched.h | 10 +++++++++- kernel/exit.c | 3 +++ kernel/fork.c | 3 ++- kernel/signal.c | 8 ++++---- 4 files changed, 18 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 2127e959e0f4..2e0646a30314 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -559,7 +559,7 @@ struct signal_struct { * Live threads maintain their own counters and add to these * in __exit_signal, except for the group leader. */ - cputime_t cutime, cstime; + cputime_t utime, stime, cutime, cstime; cputime_t gtime; cputime_t cgtime; unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw; @@ -567,6 +567,14 @@ struct signal_struct { unsigned long inblock, oublock, cinblock, coublock; struct task_io_accounting ioac; + /* + * Cumulative ns of schedule CPU time fo dead threads in the + * group, not including a zombie group leader, (This only differs + * from jiffies_to_ns(utime + stime) if sched_clock uses something + * other than jiffies.) + */ + unsigned long long sum_sched_runtime; + /* * We don't bother to synchronize most readers of this at all, * because there is no reader checking a limit that actually needs diff --git a/kernel/exit.c b/kernel/exit.c index f80dec3f1875..efd30ccf3858 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -118,6 +118,8 @@ static void __exit_signal(struct task_struct *tsk) * We won't ever get here for the group leader, since it * will have been the last reference on the signal_struct. */ + sig->utime = cputime_add(sig->utime, task_utime(tsk)); + sig->stime = cputime_add(sig->stime, task_stime(tsk)); sig->gtime = cputime_add(sig->gtime, task_gtime(tsk)); sig->min_flt += tsk->min_flt; sig->maj_flt += tsk->maj_flt; @@ -126,6 +128,7 @@ static void __exit_signal(struct task_struct *tsk) sig->inblock += task_io_get_inblock(tsk); sig->oublock += task_io_get_oublock(tsk); task_io_accounting_add(&sig->ioac, &tsk->ioac); + sig->sum_sched_runtime += tsk->se.sum_exec_runtime; sig = NULL; /* Marker for below. */ } diff --git a/kernel/fork.c b/kernel/fork.c index 242a706e7721..e8e854a04ad2 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -851,13 +851,14 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) sig->tty_old_pgrp = NULL; sig->tty = NULL; - sig->cutime = sig->cstime = cputime_zero; + sig->utime = sig->stime = sig->cutime = sig->cstime = cputime_zero; sig->gtime = cputime_zero; sig->cgtime = cputime_zero; sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0; sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0; sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0; task_io_accounting_init(&sig->ioac); + sig->sum_sched_runtime = 0; taskstats_tgid_init(sig); task_lock(current->group_leader); diff --git a/kernel/signal.c b/kernel/signal.c index b6b36768b758..2a74fe87c0dd 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1367,7 +1367,6 @@ int do_notify_parent(struct task_struct *tsk, int sig) struct siginfo info; unsigned long flags; struct sighand_struct *psig; - struct task_cputime cputime; int ret = sig; BUG_ON(sig == -1); @@ -1397,9 +1396,10 @@ int do_notify_parent(struct task_struct *tsk, int sig) info.si_uid = __task_cred(tsk)->uid; rcu_read_unlock(); - thread_group_cputime(tsk, &cputime); - info.si_utime = cputime_to_jiffies(cputime.utime); - info.si_stime = cputime_to_jiffies(cputime.stime); + info.si_utime = cputime_to_clock_t(cputime_add(tsk->utime, + tsk->signal->utime)); + info.si_stime = cputime_to_clock_t(cputime_add(tsk->stime, + tsk->signal->stime)); info.si_status = tsk->exit_code & 0x7f; if (tsk->exit_code & 0x80) -- cgit v1.2.3-70-g09d2 From 4cd4c1b40d40447fb5e7ba80746c6d7ba91d7a53 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 5 Feb 2009 12:24:16 +0100 Subject: timers: split process wide cpu clocks/timers Change the process wide cpu timers/clocks so that we: 1) don't mess up the kernel with too many threads, 2) don't have a per-cpu allocation for each process, 3) have no impact when not used. In order to accomplish this we're going to split it into two parts: - clocks; which can take all the time they want since they run from user context -- ie. sys_clock_gettime(CLOCK_PROCESS_CPUTIME_ID) - timers; which need constant time sampling but since they're explicity used, the user can pay the overhead. The clock readout will go back to a full sum of the thread group, while the timers will run of a global 'clock' that only runs when needed, so only programs that make use of the facility pay the price. Signed-off-by: Peter Zijlstra Reviewed-by: Ingo Molnar Signed-off-by: Ingo Molnar --- include/linux/init_task.h | 11 +++--- include/linux/sched.h | 54 +++++++++++++++------------ kernel/itimer.c | 4 +- kernel/posix-cpu-timers.c | 95 +++++++++++++++++++++++++++++++++++++++++++++-- kernel/sched_stats.h | 45 ++++++++++++---------- 5 files changed, 155 insertions(+), 54 deletions(-) (limited to 'include') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index ea0ea1a4c36f..e752d973fa21 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -48,12 +48,11 @@ extern struct fs_struct init_fs; .posix_timers = LIST_HEAD_INIT(sig.posix_timers), \ .cpu_timers = INIT_CPU_TIMERS(sig.cpu_timers), \ .rlim = INIT_RLIMITS, \ - .cputime = { .totals = { \ - .utime = cputime_zero, \ - .stime = cputime_zero, \ - .sum_exec_runtime = 0, \ - .lock = __SPIN_LOCK_UNLOCKED(sig.cputime.totals.lock), \ - }, }, \ + .cputimer = { \ + .cputime = INIT_CPUTIME, \ + .running = 0, \ + .lock = __SPIN_LOCK_UNLOCKED(sig.cputimer.lock), \ + }, \ } extern struct nsproxy init_nsproxy; diff --git a/include/linux/sched.h b/include/linux/sched.h index 2e0646a30314..082d7619b3a1 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -443,7 +443,6 @@ struct pacct_struct { * @utime: time spent in user mode, in &cputime_t units * @stime: time spent in kernel mode, in &cputime_t units * @sum_exec_runtime: total time spent on the CPU, in nanoseconds - * @lock: lock for fields in this struct * * This structure groups together three kinds of CPU time that are * tracked for threads and thread groups. Most things considering @@ -454,23 +453,33 @@ struct task_cputime { cputime_t utime; cputime_t stime; unsigned long long sum_exec_runtime; - spinlock_t lock; }; /* Alternate field names when used to cache expirations. */ #define prof_exp stime #define virt_exp utime #define sched_exp sum_exec_runtime +#define INIT_CPUTIME \ + (struct task_cputime) { \ + .utime = cputime_zero, \ + .stime = cputime_zero, \ + .sum_exec_runtime = 0, \ + } + /** - * struct thread_group_cputime - thread group interval timer counts - * @totals: thread group interval timers; substructure for - * uniprocessor kernel, per-cpu for SMP kernel. + * struct thread_group_cputimer - thread group interval timer counts + * @cputime: thread group interval timers. + * @running: non-zero when there are timers running and + * @cputime receives updates. + * @lock: lock for fields in this struct. * * This structure contains the version of task_cputime, above, that is - * used for thread group CPU clock calculations. + * used for thread group CPU timer calculations. */ -struct thread_group_cputime { - struct task_cputime totals; +struct thread_group_cputimer { + struct task_cputime cputime; + int running; + spinlock_t lock; }; /* @@ -519,10 +528,10 @@ struct signal_struct { cputime_t it_prof_incr, it_virt_incr; /* - * Thread group totals for process CPU clocks. - * See thread_group_cputime(), et al, for details. + * Thread group totals for process CPU timers. + * See thread_group_cputimer(), et al, for details. */ - struct thread_group_cputime cputime; + struct thread_group_cputimer cputimer; /* Earliest-expiration cache. */ struct task_cputime cputime_expires; @@ -2191,27 +2200,26 @@ static inline int spin_needbreak(spinlock_t *lock) /* * Thread group CPU time accounting. */ +void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times); static inline -void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) +void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) { - struct task_cputime *totals = &tsk->signal->cputime.totals; + struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; unsigned long flags; - spin_lock_irqsave(&totals->lock, flags); - *times = *totals; - spin_unlock_irqrestore(&totals->lock, flags); + WARN_ON(!cputimer->running); + + spin_lock_irqsave(&cputimer->lock, flags); + *times = cputimer->cputime; + spin_unlock_irqrestore(&cputimer->lock, flags); } static inline void thread_group_cputime_init(struct signal_struct *sig) { - sig->cputime.totals = (struct task_cputime){ - .utime = cputime_zero, - .stime = cputime_zero, - .sum_exec_runtime = 0, - }; - - spin_lock_init(&sig->cputime.totals.lock); + sig->cputimer.cputime = INIT_CPUTIME; + spin_lock_init(&sig->cputimer.lock); + sig->cputimer.running = 0; } static inline void thread_group_cputime_free(struct signal_struct *sig) diff --git a/kernel/itimer.c b/kernel/itimer.c index 6a5fe93dd8bd..58762f7077ec 100644 --- a/kernel/itimer.c +++ b/kernel/itimer.c @@ -62,7 +62,7 @@ int do_getitimer(int which, struct itimerval *value) struct task_cputime cputime; cputime_t utime; - thread_group_cputime(tsk, &cputime); + thread_group_cputimer(tsk, &cputime); utime = cputime.utime; if (cputime_le(cval, utime)) { /* about to fire */ cval = jiffies_to_cputime(1); @@ -82,7 +82,7 @@ int do_getitimer(int which, struct itimerval *value) struct task_cputime times; cputime_t ptime; - thread_group_cputime(tsk, ×); + thread_group_cputimer(tsk, ×); ptime = cputime_add(times.utime, times.stime); if (cputime_le(cval, ptime)) { /* about to fire */ cval = jiffies_to_cputime(1); diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index fa07da94d7be..db107c9bbc05 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -230,6 +230,37 @@ static int cpu_clock_sample(const clockid_t which_clock, struct task_struct *p, return 0; } +void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) +{ + struct sighand_struct *sighand; + struct signal_struct *sig; + struct task_struct *t; + + *times = INIT_CPUTIME; + + rcu_read_lock(); + sighand = rcu_dereference(tsk->sighand); + if (!sighand) + goto out; + + sig = tsk->signal; + + t = tsk; + do { + times->utime = cputime_add(times->utime, t->utime); + times->stime = cputime_add(times->stime, t->stime); + times->sum_exec_runtime += t->se.sum_exec_runtime; + + t = next_thread(t); + } while (t != tsk); + + times->utime = cputime_add(times->utime, sig->utime); + times->stime = cputime_add(times->stime, sig->stime); + times->sum_exec_runtime += sig->sum_sched_runtime; +out: + rcu_read_unlock(); +} + /* * Sample a process (thread group) clock for the given group_leader task. * Must be called with tasklist_lock held for reading. @@ -475,6 +506,29 @@ static void clear_dead_task(struct k_itimer *timer, union cpu_time_count now) now); } +/* + * Enable the process wide cpu timer accounting. + * + * serialized using ->sighand->siglock + */ +static void start_process_timers(struct task_struct *tsk) +{ + tsk->signal->cputimer.running = 1; + barrier(); +} + +/* + * Release the process wide timer accounting -- timer stops ticking when + * nobody cares about it. + * + * serialized using ->sighand->siglock + */ +static void stop_process_timers(struct task_struct *tsk) +{ + tsk->signal->cputimer.running = 0; + barrier(); +} + /* * Insert the timer on the appropriate list before any timers that * expire later. This must be called with the tasklist_lock held @@ -495,6 +549,9 @@ static void arm_timer(struct k_itimer *timer, union cpu_time_count now) BUG_ON(!irqs_disabled()); spin_lock(&p->sighand->siglock); + if (!CPUCLOCK_PERTHREAD(timer->it_clock)) + start_process_timers(p); + listpos = head; if (CPUCLOCK_WHICH(timer->it_clock) == CPUCLOCK_SCHED) { list_for_each_entry(next, head, entry) { @@ -987,13 +1044,15 @@ static void check_process_timers(struct task_struct *tsk, sig->rlim[RLIMIT_CPU].rlim_cur == RLIM_INFINITY && list_empty(&timers[CPUCLOCK_VIRT]) && cputime_eq(sig->it_virt_expires, cputime_zero) && - list_empty(&timers[CPUCLOCK_SCHED])) + list_empty(&timers[CPUCLOCK_SCHED])) { + stop_process_timers(tsk); return; + } /* * Collect the current process totals. */ - thread_group_cputime(tsk, &cputime); + thread_group_cputimer(tsk, &cputime); utime = cputime.utime; ptime = cputime_add(utime, cputime.stime); sum_sched_runtime = cputime.sum_exec_runtime; @@ -1259,7 +1318,7 @@ static inline int fastpath_timer_check(struct task_struct *tsk) if (!task_cputime_zero(&sig->cputime_expires)) { struct task_cputime group_sample; - thread_group_cputime(tsk, &group_sample); + thread_group_cputimer(tsk, &group_sample); if (task_cputime_expired(&group_sample, &sig->cputime_expires)) return 1; } @@ -1328,6 +1387,33 @@ void run_posix_cpu_timers(struct task_struct *tsk) } } +/* + * Sample a process (thread group) timer for the given group_leader task. + * Must be called with tasklist_lock held for reading. + */ +static int cpu_timer_sample_group(const clockid_t which_clock, + struct task_struct *p, + union cpu_time_count *cpu) +{ + struct task_cputime cputime; + + thread_group_cputimer(p, &cputime); + switch (CPUCLOCK_WHICH(which_clock)) { + default: + return -EINVAL; + case CPUCLOCK_PROF: + cpu->cpu = cputime_add(cputime.utime, cputime.stime); + break; + case CPUCLOCK_VIRT: + cpu->cpu = cputime.utime; + break; + case CPUCLOCK_SCHED: + cpu->sched = cputime.sum_exec_runtime + task_delta_exec(p); + break; + } + return 0; +} + /* * Set one of the process-wide special case CPU timers. * The tsk->sighand->siglock must be held by the caller. @@ -1341,7 +1427,8 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, struct list_head *head; BUG_ON(clock_idx == CPUCLOCK_SCHED); - cpu_clock_sample_group(clock_idx, tsk, &now); + start_process_timers(tsk); + cpu_timer_sample_group(clock_idx, tsk, &now); if (oldval) { if (!cputime_eq(*oldval, cputime_zero)) { diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h index 8ab0cef8ecab..a8f93dd374e1 100644 --- a/kernel/sched_stats.h +++ b/kernel/sched_stats.h @@ -296,19 +296,21 @@ sched_info_switch(struct task_struct *prev, struct task_struct *next) static inline void account_group_user_time(struct task_struct *tsk, cputime_t cputime) { - struct task_cputime *times; - struct signal_struct *sig; + struct thread_group_cputimer *cputimer; /* tsk == current, ensure it is safe to use ->signal */ if (unlikely(tsk->exit_state)) return; - sig = tsk->signal; - times = &sig->cputime.totals; + cputimer = &tsk->signal->cputimer; - spin_lock(×->lock); - times->utime = cputime_add(times->utime, cputime); - spin_unlock(×->lock); + if (!cputimer->running) + return; + + spin_lock(&cputimer->lock); + cputimer->cputime.utime = + cputime_add(cputimer->cputime.utime, cputime); + spin_unlock(&cputimer->lock); } /** @@ -324,19 +326,21 @@ static inline void account_group_user_time(struct task_struct *tsk, static inline void account_group_system_time(struct task_struct *tsk, cputime_t cputime) { - struct task_cputime *times; - struct signal_struct *sig; + struct thread_group_cputimer *cputimer; /* tsk == current, ensure it is safe to use ->signal */ if (unlikely(tsk->exit_state)) return; - sig = tsk->signal; - times = &sig->cputime.totals; + cputimer = &tsk->signal->cputimer; + + if (!cputimer->running) + return; - spin_lock(×->lock); - times->stime = cputime_add(times->stime, cputime); - spin_unlock(×->lock); + spin_lock(&cputimer->lock); + cputimer->cputime.stime = + cputime_add(cputimer->cputime.stime, cputime); + spin_unlock(&cputimer->lock); } /** @@ -352,7 +356,7 @@ static inline void account_group_system_time(struct task_struct *tsk, static inline void account_group_exec_runtime(struct task_struct *tsk, unsigned long long ns) { - struct task_cputime *times; + struct thread_group_cputimer *cputimer; struct signal_struct *sig; sig = tsk->signal; @@ -361,9 +365,12 @@ static inline void account_group_exec_runtime(struct task_struct *tsk, if (unlikely(!sig)) return; - times = &sig->cputime.totals; + cputimer = &sig->cputimer; + + if (!cputimer->running) + return; - spin_lock(×->lock); - times->sum_exec_runtime += ns; - spin_unlock(×->lock); + spin_lock(&cputimer->lock); + cputimer->cputime.sum_exec_runtime += ns; + spin_unlock(&cputimer->lock); } -- cgit v1.2.3-70-g09d2 From ac7b9004909d03d67016368093e81d37cae72895 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 4 Feb 2009 15:11:59 -0800 Subject: generic swap(): don't return a value from swap() The swap() macro is accidentally retuning the value of its first argument. Change it into a doesn't-return-anything macro before someone goes and relies upon this behaviour. Signed-off-by: Peter Zijlstra Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 343df9ef2412..7fa371898e3e 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -480,7 +480,8 @@ static inline char *pack_hex_byte(char *buf, u8 byte) /* * swap - swap value of @a and @b */ -#define swap(a, b) ({ typeof(a) __tmp = (a); (a) = (b); (b) = __tmp; }) +#define swap(a, b) \ + do { typeof(a) __tmp = (a); (a) = (b); (b) = __tmp; } while (0) /** * container_of - cast a member of a structure out to the containing structure -- cgit v1.2.3-70-g09d2 From 1f5e31d7e55ac7fbd4ec5e5b20c8868b0e4564c9 Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Wed, 4 Feb 2009 15:12:03 -0800 Subject: fbmem: don't call copy_from/to_user() with mutex held Avoid calling copy_from/to_user() with fb_info->lock mutex held in fbmem ioctl(). fb_mmap() is called under mm->mmap_sem (A) held, that also acquires fb_info->lock (B); fb_ioctl() takes fb_info->lock (B) and does copy_from/to_user() that might acquire mm->mmap_sem (A), causing a deadlock. NOTE: it doesn't push down the fb_info->lock in each own driver's fb_ioctl(), so there are still potential deadlocks elsewhere. Signed-off-by: Andrea Righi Cc: Dave Jones Cc: "Rafael J. Wysocki" Cc: Johannes Weiner Cc: Krzysztof Helt Cc: Harvey Harrison Cc: Stefan Richter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/fbcmap.c | 20 +++++--- drivers/video/fbmem.c | 135 +++++++++++++++++++++++++------------------------ include/linux/fb.h | 15 ++++++ 3 files changed, 98 insertions(+), 72 deletions(-) (limited to 'include') diff --git a/drivers/video/fbcmap.c b/drivers/video/fbcmap.c index 91b78e691505..f53b9f1d6aba 100644 --- a/drivers/video/fbcmap.c +++ b/drivers/video/fbcmap.c @@ -250,10 +250,6 @@ int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *info) int rc, size = cmap->len * sizeof(u16); struct fb_cmap umap; - if (cmap->start < 0 || (!info->fbops->fb_setcolreg && - !info->fbops->fb_setcmap)) - return -EINVAL; - memset(&umap, 0, sizeof(struct fb_cmap)); rc = fb_alloc_cmap(&umap, cmap->len, cmap->transp != NULL); if (rc) @@ -262,11 +258,23 @@ int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *info) copy_from_user(umap.green, cmap->green, size) || copy_from_user(umap.blue, cmap->blue, size) || (cmap->transp && copy_from_user(umap.transp, cmap->transp, size))) { - fb_dealloc_cmap(&umap); - return -EFAULT; + rc = -EFAULT; + goto out; } umap.start = cmap->start; + if (!lock_fb_info(info)) { + rc = -ENODEV; + goto out; + } + if (cmap->start < 0 || (!info->fbops->fb_setcolreg && + !info->fbops->fb_setcmap)) { + rc = -EINVAL; + goto out1; + } rc = fb_set_cmap(&umap, info); +out1: + unlock_fb_info(info); +out: fb_dealloc_cmap(&umap); return rc; } diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 756efeb91abc..cfd9dce1ce0b 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -1013,132 +1013,139 @@ static long do_fb_ioctl(struct fb_info *info, unsigned int cmd, struct fb_var_screeninfo var; struct fb_fix_screeninfo fix; struct fb_con2fbmap con2fb; + struct fb_cmap cmap_from; struct fb_cmap_user cmap; struct fb_event event; void __user *argp = (void __user *)arg; long ret = 0; - fb = info->fbops; - if (!fb) - return -ENODEV; - switch (cmd) { case FBIOGET_VSCREENINFO: - ret = copy_to_user(argp, &info->var, - sizeof(var)) ? -EFAULT : 0; + if (!lock_fb_info(info)) + return -ENODEV; + var = info->var; + unlock_fb_info(info); + + ret = copy_to_user(argp, &var, sizeof(var)) ? -EFAULT : 0; break; case FBIOPUT_VSCREENINFO: - if (copy_from_user(&var, argp, sizeof(var))) { - ret = -EFAULT; - break; - } + if (copy_from_user(&var, argp, sizeof(var))) + return -EFAULT; + if (!lock_fb_info(info)) + return -ENODEV; acquire_console_sem(); info->flags |= FBINFO_MISC_USEREVENT; ret = fb_set_var(info, &var); info->flags &= ~FBINFO_MISC_USEREVENT; release_console_sem(); - if (ret == 0 && copy_to_user(argp, &var, sizeof(var))) + unlock_fb_info(info); + if (!ret && copy_to_user(argp, &var, sizeof(var))) ret = -EFAULT; break; case FBIOGET_FSCREENINFO: - ret = copy_to_user(argp, &info->fix, - sizeof(fix)) ? -EFAULT : 0; + if (!lock_fb_info(info)) + return -ENODEV; + fix = info->fix; + unlock_fb_info(info); + + ret = copy_to_user(argp, &fix, sizeof(fix)) ? -EFAULT : 0; break; case FBIOPUTCMAP: if (copy_from_user(&cmap, argp, sizeof(cmap))) - ret = -EFAULT; - else - ret = fb_set_user_cmap(&cmap, info); + return -EFAULT; + ret = fb_set_user_cmap(&cmap, info); break; case FBIOGETCMAP: if (copy_from_user(&cmap, argp, sizeof(cmap))) - ret = -EFAULT; - else - ret = fb_cmap_to_user(&info->cmap, &cmap); + return -EFAULT; + if (!lock_fb_info(info)) + return -ENODEV; + cmap_from = info->cmap; + unlock_fb_info(info); + ret = fb_cmap_to_user(&cmap_from, &cmap); break; case FBIOPAN_DISPLAY: - if (copy_from_user(&var, argp, sizeof(var))) { - ret = -EFAULT; - break; - } + if (copy_from_user(&var, argp, sizeof(var))) + return -EFAULT; + if (!lock_fb_info(info)) + return -ENODEV; acquire_console_sem(); ret = fb_pan_display(info, &var); release_console_sem(); + unlock_fb_info(info); if (ret == 0 && copy_to_user(argp, &var, sizeof(var))) - ret = -EFAULT; + return -EFAULT; break; case FBIO_CURSOR: ret = -EINVAL; break; case FBIOGET_CON2FBMAP: if (copy_from_user(&con2fb, argp, sizeof(con2fb))) - ret = -EFAULT; - else if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) - ret = -EINVAL; - else { - con2fb.framebuffer = -1; - event.info = info; - event.data = &con2fb; - fb_notifier_call_chain(FB_EVENT_GET_CONSOLE_MAP, - &event); - ret = copy_to_user(argp, &con2fb, - sizeof(con2fb)) ? -EFAULT : 0; - } + return -EFAULT; + if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) + return -EINVAL; + con2fb.framebuffer = -1; + event.data = &con2fb; + + if (!lock_fb_info(info)) + return -ENODEV; + event.info = info; + fb_notifier_call_chain(FB_EVENT_GET_CONSOLE_MAP, &event); + unlock_fb_info(info); + + ret = copy_to_user(argp, &con2fb, sizeof(con2fb)) ? -EFAULT : 0; break; case FBIOPUT_CON2FBMAP: - if (copy_from_user(&con2fb, argp, sizeof(con2fb))) { - ret = -EFAULT; - break; - } - if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) { - ret = -EINVAL; - break; - } - if (con2fb.framebuffer < 0 || con2fb.framebuffer >= FB_MAX) { - ret = -EINVAL; - break; - } + if (copy_from_user(&con2fb, argp, sizeof(con2fb))) + return -EFAULT; + if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) + return -EINVAL; + if (con2fb.framebuffer < 0 || con2fb.framebuffer >= FB_MAX) + return -EINVAL; if (!registered_fb[con2fb.framebuffer]) request_module("fb%d", con2fb.framebuffer); if (!registered_fb[con2fb.framebuffer]) { ret = -EINVAL; break; } - event.info = info; event.data = &con2fb; + if (!lock_fb_info(info)) + return -ENODEV; + event.info = info; ret = fb_notifier_call_chain(FB_EVENT_SET_CONSOLE_MAP, &event); + unlock_fb_info(info); break; case FBIOBLANK: + if (!lock_fb_info(info)) + return -ENODEV; acquire_console_sem(); info->flags |= FBINFO_MISC_USEREVENT; ret = fb_blank(info, arg); info->flags &= ~FBINFO_MISC_USEREVENT; release_console_sem(); - break;; + unlock_fb_info(info); + break; default: - if (fb->fb_ioctl == NULL) - ret = -ENOTTY; - else + if (!lock_fb_info(info)) + return -ENODEV; + fb = info->fbops; + if (fb->fb_ioctl) ret = fb->fb_ioctl(info, cmd, arg); + else + ret = -ENOTTY; + unlock_fb_info(info); } return ret; } static long fb_ioctl(struct file *file, unsigned int cmd, unsigned long arg) -__acquires(&info->lock) -__releases(&info->lock) { struct inode *inode = file->f_path.dentry->d_inode; int fbidx = iminor(inode); - struct fb_info *info; - long ret; + struct fb_info *info = registered_fb[fbidx]; - info = registered_fb[fbidx]; - mutex_lock(&info->lock); - ret = do_fb_ioctl(info, cmd, arg); - mutex_unlock(&info->lock); - return ret; + return do_fb_ioctl(info, cmd, arg); } #ifdef CONFIG_COMPAT @@ -1257,8 +1264,6 @@ static int fb_get_fscreeninfo(struct fb_info *info, unsigned int cmd, static long fb_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) -__acquires(&info->lock) -__releases(&info->lock) { struct inode *inode = file->f_path.dentry->d_inode; int fbidx = iminor(inode); @@ -1266,7 +1271,6 @@ __releases(&info->lock) struct fb_ops *fb = info->fbops; long ret = -ENOIOCTLCMD; - mutex_lock(&info->lock); switch(cmd) { case FBIOGET_VSCREENINFO: case FBIOPUT_VSCREENINFO: @@ -1292,7 +1296,6 @@ __releases(&info->lock) ret = fb->fb_compat_ioctl(info, cmd, arg); break; } - mutex_unlock(&info->lock); return ret; } #endif diff --git a/include/linux/fb.h b/include/linux/fb.h index 818fe21257e8..31527e17076b 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -960,6 +960,21 @@ extern struct fb_info *registered_fb[FB_MAX]; extern int num_registered_fb; extern struct class *fb_class; +static inline int lock_fb_info(struct fb_info *info) +{ + mutex_lock(&info->lock); + if (!info->fbops) { + mutex_unlock(&info->lock); + return 0; + } + return 1; +} + +static inline void unlock_fb_info(struct fb_info *info) +{ + mutex_unlock(&info->lock); +} + static inline void __fb_pad_aligned_buffer(u8 *dst, u32 d_pitch, u8 *src, u32 s_pitch, u32 height) { -- cgit v1.2.3-70-g09d2 From 777c6c5f1f6e757ae49ecca2ed72d6b1f523c007 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 4 Feb 2009 15:12:14 -0800 Subject: wait: prevent exclusive waiter starvation With exclusive waiters, every process woken up through the wait queue must ensure that the next waiter down the line is woken when it has finished. Interruptible waiters don't do that when aborting due to a signal. And if an aborting waiter is concurrently woken up through the waitqueue, noone will ever wake up the next waiter. This has been observed with __wait_on_bit_lock() used by lock_page_killable(): the first contender on the queue was aborting when the actual lock holder woke it up concurrently. The aborted contender didn't acquire the lock and therefor never did an unlock followed by waking up the next waiter. Add abort_exclusive_wait() which removes the process' wait descriptor from the waitqueue, iff still queued, or wakes up the next waiter otherwise. It does so under the waitqueue lock. Racing with a wake up means the aborting process is either already woken (removed from the queue) and will wake up the next waiter, or it will remove itself from the queue and the concurrent wake up will apply to the next waiter after it. Use abort_exclusive_wait() in __wait_event_interruptible_exclusive() and __wait_on_bit_lock() when they were interrupted by other means than a wake up through the queue. [akpm@linux-foundation.org: coding-style fixes] Reported-by: Chris Mason Signed-off-by: Johannes Weiner Mentored-by: Oleg Nesterov Cc: Peter Zijlstra Cc: Matthew Wilcox Cc: Chuck Lever Cc: Nick Piggin Cc: Ingo Molnar Cc: ["after some testing"] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/wait.h | 11 ++++++++-- kernel/sched.c | 4 ++-- kernel/wait.c | 59 +++++++++++++++++++++++++++++++++++++++++++++------- 3 files changed, 63 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/wait.h b/include/linux/wait.h index ef609f842fac..a210ede73b56 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -132,6 +132,8 @@ static inline void __remove_wait_queue(wait_queue_head_t *head, list_del(&old->task_list); } +void __wake_up_common(wait_queue_head_t *q, unsigned int mode, + int nr_exclusive, int sync, void *key); void __wake_up(wait_queue_head_t *q, unsigned int mode, int nr, void *key); extern void __wake_up_locked(wait_queue_head_t *q, unsigned int mode); extern void __wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr); @@ -333,16 +335,19 @@ do { \ for (;;) { \ prepare_to_wait_exclusive(&wq, &__wait, \ TASK_INTERRUPTIBLE); \ - if (condition) \ + if (condition) { \ + finish_wait(&wq, &__wait); \ break; \ + } \ if (!signal_pending(current)) { \ schedule(); \ continue; \ } \ ret = -ERESTARTSYS; \ + abort_exclusive_wait(&wq, &__wait, \ + TASK_INTERRUPTIBLE, NULL); \ break; \ } \ - finish_wait(&wq, &__wait); \ } while (0) #define wait_event_interruptible_exclusive(wq, condition) \ @@ -431,6 +436,8 @@ extern long interruptible_sleep_on_timeout(wait_queue_head_t *q, void prepare_to_wait(wait_queue_head_t *q, wait_queue_t *wait, int state); void prepare_to_wait_exclusive(wait_queue_head_t *q, wait_queue_t *wait, int state); void finish_wait(wait_queue_head_t *q, wait_queue_t *wait); +void abort_exclusive_wait(wait_queue_head_t *q, wait_queue_t *wait, + unsigned int mode, void *key); int autoremove_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key); int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key); diff --git a/kernel/sched.c b/kernel/sched.c index 242d0d47a70d..8ee437a5ec1d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4697,8 +4697,8 @@ EXPORT_SYMBOL(default_wake_function); * started to run but is not in state TASK_RUNNING. try_to_wake_up() returns * zero in this (rare) case, and we handle it by continuing to scan the queue. */ -static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, - int nr_exclusive, int sync, void *key) +void __wake_up_common(wait_queue_head_t *q, unsigned int mode, + int nr_exclusive, int sync, void *key) { wait_queue_t *curr, *next; diff --git a/kernel/wait.c b/kernel/wait.c index cd87131f2fc2..42a2dbc181c8 100644 --- a/kernel/wait.c +++ b/kernel/wait.c @@ -91,6 +91,15 @@ prepare_to_wait_exclusive(wait_queue_head_t *q, wait_queue_t *wait, int state) } EXPORT_SYMBOL(prepare_to_wait_exclusive); +/* + * finish_wait - clean up after waiting in a queue + * @q: waitqueue waited on + * @wait: wait descriptor + * + * Sets current thread back to running state and removes + * the wait descriptor from the given waitqueue if still + * queued. + */ void finish_wait(wait_queue_head_t *q, wait_queue_t *wait) { unsigned long flags; @@ -117,6 +126,39 @@ void finish_wait(wait_queue_head_t *q, wait_queue_t *wait) } EXPORT_SYMBOL(finish_wait); +/* + * abort_exclusive_wait - abort exclusive waiting in a queue + * @q: waitqueue waited on + * @wait: wait descriptor + * @state: runstate of the waiter to be woken + * @key: key to identify a wait bit queue or %NULL + * + * Sets current thread back to running state and removes + * the wait descriptor from the given waitqueue if still + * queued. + * + * Wakes up the next waiter if the caller is concurrently + * woken up through the queue. + * + * This prevents waiter starvation where an exclusive waiter + * aborts and is woken up concurrently and noone wakes up + * the next waiter. + */ +void abort_exclusive_wait(wait_queue_head_t *q, wait_queue_t *wait, + unsigned int mode, void *key) +{ + unsigned long flags; + + __set_current_state(TASK_RUNNING); + spin_lock_irqsave(&q->lock, flags); + if (!list_empty(&wait->task_list)) + list_del_init(&wait->task_list); + else if (waitqueue_active(q)) + __wake_up_common(q, mode, 1, 0, key); + spin_unlock_irqrestore(&q->lock, flags); +} +EXPORT_SYMBOL(abort_exclusive_wait); + int autoremove_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key) { int ret = default_wake_function(wait, mode, sync, key); @@ -177,17 +219,20 @@ int __sched __wait_on_bit_lock(wait_queue_head_t *wq, struct wait_bit_queue *q, int (*action)(void *), unsigned mode) { - int ret = 0; - do { + int ret; + prepare_to_wait_exclusive(wq, &q->wait, mode); - if (test_bit(q->key.bit_nr, q->key.flags)) { - if ((ret = (*action)(q->key.flags))) - break; - } + if (!test_bit(q->key.bit_nr, q->key.flags)) + continue; + ret = action(q->key.flags); + if (!ret) + continue; + abort_exclusive_wait(wq, &q->wait, mode, &q->key); + return ret; } while (test_and_set_bit(q->key.bit_nr, q->key.flags)); finish_wait(wq, &q->wait); - return ret; + return 0; } EXPORT_SYMBOL(__wait_on_bit_lock); -- cgit v1.2.3-70-g09d2 From fe86175bce50bc3d65ff09c287fed955c4da1eb3 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 4 Feb 2009 15:12:20 -0800 Subject: atyfb: fix CONFIG_ namespace violations Fix namespace violations by changing non-kconfig CONFIG_ names to CNFG_*. Fixes breakage in staging/, which adds a real CONFIG_PANEL. Signed-off-by: Randy Dunlap Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/aty/aty128fb.c | 8 ++++---- drivers/video/aty/atyfb_base.c | 22 +++++++++++----------- drivers/video/aty/radeon_base.c | 10 +++++----- drivers/video/aty/radeon_pm.c | 18 +++++++++--------- include/video/aty128.h | 4 ++-- include/video/mach64.h | 24 ++++++++++++------------ include/video/radeon.h | 18 +++++++++--------- 7 files changed, 52 insertions(+), 52 deletions(-) (limited to 'include') diff --git a/drivers/video/aty/aty128fb.c b/drivers/video/aty/aty128fb.c index db16112cf197..fb2b0f5b23bd 100644 --- a/drivers/video/aty/aty128fb.c +++ b/drivers/video/aty/aty128fb.c @@ -1475,7 +1475,7 @@ static int aty128fb_set_par(struct fb_info *info) aty128_set_pll(&par->pll, par); aty128_set_fifo(&par->fifo_reg, par); - config = aty_ld_le32(CONFIG_CNTL) & ~3; + config = aty_ld_le32(CNFG_CNTL) & ~3; #if defined(__BIG_ENDIAN) if (par->crtc.bpp == 32) @@ -1484,7 +1484,7 @@ static int aty128fb_set_par(struct fb_info *info) config |= 1; /* make aperture do 16 bit swapping */ #endif - aty_st_le32(CONFIG_CNTL, config); + aty_st_le32(CNFG_CNTL, config); aty_st_8(CRTC_EXT_CNTL + 1, 0); /* turn the video back on */ info->fix.line_length = (par->crtc.vxres * par->crtc.bpp) >> 3; @@ -1875,7 +1875,7 @@ static int __devinit aty128_init(struct pci_dev *pdev, const struct pci_device_i u32 dac; /* Get the chip revision */ - chip_rev = (aty_ld_le32(CONFIG_CNTL) >> 16) & 0x1F; + chip_rev = (aty_ld_le32(CNFG_CNTL) >> 16) & 0x1F; strcpy(video_card, "Rage128 XX "); video_card[8] = ent->device >> 8; @@ -2057,7 +2057,7 @@ static int __devinit aty128_probe(struct pci_dev *pdev, const struct pci_device_ /* Grab memory size from the card */ // How does this relate to the resource length from the PCI hardware? - par->vram_size = aty_ld_le32(CONFIG_MEMSIZE) & 0x03FFFFFF; + par->vram_size = aty_ld_le32(CNFG_MEMSIZE) & 0x03FFFFFF; /* Virtualize the framebuffer */ info->screen_base = ioremap(fb_addr, par->vram_size); diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c index cc6b470073da..1d6e16d346a5 100644 --- a/drivers/video/aty/atyfb_base.c +++ b/drivers/video/aty/atyfb_base.c @@ -135,7 +135,7 @@ #if defined(CONFIG_PM) || defined(CONFIG_PMAC_BACKLIGHT) || \ defined (CONFIG_FB_ATY_GENERIC_LCD) || defined(CONFIG_FB_ATY_BACKLIGHT) static const u32 lt_lcd_regs[] = { - CONFIG_PANEL_LG, + CNFG_PANEL_LG, LCD_GEN_CNTL_LG, DSTN_CONTROL_LG, HFB_PITCH_ADDR_LG, @@ -446,7 +446,7 @@ static int __devinit correct_chipset(struct atyfb_par *par) par->pll_limits.ecp_max = aty_chips[i].ecp_max; par->features = aty_chips[i].features; - chip_id = aty_ld_le32(CONFIG_CHIP_ID, par); + chip_id = aty_ld_le32(CNFG_CHIP_ID, par); type = chip_id & CFG_CHIP_TYPE; rev = (chip_id & CFG_CHIP_REV) >> 24; @@ -629,7 +629,7 @@ static void aty_get_crtc(const struct atyfb_par *par, struct crtc *crtc) crtc->lcd_index = aty_ld_le32(LCD_INDEX, par); aty_st_le32(LCD_INDEX, crtc->lcd_index, par); } - crtc->lcd_config_panel = aty_ld_lcd(CONFIG_PANEL, par); + crtc->lcd_config_panel = aty_ld_lcd(CNFG_PANEL, par); crtc->lcd_gen_cntl = aty_ld_lcd(LCD_GEN_CNTL, par); @@ -676,7 +676,7 @@ static void aty_set_crtc(const struct atyfb_par *par, const struct crtc *crtc) aty_st_le32(CRTC_GEN_CNTL, crtc->gen_cntl & ~(CRTC_EXT_DISP_EN | CRTC_EN), par); /* update non-shadow registers first */ - aty_st_lcd(CONFIG_PANEL, crtc->lcd_config_panel, par); + aty_st_lcd(CNFG_PANEL, crtc->lcd_config_panel, par); aty_st_lcd(LCD_GEN_CNTL, crtc->lcd_gen_cntl & ~(CRTC_RW_SELECT | SHADOW_EN | SHADOW_RW_EN), par); @@ -858,7 +858,7 @@ static int aty_var_to_crtc(const struct fb_info *info, if (!M64_HAS(MOBIL_BUS)) crtc->lcd_index |= CRTC2_DISPLAY_DIS; - crtc->lcd_config_panel = aty_ld_lcd(CONFIG_PANEL, par) | 0x4000; + crtc->lcd_config_panel = aty_ld_lcd(CNFG_PANEL, par) | 0x4000; crtc->lcd_gen_cntl = aty_ld_lcd(LCD_GEN_CNTL, par) & ~CRTC_RW_SELECT; crtc->lcd_gen_cntl &= @@ -2254,7 +2254,7 @@ static int __devinit aty_init(struct fb_info *info) if (!M64_HAS(INTEGRATED)) { u32 stat0; u8 dac_type, dac_subtype, clk_type; - stat0 = aty_ld_le32(CONFIG_STAT0, par); + stat0 = aty_ld_le32(CNFG_STAT0, par); par->bus_type = (stat0 >> 0) & 0x07; par->ram_type = (stat0 >> 3) & 0x07; ramname = aty_gx_ram[par->ram_type]; @@ -2324,7 +2324,7 @@ static int __devinit aty_init(struct fb_info *info) par->dac_ops = &aty_dac_ct; par->pll_ops = &aty_pll_ct; par->bus_type = PCI; - par->ram_type = (aty_ld_le32(CONFIG_STAT0, par) & 0x07); + par->ram_type = (aty_ld_le32(CNFG_STAT0, par) & 0x07); ramname = aty_ct_ram[par->ram_type]; /* for many chips, the mclk is 67 MHz for SDRAM, 63 MHz otherwise */ if (par->pll_limits.mclk == 67 && par->ram_type < SDRAM) @@ -2433,7 +2433,7 @@ static int __devinit aty_init(struct fb_info *info) } if (M64_HAS(MAGIC_VRAM_SIZE)) { - if (aty_ld_le32(CONFIG_STAT1, par) & 0x40000000) + if (aty_ld_le32(CNFG_STAT1, par) & 0x40000000) info->fix.smem_len += 0x400000; } @@ -2946,7 +2946,7 @@ static int __devinit atyfb_setup_sparc(struct pci_dev *pdev, * Fix PROMs idea of MEM_CNTL settings... */ mem = aty_ld_le32(MEM_CNTL, par); - chip_id = aty_ld_le32(CONFIG_CHIP_ID, par); + chip_id = aty_ld_le32(CNFG_CHIP_ID, par); if (((chip_id & CFG_CHIP_TYPE) == VT_CHIP_ID) && !((chip_id >> 24) & 1)) { switch (mem & 0x0f) { case 3: @@ -2964,7 +2964,7 @@ static int __devinit atyfb_setup_sparc(struct pci_dev *pdev, default: break; } - if ((aty_ld_le32(CONFIG_STAT0, par) & 7) >= SDRAM) + if ((aty_ld_le32(CNFG_STAT0, par) & 7) >= SDRAM) mem &= ~(0x00700000); } mem &= ~(0xcf80e000); /* Turn off all undocumented bits. */ @@ -3572,7 +3572,7 @@ static int __init atyfb_atari_probe(void) } /* Fake pci_id for correct_chipset() */ - switch (aty_ld_le32(CONFIG_CHIP_ID, par) & CFG_CHIP_TYPE) { + switch (aty_ld_le32(CNFG_CHIP_ID, par) & CFG_CHIP_TYPE) { case 0x00d7: par->pci_id = PCI_CHIP_MACH64GX; break; diff --git a/drivers/video/aty/radeon_base.c b/drivers/video/aty/radeon_base.c index d0f1a7fc2c9d..16bb7e3c0310 100644 --- a/drivers/video/aty/radeon_base.c +++ b/drivers/video/aty/radeon_base.c @@ -1936,8 +1936,8 @@ static void fixup_memory_mappings(struct radeonfb_info *rinfo) OUTREG(CRTC_GEN_CNTL, save_crtc_gen_cntl | CRTC_DISP_REQ_EN_B); mdelay(100); - aper_base = INREG(CONFIG_APER_0_BASE); - aper_size = INREG(CONFIG_APER_SIZE); + aper_base = INREG(CNFG_APER_0_BASE); + aper_size = INREG(CNFG_APER_SIZE); #ifdef SET_MC_FB_FROM_APERTURE /* Set framebuffer to be at the same address as set in PCI BAR */ @@ -2024,11 +2024,11 @@ static void radeon_identify_vram(struct radeonfb_info *rinfo) ~CRTC_H_CUTOFF_ACTIVE_EN); } } else { - tmp = INREG(CONFIG_MEMSIZE); + tmp = INREG(CNFG_MEMSIZE); } /* mem size is bits [28:0], mask off the rest */ - rinfo->video_ram = tmp & CONFIG_MEMSIZE_MASK; + rinfo->video_ram = tmp & CNFG_MEMSIZE_MASK; /* * Hack to get around some busted production M6's @@ -2228,7 +2228,7 @@ static int __devinit radeonfb_pci_register (struct pci_dev *pdev, */ rinfo->errata = 0; if (rinfo->family == CHIP_FAMILY_R300 && - (INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK) + (INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK) == CFG_ATI_REV_A11) rinfo->errata |= CHIP_ERRATA_R300_CG; diff --git a/drivers/video/aty/radeon_pm.c b/drivers/video/aty/radeon_pm.c index 675abdafc2d8..c4ac2a032fcb 100644 --- a/drivers/video/aty/radeon_pm.c +++ b/drivers/video/aty/radeon_pm.c @@ -333,7 +333,7 @@ static void radeon_pm_enable_dynamic_mode(struct radeonfb_info *rinfo) if (!rinfo->has_CRTC2) { tmp = INPLL(pllSCLK_CNTL); - if ((INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK) > CFG_ATI_REV_A13) + if ((INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK) > CFG_ATI_REV_A13) tmp &= ~(SCLK_CNTL__FORCE_CP | SCLK_CNTL__FORCE_RB); tmp &= ~(SCLK_CNTL__FORCE_HDP | SCLK_CNTL__FORCE_DISP1 | SCLK_CNTL__FORCE_TOP | SCLK_CNTL__FORCE_SE | @@ -468,9 +468,9 @@ static void radeon_pm_enable_dynamic_mode(struct radeonfb_info *rinfo) /*RAGE_6::A11 A12 A12N1 A13, RV250::A11 A12, R300*/ if ((rinfo->family == CHIP_FAMILY_RV250 && - ((INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13)) || + ((INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13)) || ((rinfo->family == CHIP_FAMILY_RV100) && - ((INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK) <= CFG_ATI_REV_A13))) { + ((INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK) <= CFG_ATI_REV_A13))) { tmp |= SCLK_CNTL__FORCE_CP; tmp |= SCLK_CNTL__FORCE_VIP; } @@ -486,7 +486,7 @@ static void radeon_pm_enable_dynamic_mode(struct radeonfb_info *rinfo) /* RV200::A11 A12 RV250::A11 A12 */ if (((rinfo->family == CHIP_FAMILY_RV200) || (rinfo->family == CHIP_FAMILY_RV250)) && - ((INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13)) + ((INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13)) tmp |= SCLK_MORE_CNTL__FORCEON; OUTPLL(pllSCLK_MORE_CNTL, tmp); @@ -497,7 +497,7 @@ static void radeon_pm_enable_dynamic_mode(struct radeonfb_info *rinfo) /* RV200::A11 A12, RV250::A11 A12 */ if (((rinfo->family == CHIP_FAMILY_RV200) || (rinfo->family == CHIP_FAMILY_RV250)) && - ((INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13)) { + ((INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13)) { tmp = INPLL(pllPLL_PWRMGT_CNTL); tmp |= PLL_PWRMGT_CNTL__TCL_BYPASS_DISABLE; OUTPLL(pllPLL_PWRMGT_CNTL, tmp); @@ -702,7 +702,7 @@ static void radeon_pm_restore_regs(struct radeonfb_info *rinfo) OUTREG(DISPLAY_BASE_ADDR, rinfo->save_regs[31]); OUTREG(MC_AGP_LOCATION, rinfo->save_regs[32]); OUTREG(CRTC2_DISPLAY_BASE_ADDR, rinfo->save_regs[33]); - OUTREG(CONFIG_MEMSIZE, rinfo->video_ram); + OUTREG(CNFG_MEMSIZE, rinfo->video_ram); OUTREG(DISP_MISC_CNTL, rinfo->save_regs[9]); OUTREG(DISP_PWR_MAN, rinfo->save_regs[10]); @@ -1723,7 +1723,7 @@ static void radeon_reinitialize_M10(struct radeonfb_info *rinfo) OUTREG(CRTC2_DISPLAY_BASE_ADDR, rinfo->save_regs[33]); OUTREG(MC_FB_LOCATION, rinfo->save_regs[30]); OUTREG(OV0_BASE_ADDR, rinfo->save_regs[80]); - OUTREG(CONFIG_MEMSIZE, rinfo->video_ram); + OUTREG(CNFG_MEMSIZE, rinfo->video_ram); OUTREG(BUS_CNTL, rinfo->save_regs[36]); OUTREG(BUS_CNTL1, rinfo->save_regs[14]); OUTREG(MPP_TB_CONFIG, rinfo->save_regs[37]); @@ -1961,7 +1961,7 @@ static void radeon_pm_m9p_reconfigure_mc(struct radeonfb_info *rinfo) OUTMC(rinfo, ixMC_CHP_IO_CNTL_B1, rinfo->save_regs[68] /*0x141555ff*/); OUTMC(rinfo, ixMC_IMP_CNTL_0, rinfo->save_regs[71] /*0x00009249*/); OUTREG(MC_IND_INDEX, 0); - OUTREG(CONFIG_MEMSIZE, rinfo->video_ram); + OUTREG(CNFG_MEMSIZE, rinfo->video_ram); mdelay(20); } @@ -2361,7 +2361,7 @@ static void radeon_reinitialize_QW(struct radeonfb_info *rinfo) OUTMC(rinfo, ixMC_IMP_CNTL_0, 0x00009249); OUTREG(MC_IND_INDEX, 0); - OUTREG(CONFIG_MEMSIZE, rinfo->video_ram); + OUTREG(CNFG_MEMSIZE, rinfo->video_ram); radeon_pm_full_reset_sdram(rinfo); diff --git a/include/video/aty128.h b/include/video/aty128.h index 7079beb005e8..51ac69f05bdc 100644 --- a/include/video/aty128.h +++ b/include/video/aty128.h @@ -21,9 +21,9 @@ #define I2C_CNTL_1 0x0094 #define PALETTE_INDEX 0x00b0 #define PALETTE_DATA 0x00b4 -#define CONFIG_CNTL 0x00e0 +#define CNFG_CNTL 0x00e0 #define GEN_RESET_CNTL 0x00f0 -#define CONFIG_MEMSIZE 0x00f8 +#define CNFG_MEMSIZE 0x00f8 #define MEM_CNTL 0x0140 #define MEM_POWER_MISC 0x015c #define AGP_BASE 0x0170 diff --git a/include/video/mach64.h b/include/video/mach64.h index a8332e528ec1..89e91c0cb737 100644 --- a/include/video/mach64.h +++ b/include/video/mach64.h @@ -103,7 +103,7 @@ #define CUR_HORZ_VERT_OFF 0x0070 /* Dword offset 0_1C */ #define CUR2_HORZ_VERT_OFF 0x0070 /* Dword offset 0_1C */ -#define CONFIG_PANEL_LG 0x0074 /* Dword offset 0_1D (LG) */ +#define CNFG_PANEL_LG 0x0074 /* Dword offset 0_1D (LG) */ /* General I/O Control */ #define GP_IO 0x0078 /* Dword offset 0_1E */ @@ -146,8 +146,8 @@ #define CLOCK_SEL_CNTL 0x0090 /* Dword offset 0_24 */ /* Configuration */ -#define CONFIG_STAT1 0x0094 /* Dword offset 0_25 */ -#define CONFIG_STAT2 0x0098 /* Dword offset 0_26 */ +#define CNFG_STAT1 0x0094 /* Dword offset 0_25 */ +#define CNFG_STAT2 0x0098 /* Dword offset 0_26 */ /* Bus Control */ #define BUS_CNTL 0x00A0 /* Dword offset 0_28 */ @@ -190,9 +190,9 @@ #define POWER_MANAGEMENT_LG 0x00D8 /* Dword offset 0_36 (LG) */ /* Configuration */ -#define CONFIG_CNTL 0x00DC /* Dword offset 0_37 (CT, ET, VT) */ -#define CONFIG_CHIP_ID 0x00E0 /* Dword offset 0_38 */ -#define CONFIG_STAT0 0x00E4 /* Dword offset 0_39 */ +#define CNFG_CNTL 0x00DC /* Dword offset 0_37 (CT, ET, VT) */ +#define CNFG_CHIP_ID 0x00E0 /* Dword offset 0_38 */ +#define CNFG_STAT0 0x00E4 /* Dword offset 0_39 */ /* Test and Debug */ #define CRC_SIG 0x00E8 /* Dword offset 0_3A */ @@ -851,17 +851,17 @@ #define PLL_YCLK_CNTL 0x29 #define PM_DYN_CLK_CNTL 0x2A -/* CONFIG_CNTL register constants */ +/* CNFG_CNTL register constants */ #define APERTURE_4M_ENABLE 1 #define APERTURE_8M_ENABLE 2 #define VGA_APERTURE_ENABLE 4 -/* CONFIG_STAT0 register constants (GX, CX) */ +/* CNFG_STAT0 register constants (GX, CX) */ #define CFG_BUS_TYPE 0x00000007 #define CFG_MEM_TYPE 0x00000038 #define CFG_INIT_DAC_TYPE 0x00000e00 -/* CONFIG_STAT0 register constants (CT, ET, VT) */ +/* CNFG_STAT0 register constants (CT, ET, VT) */ #define CFG_MEM_TYPE_xT 0x00000007 #define ISA 0 @@ -942,7 +942,7 @@ #define PCI_ATI_VENDOR_ID 0x1002 -/* CONFIG_CHIP_ID register constants */ +/* CNFG_CHIP_ID register constants */ #define CFG_CHIP_TYPE 0x0000FFFF #define CFG_CHIP_CLASS 0x00FF0000 #define CFG_CHIP_REV 0xFF000000 @@ -951,7 +951,7 @@ #define CFG_CHIP_MINOR 0xC0000000 -/* Chip IDs read from CONFIG_CHIP_ID */ +/* Chip IDs read from CNFG_CHIP_ID */ /* mach64GX family */ #define GX_CHIP_ID 0xD7 /* mach64GX (ATI888GX00) */ @@ -1254,7 +1254,7 @@ #define CRTC2_DISPLAY_DIS 0x00000400 /* LCD register indices */ -#define CONFIG_PANEL 0x00 +#define CNFG_PANEL 0x00 #define LCD_GEN_CNTL 0x01 #define DSTN_CONTROL 0x02 #define HFB_PITCH_ADDR 0x03 diff --git a/include/video/radeon.h b/include/video/radeon.h index 1cd09cc5b169..e072b16b39ab 100644 --- a/include/video/radeon.h +++ b/include/video/radeon.h @@ -11,13 +11,13 @@ #define HI_STAT 0x004C #define BUS_CNTL1 0x0034 #define I2C_CNTL_1 0x0094 -#define CONFIG_CNTL 0x00E0 -#define CONFIG_MEMSIZE 0x00F8 -#define CONFIG_APER_0_BASE 0x0100 -#define CONFIG_APER_1_BASE 0x0104 -#define CONFIG_APER_SIZE 0x0108 -#define CONFIG_REG_1_BASE 0x010C -#define CONFIG_REG_APER_SIZE 0x0110 +#define CNFG_CNTL 0x00E0 +#define CNFG_MEMSIZE 0x00F8 +#define CNFG_APER_0_BASE 0x0100 +#define CNFG_APER_1_BASE 0x0104 +#define CNFG_APER_SIZE 0x0108 +#define CNFG_REG_1_BASE 0x010C +#define CNFG_REG_APER_SIZE 0x0110 #define PAD_AGPINPUT_DELAY 0x0164 #define PAD_CTLR_STRENGTH 0x0168 #define PAD_CTLR_UPDATE 0x016C @@ -509,7 +509,7 @@ /* CLOCK_CNTL_INDEX bit constants */ #define PLL_WR_EN 0x00000080 -/* CONFIG_CNTL bit constants */ +/* CNFG_CNTL bit constants */ #define CFG_VGA_RAM_EN 0x00000100 #define CFG_ATI_REV_ID_MASK (0xf << 16) #define CFG_ATI_REV_A11 (0 << 16) @@ -980,7 +980,7 @@ /* masks */ -#define CONFIG_MEMSIZE_MASK 0x1f000000 +#define CNFG_MEMSIZE_MASK 0x1f000000 #define MEM_CFG_TYPE 0x40000000 #define DST_OFFSET_MASK 0x003fffff #define DST_PITCH_MASK 0x3fc00000 -- cgit v1.2.3-70-g09d2 From 7d8e23df69820e6be42bcc41d441f4860e8c76f7 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 6 Feb 2009 14:57:51 +0100 Subject: timers: split process wide cpu clocks/timers, remove spurious warning Mike Galbraith reported that the new warning in thread_group_cputimer() triggers en masse with Amarok running. Oleg Nesterov observed: Can't fastpath_timer_check()->thread_group_cputimer() have the false warning too? Suppose we had the timer, then posix_cpu_timer_del() removes this timer, but task_cputime_zero(&sig->cputime_expires) still not true. Remove the spurious debug warning. Reported-by: Mike Galbraith Explained-by: Oleg Nesterov Signed-off-by: Ingo Molnar --- include/linux/sched.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 082d7619b3a1..79392916d6c9 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2208,8 +2208,6 @@ void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; unsigned long flags; - WARN_ON(!cputimer->running); - spin_lock_irqsave(&cputimer->lock, flags); *times = cputimer->cputime; spin_unlock_irqrestore(&cputimer->lock, flags); -- cgit v1.2.3-70-g09d2 From d96f94c604453f87fe24154b87e1e9a3a72511f8 Mon Sep 17 00:00:00 2001 From: "Pallipadi, Venkatesh" Date: Mon, 2 Feb 2009 11:57:18 -0800 Subject: ACPI: Enable bit 11 in _PDC to advertise hw coord Bit 11 in intel PDC definitions is meant for OS capability to handle hardware coordination of P-states. In Linux we have always supported hwardware coordination of P-states. Just let the BIOSes know that we support it, by setting this bit. Some BIOSes use this bit to choose between hardware or software coordination and without this change below, BIOSes switch to software coordination, which is not very optimal in terms of power consumption and extra wakeups from idle. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Len Brown --- include/acpi/pdc_intel.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/acpi/pdc_intel.h b/include/acpi/pdc_intel.h index e72bfdd887f9..552637b0d051 100644 --- a/include/acpi/pdc_intel.h +++ b/include/acpi/pdc_intel.h @@ -14,6 +14,7 @@ #define ACPI_PDC_SMP_T_SWCOORD (0x0080) #define ACPI_PDC_C_C1_FFH (0x0100) #define ACPI_PDC_C_C2C3_FFH (0x0200) +#define ACPI_PDC_SMP_P_HWCOORD (0x0800) #define ACPI_PDC_EST_CAPABILITY_SMP (ACPI_PDC_SMP_C1PT | \ ACPI_PDC_C_C1_HALT | \ @@ -22,6 +23,7 @@ #define ACPI_PDC_EST_CAPABILITY_SWSMP (ACPI_PDC_SMP_C1PT | \ ACPI_PDC_C_C1_HALT | \ ACPI_PDC_SMP_P_SWCOORD | \ + ACPI_PDC_SMP_P_HWCOORD | \ ACPI_PDC_P_FFH) #define ACPI_PDC_C_CAPABILITY_SMP (ACPI_PDC_SMP_C2C3 | \ -- cgit v1.2.3-70-g09d2 From b4bd07c20ba0c1fa7ad09ba257e0a5cfc2bf6bb3 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 6 Feb 2009 22:06:43 -0800 Subject: net_dma: call dmaengine_get only if NET_DMA enabled Based upon a patch from Atsushi Nemoto -------------------- The commit 649274d993212e7c23c0cb734572c2311c200872 ("net_dma: acquire/release dma channels on ifup/ifdown") added unconditional call of dmaengine_get() to net_dma. The API should be called only if NET_DMA was enabled. -------------------- Signed-off-by: David S. Miller Acked-by: Dan Williams --- include/linux/dmaengine.h | 12 ++++++++++++ net/core/dev.c | 4 ++-- 2 files changed, 14 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index 3e0f64c335c8..3e68469c1885 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -282,6 +282,18 @@ static inline void dmaengine_put(void) } #endif +#ifdef CONFIG_NET_DMA +#define net_dmaengine_get() dmaengine_get() +#define net_dmaengine_put() dmaengine_put() +#else +static inline void net_dmaengine_get(void) +{ +} +static inline void net_dmaengine_put(void) +{ +} +#endif + dma_cookie_t dma_async_memcpy_buf_to_buf(struct dma_chan *chan, void *dest, void *src, size_t len); dma_cookie_t dma_async_memcpy_buf_to_pg(struct dma_chan *chan, diff --git a/net/core/dev.c b/net/core/dev.c index 5379b0c1190a..a17e00662363 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1090,7 +1090,7 @@ int dev_open(struct net_device *dev) /* * Enable NET_DMA */ - dmaengine_get(); + net_dmaengine_get(); /* * Initialize multicasting status @@ -1172,7 +1172,7 @@ int dev_close(struct net_device *dev) /* * Shutdown NET_DMA */ - dmaengine_put(); + net_dmaengine_put(); return 0; } -- cgit v1.2.3-70-g09d2 From 7f9a50a5b89b87f8e754f59ae9968da28be618a5 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Sat, 7 Feb 2009 18:15:56 +1030 Subject: module: remove over-zealous check in __module_get() Impact: fix spurious BUG_ON() triggered under load module_refcount() isn't reliable outside stop_machine(), as demonstrated by Karsten Keil , networking can trigger it under load (an inc on one cpu and dec on another while module_refcount() is tallying can give false results, for example). Almost noone should be using __module_get, but that's another issue. Cc: Karsten Keil Signed-off-by: Rusty Russell Signed-off-by: Linus Torvalds --- include/linux/module.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include') diff --git a/include/linux/module.h b/include/linux/module.h index f3b8329eb5b8..145a75528cc1 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -407,7 +407,6 @@ static inline local_t *__module_ref_addr(struct module *mod, int cpu) static inline void __module_get(struct module *module) { if (module) { - BUG_ON(module_refcount(module) == 0); local_inc(__module_ref_addr(module, get_cpu())); put_cpu(); } -- cgit v1.2.3-70-g09d2 From 0f973f27888e4664b253ab2cf69c67c2eb80ab1b Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Mon, 26 Jan 2009 17:10:45 -0800 Subject: drm/i915: add fence register management to execbuf Adds code to set up fence registers at execbuf time on pre-965 chips as necessary. Also fixes up a few bugs in the pre-965 tile register support (get_order != ffs). The number of fences available to the kernel defaults to the hw limit minus 3 (for legacy X front/back/depth), but a new parameter allows userspace to override that as needed. Signed-off-by: Jesse Barnes Signed-off-by: Eric Anholt Signed-off-by: Dave Airlie --- drivers/gpu/drm/i915/i915_dma.c | 10 ++++ drivers/gpu/drm/i915/i915_drv.h | 6 +++ drivers/gpu/drm/i915/i915_gem.c | 56 +++++++++++++++------- drivers/gpu/drm/i915/i915_gem_tiling.c | 88 +++++++++++++++++++++++++++++++++- drivers/gpu/drm/i915/i915_reg.h | 4 +- include/drm/i915_drm.h | 2 + 6 files changed, 146 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 1e01e7847155..cc0adb428cee 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -731,6 +731,9 @@ static int i915_getparam(struct drm_device *dev, void *data, case I915_PARAM_HAS_GEM: value = dev_priv->has_gem; break; + case I915_PARAM_NUM_FENCES_AVAIL: + value = dev_priv->num_fence_regs - dev_priv->fence_reg_start; + break; default: DRM_ERROR("Unknown parameter %d\n", param->param); return -EINVAL; @@ -764,6 +767,13 @@ static int i915_setparam(struct drm_device *dev, void *data, case I915_SETPARAM_ALLOW_BATCHBUFFER: dev_priv->allow_batchbuffer = param->value; break; + case I915_SETPARAM_NUM_USED_FENCES: + if (param->value > dev_priv->num_fence_regs || + param->value < 0) + return -EINVAL; + /* Userspace can use first N regs */ + dev_priv->fence_reg_start = param->value; + break; default: DRM_ERROR("unknown parameter %d\n", param->param); return -EINVAL; diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index f471d218b89a..a70bf77290fc 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -602,6 +602,7 @@ int i915_gem_init_object(struct drm_gem_object *obj); void i915_gem_free_object(struct drm_gem_object *obj); int i915_gem_object_pin(struct drm_gem_object *obj, uint32_t alignment); void i915_gem_object_unpin(struct drm_gem_object *obj); +int i915_gem_object_unbind(struct drm_gem_object *obj); void i915_gem_lastclose(struct drm_device *dev); uint32_t i915_get_gem_seqno(struct drm_device *dev); void i915_gem_retire_requests(struct drm_device *dev); @@ -785,6 +786,11 @@ extern int i915_wait_ring(struct drm_device * dev, int n, const char *caller); IS_I945GM(dev) || IS_I965GM(dev) || IS_GM45(dev)) #define I915_NEED_GFX_HWS(dev) (IS_G33(dev) || IS_GM45(dev) || IS_G4X(dev)) +/* With the 945 and later, Y tiling got adjusted so that it was 32 128-byte + * rows, which changed the alignment requirements and fence programming. + */ +#define HAS_128_BYTE_Y_TILING(dev) (IS_I9XX(dev) && !(IS_I915G(dev) || \ + IS_I915GM(dev))) #define SUPPORTS_INTEGRATED_HDMI(dev) (IS_G4X(dev)) #define PRIMARY_RINGBUFFER_SIZE (128*1024) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index e1f831f166ca..6a9e3a875083 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -52,7 +52,7 @@ static void i915_gem_object_free_page_list(struct drm_gem_object *obj); static int i915_gem_object_wait_rendering(struct drm_gem_object *obj); static int i915_gem_object_bind_to_gtt(struct drm_gem_object *obj, unsigned alignment); -static int i915_gem_object_get_fence_reg(struct drm_gem_object *obj); +static int i915_gem_object_get_fence_reg(struct drm_gem_object *obj, bool write); static void i915_gem_clear_fence_reg(struct drm_gem_object *obj); static int i915_gem_evict_something(struct drm_device *dev); static int i915_gem_phys_pwrite(struct drm_device *dev, struct drm_gem_object *obj, @@ -567,6 +567,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) pgoff_t page_offset; unsigned long pfn; int ret = 0; + bool write = !!(vmf->flags & FAULT_FLAG_WRITE); /* We don't use vmf->pgoff since that has the fake offset */ page_offset = ((unsigned long)vmf->virtual_address - vma->vm_start) >> @@ -586,7 +587,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) /* Need a new fence register? */ if (obj_priv->fence_reg == I915_FENCE_REG_NONE && obj_priv->tiling_mode != I915_TILING_NONE) { - ret = i915_gem_object_get_fence_reg(obj); + ret = i915_gem_object_get_fence_reg(obj, write); if (ret != 0) return VM_FAULT_SIGBUS; } @@ -1214,7 +1215,7 @@ i915_gem_object_wait_rendering(struct drm_gem_object *obj) /** * Unbinds an object from the GTT aperture. */ -static int +int i915_gem_object_unbind(struct drm_gem_object *obj) { struct drm_device *dev = obj->dev; @@ -1448,21 +1449,26 @@ static void i915_write_fence_reg(struct drm_i915_fence_reg *reg) drm_i915_private_t *dev_priv = dev->dev_private; struct drm_i915_gem_object *obj_priv = obj->driver_private; int regnum = obj_priv->fence_reg; + int tile_width; uint32_t val; uint32_t pitch_val; if ((obj_priv->gtt_offset & ~I915_FENCE_START_MASK) || (obj_priv->gtt_offset & (obj->size - 1))) { - WARN(1, "%s: object not 1M or size aligned\n", __func__); + WARN(1, "%s: object 0x%08x not 1M or size (0x%x) aligned\n", + __func__, obj_priv->gtt_offset, obj->size); return; } - if (obj_priv->tiling_mode == I915_TILING_Y && (IS_I945G(dev) || - IS_I945GM(dev) || - IS_G33(dev))) - pitch_val = (obj_priv->stride / 128) - 1; + if (obj_priv->tiling_mode == I915_TILING_Y && + HAS_128_BYTE_Y_TILING(dev)) + tile_width = 128; else - pitch_val = (obj_priv->stride / 512) - 1; + tile_width = 512; + + /* Note: pitch better be a power of two tile widths */ + pitch_val = obj_priv->stride / tile_width; + pitch_val = ffs(pitch_val) - 1; val = obj_priv->gtt_offset; if (obj_priv->tiling_mode == I915_TILING_Y) @@ -1486,7 +1492,8 @@ static void i830_write_fence_reg(struct drm_i915_fence_reg *reg) if ((obj_priv->gtt_offset & ~I915_FENCE_START_MASK) || (obj_priv->gtt_offset & (obj->size - 1))) { - WARN(1, "%s: object not 1M or size aligned\n", __func__); + WARN(1, "%s: object 0x%08x not 1M or size aligned\n", + __func__, obj_priv->gtt_offset); return; } @@ -1506,6 +1513,7 @@ static void i830_write_fence_reg(struct drm_i915_fence_reg *reg) /** * i915_gem_object_get_fence_reg - set up a fence reg for an object * @obj: object to map through a fence reg + * @write: object is about to be written * * When mapping objects through the GTT, userspace wants to be able to write * to them without having to worry about swizzling if the object is tiled. @@ -1517,7 +1525,7 @@ static void i830_write_fence_reg(struct drm_i915_fence_reg *reg) * and tiling format. */ static int -i915_gem_object_get_fence_reg(struct drm_gem_object *obj) +i915_gem_object_get_fence_reg(struct drm_gem_object *obj, bool write) { struct drm_device *dev = obj->dev; struct drm_i915_private *dev_priv = dev->dev_private; @@ -1530,12 +1538,18 @@ i915_gem_object_get_fence_reg(struct drm_gem_object *obj) WARN(1, "allocating a fence for non-tiled object?\n"); break; case I915_TILING_X: - WARN(obj_priv->stride & (512 - 1), - "object is X tiled but has non-512B pitch\n"); + if (!obj_priv->stride) + return -EINVAL; + WARN((obj_priv->stride & (512 - 1)), + "object 0x%08x is X tiled but has non-512B pitch\n", + obj_priv->gtt_offset); break; case I915_TILING_Y: - WARN(obj_priv->stride & (128 - 1), - "object is Y tiled but has non-128B pitch\n"); + if (!obj_priv->stride) + return -EINVAL; + WARN((obj_priv->stride & (128 - 1)), + "object 0x%08x is Y tiled but has non-128B pitch\n", + obj_priv->gtt_offset); break; } @@ -1637,7 +1651,7 @@ i915_gem_object_bind_to_gtt(struct drm_gem_object *obj, unsigned alignment) if (dev_priv->mm.suspended) return -EBUSY; if (alignment == 0) - alignment = PAGE_SIZE; + alignment = i915_gem_get_gtt_alignment(obj); if (alignment & (PAGE_SIZE - 1)) { DRM_ERROR("Invalid object alignment requested %u\n", alignment); return -EINVAL; @@ -2658,6 +2672,14 @@ i915_gem_object_pin(struct drm_gem_object *obj, uint32_t alignment) DRM_ERROR("Failure to bind: %d", ret); return ret; } + /* + * Pre-965 chips need a fence register set up in order to + * properly handle tiled surfaces. + */ + if (!IS_I965G(dev) && + obj_priv->fence_reg == I915_FENCE_REG_NONE && + obj_priv->tiling_mode != I915_TILING_NONE) + i915_gem_object_get_fence_reg(obj, true); } obj_priv->pin_count++; @@ -3297,7 +3319,7 @@ i915_gem_load(struct drm_device *dev) /* Old X drivers will take 0-2 for front, back, depth buffers */ dev_priv->fence_reg_start = 3; - if (IS_I965G(dev)) + if (IS_I965G(dev) || IS_I945G(dev) || IS_I945GM(dev) || IS_G33(dev)) dev_priv->num_fence_regs = 16; else dev_priv->num_fence_regs = 8; diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c index 241f39b7f460..2534c792808e 100644 --- a/drivers/gpu/drm/i915/i915_gem_tiling.c +++ b/drivers/gpu/drm/i915/i915_gem_tiling.c @@ -173,6 +173,73 @@ i915_gem_detect_bit_6_swizzle(struct drm_device *dev) dev_priv->mm.bit_6_swizzle_y = swizzle_y; } + +/** + * Returns the size of the fence for a tiled object of the given size. + */ +static int +i915_get_fence_size(struct drm_device *dev, int size) +{ + int i; + int start; + + if (IS_I965G(dev)) { + /* The 965 can have fences at any page boundary. */ + return ALIGN(size, 4096); + } else { + /* Align the size to a power of two greater than the smallest + * fence size. + */ + if (IS_I9XX(dev)) + start = 1024 * 1024; + else + start = 512 * 1024; + + for (i = start; i < size; i <<= 1) + ; + + return i; + } +} + +/* Check pitch constriants for all chips & tiling formats */ +static bool +i915_tiling_ok(struct drm_device *dev, int stride, int size, int tiling_mode) +{ + int tile_width; + + /* Linear is always fine */ + if (tiling_mode == I915_TILING_NONE) + return true; + + if (tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev)) + tile_width = 128; + else + tile_width = 512; + + /* 965+ just needs multiples of tile width */ + if (IS_I965G(dev)) { + if (stride & (tile_width - 1)) + return false; + return true; + } + + /* Pre-965 needs power of two tile widths */ + if (stride < tile_width) + return false; + + if (stride & (stride - 1)) + return false; + + /* We don't handle the aperture area covered by the fence being bigger + * than the object size. + */ + if (i915_get_fence_size(dev, size) != size) + return false; + + return true; +} + /** * Sets the tiling mode of an object, returning the required swizzling of * bit 6 of addresses in the object. @@ -191,6 +258,9 @@ i915_gem_set_tiling(struct drm_device *dev, void *data, return -EINVAL; obj_priv = obj->driver_private; + if (!i915_tiling_ok(dev, args->stride, obj->size, args->tiling_mode)) + return -EINVAL; + mutex_lock(&dev->struct_mutex); if (args->tiling_mode == I915_TILING_NONE) { @@ -207,7 +277,23 @@ i915_gem_set_tiling(struct drm_device *dev, void *data, args->swizzle_mode = I915_BIT_6_SWIZZLE_NONE; } } - obj_priv->tiling_mode = args->tiling_mode; + if (args->tiling_mode != obj_priv->tiling_mode) { + int ret; + + /* Unbind the object, as switching tiling means we're + * switching the cache organization due to fencing, probably. + */ + ret = i915_gem_object_unbind(obj); + if (ret != 0) { + WARN(ret != -ERESTARTSYS, + "failed to unbind object for tiling switch"); + args->tiling_mode = obj_priv->tiling_mode; + mutex_unlock(&dev->struct_mutex); + + return ret; + } + obj_priv->tiling_mode = args->tiling_mode; + } obj_priv->stride = args->stride; mutex_unlock(&dev->struct_mutex); diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 273162579e1b..928e00462570 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -186,12 +186,12 @@ #define FENCE_REG_830_0 0x2000 #define I830_FENCE_START_MASK 0x07f80000 #define I830_FENCE_TILING_Y_SHIFT 12 -#define I830_FENCE_SIZE_BITS(size) ((get_order(size >> 19) - 1) << 8) +#define I830_FENCE_SIZE_BITS(size) ((ffs((size) >> 19) - 1) << 8) #define I830_FENCE_PITCH_SHIFT 4 #define I830_FENCE_REG_VALID (1<<0) #define I915_FENCE_START_MASK 0x0ff00000 -#define I915_FENCE_SIZE_BITS(size) ((get_order(size >> 20) - 1) << 8) +#define I915_FENCE_SIZE_BITS(size) ((ffs((size) >> 20) - 1) << 8) #define FENCE_REG_965_0 0x03000 #define I965_FENCE_PITCH_SHIFT 2 diff --git a/include/drm/i915_drm.h b/include/drm/i915_drm.h index b3bcf72dc656..912cd52db965 100644 --- a/include/drm/i915_drm.h +++ b/include/drm/i915_drm.h @@ -261,6 +261,7 @@ typedef struct drm_i915_irq_wait { #define I915_PARAM_LAST_DISPATCH 3 #define I915_PARAM_CHIPSET_ID 4 #define I915_PARAM_HAS_GEM 5 +#define I915_PARAM_NUM_FENCES_AVAIL 6 typedef struct drm_i915_getparam { int param; @@ -272,6 +273,7 @@ typedef struct drm_i915_getparam { #define I915_SETPARAM_USE_MI_BATCHBUFFER_START 1 #define I915_SETPARAM_TEX_LRU_LOG_GRANULARITY 2 #define I915_SETPARAM_ALLOW_BATCHBUFFER 3 +#define I915_SETPARAM_NUM_USED_FENCES 4 typedef struct drm_i915_setparam { int param; -- cgit v1.2.3-70-g09d2 From 766ccb9ed406c230d13c145def08ebea1b932982 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Tue, 20 Jan 2009 15:31:31 +0100 Subject: async: Rename _special -> _domain for clarity. Rename the async_*_special() functions to async_*_domain(), which describes the purpose of these functions much better. [Broke up long lines to silence checkpatch] Signed-off-by: Cornelia Huck Signed-off-by: Arjan van de Ven --- fs/super.c | 4 ++-- include/linux/async.h | 8 +++++--- kernel/async.c | 41 ++++++++++++++++++++++------------------- 3 files changed, 29 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/fs/super.c b/fs/super.c index 645e5403f2a0..61dce001dd57 100644 --- a/fs/super.c +++ b/fs/super.c @@ -301,7 +301,7 @@ void generic_shutdown_super(struct super_block *sb) /* * wait for asynchronous fs operations to finish before going further */ - async_synchronize_full_special(&sb->s_async_list); + async_synchronize_full_domain(&sb->s_async_list); /* bad name - it should be evict_inodes() */ invalidate_inodes(sb); @@ -470,7 +470,7 @@ restart: sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); - async_synchronize_full_special(&sb->s_async_list); + async_synchronize_full_domain(&sb->s_async_list); if (sb->s_root && (wait || sb->s_dirt)) sb->s_op->sync_fs(sb, wait); up_read(&sb->s_umount); diff --git a/include/linux/async.h b/include/linux/async.h index c4ecacd0b327..68a9530196f2 100644 --- a/include/linux/async.h +++ b/include/linux/async.h @@ -17,9 +17,11 @@ typedef u64 async_cookie_t; typedef void (async_func_ptr) (void *data, async_cookie_t cookie); extern async_cookie_t async_schedule(async_func_ptr *ptr, void *data); -extern async_cookie_t async_schedule_special(async_func_ptr *ptr, void *data, struct list_head *list); +extern async_cookie_t async_schedule_domain(async_func_ptr *ptr, void *data, + struct list_head *list); extern void async_synchronize_full(void); -extern void async_synchronize_full_special(struct list_head *list); +extern void async_synchronize_full_domain(struct list_head *list); extern void async_synchronize_cookie(async_cookie_t cookie); -extern void async_synchronize_cookie_special(async_cookie_t cookie, struct list_head *list); +extern void async_synchronize_cookie_domain(async_cookie_t cookie, + struct list_head *list); diff --git a/kernel/async.c b/kernel/async.c index b5f0d4b94937..e23399d88bac 100644 --- a/kernel/async.c +++ b/kernel/async.c @@ -224,22 +224,23 @@ async_cookie_t async_schedule(async_func_ptr *ptr, void *data) EXPORT_SYMBOL_GPL(async_schedule); /** - * async_schedule_special - schedule a function for asynchronous execution with a special running queue + * async_schedule_domain - schedule a function for asynchronous execution within a certain domain * @ptr: function to execute asynchronously * @data: data pointer to pass to the function - * @running: list head to add to while running + * @running: running list for the domain * * Returns an async_cookie_t that may be used for checkpointing later. - * @running may be used in the async_synchronize_*_special() functions - * to wait on a special running queue rather than on the global running - * queue. + * @running may be used in the async_synchronize_*_domain() functions + * to wait within a certain synchronization domain rather than globally. + * A synchronization domain is specified via the running queue @running to use. * Note: This function may be called from atomic or non-atomic contexts. */ -async_cookie_t async_schedule_special(async_func_ptr *ptr, void *data, struct list_head *running) +async_cookie_t async_schedule_domain(async_func_ptr *ptr, void *data, + struct list_head *running) { return __async_schedule(ptr, data, running); } -EXPORT_SYMBOL_GPL(async_schedule_special); +EXPORT_SYMBOL_GPL(async_schedule_domain); /** * async_synchronize_full - synchronize all asynchronous function calls @@ -255,27 +256,29 @@ void async_synchronize_full(void) EXPORT_SYMBOL_GPL(async_synchronize_full); /** - * async_synchronize_full_special - synchronize all asynchronous function calls for a running list + * async_synchronize_full_domain - synchronize all asynchronous function within a certain domain * @list: running list to synchronize on * - * This function waits until all asynchronous function calls for the running - * list @list have been done. + * This function waits until all asynchronous function calls for the + * synchronization domain specified by the running list @list have been done. */ -void async_synchronize_full_special(struct list_head *list) +void async_synchronize_full_domain(struct list_head *list) { - async_synchronize_cookie_special(next_cookie, list); + async_synchronize_cookie_domain(next_cookie, list); } -EXPORT_SYMBOL_GPL(async_synchronize_full_special); +EXPORT_SYMBOL_GPL(async_synchronize_full_domain); /** - * async_synchronize_cookie_special - synchronize asynchronous function calls on a running list with cookie checkpointing + * async_synchronize_cookie_domain - synchronize asynchronous function calls within a certain domain with cookie checkpointing * @cookie: async_cookie_t to use as checkpoint * @running: running list to synchronize on * - * This function waits until all asynchronous function calls for the running - * list @list submitted prior to @cookie have been done. + * This function waits until all asynchronous function calls for the + * synchronization domain specified by the running list @list submitted + * prior to @cookie have been done. */ -void async_synchronize_cookie_special(async_cookie_t cookie, struct list_head *running) +void async_synchronize_cookie_domain(async_cookie_t cookie, + struct list_head *running) { ktime_t starttime, delta, endtime; @@ -295,7 +298,7 @@ void async_synchronize_cookie_special(async_cookie_t cookie, struct list_head *r (long long)ktime_to_ns(delta) >> 10); } } -EXPORT_SYMBOL_GPL(async_synchronize_cookie_special); +EXPORT_SYMBOL_GPL(async_synchronize_cookie_domain); /** * async_synchronize_cookie - synchronize asynchronous function calls with cookie checkpointing @@ -306,7 +309,7 @@ EXPORT_SYMBOL_GPL(async_synchronize_cookie_special); */ void async_synchronize_cookie(async_cookie_t cookie) { - async_synchronize_cookie_special(cookie, &async_running); + async_synchronize_cookie_domain(cookie, &async_running); } EXPORT_SYMBOL_GPL(async_synchronize_cookie); -- cgit v1.2.3-70-g09d2 From 704126ad81b8cb7d3d70adb9ecb143f4d3fb38af Mon Sep 17 00:00:00 2001 From: Yu Zhao Date: Sun, 4 Jan 2009 16:28:52 +0800 Subject: VT-d: handle Invalidation Queue Error to avoid system hang When hardware detects any error with a descriptor from the invalidation queue, it stops fetching new descriptors from the queue until software clears the Invalidation Queue Error bit in the Fault Status register. Following fix handles the IQE so the kernel won't be trapped in an infinite loop. Signed-off-by: Yu Zhao Signed-off-by: David Woodhouse --- drivers/pci/dmar.c | 61 ++++++++++++++++++++++++++++++++------------ drivers/pci/intr_remapping.c | 21 ++++++++------- include/linux/intel-iommu.h | 3 ++- 3 files changed, 59 insertions(+), 26 deletions(-) (limited to 'include') diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c index 2b4162d9ca30..8d3e9c261061 100644 --- a/drivers/pci/dmar.c +++ b/drivers/pci/dmar.c @@ -573,19 +573,49 @@ static inline void reclaim_free_desc(struct q_inval *qi) } } +static int qi_check_fault(struct intel_iommu *iommu, int index) +{ + u32 fault; + int head; + struct q_inval *qi = iommu->qi; + int wait_index = (index + 1) % QI_LENGTH; + + fault = readl(iommu->reg + DMAR_FSTS_REG); + + /* + * If IQE happens, the head points to the descriptor associated + * with the error. No new descriptors are fetched until the IQE + * is cleared. + */ + if (fault & DMA_FSTS_IQE) { + head = readl(iommu->reg + DMAR_IQH_REG); + if ((head >> 4) == index) { + memcpy(&qi->desc[index], &qi->desc[wait_index], + sizeof(struct qi_desc)); + __iommu_flush_cache(iommu, &qi->desc[index], + sizeof(struct qi_desc)); + writel(DMA_FSTS_IQE, iommu->reg + DMAR_FSTS_REG); + return -EINVAL; + } + } + + return 0; +} + /* * Submit the queued invalidation descriptor to the remapping * hardware unit and wait for its completion. */ -void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu) +int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu) { + int rc = 0; struct q_inval *qi = iommu->qi; struct qi_desc *hw, wait_desc; int wait_index, index; unsigned long flags; if (!qi) - return; + return 0; hw = qi->desc; @@ -603,7 +633,8 @@ void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu) hw[index] = *desc; - wait_desc.low = QI_IWD_STATUS_DATA(2) | QI_IWD_STATUS_WRITE | QI_IWD_TYPE; + wait_desc.low = QI_IWD_STATUS_DATA(QI_DONE) | + QI_IWD_STATUS_WRITE | QI_IWD_TYPE; wait_desc.high = virt_to_phys(&qi->desc_status[wait_index]); hw[wait_index] = wait_desc; @@ -614,13 +645,11 @@ void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu) qi->free_head = (qi->free_head + 2) % QI_LENGTH; qi->free_cnt -= 2; - spin_lock(&iommu->register_lock); /* * update the HW tail register indicating the presence of * new descriptors. */ writel(qi->free_head << 4, iommu->reg + DMAR_IQT_REG); - spin_unlock(&iommu->register_lock); while (qi->desc_status[wait_index] != QI_DONE) { /* @@ -630,15 +659,21 @@ void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu) * a deadlock where the interrupt context can wait indefinitely * for free slots in the queue. */ + rc = qi_check_fault(iommu, index); + if (rc) + goto out; + spin_unlock(&qi->q_lock); cpu_relax(); spin_lock(&qi->q_lock); } - - qi->desc_status[index] = QI_DONE; +out: + qi->desc_status[index] = qi->desc_status[wait_index] = QI_DONE; reclaim_free_desc(qi); spin_unlock_irqrestore(&qi->q_lock, flags); + + return rc; } /* @@ -651,13 +686,13 @@ void qi_global_iec(struct intel_iommu *iommu) desc.low = QI_IEC_TYPE; desc.high = 0; + /* should never fail */ qi_submit_sync(&desc, iommu); } int qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid, u8 fm, u64 type, int non_present_entry_flush) { - struct qi_desc desc; if (non_present_entry_flush) { @@ -671,10 +706,7 @@ int qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid, u8 fm, | QI_CC_GRAN(type) | QI_CC_TYPE; desc.high = 0; - qi_submit_sync(&desc, iommu); - - return 0; - + return qi_submit_sync(&desc, iommu); } int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr, @@ -704,10 +736,7 @@ int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr, desc.high = QI_IOTLB_ADDR(addr) | QI_IOTLB_IH(ih) | QI_IOTLB_AM(size_order); - qi_submit_sync(&desc, iommu); - - return 0; - + return qi_submit_sync(&desc, iommu); } /* diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c index f78371b22529..45effc5726c0 100644 --- a/drivers/pci/intr_remapping.c +++ b/drivers/pci/intr_remapping.c @@ -207,7 +207,7 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count) return index; } -static void qi_flush_iec(struct intel_iommu *iommu, int index, int mask) +static int qi_flush_iec(struct intel_iommu *iommu, int index, int mask) { struct qi_desc desc; @@ -215,7 +215,7 @@ static void qi_flush_iec(struct intel_iommu *iommu, int index, int mask) | QI_IEC_SELECTIVE; desc.high = 0; - qi_submit_sync(&desc, iommu); + return qi_submit_sync(&desc, iommu); } int map_irq_to_irte_handle(int irq, u16 *sub_handle) @@ -283,6 +283,7 @@ int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index) int modify_irte(int irq, struct irte *irte_modified) { + int rc; int index; struct irte *irte; struct intel_iommu *iommu; @@ -303,14 +304,15 @@ int modify_irte(int irq, struct irte *irte_modified) set_64bit((unsigned long *)irte, irte_modified->low | (1 << 1)); __iommu_flush_cache(iommu, irte, sizeof(*irte)); - qi_flush_iec(iommu, index, 0); - + rc = qi_flush_iec(iommu, index, 0); spin_unlock(&irq_2_ir_lock); - return 0; + + return rc; } int flush_irte(int irq) { + int rc; int index; struct intel_iommu *iommu; struct irq_2_iommu *irq_iommu; @@ -326,10 +328,10 @@ int flush_irte(int irq) index = irq_iommu->irte_index + irq_iommu->sub_handle; - qi_flush_iec(iommu, index, irq_iommu->irte_mask); + rc = qi_flush_iec(iommu, index, irq_iommu->irte_mask); spin_unlock(&irq_2_ir_lock); - return 0; + return rc; } struct intel_iommu *map_ioapic_to_ir(int apic) @@ -355,6 +357,7 @@ struct intel_iommu *map_dev_to_ir(struct pci_dev *dev) int free_irte(int irq) { + int rc = 0; int index, i; struct irte *irte; struct intel_iommu *iommu; @@ -375,7 +378,7 @@ int free_irte(int irq) if (!irq_iommu->sub_handle) { for (i = 0; i < (1 << irq_iommu->irte_mask); i++) set_64bit((unsigned long *)irte, 0); - qi_flush_iec(iommu, index, irq_iommu->irte_mask); + rc = qi_flush_iec(iommu, index, irq_iommu->irte_mask); } irq_iommu->iommu = NULL; @@ -385,7 +388,7 @@ int free_irte(int irq) spin_unlock(&irq_2_ir_lock); - return 0; + return rc; } static void iommu_set_intr_remapping(struct intel_iommu *iommu, int mode) diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h index c4f6c101dbcd..d2e3cbfba14f 100644 --- a/include/linux/intel-iommu.h +++ b/include/linux/intel-iommu.h @@ -194,6 +194,7 @@ static inline void dmar_writeq(void __iomem *addr, u64 val) /* FSTS_REG */ #define DMA_FSTS_PPF ((u32)2) #define DMA_FSTS_PFO ((u32)1) +#define DMA_FSTS_IQE (1 << 4) #define dma_fsts_fault_record_index(s) (((s) >> 8) & 0xff) /* FRCD_REG, 32 bits access */ @@ -328,7 +329,7 @@ extern int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr, unsigned int size_order, u64 type, int non_present_entry_flush); -extern void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu); +extern int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu); extern void *intel_alloc_coherent(struct device *, size_t, dma_addr_t *, gfp_t); extern void intel_free_coherent(struct device *, size_t, void *, dma_addr_t); -- cgit v1.2.3-70-g09d2 From a5ef7ca0e2636bad0ccd07b996d775348ae2b65e Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Sun, 8 Feb 2009 17:39:58 -0500 Subject: x86: spinlocks: define dummy __raw_spin_is_contended Architectures other than mips and x86 are not using ticket spinlocks. Therefore, the contention on the lock is meaningless, since there is nobody known to be waiting on it (arguably /fairly/ unfair locks). Dummy it out to return 0 on other architectures. Signed-off-by: Kyle McMartin Acked-by: Ralf Baechle Acked-by: Ingo Molnar Signed-off-by: Linus Torvalds --- arch/mips/include/asm/spinlock.h | 1 + arch/x86/include/asm/paravirt.h | 1 + arch/x86/include/asm/spinlock.h | 1 + include/linux/spinlock.h | 5 +++++ 4 files changed, 8 insertions(+) (limited to 'include') diff --git a/arch/mips/include/asm/spinlock.h b/arch/mips/include/asm/spinlock.h index 1a1f320c30d8..0884947ebe27 100644 --- a/arch/mips/include/asm/spinlock.h +++ b/arch/mips/include/asm/spinlock.h @@ -51,6 +51,7 @@ static inline int __raw_spin_is_contended(raw_spinlock_t *lock) return (((counters >> 14) - counters) & 0x1fff) > 1; } +#define __raw_spin_is_contended __raw_spin_is_contended static inline void __raw_spin_lock(raw_spinlock_t *lock) { diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index ba3e2ff6aedc..c09a14127584 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -1402,6 +1402,7 @@ static inline int __raw_spin_is_contended(struct raw_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_is_contended, lock); } +#define __raw_spin_is_contended __raw_spin_is_contended static __always_inline void __raw_spin_lock(struct raw_spinlock *lock) { diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h index d17c91981da2..8247e94ac6b1 100644 --- a/arch/x86/include/asm/spinlock.h +++ b/arch/x86/include/asm/spinlock.h @@ -245,6 +245,7 @@ static inline int __raw_spin_is_contended(raw_spinlock_t *lock) { return __ticket_spin_is_contended(lock); } +#define __raw_spin_is_contended __raw_spin_is_contended static __always_inline void __raw_spin_lock(raw_spinlock_t *lock) { diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index e0c0fccced46..a0c66a2e00ad 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -124,7 +124,12 @@ do { \ #ifdef CONFIG_GENERIC_LOCKBREAK #define spin_is_contended(lock) ((lock)->break_lock) #else + +#ifdef __raw_spin_is_contended #define spin_is_contended(lock) __raw_spin_is_contended(&(lock)->raw_lock) +#else +#define spin_is_contended(lock) (((void)(lock), 0)) +#endif /*__raw_spin_is_contended*/ #endif /** -- cgit v1.2.3-70-g09d2 From 43a990765a9e874350bae1009366d00809dbc9d8 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 10 Feb 2009 00:00:22 +0100 Subject: sound: Remove OSSlib stuff from linux/soundcard.h Removed OSSlib stuff from linux/soundcard.h to fix the warnings for 'make headers_check'. This patch breaks building against OSSlib with the kernel headers instead of its own headers. It should still work with any version of the library from the 2003 onwards which provide their own headers for the latest interface. Signed-off-by: Arnd Bergmann Cc: Jaswinder Singh Rajput Signed-off-by: Takashi Iwai --- include/linux/soundcard.h | 74 +++++++++++++++-------------------------------- 1 file changed, 23 insertions(+), 51 deletions(-) (limited to 'include') diff --git a/include/linux/soundcard.h b/include/linux/soundcard.h index 523d069c862c..1904afedb82f 100644 --- a/include/linux/soundcard.h +++ b/include/linux/soundcard.h @@ -1045,50 +1045,36 @@ typedef struct mixer_vol_table { */ #define LOCL_STARTAUDIO 1 -#if (!defined(__KERNEL__) && !defined(KERNEL) && !defined(INKERNEL) && !defined(_KERNEL)) || defined(USE_SEQ_MACROS) +#if !defined(__KERNEL__) || defined(USE_SEQ_MACROS) /* * Some convenience macros to simplify programming of the * /dev/sequencer interface * - * These macros define the API which should be used when possible. + * This is a legacy interface for applications written against + * the OSSlib-3.8 style interface. It is no longer possible + * to actually link against OSSlib with this header, but we + * still provide these macros for programs using them. + * + * If you want to use OSSlib, it is recommended that you get + * the GPL version of OSS-4.x and build against that version + * of the header. + * + * We redefine the extern keyword so that make headers_check + * does not complain about SEQ_USE_EXTBUF. */ #define SEQ_DECLAREBUF() SEQ_USE_EXTBUF() void seqbuf_dump(void); /* This function must be provided by programs */ -extern int OSS_init(int seqfd, int buflen); -extern void OSS_seqbuf_dump(int fd, unsigned char *buf, int buflen); -extern void OSS_seq_advbuf(int len, int fd, unsigned char *buf, int buflen); -extern void OSS_seq_needbuf(int len, int fd, unsigned char *buf, int buflen); -extern void OSS_patch_caching(int dev, int chn, int patch, - int fd, unsigned char *buf, int buflen); -extern void OSS_drum_caching(int dev, int chn, int patch, - int fd, unsigned char *buf, int buflen); -extern void OSS_write_patch(int fd, unsigned char *buf, int len); -extern int OSS_write_patch2(int fd, unsigned char *buf, int len); - #define SEQ_PM_DEFINES int __foo_bar___ -#ifdef OSSLIB -# define SEQ_USE_EXTBUF() \ - extern unsigned char *_seqbuf; \ - extern int _seqbuflen;extern int _seqbufptr -# define SEQ_DEFINEBUF(len) SEQ_USE_EXTBUF();static int _requested_seqbuflen=len -# define _SEQ_ADVBUF(len) OSS_seq_advbuf(len, seqfd, _seqbuf, _seqbuflen) -# define _SEQ_NEEDBUF(len) OSS_seq_needbuf(len, seqfd, _seqbuf, _seqbuflen) -# define SEQ_DUMPBUF() OSS_seqbuf_dump(seqfd, _seqbuf, _seqbuflen) - -# define SEQ_LOAD_GMINSTR(dev, instr) \ - OSS_patch_caching(dev, -1, instr, seqfd, _seqbuf, _seqbuflen) -# define SEQ_LOAD_GMDRUM(dev, drum) \ - OSS_drum_caching(dev, -1, drum, seqfd, _seqbuf, _seqbuflen) -#else /* !OSSLIB */ - -# define SEQ_LOAD_GMINSTR(dev, instr) -# define SEQ_LOAD_GMDRUM(dev, drum) - -# define SEQ_USE_EXTBUF() \ - extern unsigned char _seqbuf[]; \ - extern int _seqbuflen;extern int _seqbufptr + +#define SEQ_LOAD_GMINSTR(dev, instr) +#define SEQ_LOAD_GMDRUM(dev, drum) + +#define _SEQ_EXTERN extern +#define SEQ_USE_EXTBUF() \ + _SEQ_EXTERN unsigned char _seqbuf[]; \ + _SEQ_EXTERN int _seqbuflen; _SEQ_EXTERN int _seqbufptr #ifndef USE_SIMPLE_MACROS /* Sample seqbuf_dump() implementation: @@ -1131,7 +1117,6 @@ extern int OSS_write_patch2(int fd, unsigned char *buf, int len); */ #define _SEQ_NEEDBUF(len) /* empty */ #endif -#endif /* !OSSLIB */ #define SEQ_VOLUME_MODE(dev, mode) {_SEQ_NEEDBUF(8);\ _seqbuf[_seqbufptr] = SEQ_EXTENDED;\ @@ -1215,14 +1200,8 @@ extern int OSS_write_patch2(int fd, unsigned char *buf, int len); _CHN_COMMON(dev, MIDI_CHN_PRESSURE, chn, pressure, 0, 0) #define SEQ_SET_PATCH SEQ_PGM_CHANGE -#ifdef OSSLIB -# define SEQ_PGM_CHANGE(dev, chn, patch) \ - {OSS_patch_caching(dev, chn, patch, seqfd, _seqbuf, _seqbuflen); \ - _CHN_COMMON(dev, MIDI_PGM_CHANGE, chn, patch, 0, 0);} -#else -# define SEQ_PGM_CHANGE(dev, chn, patch) \ +#define SEQ_PGM_CHANGE(dev, chn, patch) \ _CHN_COMMON(dev, MIDI_PGM_CHANGE, chn, patch, 0, 0) -#endif #define SEQ_CONTROL(dev, chn, controller, value) \ _CHN_COMMON(dev, MIDI_CTL_CHANGE, chn, controller, 0, value) @@ -1300,19 +1279,12 @@ extern int OSS_write_patch2(int fd, unsigned char *buf, int len); /* * Patch loading. */ -#ifdef OSSLIB -# define SEQ_WRPATCH(patchx, len) \ - OSS_write_patch(seqfd, (char*)(patchx), len) -# define SEQ_WRPATCH2(patchx, len) \ - OSS_write_patch2(seqfd, (char*)(patchx), len) -#else -# define SEQ_WRPATCH(patchx, len) \ +#define SEQ_WRPATCH(patchx, len) \ {if (_seqbufptr) SEQ_DUMPBUF();\ if (write(seqfd, (char*)(patchx), len)==-1) \ perror("Write patch: /dev/sequencer");} -# define SEQ_WRPATCH2(patchx, len) \ +#define SEQ_WRPATCH2(patchx, len) \ (SEQ_DUMPBUF(), write(seqfd, (char*)(patchx), len)) -#endif #endif #endif -- cgit v1.2.3-70-g09d2 From 7f5aa215088b817add9c71914b83650bdd49f8a9 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 10 Feb 2009 11:15:34 -0500 Subject: jbd2: Avoid possible NULL dereference in jbd2_journal_begin_ordered_truncate() If we race with commit code setting i_transaction to NULL, we could possibly dereference it. Proper locking requires the journal pointer (to access journal->j_list_lock), which we don't have. So we have to change the prototype of the function so that filesystem passes us the journal pointer. Also add a more detailed comment about why the function jbd2_journal_begin_ordered_truncate() does what it does and how it should be used. Thanks to Dan Carpenter for pointing to the suspitious code. Signed-off-by: Jan Kara Signed-off-by: "Theodore Ts'o" Acked-by: Joel Becker CC: linux-ext4@vger.kernel.org CC: ocfs2-devel@oss.oracle.com CC: mfasheh@suse.de CC: Dan Carpenter --- fs/ext4/inode.c | 6 ++++-- fs/jbd2/transaction.c | 42 +++++++++++++++++++++++++++++++----------- fs/ocfs2/journal.h | 6 ++++-- include/linux/jbd2.h | 3 ++- 4 files changed, 41 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 03ba20be1329..658c4a7f2578 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -47,8 +47,10 @@ static inline int ext4_begin_ordered_truncate(struct inode *inode, loff_t new_size) { - return jbd2_journal_begin_ordered_truncate(&EXT4_I(inode)->jinode, - new_size); + return jbd2_journal_begin_ordered_truncate( + EXT4_SB(inode->i_sb)->s_journal, + &EXT4_I(inode)->jinode, + new_size); } static void ext4_invalidatepage(struct page *page, unsigned long offset); diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 46b4e347ed7d..28ce21d8598e 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -2129,26 +2129,46 @@ done: } /* - * This function must be called when inode is journaled in ordered mode - * before truncation happens. It starts writeout of truncated part in - * case it is in the committing transaction so that we stand to ordered - * mode consistency guarantees. + * File truncate and transaction commit interact with each other in a + * non-trivial way. If a transaction writing data block A is + * committing, we cannot discard the data by truncate until we have + * written them. Otherwise if we crashed after the transaction with + * write has committed but before the transaction with truncate has + * committed, we could see stale data in block A. This function is a + * helper to solve this problem. It starts writeout of the truncated + * part in case it is in the committing transaction. + * + * Filesystem code must call this function when inode is journaled in + * ordered mode before truncation happens and after the inode has been + * placed on orphan list with the new inode size. The second condition + * avoids the race that someone writes new data and we start + * committing the transaction after this function has been called but + * before a transaction for truncate is started (and furthermore it + * allows us to optimize the case where the addition to orphan list + * happens in the same transaction as write --- we don't have to write + * any data in such case). */ -int jbd2_journal_begin_ordered_truncate(struct jbd2_inode *inode, +int jbd2_journal_begin_ordered_truncate(journal_t *journal, + struct jbd2_inode *jinode, loff_t new_size) { - journal_t *journal; - transaction_t *commit_trans; + transaction_t *inode_trans, *commit_trans; int ret = 0; - if (!inode->i_transaction && !inode->i_next_transaction) + /* This is a quick check to avoid locking if not necessary */ + if (!jinode->i_transaction) goto out; - journal = inode->i_transaction->t_journal; + /* Locks are here just to force reading of recent values, it is + * enough that the transaction was not committing before we started + * a transaction adding the inode to orphan list */ spin_lock(&journal->j_state_lock); commit_trans = journal->j_committing_transaction; spin_unlock(&journal->j_state_lock); - if (inode->i_transaction == commit_trans) { - ret = filemap_fdatawrite_range(inode->i_vfs_inode->i_mapping, + spin_lock(&journal->j_list_lock); + inode_trans = jinode->i_transaction; + spin_unlock(&journal->j_list_lock); + if (inode_trans == commit_trans) { + ret = filemap_fdatawrite_range(jinode->i_vfs_inode->i_mapping, new_size, LLONG_MAX); if (ret) jbd2_journal_abort(journal, ret); diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index 3c3532e1307c..172850a9a12a 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -513,8 +513,10 @@ static inline int ocfs2_jbd2_file_inode(handle_t *handle, struct inode *inode) static inline int ocfs2_begin_ordered_truncate(struct inode *inode, loff_t new_size) { - return jbd2_journal_begin_ordered_truncate(&OCFS2_I(inode)->ip_jinode, - new_size); + return jbd2_journal_begin_ordered_truncate( + OCFS2_SB(inode->i_sb)->journal->j_journal, + &OCFS2_I(inode)->ip_jinode, + new_size); } #endif /* OCFS2_JOURNAL_H */ diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index b28b37eb11c6..4d248b3f1323 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1150,7 +1150,8 @@ extern int jbd2_journal_clear_err (journal_t *); extern int jbd2_journal_bmap(journal_t *, unsigned long, unsigned long long *); extern int jbd2_journal_force_commit(journal_t *); extern int jbd2_journal_file_inode(handle_t *handle, struct jbd2_inode *inode); -extern int jbd2_journal_begin_ordered_truncate(struct jbd2_inode *inode, loff_t new_size); +extern int jbd2_journal_begin_ordered_truncate(journal_t *journal, + struct jbd2_inode *inode, loff_t new_size); extern void jbd2_journal_init_jbd_inode(struct jbd2_inode *jinode, struct inode *inode); extern void jbd2_journal_release_jbd_inode(journal_t *journal, struct jbd2_inode *jinode); -- cgit v1.2.3-70-g09d2 From 5a6fe125950676015f5108fb71b2a67441755003 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Tue, 10 Feb 2009 14:02:27 +0000 Subject: Do not account for the address space used by hugetlbfs using VM_ACCOUNT When overcommit is disabled, the core VM accounts for pages used by anonymous shared, private mappings and special mappings. It keeps track of VMAs that should be accounted for with VM_ACCOUNT and VMAs that never had a reserve with VM_NORESERVE. Overcommit for hugetlbfs is much riskier than overcommit for base pages due to contiguity requirements. It avoids overcommiting on both shared and private mappings using reservation counters that are checked and updated during mmap(). This ensures (within limits) that hugepages exist in the future when faults occurs or it is too easy to applications to be SIGKILLed. As hugetlbfs makes its own reservations of a different unit to the base page size, VM_ACCOUNT should never be set. Even if the units were correct, we would double account for the usage in the core VM and hugetlbfs. VM_NORESERVE may be set because an application can request no reserves be made for hugetlbfs at the risk of getting killed later. With commit fc8744adc870a8d4366908221508bb113d8b72ee, VM_NORESERVE and VM_ACCOUNT are getting unconditionally set for hugetlbfs-backed mappings. This breaks the accounting for both the core VM and hugetlbfs, can trigger an OOM storm when hugepage pools are too small lockups and corrupted counters otherwise are used. This patch brings hugetlbfs more in line with how the core VM treats VM_NORESERVE but prevents VM_ACCOUNT being set. Signed-off-by: Mel Gorman Signed-off-by: Linus Torvalds --- fs/hugetlbfs/inode.c | 8 +++++--- include/linux/hugetlb.h | 5 +++-- include/linux/mm.h | 3 +-- ipc/shm.c | 8 +++++--- mm/fremap.c | 2 +- mm/hugetlb.c | 39 +++++++++++++++++++++++++-------------- mm/mmap.c | 38 ++++++++++++++++++++++---------------- mm/mprotect.c | 5 +++-- 8 files changed, 65 insertions(+), 43 deletions(-) (limited to 'include') diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 6903d37af037..9b800d97a687 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -108,7 +108,8 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) if (hugetlb_reserve_pages(inode, vma->vm_pgoff >> huge_page_order(h), - len >> huge_page_shift(h), vma)) + len >> huge_page_shift(h), vma, + vma->vm_flags)) goto out; ret = 0; @@ -947,7 +948,7 @@ static int can_do_hugetlb_shm(void) can_do_mlock()); } -struct file *hugetlb_file_setup(const char *name, size_t size) +struct file *hugetlb_file_setup(const char *name, size_t size, int acctflag) { int error = -ENOMEM; struct file *file; @@ -981,7 +982,8 @@ struct file *hugetlb_file_setup(const char *name, size_t size) error = -ENOMEM; if (hugetlb_reserve_pages(inode, 0, - size >> huge_page_shift(hstate_inode(inode)), NULL)) + size >> huge_page_shift(hstate_inode(inode)), NULL, + acctflag)) goto out_inode; d_instantiate(dentry, inode); diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index f1d2fba19ea0..af09660001c7 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -33,7 +33,8 @@ unsigned long hugetlb_total_pages(void); int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, int write_access); int hugetlb_reserve_pages(struct inode *inode, long from, long to, - struct vm_area_struct *vma); + struct vm_area_struct *vma, + int acctflags); void hugetlb_unreserve_pages(struct inode *inode, long offset, long freed); extern unsigned long hugepages_treat_as_movable; @@ -138,7 +139,7 @@ static inline struct hugetlbfs_sb_info *HUGETLBFS_SB(struct super_block *sb) extern const struct file_operations hugetlbfs_file_operations; extern struct vm_operations_struct hugetlb_vm_ops; -struct file *hugetlb_file_setup(const char *name, size_t); +struct file *hugetlb_file_setup(const char *name, size_t, int); int hugetlb_get_quota(struct address_space *mapping, long delta); void hugetlb_put_quota(struct address_space *mapping, long delta); diff --git a/include/linux/mm.h b/include/linux/mm.h index e8ddc98b8405..323561582c10 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1129,8 +1129,7 @@ extern unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, unsigned long flag, unsigned long pgoff); extern unsigned long mmap_region(struct file *file, unsigned long addr, unsigned long len, unsigned long flags, - unsigned int vm_flags, unsigned long pgoff, - int accountable); + unsigned int vm_flags, unsigned long pgoff); static inline unsigned long do_mmap(struct file *file, unsigned long addr, unsigned long len, unsigned long prot, diff --git a/ipc/shm.c b/ipc/shm.c index f8f69fad3a27..05d51d2a792c 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -340,6 +340,7 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params) struct file * file; char name[13]; int id; + int acctflag = 0; if (size < SHMMIN || size > ns->shm_ctlmax) return -EINVAL; @@ -364,11 +365,12 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params) sprintf (name, "SYSV%08x", key); if (shmflg & SHM_HUGETLB) { - /* hugetlb_file_setup takes care of mlock user accounting */ - file = hugetlb_file_setup(name, size); + /* hugetlb_file_setup applies strict accounting */ + if (shmflg & SHM_NORESERVE) + acctflag = VM_NORESERVE; + file = hugetlb_file_setup(name, size, acctflag); shp->mlock_user = current_user(); } else { - int acctflag = 0; /* * Do not allow no accounting for OVERCOMMIT_NEVER, even * if it's asked for. diff --git a/mm/fremap.c b/mm/fremap.c index 736ba7f3306a..b6ec85abbb39 100644 --- a/mm/fremap.c +++ b/mm/fremap.c @@ -198,7 +198,7 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size, flags &= MAP_NONBLOCK; get_file(file); addr = mmap_region(file, start, size, - flags, vma->vm_flags, pgoff, 1); + flags, vma->vm_flags, pgoff); fput(file); if (IS_ERR_VALUE(addr)) { err = addr; diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 618e98304080..207464209546 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -2269,14 +2269,12 @@ void hugetlb_change_protection(struct vm_area_struct *vma, int hugetlb_reserve_pages(struct inode *inode, long from, long to, - struct vm_area_struct *vma) + struct vm_area_struct *vma, + int acctflag) { - long ret, chg; + long ret = 0, chg; struct hstate *h = hstate_inode(inode); - if (vma && vma->vm_flags & VM_NORESERVE) - return 0; - /* * Shared mappings base their reservation on the number of pages that * are already allocated on behalf of the file. Private mappings need @@ -2285,22 +2283,25 @@ int hugetlb_reserve_pages(struct inode *inode, */ if (!vma || vma->vm_flags & VM_SHARED) chg = region_chg(&inode->i_mapping->private_list, from, to); - else { - struct resv_map *resv_map = resv_map_alloc(); - if (!resv_map) - return -ENOMEM; - + else chg = to - from; - set_vma_resv_map(vma, resv_map); - set_vma_resv_flags(vma, HPAGE_RESV_OWNER); - } - if (chg < 0) return chg; if (hugetlb_get_quota(inode->i_mapping, chg)) return -ENOSPC; + + /* + * Only apply hugepage reservation if asked. We still have to + * take the filesystem quota because it is an upper limit + * defined for the mount and not necessarily memory as a whole + */ + if (acctflag & VM_NORESERVE) { + reset_vma_resv_huge_pages(vma); + return 0; + } + ret = hugetlb_acct_memory(h, chg); if (ret < 0) { hugetlb_put_quota(inode->i_mapping, chg); @@ -2308,6 +2309,16 @@ int hugetlb_reserve_pages(struct inode *inode, } if (!vma || vma->vm_flags & VM_SHARED) region_add(&inode->i_mapping->private_list, from, to); + else { + struct resv_map *resv_map = resv_map_alloc(); + + if (!resv_map) + return -ENOMEM; + + set_vma_resv_map(vma, resv_map); + set_vma_resv_flags(vma, HPAGE_RESV_OWNER); + } + return 0; } diff --git a/mm/mmap.c b/mm/mmap.c index 214b6a258eeb..eb1270bebe67 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -918,7 +918,6 @@ unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, struct inode *inode; unsigned int vm_flags; int error; - int accountable = 1; unsigned long reqprot = prot; /* @@ -1019,8 +1018,6 @@ unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, return -EPERM; vm_flags &= ~VM_MAYEXEC; } - if (is_file_hugepages(file)) - accountable = 0; if (!file->f_op || !file->f_op->mmap) return -ENODEV; @@ -1053,8 +1050,7 @@ unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, if (error) return error; - return mmap_region(file, addr, len, flags, vm_flags, pgoff, - accountable); + return mmap_region(file, addr, len, flags, vm_flags, pgoff); } EXPORT_SYMBOL(do_mmap_pgoff); @@ -1092,17 +1088,23 @@ int vma_wants_writenotify(struct vm_area_struct *vma) /* * We account for memory if it's a private writeable mapping, - * and VM_NORESERVE wasn't set. + * not hugepages and VM_NORESERVE wasn't set. */ -static inline int accountable_mapping(unsigned int vm_flags) +static inline int accountable_mapping(struct file *file, unsigned int vm_flags) { + /* + * hugetlb has its own accounting separate from the core VM + * VM_HUGETLB may not be set yet so we cannot check for that flag. + */ + if (file && is_file_hugepages(file)) + return 0; + return (vm_flags & (VM_NORESERVE | VM_SHARED | VM_WRITE)) == VM_WRITE; } unsigned long mmap_region(struct file *file, unsigned long addr, unsigned long len, unsigned long flags, - unsigned int vm_flags, unsigned long pgoff, - int accountable) + unsigned int vm_flags, unsigned long pgoff) { struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *prev; @@ -1128,18 +1130,22 @@ munmap_back: /* * Set 'VM_NORESERVE' if we should not account for the - * memory use of this mapping. We only honor MAP_NORESERVE - * if we're allowed to overcommit memory. + * memory use of this mapping. */ - if ((flags & MAP_NORESERVE) && sysctl_overcommit_memory != OVERCOMMIT_NEVER) - vm_flags |= VM_NORESERVE; - if (!accountable) - vm_flags |= VM_NORESERVE; + if ((flags & MAP_NORESERVE)) { + /* We honor MAP_NORESERVE if allowed to overcommit */ + if (sysctl_overcommit_memory != OVERCOMMIT_NEVER) + vm_flags |= VM_NORESERVE; + + /* hugetlb applies strict overcommit unless MAP_NORESERVE */ + if (file && is_file_hugepages(file)) + vm_flags |= VM_NORESERVE; + } /* * Private writable mapping: check memory availability */ - if (accountable_mapping(vm_flags)) { + if (accountable_mapping(file, vm_flags)) { charged = len >> PAGE_SHIFT; if (security_vm_enough_memory(charged)) return -ENOMEM; diff --git a/mm/mprotect.c b/mm/mprotect.c index abe2694e13f4..258197b76fb4 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -151,10 +151,11 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev, /* * If we make a private mapping writable we increase our commit; * but (without finer accounting) cannot reduce our commit if we - * make it unwritable again. + * make it unwritable again. hugetlb mapping were accounted for + * even if read-only so there is no need to account for them here */ if (newflags & VM_WRITE) { - if (!(oldflags & (VM_ACCOUNT|VM_WRITE| + if (!(oldflags & (VM_ACCOUNT|VM_WRITE|VM_HUGETLB| VM_SHARED|VM_NORESERVE))) { charged = nrpages; if (security_vm_enough_memory(charged)) -- cgit v1.2.3-70-g09d2 From 1db8508cf483dc1ecf66141f90a7c03659d69512 Mon Sep 17 00:00:00 2001 From: Stefan Richter Date: Tue, 10 Feb 2009 23:27:32 +0100 Subject: hugetlbfs: fix build failure with !CONFIG_HUGETLBFS Fix regression due to 5a6fe125950676015f5108fb71b2a67441755003, "Do not account for the address space used by hugetlbfs using VM_ACCOUNT" which added an argument to the function hugetlb_file_setup() but not to the macro hugetlb_file_setup(). Reported-by: Chris Clayton Signed-off-by: Stefan Richter Acked-by: Mel Gorman Signed-off-by: Linus Torvalds --- include/linux/hugetlb.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index af09660001c7..03be7f29ca01 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -159,9 +159,9 @@ static inline void set_file_hugepages(struct file *file) } #else /* !CONFIG_HUGETLBFS */ -#define is_file_hugepages(file) 0 -#define set_file_hugepages(file) BUG() -#define hugetlb_file_setup(name,size) ERR_PTR(-ENOSYS) +#define is_file_hugepages(file) 0 +#define set_file_hugepages(file) BUG() +#define hugetlb_file_setup(name,size,acctflag) ERR_PTR(-ENOSYS) #endif /* !CONFIG_HUGETLBFS */ -- cgit v1.2.3-70-g09d2 From e672f7db767156bf71adf9c592cfe81b339523d6 Mon Sep 17 00:00:00 2001 From: Chuck Ebbert Date: Tue, 10 Feb 2009 17:18:17 -0800 Subject: pkt_sched: type should be __u32 in header Using u32 in this header breaks the build of iptables. Signed-off-by: Chuck Ebbert Signed-off-by: David S. Miller --- include/linux/pkt_sched.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h index b2648e8e4987..d51a2b3e221e 100644 --- a/include/linux/pkt_sched.h +++ b/include/linux/pkt_sched.h @@ -515,7 +515,7 @@ enum struct tc_drr_stats { - u32 deficit; + __u32 deficit; }; #endif -- cgit v1.2.3-70-g09d2 From 3fccfd67df79c6351a156eb25a7a514e5f39c4d9 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 10 Feb 2009 16:37:31 +0100 Subject: timers: split process wide cpu clocks/timers, fix To decrease the chance of a missed enable, always enable the timer when we sample it, we'll always disable it when we find that there are no active timers in the jiffy tick. This fixes a flood of warnings reported by Mike Galbraith. Reported-by: Mike Galbraith Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 + kernel/posix-cpu-timers.c | 42 ++++++++++++++---------------------------- 2 files changed, 15 insertions(+), 28 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 79392916d6c9..5d10fa0b6002 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2209,6 +2209,7 @@ void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) unsigned long flags; spin_lock_irqsave(&cputimer->lock, flags); + cputimer->running = 1; *times = cputimer->cputime; spin_unlock_irqrestore(&cputimer->lock, flags); } diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index db107c9bbc05..e5d7bfdfa7d4 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -488,7 +488,7 @@ void posix_cpu_timers_exit_group(struct task_struct *tsk) { struct task_cputime cputime; - thread_group_cputime(tsk, &cputime); + thread_group_cputimer(tsk, &cputime); cleanup_timers(tsk->signal->cpu_timers, cputime.utime, cputime.stime, cputime.sum_exec_runtime); } @@ -506,29 +506,6 @@ static void clear_dead_task(struct k_itimer *timer, union cpu_time_count now) now); } -/* - * Enable the process wide cpu timer accounting. - * - * serialized using ->sighand->siglock - */ -static void start_process_timers(struct task_struct *tsk) -{ - tsk->signal->cputimer.running = 1; - barrier(); -} - -/* - * Release the process wide timer accounting -- timer stops ticking when - * nobody cares about it. - * - * serialized using ->sighand->siglock - */ -static void stop_process_timers(struct task_struct *tsk) -{ - tsk->signal->cputimer.running = 0; - barrier(); -} - /* * Insert the timer on the appropriate list before any timers that * expire later. This must be called with the tasklist_lock held @@ -549,9 +526,6 @@ static void arm_timer(struct k_itimer *timer, union cpu_time_count now) BUG_ON(!irqs_disabled()); spin_lock(&p->sighand->siglock); - if (!CPUCLOCK_PERTHREAD(timer->it_clock)) - start_process_timers(p); - listpos = head; if (CPUCLOCK_WHICH(timer->it_clock) == CPUCLOCK_SCHED) { list_for_each_entry(next, head, entry) { @@ -1021,6 +995,19 @@ static void check_thread_timers(struct task_struct *tsk, } } +static void stop_process_timers(struct task_struct *tsk) +{ + struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; + unsigned long flags; + + if (!cputimer->running) + return; + + spin_lock_irqsave(&cputimer->lock, flags); + cputimer->running = 0; + spin_unlock_irqrestore(&cputimer->lock, flags); +} + /* * Check for any per-thread CPU timers that have fired and move them * off the tsk->*_timers list onto the firing list. Per-thread timers @@ -1427,7 +1414,6 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, struct list_head *head; BUG_ON(clock_idx == CPUCLOCK_SCHED); - start_process_timers(tsk); cpu_timer_sample_group(clock_idx, tsk, &now); if (oldval) { -- cgit v1.2.3-70-g09d2 From 4da94d49b2ecb0a26e716a8811c3ecc542c2a65d Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 11 Feb 2009 11:30:27 +0100 Subject: timers: fix TIMER_ABSTIME for process wide cpu timers The POSIX timer interface allows for absolute time expiry values through the TIMER_ABSTIME flag, therefore we have to synchronize the timer to the clock every time we start it. Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- include/linux/sched.h | 13 +------------ kernel/posix-cpu-timers.c | 34 ++++++++++++++++++++++++++++++++++ 2 files changed, 35 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 5d10fa0b6002..8981e52c714f 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2201,18 +2201,7 @@ static inline int spin_needbreak(spinlock_t *lock) * Thread group CPU time accounting. */ void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times); - -static inline -void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) -{ - struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; - unsigned long flags; - - spin_lock_irqsave(&cputimer->lock, flags); - cputimer->running = 1; - *times = cputimer->cputime; - spin_unlock_irqrestore(&cputimer->lock, flags); -} +void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times); static inline void thread_group_cputime_init(struct signal_struct *sig) { diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index e5d7bfdfa7d4..2313a4cc14ea 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -261,6 +261,40 @@ out: rcu_read_unlock(); } +static void update_gt_cputime(struct task_cputime *a, struct task_cputime *b) +{ + if (cputime_gt(b->utime, a->utime)) + a->utime = b->utime; + + if (cputime_gt(b->stime, a->stime)) + a->stime = b->stime; + + if (b->sum_exec_runtime > a->sum_exec_runtime) + a->sum_exec_runtime = b->sum_exec_runtime; +} + +void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) +{ + struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; + struct task_cputime sum; + unsigned long flags; + + spin_lock_irqsave(&cputimer->lock, flags); + if (!cputimer->running) { + cputimer->running = 1; + /* + * The POSIX timer interface allows for absolute time expiry + * values through the TIMER_ABSTIME flag, therefore we have + * to synchronize the timer to the clock every time we start + * it. + */ + thread_group_cputime(tsk, &sum); + update_gt_cputime(&cputimer->cputime, &sum); + } + *times = cputimer->cputime; + spin_unlock_irqrestore(&cputimer->lock, flags); +} + /* * Sample a process (thread group) clock for the given group_leader task. * Must be called with tasklist_lock held for reading. -- cgit v1.2.3-70-g09d2 From 9f339e7028e2855717af3193c938f9960ad13b38 Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Wed, 11 Feb 2009 15:10:27 +0100 Subject: x86, ptrace, mm: fix double-free on race Ptrace_detach() races with __ptrace_unlink() if the traced task is reaped while detaching. This might cause a double-free of the BTS buffer. Change the ptrace_detach() path to only do the memory accounting in ptrace_bts_detach() and leave the buffer free to ptrace_bts_untrace() which will be called from __ptrace_unlink(). The fix follows a proposal from Oleg Nesterov. Reported-by: Oleg Nesterov Signed-off-by: Markus Metzger Signed-off-by: Ingo Molnar --- arch/x86/kernel/ptrace.c | 16 ++++++++++------ include/linux/mm.h | 1 + mm/mlock.c | 7 ++++++- 3 files changed, 17 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 0a5df5f82fb9..5a4c23d89892 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -810,12 +810,16 @@ static void ptrace_bts_untrace(struct task_struct *child) static void ptrace_bts_detach(struct task_struct *child) { - if (unlikely(child->bts)) { - ds_release_bts(child->bts); - child->bts = NULL; - - ptrace_bts_free_buffer(child); - } + /* + * Ptrace_detach() races with ptrace_untrace() in case + * the child dies and is reaped by another thread. + * + * We only do the memory accounting at this point and + * leave the buffer deallocation and the bts tracer + * release to ptrace_bts_untrace() which will be called + * later on with tasklist_lock held. + */ + release_locked_buffer(child->bts_buffer, child->bts_size); } #else static inline void ptrace_bts_fork(struct task_struct *tsk) {} diff --git a/include/linux/mm.h b/include/linux/mm.h index e8ddc98b8405..3d7fb44d7d7e 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1305,5 +1305,6 @@ void vmemmap_populate_print_last(void); extern void *alloc_locked_buffer(size_t size); extern void free_locked_buffer(void *buffer, size_t size); +extern void release_locked_buffer(void *buffer, size_t size); #endif /* __KERNEL__ */ #endif /* _LINUX_MM_H */ diff --git a/mm/mlock.c b/mm/mlock.c index 028ec482fdd4..2b57f7e60390 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -657,7 +657,7 @@ void *alloc_locked_buffer(size_t size) return buffer; } -void free_locked_buffer(void *buffer, size_t size) +void release_locked_buffer(void *buffer, size_t size) { unsigned long pgsz = PAGE_ALIGN(size) >> PAGE_SHIFT; @@ -667,6 +667,11 @@ void free_locked_buffer(void *buffer, size_t size) current->mm->locked_vm -= pgsz; up_write(¤t->mm->mmap_sem); +} + +void free_locked_buffer(void *buffer, size_t size) +{ + release_locked_buffer(buffer, size); kfree(buffer); } -- cgit v1.2.3-70-g09d2 From cfebe563bd0a3ff97e1bc167123120d59c7a84db Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 11 Feb 2009 13:04:36 -0800 Subject: cgroups: fix lockdep subclasses overflow I enabled all cgroup subsystems when compiling kernel, and then: # mount -t cgroup -o net_cls xxx /mnt # mkdir /mnt/0 This showed up immediately: BUG: MAX_LOCKDEP_SUBCLASSES too low! turning off the locking correctness validator. It's caused by the cgroup hierarchy lock: for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; if (ss->root == root) mutex_lock_nested(&ss->hierarchy_mutex, i); } Now we have 9 cgroup subsystems, and the above 'i' for net_cls is 8, but MAX_LOCKDEP_SUBCLASSES is 8. This patch uses different lockdep keys for different subsystems. Signed-off-by: Li Zefan Acked-by: Paul Menage Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cgroup.h | 1 + kernel/cgroup.c | 3 ++- 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index e4e8e117d27d..499900d0cee7 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -378,6 +378,7 @@ struct cgroup_subsys { * - initiating hotplug events */ struct mutex hierarchy_mutex; + struct lock_class_key subsys_key; /* * Link to parent, and list entry in parent's children. diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 5a54ff42874e..e14db9c089b9 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -2351,7 +2351,7 @@ static void cgroup_lock_hierarchy(struct cgroupfs_root *root) for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; if (ss->root == root) - mutex_lock_nested(&ss->hierarchy_mutex, i); + mutex_lock(&ss->hierarchy_mutex); } } @@ -2637,6 +2637,7 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss) BUG_ON(!list_empty(&init_task.tasks)); mutex_init(&ss->hierarchy_mutex); + lockdep_set_class(&ss->hierarchy_mutex, &ss->subsys_key); ss->active = 1; } -- cgit v1.2.3-70-g09d2 From 6c5979631b4b03c9288776562c18036765e398c1 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 11 Feb 2009 13:04:38 -0800 Subject: syscall define: fix uml compile bug With the new system call defines we get this on uml: arch/um/sys-i386/built-in.o: In function `sys_call_table': (.rodata+0x308): undefined reference to `sys_sigprocmask' Reason for this is that uml passes the preprocessor option -Dsigprocmask=kernel_sigprocmask to gcc when compiling the kernel. This causes SYSCALL_DEFINE3(sigprocmask, ...) to be expanded to SYSCALL_DEFINEx(3, kernel_sigprocmask, ...) and finally to a system call named sys_kernel_sigprocmask. However sys_sigprocmask is missing because of this. To avoid macro expansion for the system call name just concatenate the name at first define instead of carrying it through severel levels. This was pointed out by Al Viro. Signed-off-by: Heiko Carstens Cc: Geert Uytterhoeven Cc: Al Viro Reviewed-by: WANG Cong Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/syscalls.h | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 0eda02ff2414..f9f900cfd066 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -95,13 +95,13 @@ struct old_linux_dirent; #define __SC_TEST5(t5, a5, ...) __SC_TEST(t5); __SC_TEST4(__VA_ARGS__) #define __SC_TEST6(t6, a6, ...) __SC_TEST(t6); __SC_TEST5(__VA_ARGS__) -#define SYSCALL_DEFINE0(name) asmlinkage long sys_##name(void) -#define SYSCALL_DEFINE1(...) SYSCALL_DEFINEx(1, __VA_ARGS__) -#define SYSCALL_DEFINE2(...) SYSCALL_DEFINEx(2, __VA_ARGS__) -#define SYSCALL_DEFINE3(...) SYSCALL_DEFINEx(3, __VA_ARGS__) -#define SYSCALL_DEFINE4(...) SYSCALL_DEFINEx(4, __VA_ARGS__) -#define SYSCALL_DEFINE5(...) SYSCALL_DEFINEx(5, __VA_ARGS__) -#define SYSCALL_DEFINE6(...) SYSCALL_DEFINEx(6, __VA_ARGS__) +#define SYSCALL_DEFINE0(name) asmlinkage long sys_##name(void) +#define SYSCALL_DEFINE1(name, ...) SYSCALL_DEFINEx(1, _##name, __VA_ARGS__) +#define SYSCALL_DEFINE2(name, ...) SYSCALL_DEFINEx(2, _##name, __VA_ARGS__) +#define SYSCALL_DEFINE3(name, ...) SYSCALL_DEFINEx(3, _##name, __VA_ARGS__) +#define SYSCALL_DEFINE4(name, ...) SYSCALL_DEFINEx(4, _##name, __VA_ARGS__) +#define SYSCALL_DEFINE5(name, ...) SYSCALL_DEFINEx(5, _##name, __VA_ARGS__) +#define SYSCALL_DEFINE6(name, ...) SYSCALL_DEFINEx(6, _##name, __VA_ARGS__) #ifdef CONFIG_PPC64 #define SYSCALL_ALIAS(alias, name) \ @@ -121,21 +121,21 @@ struct old_linux_dirent; #define SYSCALL_DEFINE(name) static inline long SYSC_##name #define SYSCALL_DEFINEx(x, name, ...) \ - asmlinkage long sys_##name(__SC_DECL##x(__VA_ARGS__)); \ - static inline long SYSC_##name(__SC_DECL##x(__VA_ARGS__)); \ - asmlinkage long SyS_##name(__SC_LONG##x(__VA_ARGS__)) \ + asmlinkage long sys##name(__SC_DECL##x(__VA_ARGS__)); \ + static inline long SYSC##name(__SC_DECL##x(__VA_ARGS__)); \ + asmlinkage long SyS##name(__SC_LONG##x(__VA_ARGS__)) \ { \ __SC_TEST##x(__VA_ARGS__); \ - return (long) SYSC_##name(__SC_CAST##x(__VA_ARGS__)); \ + return (long) SYSC##name(__SC_CAST##x(__VA_ARGS__)); \ } \ - SYSCALL_ALIAS(sys_##name, SyS_##name); \ - static inline long SYSC_##name(__SC_DECL##x(__VA_ARGS__)) + SYSCALL_ALIAS(sys##name, SyS##name); \ + static inline long SYSC##name(__SC_DECL##x(__VA_ARGS__)) #else /* CONFIG_HAVE_SYSCALL_WRAPPERS */ #define SYSCALL_DEFINE(name) asmlinkage long sys_##name #define SYSCALL_DEFINEx(x, name, ...) \ - asmlinkage long sys_##name(__SC_DECL##x(__VA_ARGS__)) + asmlinkage long sys##name(__SC_DECL##x(__VA_ARGS__)) #endif /* CONFIG_HAVE_SYSCALL_WRAPPERS */ -- cgit v1.2.3-70-g09d2 From 1d93e52eb48df986a3c4d5ad8a520bf1f6837367 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 11 Feb 2009 08:47:19 -0700 Subject: dmaengine: update kerneldoc Some of the kerneldoc comments in the dmaengine header describe already removed structure members. Remove them. Also add a short description for dma_device->device_is_tx_complete. Signed-off-by: Johannes Weiner Signed-off-by: Dan Williams --- include/linux/dmaengine.h | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index 3e68469c1885..087e79acf8c7 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -97,7 +97,6 @@ typedef struct { DECLARE_BITMAP(bits, DMA_TX_TYPE_END); } dma_cap_mask_t; /** * struct dma_chan_percpu - the per-CPU part of struct dma_chan - * @refcount: local_t used for open-coded "bigref" counting * @memcpy_count: transaction counter * @bytes_transferred: byte counter */ @@ -114,9 +113,6 @@ struct dma_chan_percpu { * @cookie: last cookie value returned to client * @chan_id: channel ID for sysfs * @dev: class device for sysfs - * @refcount: kref, used in "bigref" slow-mode - * @slow_ref: indicates that the DMA channel is free - * @rcu: the DMA channel's RCU head * @device_node: used to add this to the device chan list * @local: per-cpu pointer to a struct dma_chan_percpu * @client-count: how many clients are using this channel @@ -211,8 +207,6 @@ struct dma_async_tx_descriptor { * @global_node: list_head for global dma_device_list * @cap_mask: one or more dma_capability flags * @max_xor: maximum number of xor sources, 0 if no capability - * @refcount: reference count - * @done: IO completion struct * @dev_id: unique device ID * @dev: struct device reference for dma mapping api * @device_alloc_chan_resources: allocate resources and return the @@ -225,6 +219,7 @@ struct dma_async_tx_descriptor { * @device_prep_dma_interrupt: prepares an end of chain interrupt operation * @device_prep_slave_sg: prepares a slave dma operation * @device_terminate_all: terminate all pending operations + * @device_is_tx_complete: poll for transaction completion * @device_issue_pending: push pending transactions to hardware */ struct dma_device { -- cgit v1.2.3-70-g09d2 From 99709372736a216f99eb32b76fba835a2bfc93a8 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 12 Feb 2009 16:43:17 -0800 Subject: net: don't use in_atomic() in gfp_any() The problem is that in_atomic() will return false inside spinlocks if CONFIG_PREEMPT=n. This will lead to deadlockable GFP_KERNEL allocations from spinlocked regions. Secondly, if CONFIG_PREEMPT=y, this bug solves itself because networking will instead use GFP_ATOMIC from this callsite. Hence we won't get the might_sleep() debugging warnings which would have informed us of the buggy callsites. Solve both these problems by switching to in_interrupt(). Now, if someone runs a gfp_any() allocation from inside spinlock we will get the warning if CONFIG_PREEMPT=y. I reviewed all callsites and most of them were too complex for my little brain and none of them documented their interface requirements. I have no idea what this patch will do. Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- include/net/sock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 5a3a151bd730..ce3b5b622683 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1308,7 +1308,7 @@ static inline int sock_writeable(const struct sock *sk) static inline gfp_t gfp_any(void) { - return in_atomic() ? GFP_ATOMIC : GFP_KERNEL; + return in_softirq() ? GFP_ATOMIC : GFP_KERNEL; } static inline long sock_rcvtimeo(const struct sock *sk, int noblock) -- cgit v1.2.3-70-g09d2 From 7a0eb1960e8ddcb68ea631caf16815485af0e228 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 19 Jan 2009 14:57:52 +0200 Subject: KVM: Avoid using CONFIG_ in userspace visible headers Kconfig symbols are not available in userspace, and are not stripped by headers-install. Avoid their use by adding #defines in to suit each architecture. Signed-off-by: Avi Kivity --- arch/ia64/include/asm/kvm.h | 4 ++++ arch/x86/include/asm/kvm.h | 7 +++++++ include/linux/kvm.h | 10 +++++----- 3 files changed, 16 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/arch/ia64/include/asm/kvm.h b/arch/ia64/include/asm/kvm.h index 68aa6da807c1..bfa86b6af7cd 100644 --- a/arch/ia64/include/asm/kvm.h +++ b/arch/ia64/include/asm/kvm.h @@ -25,6 +25,10 @@ #include +/* Select x86 specific features in */ +#define __KVM_HAVE_IOAPIC +#define __KVM_HAVE_DEVICE_ASSIGNMENT + /* Architectural interrupt line count. */ #define KVM_NR_INTERRUPTS 256 diff --git a/arch/x86/include/asm/kvm.h b/arch/x86/include/asm/kvm.h index d2e3bf3608af..886c9402ec45 100644 --- a/arch/x86/include/asm/kvm.h +++ b/arch/x86/include/asm/kvm.h @@ -9,6 +9,13 @@ #include #include +/* Select x86 specific features in */ +#define __KVM_HAVE_PIT +#define __KVM_HAVE_IOAPIC +#define __KVM_HAVE_DEVICE_ASSIGNMENT +#define __KVM_HAVE_MSI +#define __KVM_HAVE_USER_NMI + /* Architectural interrupt line count. */ #define KVM_NR_INTERRUPTS 256 diff --git a/include/linux/kvm.h b/include/linux/kvm.h index 5715f1907601..0424326f1679 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -58,10 +58,10 @@ struct kvm_irqchip { __u32 pad; union { char dummy[512]; /* reserving space */ -#ifdef CONFIG_X86 +#ifdef __KVM_HAVE_PIT struct kvm_pic_state pic; #endif -#if defined(CONFIG_X86) || defined(CONFIG_IA64) +#ifdef __KVM_HAVE_IOAPIC struct kvm_ioapic_state ioapic; #endif } chip; @@ -384,16 +384,16 @@ struct kvm_trace_rec { #define KVM_CAP_MP_STATE 14 #define KVM_CAP_COALESCED_MMIO 15 #define KVM_CAP_SYNC_MMU 16 /* Changes to host mmap are reflected in guest */ -#if defined(CONFIG_X86)||defined(CONFIG_IA64) +#ifdef __KVM_HAVE_DEVICE_ASSIGNMENT #define KVM_CAP_DEVICE_ASSIGNMENT 17 #endif #define KVM_CAP_IOMMU 18 -#if defined(CONFIG_X86) +#ifdef __KVM_HAVE_MSI #define KVM_CAP_DEVICE_MSI 20 #endif /* Bug in KVM_SET_USER_MEMORY_REGION fixed: */ #define KVM_CAP_DESTROY_MEMORY_REGION_WORKS 21 -#if defined(CONFIG_X86) +#ifdef __KVM_HAVE_USER_NMI #define KVM_CAP_USER_NMI 22 #endif -- cgit v1.2.3-70-g09d2 From ad8ba2cd44d4d39fb3fe55d5dcc565b19fc3a7fb Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Tue, 6 Jan 2009 10:03:02 +0800 Subject: KVM: Add kvm_arch_sync_events to sync with asynchronize events kvm_arch_sync_events is introduced to quiet down all other events may happen contemporary with VM destroy process, like IRQ handler and work struct for assigned device. For kvm_arch_sync_events is called at the very beginning of kvm_destroy_vm(), so the state of KVM here is legal and can provide a environment to quiet down other events. Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/ia64/kvm/kvm-ia64.c | 4 ++++ arch/powerpc/kvm/powerpc.c | 4 ++++ arch/s390/kvm/kvm-s390.c | 4 ++++ arch/x86/kvm/x86.c | 4 ++++ include/linux/kvm_host.h | 1 + virt/kvm/kvm_main.c | 1 + 6 files changed, 18 insertions(+) (limited to 'include') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 4e586f6110aa..28f982045f29 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1337,6 +1337,10 @@ static void kvm_release_vm_pages(struct kvm *kvm) } } +void kvm_arch_sync_events(struct kvm *kvm) +{ +} + void kvm_arch_destroy_vm(struct kvm *kvm) { kvm_iommu_unmap_guest(kvm); diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index 2822c8ccfaaf..5f81256287f5 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -125,6 +125,10 @@ static void kvmppc_free_vcpus(struct kvm *kvm) } } +void kvm_arch_sync_events(struct kvm *kvm) +{ +} + void kvm_arch_destroy_vm(struct kvm *kvm) { kvmppc_free_vcpus(kvm); diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index be8497186b96..0d33893e1e89 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -212,6 +212,10 @@ static void kvm_free_vcpus(struct kvm *kvm) } } +void kvm_arch_sync_events(struct kvm *kvm) +{ +} + void kvm_arch_destroy_vm(struct kvm *kvm) { kvm_free_vcpus(kvm); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index cc17546a2406..b0fc079f1bee 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -4127,6 +4127,10 @@ static void kvm_free_vcpus(struct kvm *kvm) } +void kvm_arch_sync_events(struct kvm *kvm) +{ +} + void kvm_arch_destroy_vm(struct kvm *kvm) { kvm_free_all_assigned_devices(kvm); diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index ec49d0be7f52..bf6f703642fc 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -285,6 +285,7 @@ void kvm_free_physmem(struct kvm *kvm); struct kvm *kvm_arch_create_vm(void); void kvm_arch_destroy_vm(struct kvm *kvm); void kvm_free_all_assigned_devices(struct kvm *kvm); +void kvm_arch_sync_events(struct kvm *kvm); int kvm_cpu_get_interrupt(struct kvm_vcpu *v); int kvm_cpu_has_interrupt(struct kvm_vcpu *v); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 0b6f2f71271f..68e3f1ec1674 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -891,6 +891,7 @@ static void kvm_destroy_vm(struct kvm *kvm) { struct mm_struct *mm = kvm->mm; + kvm_arch_sync_events(kvm); spin_lock(&kvm_lock); list_del(&kvm->vm_list); spin_unlock(&kvm_lock); -- cgit v1.2.3-70-g09d2 From 2f6097129af5625db2fb5e601f89e5bf55cd1dcd Mon Sep 17 00:00:00 2001 From: Roel Kluin Date: Mon, 16 Feb 2009 12:49:16 +0000 Subject: FRV: __pte_to_swp_entry doesn't expand correctly The macro doesn't expand correctly when its parameter isn't 'pte'. Signed-off-by: Roel Kluin Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- include/asm-frv/pgtable.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-frv/pgtable.h b/include/asm-frv/pgtable.h index 83c51aba534b..e16fdb1f4f4f 100644 --- a/include/asm-frv/pgtable.h +++ b/include/asm-frv/pgtable.h @@ -478,7 +478,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) #define __swp_type(x) (((x).val >> 2) & 0x1f) #define __swp_offset(x) ((x).val >> 8) #define __swp_entry(type, offset) ((swp_entry_t) { ((type) << 2) | ((offset) << 8) }) -#define __pte_to_swp_entry(pte) ((swp_entry_t) { (pte).pte }) +#define __pte_to_swp_entry(_pte) ((swp_entry_t) { (_pte).pte }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) static inline int pte_file(pte_t pte) -- cgit v1.2.3-70-g09d2 From 5955c7a2cfb6a35429adea5dc480002b15ca8cfc Mon Sep 17 00:00:00 2001 From: Zlatko Calusic Date: Wed, 18 Feb 2009 01:33:34 +0100 Subject: Add support for VT6415 PCIE PATA IDE Host Controller Signed-off-by: Zlatko Calusic Signed-off-by: Linus Torvalds --- drivers/ata/pata_via.c | 4 +++- include/linux/pci_ids.h | 1 + 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c index 79a6c9a0b721..ba556d3e6963 100644 --- a/drivers/ata/pata_via.c +++ b/drivers/ata/pata_via.c @@ -110,7 +110,8 @@ static const struct via_isa_bridge { { "vt8237s", PCI_DEVICE_ID_VIA_8237S, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8251", PCI_DEVICE_ID_VIA_8251, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "cx700", PCI_DEVICE_ID_VIA_CX700, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST | VIA_SATA_PATA }, - { "vt6410", PCI_DEVICE_ID_VIA_6410, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST | VIA_NO_ENABLES}, + { "vt6410", PCI_DEVICE_ID_VIA_6410, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST | VIA_NO_ENABLES }, + { "vt6415", PCI_DEVICE_ID_VIA_6415, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST | VIA_NO_ENABLES }, { "vt8237a", PCI_DEVICE_ID_VIA_8237A, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8237", PCI_DEVICE_ID_VIA_8237, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8235", PCI_DEVICE_ID_VIA_8235, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, @@ -593,6 +594,7 @@ static int via_reinit_one(struct pci_dev *pdev) #endif static const struct pci_device_id via[] = { + { PCI_VDEVICE(VIA, 0x0415), }, { PCI_VDEVICE(VIA, 0x0571), }, { PCI_VDEVICE(VIA, 0x0581), }, { PCI_VDEVICE(VIA, 0x1571), }, diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 52a9fe08451c..918391b4b109 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1312,6 +1312,7 @@ #define PCI_DEVICE_ID_VIA_VT3351 0x0351 #define PCI_DEVICE_ID_VIA_VT3364 0x0364 #define PCI_DEVICE_ID_VIA_8371_0 0x0391 +#define PCI_DEVICE_ID_VIA_6415 0x0415 #define PCI_DEVICE_ID_VIA_8501_0 0x0501 #define PCI_DEVICE_ID_VIA_82C561 0x0561 #define PCI_DEVICE_ID_VIA_82C586_1 0x0571 -- cgit v1.2.3-70-g09d2 From 92a0acce186cde8ead56c6915d9479773673ea1a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 17 Feb 2009 21:24:05 -0800 Subject: net: Kill skb_truesize_check(), it only catches false-positives. A long time ago we had bugs, primarily in TCP, where we would modify skb->truesize (for TSO queue collapsing) in ways which would corrupt the socket memory accounting. skb_truesize_check() was added in order to try and catch this error more systematically. However this debugging check has morphed into a Frankenstein of sorts and these days it does nothing other than catch false-positives. Signed-off-by: David S. Miller --- include/linux/skbuff.h | 9 --------- include/net/sock.h | 1 - net/core/skbuff.c | 8 -------- net/core/sock.c | 1 - 4 files changed, 19 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index cf2cb50f77d1..9dcf956ad18a 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -416,15 +416,6 @@ extern void skb_over_panic(struct sk_buff *skb, int len, void *here); extern void skb_under_panic(struct sk_buff *skb, int len, void *here); -extern void skb_truesize_bug(struct sk_buff *skb); - -static inline void skb_truesize_check(struct sk_buff *skb) -{ - int len = sizeof(struct sk_buff) + skb->len; - - if (unlikely((int)skb->truesize < len)) - skb_truesize_bug(skb); -} extern int skb_append_datato_frags(struct sock *sk, struct sk_buff *skb, int getfrag(void *from, char *to, int offset, diff --git a/include/net/sock.h b/include/net/sock.h index ce3b5b622683..eefeeaf7fc46 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -860,7 +860,6 @@ static inline void sk_mem_uncharge(struct sock *sk, int size) static inline void sk_wmem_free_skb(struct sock *sk, struct sk_buff *skb) { - skb_truesize_check(skb); sock_set_flag(sk, SOCK_QUEUE_SHRUNK); sk->sk_wmem_queued -= skb->truesize; sk_mem_uncharge(sk, skb->truesize); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index da74b844f4ea..c6a6b166f8d6 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -143,14 +143,6 @@ void skb_under_panic(struct sk_buff *skb, int sz, void *here) BUG(); } -void skb_truesize_bug(struct sk_buff *skb) -{ - WARN(net_ratelimit(), KERN_ERR "SKB BUG: Invalid truesize (%u) " - "len=%u, sizeof(sk_buff)=%Zd\n", - skb->truesize, skb->len, sizeof(struct sk_buff)); -} -EXPORT_SYMBOL(skb_truesize_bug); - /* Allocate a new skbuff. We do this ourselves so we can fill in a few * 'private' fields and also do memory statistics to find all the * [BEEP] leaks. diff --git a/net/core/sock.c b/net/core/sock.c index 6f2e1337975d..6e4f14d1ef81 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1137,7 +1137,6 @@ void sock_rfree(struct sk_buff *skb) { struct sock *sk = skb->sk; - skb_truesize_check(skb); atomic_sub(skb->truesize, &sk->sk_rmem_alloc); sk_mem_uncharge(skb->sk, skb->truesize); } -- cgit v1.2.3-70-g09d2 From 93dbb393503d53cd226e5e1f0088fe8f4dbaa2b8 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 16 Feb 2009 10:25:40 +0100 Subject: block: fix bad definition of BIO_RW_SYNC We can't OR shift values, so get rid of BIO_RW_SYNC and use BIO_RW_SYNCIO and BIO_RW_UNPLUG explicitly. This brings back the behaviour from before 213d9417fec62ef4c3675621b9364a667954d4dd. Signed-off-by: Jens Axboe --- block/blktrace.c | 2 +- drivers/md/dm-io.c | 2 +- drivers/md/dm-kcopyd.c | 2 +- drivers/md/md.c | 4 ++-- include/linux/bio.h | 2 -- include/linux/blktrace_api.h | 1 + include/linux/fs.h | 6 +++--- kernel/power/swap.c | 5 +++-- mm/page_io.c | 2 +- 9 files changed, 13 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/block/blktrace.c b/block/blktrace.c index 39cc3bfe56e4..7cf9d1ff45a0 100644 --- a/block/blktrace.c +++ b/block/blktrace.c @@ -142,7 +142,7 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, what |= ddir_act[rw & WRITE]; what |= MASK_TC_BIT(rw, BARRIER); - what |= MASK_TC_BIT(rw, SYNC); + what |= MASK_TC_BIT(rw, SYNCIO); what |= MASK_TC_BIT(rw, AHEAD); what |= MASK_TC_BIT(rw, META); what |= MASK_TC_BIT(rw, DISCARD); diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c index a34338567a2a..f14813be4eff 100644 --- a/drivers/md/dm-io.c +++ b/drivers/md/dm-io.c @@ -328,7 +328,7 @@ static void dispatch_io(int rw, unsigned int num_regions, struct dpages old_pages = *dp; if (sync) - rw |= (1 << BIO_RW_SYNC); + rw |= (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG); /* * For multiple regions we need to be careful to rewind diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c index 3073618269ea..0a225da21272 100644 --- a/drivers/md/dm-kcopyd.c +++ b/drivers/md/dm-kcopyd.c @@ -344,7 +344,7 @@ static int run_io_job(struct kcopyd_job *job) { int r; struct dm_io_request io_req = { - .bi_rw = job->rw | (1 << BIO_RW_SYNC), + .bi_rw = job->rw | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG), .mem.type = DM_IO_PAGE_LIST, .mem.ptr.pl = job->pages, .mem.offset = job->offset, diff --git a/drivers/md/md.c b/drivers/md/md.c index 4495104f6c9f..03b4cd0a6344 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -474,7 +474,7 @@ void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev, * causes ENOTSUPP, we allocate a spare bio... */ struct bio *bio = bio_alloc(GFP_NOIO, 1); - int rw = (1<bi_bdev = rdev->bdev; bio->bi_sector = sector; @@ -531,7 +531,7 @@ int sync_page_io(struct block_device *bdev, sector_t sector, int size, struct completion event; int ret; - rw |= (1 << BIO_RW_SYNC); + rw |= (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG); bio->bi_bdev = bdev; bio->bi_sector = sector; diff --git a/include/linux/bio.h b/include/linux/bio.h index 2aa283ab062b..1b16108a5417 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -171,8 +171,6 @@ struct bio { #define BIO_RW_FAILFAST_TRANSPORT 8 #define BIO_RW_FAILFAST_DRIVER 9 -#define BIO_RW_SYNC (BIO_RW_SYNCIO | BIO_RW_UNPLUG) - #define bio_rw_flagged(bio, flag) ((bio)->bi_rw & (1 << (flag))) /* diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h index 25379cba2370..6e915878e88c 100644 --- a/include/linux/blktrace_api.h +++ b/include/linux/blktrace_api.h @@ -15,6 +15,7 @@ enum blktrace_cat { BLK_TC_WRITE = 1 << 1, /* writes */ BLK_TC_BARRIER = 1 << 2, /* barrier */ BLK_TC_SYNC = 1 << 3, /* sync IO */ + BLK_TC_SYNCIO = BLK_TC_SYNC, BLK_TC_QUEUE = 1 << 4, /* queueing/merging */ BLK_TC_REQUEUE = 1 << 5, /* requeueing */ BLK_TC_ISSUE = 1 << 6, /* issue */ diff --git a/include/linux/fs.h b/include/linux/fs.h index 6022f44043f2..67857dc16365 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -87,10 +87,10 @@ struct inodes_stat_t { #define WRITE 1 #define READA 2 /* read-ahead - don't block if no resources */ #define SWRITE 3 /* for ll_rw_block() - wait for buffer lock */ -#define READ_SYNC (READ | (1 << BIO_RW_SYNC)) +#define READ_SYNC (READ | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG)) #define READ_META (READ | (1 << BIO_RW_META)) -#define WRITE_SYNC (WRITE | (1 << BIO_RW_SYNC)) -#define SWRITE_SYNC (SWRITE | (1 << BIO_RW_SYNC)) +#define WRITE_SYNC (WRITE | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG)) +#define SWRITE_SYNC (SWRITE | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG)) #define WRITE_BARRIER (WRITE | (1 << BIO_RW_BARRIER)) #define DISCARD_NOBARRIER (1 << BIO_RW_DISCARD) #define DISCARD_BARRIER ((1 << BIO_RW_DISCARD) | (1 << BIO_RW_BARRIER)) diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 6da14358537c..505f319e489c 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -60,6 +60,7 @@ static struct block_device *resume_bdev; static int submit(int rw, pgoff_t page_off, struct page *page, struct bio **bio_chain) { + const int bio_rw = rw | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG); struct bio *bio; bio = bio_alloc(__GFP_WAIT | __GFP_HIGH, 1); @@ -80,7 +81,7 @@ static int submit(int rw, pgoff_t page_off, struct page *page, bio_get(bio); if (bio_chain == NULL) { - submit_bio(rw | (1 << BIO_RW_SYNC), bio); + submit_bio(bio_rw, bio); wait_on_page_locked(page); if (rw == READ) bio_set_pages_dirty(bio); @@ -90,7 +91,7 @@ static int submit(int rw, pgoff_t page_off, struct page *page, get_page(page); /* These pages are freed later */ bio->bi_private = *bio_chain; *bio_chain = bio; - submit_bio(rw | (1 << BIO_RW_SYNC), bio); + submit_bio(bio_rw, bio); } return 0; } diff --git a/mm/page_io.c b/mm/page_io.c index dc6ce0afbded..3023c475e041 100644 --- a/mm/page_io.c +++ b/mm/page_io.c @@ -111,7 +111,7 @@ int swap_writepage(struct page *page, struct writeback_control *wbc) goto out; } if (wbc->sync_mode == WB_SYNC_ALL) - rw |= (1 << BIO_RW_SYNC); + rw |= (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG); count_vm_event(PSWPOUT); set_page_writeback(page); unlock_page(page); -- cgit v1.2.3-70-g09d2 From 5ca431f9ae8db8c6edb9c64bebe6d6521077afd6 Mon Sep 17 00:00:00 2001 From: Eric Leblond Date: Wed, 18 Feb 2009 15:29:23 +0100 Subject: netfilter: nfnetlink_log: fix per-rule qthreshold override In NFLOG the per-rule qthreshold should overrides per-instance only it is set. With current code, the per-rule qthreshold is 1 if not set and it overrides the per-instance qthreshold. This patch modifies the default xt_NFLOG threshold from 1 to 0. Thus a value of 0 means there is no per-rule setting and the instance parameter has to apply. Signed-off-by: Eric Leblond Signed-off-by: Patrick McHardy --- include/linux/netfilter/xt_NFLOG.h | 2 +- net/netfilter/nfnetlink_log.c | 6 ++++-- 2 files changed, 5 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/xt_NFLOG.h b/include/linux/netfilter/xt_NFLOG.h index cdcd0ed58f7a..4b36aeb46a10 100644 --- a/include/linux/netfilter/xt_NFLOG.h +++ b/include/linux/netfilter/xt_NFLOG.h @@ -2,7 +2,7 @@ #define _XT_NFLOG_TARGET #define XT_NFLOG_DEFAULT_GROUP 0x1 -#define XT_NFLOG_DEFAULT_THRESHOLD 1 +#define XT_NFLOG_DEFAULT_THRESHOLD 0 #define XT_NFLOG_MASK 0x0 diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index fa49dc7fe100..580b837e44dd 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -590,8 +590,10 @@ nfulnl_log_packet(u_int8_t pf, qthreshold = inst->qthreshold; /* per-rule qthreshold overrides per-instance */ - if (qthreshold > li->u.ulog.qthreshold) - qthreshold = li->u.ulog.qthreshold; + if (li->u.ulog.qthreshold) + if (qthreshold > li->u.ulog.qthreshold) + qthreshold = li->u.ulog.qthreshold; + switch (inst->copy_mode) { case NFULNL_COPY_META: -- cgit v1.2.3-70-g09d2 From 5962fc6d5fff09c8e6fb8cadcb18327a0f4277f7 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 18 Feb 2009 15:30:34 +0100 Subject: netfilter: nf_conntrack: don't try to deliver events for untracked connections The untracked conntrack actually does usually have events marked for delivery as its not special-cased in that part of the code. Skip the actual delivery since it impacts performance noticeably. Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_core.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index e78afe7f28e3..c25068e38516 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -59,7 +59,7 @@ static inline int nf_conntrack_confirm(struct sk_buff *skb) struct nf_conn *ct = (struct nf_conn *)skb->nfct; int ret = NF_ACCEPT; - if (ct) { + if (ct && ct != &nf_conntrack_untracked) { if (!nf_ct_is_confirmed(ct) && !nf_ct_is_dying(ct)) ret = __nf_conntrack_confirm(skb); nf_ct_deliver_cached_events(ct); -- cgit v1.2.3-70-g09d2 From c296861291669f305deef19b78042330d7135017 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 18 Feb 2009 14:48:12 -0800 Subject: vmalloc: add __get_vm_area_caller() We have get_vm_area_caller() and __get_vm_area() but not __get_vm_area_caller() On powerpc, I use __get_vm_area() to separate the ranges of addresses given to vmalloc vs. ioremap (various good reasons for that) so in order to be able to implement the new caller tracking in /proc/vmallocinfo, I need a "_caller" variant of it. (akpm: needed for ongoing powerpc development, so merge it early) [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Benjamin Herrenschmidt Reviewed-by: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/vmalloc.h | 4 ++++ mm/vmalloc.c | 8 ++++++++ 2 files changed, 12 insertions(+) (limited to 'include') diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h index 506e7620a986..9c0890c7a06a 100644 --- a/include/linux/vmalloc.h +++ b/include/linux/vmalloc.h @@ -84,6 +84,10 @@ extern struct vm_struct *get_vm_area_caller(unsigned long size, unsigned long flags, void *caller); extern struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags, unsigned long start, unsigned long end); +extern struct vm_struct *__get_vm_area_caller(unsigned long size, + unsigned long flags, + unsigned long start, unsigned long end, + void *caller); extern struct vm_struct *get_vm_area_node(unsigned long size, unsigned long flags, int node, gfp_t gfp_mask); diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 75f49d312e8c..4dd2636d0b92 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -1106,6 +1106,14 @@ struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags, } EXPORT_SYMBOL_GPL(__get_vm_area); +struct vm_struct *__get_vm_area_caller(unsigned long size, unsigned long flags, + unsigned long start, unsigned long end, + void *caller) +{ + return __get_vm_area_node(size, flags, start, end, -1, GFP_KERNEL, + caller); +} + /** * get_vm_area - reserve a contiguous kernel virtual area * @size: size of the area -- cgit v1.2.3-70-g09d2 From 55ec82176eca52e4e0530a82a0eb59160a1a95a1 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Wed, 18 Feb 2009 14:48:15 -0800 Subject: vfs: separate FMODE_PREAD/FMODE_PWRITE into separate flags Separate FMODE_PREAD and FMODE_PWRITE into separate flags to reflect the reality that the read and write paths may have independent restrictions. A git grep verifies that these flags are always cleared together so this new behavior will only apply to interfaces that change to clear flags individually. This is required for "seq_file: properly cope with pread", a post-2.6.25 regression fix. [akpm@linux-foundation.org: add comment] Signed-off-by: Paul Turner Cc: Eric Biederman Cc: Alexey Dobriyan Cc: Al Viro Cc: [2.6.25.x, 2.6.26.x, 2.6.27.x, 2.6.28.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/fs.h b/include/linux/fs.h index 6022f44043f2..5852bd6afbe4 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -54,24 +54,30 @@ struct inodes_stat_t { #define MAY_ACCESS 16 #define MAY_OPEN 32 +/* + * flags in file.f_mode. Note that FMODE_READ and FMODE_WRITE must correspond + * to O_WRONLY and O_RDWR via the strange trick in __dentry_open() + */ + /* file is open for reading */ #define FMODE_READ ((__force fmode_t)1) /* file is open for writing */ #define FMODE_WRITE ((__force fmode_t)2) /* file is seekable */ #define FMODE_LSEEK ((__force fmode_t)4) -/* file can be accessed using pread/pwrite */ +/* file can be accessed using pread */ #define FMODE_PREAD ((__force fmode_t)8) -#define FMODE_PWRITE FMODE_PREAD /* These go hand in hand */ +/* file can be accessed using pwrite */ +#define FMODE_PWRITE ((__force fmode_t)16) /* File is opened for execution with sys_execve / sys_uselib */ -#define FMODE_EXEC ((__force fmode_t)16) +#define FMODE_EXEC ((__force fmode_t)32) /* File is opened with O_NDELAY (only set for block devices) */ -#define FMODE_NDELAY ((__force fmode_t)32) +#define FMODE_NDELAY ((__force fmode_t)64) /* File is opened with O_EXCL (only set for block devices) */ -#define FMODE_EXCL ((__force fmode_t)64) +#define FMODE_EXCL ((__force fmode_t)128) /* File is opened using open(.., 3, ..) and is writeable only for ioctls (specialy hack for floppy.c) */ -#define FMODE_WRITE_IOCTL ((__force fmode_t)128) +#define FMODE_WRITE_IOCTL ((__force fmode_t)256) /* * Don't update ctime and mtime. -- cgit v1.2.3-70-g09d2 From 8f19d472935c83d823fa4cf02bcc0a7b9952db30 Mon Sep 17 00:00:00 2001 From: Eric Biederman Date: Wed, 18 Feb 2009 14:48:16 -0800 Subject: seq_file: properly cope with pread Currently seq_read assumes that the offset passed to it is always the offset it passed to user space. In the case pread this assumption is broken and we do the wrong thing when presented with pread. To solve this I introduce an offset cache inside of struct seq_file so we know where our logical file position is. Then in seq_read if we try to read from another offset we reset our data structures and attempt to go to the offset user space wanted. [akpm@linux-foundation.org: restore FMODE_PWRITE] [pjt@google.com: seq_open needs its fmode opened up to take advantage of this] Signed-off-by: Eric Biederman Cc: Alexey Dobriyan Cc: Al Viro Cc: Paul Turner Cc: [2.6.25.x, 2.6.26.x, 2.6.27.x, 2.6.28.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/seq_file.c | 36 ++++++++++++++++++++++++++++++++---- include/linux/seq_file.h | 1 + 2 files changed, 33 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/fs/seq_file.c b/fs/seq_file.c index 5267098532bf..a1a4cfe19210 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -48,8 +48,16 @@ int seq_open(struct file *file, const struct seq_operations *op) */ file->f_version = 0; - /* SEQ files support lseek, but not pread/pwrite */ - file->f_mode &= ~(FMODE_PREAD | FMODE_PWRITE); + /* + * seq_files support lseek() and pread(). They do not implement + * write() at all, but we clear FMODE_PWRITE here for historical + * reasons. + * + * If a client of seq_files a) implements file.write() and b) wishes to + * support pwrite() then that client will need to implement its own + * file.open() which calls seq_open() and then sets FMODE_PWRITE. + */ + file->f_mode &= ~FMODE_PWRITE; return 0; } EXPORT_SYMBOL(seq_open); @@ -131,6 +139,22 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos) int err = 0; mutex_lock(&m->lock); + + /* Don't assume *ppos is where we left it */ + if (unlikely(*ppos != m->read_pos)) { + m->read_pos = *ppos; + while ((err = traverse(m, *ppos)) == -EAGAIN) + ; + if (err) { + /* With prejudice... */ + m->read_pos = 0; + m->version = 0; + m->index = 0; + m->count = 0; + goto Done; + } + } + /* * seq_file->op->..m_start/m_stop/m_next may do special actions * or optimisations based on the file->f_version, so we want to @@ -230,8 +254,10 @@ Fill: Done: if (!copied) copied = err; - else + else { *ppos += copied; + m->read_pos += copied; + } file->f_version = m->version; mutex_unlock(&m->lock); return copied; @@ -266,16 +292,18 @@ loff_t seq_lseek(struct file *file, loff_t offset, int origin) if (offset < 0) break; retval = offset; - if (offset != file->f_pos) { + if (offset != m->read_pos) { while ((retval=traverse(m, offset)) == -EAGAIN) ; if (retval) { /* with extreme prejudice... */ file->f_pos = 0; + m->read_pos = 0; m->version = 0; m->index = 0; m->count = 0; } else { + m->read_pos = offset; retval = file->f_pos = offset; } } diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h index 40ea5058c2ec..f616f31576d7 100644 --- a/include/linux/seq_file.h +++ b/include/linux/seq_file.h @@ -19,6 +19,7 @@ struct seq_file { size_t from; size_t count; loff_t index; + loff_t read_pos; u64 version; struct mutex lock; const struct seq_operations *op; -- cgit v1.2.3-70-g09d2 From 610d18f4128ebbd88845d0fc60cce67b49af881e Mon Sep 17 00:00:00 2001 From: Davide Libenzi Date: Wed, 18 Feb 2009 14:48:18 -0800 Subject: timerfd: add flags check As requested by Michael, add a missing check for valid flags in timerfd_settime(), and make it return EINVAL in case some extra bits are set. Michael said: If this is to be any use to userland apps that want to check flag support (perhaps it is too late already), then the sooner we get it into the kernel the better: 2.6.29 would be good; earlier stables as well would be even better. [akpm@linux-foundation.org: remove unused TFD_FLAGS_SET] Acked-by: Michael Kerrisk Signed-off-by: Davide Libenzi Cc: [2.6.27.x, 2.6.28.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/timerfd.c | 12 ++++++------ include/linux/timerfd.h | 16 ++++++++++++---- 2 files changed, 18 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/fs/timerfd.c b/fs/timerfd.c index 6a123b8ff3f5..b042bd7034b1 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -186,10 +186,9 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags) BUILD_BUG_ON(TFD_CLOEXEC != O_CLOEXEC); BUILD_BUG_ON(TFD_NONBLOCK != O_NONBLOCK); - if (flags & ~(TFD_CLOEXEC | TFD_NONBLOCK)) - return -EINVAL; - if (clockid != CLOCK_MONOTONIC && - clockid != CLOCK_REALTIME) + if ((flags & ~TFD_CREATE_FLAGS) || + (clockid != CLOCK_MONOTONIC && + clockid != CLOCK_REALTIME)) return -EINVAL; ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); @@ -201,7 +200,7 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags) hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS); ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx, - flags & (O_CLOEXEC | O_NONBLOCK)); + flags & TFD_SHARED_FCNTL_FLAGS); if (ufd < 0) kfree(ctx); @@ -219,7 +218,8 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags, if (copy_from_user(&ktmr, utmr, sizeof(ktmr))) return -EFAULT; - if (!timespec_valid(&ktmr.it_value) || + if ((flags & ~TFD_SETTIME_FLAGS) || + !timespec_valid(&ktmr.it_value) || !timespec_valid(&ktmr.it_interval)) return -EINVAL; diff --git a/include/linux/timerfd.h b/include/linux/timerfd.h index 86cb0501d3e2..2d0792983f8c 100644 --- a/include/linux/timerfd.h +++ b/include/linux/timerfd.h @@ -11,13 +11,21 @@ /* For O_CLOEXEC and O_NONBLOCK */ #include -/* Flags for timerfd_settime. */ +/* + * CAREFUL: Check include/asm-generic/fcntl.h when defining + * new flags, since they might collide with O_* ones. We want + * to re-use O_* flags that couldn't possibly have a meaning + * from eventfd, in order to leave a free define-space for + * shared O_* flags. + */ #define TFD_TIMER_ABSTIME (1 << 0) - -/* Flags for timerfd_create. */ #define TFD_CLOEXEC O_CLOEXEC #define TFD_NONBLOCK O_NONBLOCK +#define TFD_SHARED_FCNTL_FLAGS (TFD_CLOEXEC | TFD_NONBLOCK) +/* Flags for timerfd_create. */ +#define TFD_CREATE_FLAGS TFD_SHARED_FCNTL_FLAGS +/* Flags for timerfd_settime. */ +#define TFD_SETTIME_FLAGS TFD_TIMER_ABSTIME #endif /* _LINUX_TIMERFD_H */ - -- cgit v1.2.3-70-g09d2 From 1cf6e7d83bf334cc5916137862c920a97aabc018 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 18 Feb 2009 14:48:18 -0800 Subject: mm: task dirty accounting fix YAMAMOTO-san noticed that task_dirty_inc doesn't seem to be called properly for cases where set_page_dirty is not used to dirty a page (eg. mark_buffer_dirty). Additionally, there is some inconsistency about when task_dirty_inc is called. It is used for dirty balancing, however it even gets called for __set_page_dirty_no_writeback. So rather than increment it in a set_page_dirty wrapper, move it down to exactly where the dirty page accounting stats are incremented. Cc: YAMAMOTO Takashi Signed-off-by: Nick Piggin Acked-by: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 1 + include/linux/mm.h | 1 + mm/page-writeback.c | 13 +++---------- 3 files changed, 5 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/fs/buffer.c b/fs/buffer.c index 665d446b25bc..ff4d1cdd779b 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -777,6 +777,7 @@ static int __set_page_dirty(struct page *page, __inc_zone_page_state(page, NR_FILE_DIRTY); __inc_bdi_stat(mapping->backing_dev_info, BDI_RECLAIMABLE); + task_dirty_inc(current); task_io_account_write(PAGE_CACHE_SIZE); } radix_tree_tag_set(&mapping->page_tree, diff --git a/include/linux/mm.h b/include/linux/mm.h index 7dc04ff5ab89..10074212a35b 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1159,6 +1159,7 @@ extern int filemap_fault(struct vm_area_struct *, struct vm_fault *); /* mm/page-writeback.c */ int write_one_page(struct page *page, int wait); +void task_dirty_inc(struct task_struct *tsk); /* readahead.c */ #define VM_MAX_READAHEAD 128 /* kbytes */ diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 3c84128596ba..74dc57c74349 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -240,7 +240,7 @@ void bdi_writeout_inc(struct backing_dev_info *bdi) } EXPORT_SYMBOL_GPL(bdi_writeout_inc); -static inline void task_dirty_inc(struct task_struct *tsk) +void task_dirty_inc(struct task_struct *tsk) { prop_inc_single(&vm_dirties, &tsk->dirties); } @@ -1230,6 +1230,7 @@ int __set_page_dirty_nobuffers(struct page *page) __inc_zone_page_state(page, NR_FILE_DIRTY); __inc_bdi_stat(mapping->backing_dev_info, BDI_RECLAIMABLE); + task_dirty_inc(current); task_io_account_write(PAGE_CACHE_SIZE); } radix_tree_tag_set(&mapping->page_tree, @@ -1262,7 +1263,7 @@ EXPORT_SYMBOL(redirty_page_for_writepage); * If the mapping doesn't provide a set_page_dirty a_op, then * just fall through and assume that it wants buffer_heads. */ -static int __set_page_dirty(struct page *page) +int set_page_dirty(struct page *page) { struct address_space *mapping = page_mapping(page); @@ -1280,14 +1281,6 @@ static int __set_page_dirty(struct page *page) } return 0; } - -int set_page_dirty(struct page *page) -{ - int ret = __set_page_dirty(page); - if (ret) - task_dirty_inc(current); - return ret; -} EXPORT_SYMBOL(set_page_dirty); /* -- cgit v1.2.3-70-g09d2 From 287d859222e0adbc67666a6154aaf42d7d5bbb54 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Wed, 18 Feb 2009 14:48:26 -0800 Subject: atmel-mci: fix initialization of dma slave data The conversion of atmel-mci to dma_request_channel missed the initialization of the channel dma_slave information. The filter_fn passed to dma_request_channel is responsible for initializing the channel's private data. This implementation has the additional benefit of enabling a generic client-channel data passing mechanism. Reviewed-by: Atsushi Nemoto Signed-off-by: Dan Williams Acked-by: Haavard Skinnemoen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/dma/dmaengine.c | 2 ++ drivers/dma/dw_dmac.c | 5 ++--- drivers/dma/dw_dmac_regs.h | 2 -- drivers/mmc/host/atmel-mci.c | 5 +++-- include/linux/dmaengine.h | 2 ++ 5 files changed, 9 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c index a58993011edb..280a9d263eb3 100644 --- a/drivers/dma/dmaengine.c +++ b/drivers/dma/dmaengine.c @@ -518,6 +518,7 @@ struct dma_chan *__dma_request_channel(dma_cap_mask_t *mask, dma_filter_fn fn, v dma_chan_name(chan), err); else break; + chan->private = NULL; chan = NULL; } } @@ -536,6 +537,7 @@ void dma_release_channel(struct dma_chan *chan) WARN_ONCE(chan->client_count != 1, "chan reference count %d != 1\n", chan->client_count); dma_chan_put(chan); + chan->private = NULL; mutex_unlock(&dma_list_mutex); } EXPORT_SYMBOL_GPL(dma_release_channel); diff --git a/drivers/dma/dw_dmac.c b/drivers/dma/dw_dmac.c index 6b702cc46b3d..a97c07eef7ec 100644 --- a/drivers/dma/dw_dmac.c +++ b/drivers/dma/dw_dmac.c @@ -560,7 +560,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl, unsigned long flags) { struct dw_dma_chan *dwc = to_dw_dma_chan(chan); - struct dw_dma_slave *dws = dwc->dws; + struct dw_dma_slave *dws = chan->private; struct dw_desc *prev; struct dw_desc *first; u32 ctllo; @@ -790,7 +790,7 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan) cfghi = DWC_CFGH_FIFO_MODE; cfglo = 0; - dws = dwc->dws; + dws = chan->private; if (dws) { /* * We need controller-specific data to set up slave @@ -866,7 +866,6 @@ static void dwc_free_chan_resources(struct dma_chan *chan) spin_lock_bh(&dwc->lock); list_splice_init(&dwc->free_list, &list); dwc->descs_allocated = 0; - dwc->dws = NULL; /* Disable interrupts */ channel_clear_bit(dw, MASK.XFER, dwc->mask); diff --git a/drivers/dma/dw_dmac_regs.h b/drivers/dma/dw_dmac_regs.h index 00fdd187bb0c..b252b202c5cf 100644 --- a/drivers/dma/dw_dmac_regs.h +++ b/drivers/dma/dw_dmac_regs.h @@ -139,8 +139,6 @@ struct dw_dma_chan { struct list_head queue; struct list_head free_list; - struct dw_dma_slave *dws; - unsigned int descs_allocated; }; diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c index 76bfe16c09b1..2b1196e6142c 100644 --- a/drivers/mmc/host/atmel-mci.c +++ b/drivers/mmc/host/atmel-mci.c @@ -1548,9 +1548,10 @@ static bool filter(struct dma_chan *chan, void *slave) { struct dw_dma_slave *dws = slave; - if (dws->dma_dev == chan->device->dev) + if (dws->dma_dev == chan->device->dev) { + chan->private = dws; return true; - else + } else return false; } #endif diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index 3e68469c1885..f0413845f20e 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -121,6 +121,7 @@ struct dma_chan_percpu { * @local: per-cpu pointer to a struct dma_chan_percpu * @client-count: how many clients are using this channel * @table_count: number of appearances in the mem-to-mem allocation table + * @private: private data for certain client-channel associations */ struct dma_chan { struct dma_device *device; @@ -134,6 +135,7 @@ struct dma_chan { struct dma_chan_percpu *local; int client_count; int table_count; + void *private; }; /** -- cgit v1.2.3-70-g09d2 From f2dbcfa738368c8a40d4a5f0b65dc9879577cb21 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Wed, 18 Feb 2009 14:48:32 -0800 Subject: mm: clean up for early_pfn_to_nid() What's happening is that the assertion in mm/page_alloc.c:move_freepages() is triggering: BUG_ON(page_zone(start_page) != page_zone(end_page)); Once I knew this is what was happening, I added some annotations: if (unlikely(page_zone(start_page) != page_zone(end_page))) { printk(KERN_ERR "move_freepages: Bogus zones: " "start_page[%p] end_page[%p] zone[%p]\n", start_page, end_page, zone); printk(KERN_ERR "move_freepages: " "start_zone[%p] end_zone[%p]\n", page_zone(start_page), page_zone(end_page)); printk(KERN_ERR "move_freepages: " "start_pfn[0x%lx] end_pfn[0x%lx]\n", page_to_pfn(start_page), page_to_pfn(end_page)); printk(KERN_ERR "move_freepages: " "start_nid[%d] end_nid[%d]\n", page_to_nid(start_page), page_to_nid(end_page)); ... And here's what I got: move_freepages: Bogus zones: start_page[2207d0000] end_page[2207dffc0] zone[fffff8103effcb00] move_freepages: start_zone[fffff8103effcb00] end_zone[fffff8003fffeb00] move_freepages: start_pfn[0x81f600] end_pfn[0x81f7ff] move_freepages: start_nid[1] end_nid[0] My memory layout on this box is: [ 0.000000] Zone PFN ranges: [ 0.000000] Normal 0x00000000 -> 0x0081ff5d [ 0.000000] Movable zone start PFN for each node [ 0.000000] early_node_map[8] active PFN ranges [ 0.000000] 0: 0x00000000 -> 0x00020000 [ 0.000000] 1: 0x00800000 -> 0x0081f7ff [ 0.000000] 1: 0x0081f800 -> 0x0081fe50 [ 0.000000] 1: 0x0081fed1 -> 0x0081fed8 [ 0.000000] 1: 0x0081feda -> 0x0081fedb [ 0.000000] 1: 0x0081fedd -> 0x0081fee5 [ 0.000000] 1: 0x0081fee7 -> 0x0081ff51 [ 0.000000] 1: 0x0081ff59 -> 0x0081ff5d So it's a block move in that 0x81f600-->0x81f7ff region which triggers the problem. This patch: Declaration of early_pfn_to_nid() is scattered over per-arch include files, and it seems it's complicated to know when the declaration is used. I think it makes fix-for-memmap-init not easy. This patch moves all declaration to include/linux/mm.h After this, if !CONFIG_NODES_POPULATES_NODE_MAP && !CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID -> Use static definition in include/linux/mm.h else if !CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID -> Use generic definition in mm/page_alloc.c else -> per-arch back end function will be called. Signed-off-by: KAMEZAWA Hiroyuki Tested-by: KOSAKI Motohiro Reported-by: David Miller Cc: Mel Gorman Cc: Heiko Carstens Cc: [2.6.25.x, 2.6.26.x, 2.6.27.x, 2.6.28.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/include/asm/mmzone.h | 4 ---- arch/ia64/mm/numa.c | 2 +- arch/x86/include/asm/mmzone_32.h | 2 -- arch/x86/include/asm/mmzone_64.h | 2 -- arch/x86/mm/numa_64.c | 2 +- include/linux/mm.h | 19 ++++++++++++++++--- mm/page_alloc.c | 8 +++++++- 7 files changed, 25 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/arch/ia64/include/asm/mmzone.h b/arch/ia64/include/asm/mmzone.h index 34efe88eb849..f2ca32069b3f 100644 --- a/arch/ia64/include/asm/mmzone.h +++ b/arch/ia64/include/asm/mmzone.h @@ -31,10 +31,6 @@ static inline int pfn_to_nid(unsigned long pfn) #endif } -#ifdef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID -extern int early_pfn_to_nid(unsigned long pfn); -#endif - #ifdef CONFIG_IA64_DIG /* DIG systems are small */ # define MAX_PHYSNODE_ID 8 # define NR_NODE_MEMBLKS (MAX_NUMNODES * 8) diff --git a/arch/ia64/mm/numa.c b/arch/ia64/mm/numa.c index b73bf1838e57..5061c3fb6796 100644 --- a/arch/ia64/mm/numa.c +++ b/arch/ia64/mm/numa.c @@ -58,7 +58,7 @@ paddr_to_nid(unsigned long paddr) * SPARSEMEM to allocate the SPARSEMEM sectionmap on the NUMA node where * the section resides. */ -int early_pfn_to_nid(unsigned long pfn) +int __meminit __early_pfn_to_nid(unsigned long pfn) { int i, section = pfn >> PFN_SECTION_SHIFT, ssec, esec; diff --git a/arch/x86/include/asm/mmzone_32.h b/arch/x86/include/asm/mmzone_32.h index 07f1af494ca5..105fb90a0635 100644 --- a/arch/x86/include/asm/mmzone_32.h +++ b/arch/x86/include/asm/mmzone_32.h @@ -32,8 +32,6 @@ static inline void get_memcfg_numa(void) get_memcfg_numa_flat(); } -extern int early_pfn_to_nid(unsigned long pfn); - extern void resume_map_numa_kva(pgd_t *pgd); #else /* !CONFIG_NUMA */ diff --git a/arch/x86/include/asm/mmzone_64.h b/arch/x86/include/asm/mmzone_64.h index a5b3817d4b9e..a29f48c2a322 100644 --- a/arch/x86/include/asm/mmzone_64.h +++ b/arch/x86/include/asm/mmzone_64.h @@ -40,8 +40,6 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr) #define node_end_pfn(nid) (NODE_DATA(nid)->node_start_pfn + \ NODE_DATA(nid)->node_spanned_pages) -extern int early_pfn_to_nid(unsigned long pfn); - #ifdef CONFIG_NUMA_EMU #define FAKE_NODE_MIN_SIZE (64 * 1024 * 1024) #define FAKE_NODE_MIN_HASH_MASK (~(FAKE_NODE_MIN_SIZE - 1UL)) diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index 71a14f89f89e..f3516da035d1 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -145,7 +145,7 @@ int __init compute_hash_shift(struct bootnode *nodes, int numnodes, return shift; } -int early_pfn_to_nid(unsigned long pfn) +int __meminit __early_pfn_to_nid(unsigned long pfn) { return phys_to_nid(pfn << PAGE_SHIFT); } diff --git a/include/linux/mm.h b/include/linux/mm.h index 10074212a35b..065cdf8c09fb 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1041,10 +1041,23 @@ extern void free_bootmem_with_active_regions(int nid, typedef int (*work_fn_t)(unsigned long, unsigned long, void *); extern void work_with_active_regions(int nid, work_fn_t work_fn, void *data); extern void sparse_memory_present_with_active_regions(int nid); -#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID -extern int early_pfn_to_nid(unsigned long pfn); -#endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */ #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */ + +#if !defined(CONFIG_ARCH_POPULATES_NODE_MAP) && \ + !defined(CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID) +static inline int __early_pfn_to_nid(unsigned long pfn) +{ + return 0; +} +#else +/* please see mm/page_alloc.c */ +extern int __meminit early_pfn_to_nid(unsigned long pfn); +#ifdef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID +/* there is a per-arch backend function. */ +extern int __meminit __early_pfn_to_nid(unsigned long pfn); +#endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */ +#endif + extern void set_dma_reserve(unsigned long new_dma_reserve); extern void memmap_init_zone(unsigned long, int, unsigned long, unsigned long, enum memmap_context); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 5675b3073854..c5dd74602efc 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2989,7 +2989,7 @@ static int __meminit next_active_region_index_in_nid(int index, int nid) * was used and there are no special requirements, this is a convenient * alternative */ -int __meminit early_pfn_to_nid(unsigned long pfn) +int __meminit __early_pfn_to_nid(unsigned long pfn) { int i; @@ -3005,6 +3005,12 @@ int __meminit early_pfn_to_nid(unsigned long pfn) } #endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */ +int __meminit early_pfn_to_nid(unsigned long pfn) +{ + return __early_pfn_to_nid(pfn); +} + + /* Basic iterator support to walk early_node_map[] */ #define for_each_active_range_index_in_nid(i, nid) \ for (i = first_active_region_index_in_nid(nid); i != -1; \ -- cgit v1.2.3-70-g09d2 From cc2559bccc72767cb446f79b071d96c30c26439b Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Wed, 18 Feb 2009 14:48:33 -0800 Subject: mm: fix memmap init for handling memory hole Now, early_pfn_in_nid(PFN, NID) may returns false if PFN is a hole. and memmap initialization was not done. This was a trouble for sparc boot. To fix this, the PFN should be initialized and marked as PG_reserved. This patch changes early_pfn_in_nid() return true if PFN is a hole. Signed-off-by: KAMEZAWA Hiroyuki Reported-by: David Miller Tested-by: KOSAKI Motohiro Cc: Mel Gorman Cc: Heiko Carstens Cc: [2.6.25.x, 2.6.26.x, 2.6.27.x, 2.6.28.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/mm/numa.c | 2 +- include/linux/mmzone.h | 2 +- mm/page_alloc.c | 23 ++++++++++++++++++++--- 3 files changed, 22 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/arch/ia64/mm/numa.c b/arch/ia64/mm/numa.c index 5061c3fb6796..3efea7d0a351 100644 --- a/arch/ia64/mm/numa.c +++ b/arch/ia64/mm/numa.c @@ -70,7 +70,7 @@ int __meminit __early_pfn_to_nid(unsigned long pfn) return node_memblk[i].nid; } - return 0; + return -1; } #ifdef CONFIG_MEMORY_HOTPLUG diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 09c14e213b63..1aca6cebbb78 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -1071,7 +1071,7 @@ void sparse_init(void); #endif /* CONFIG_SPARSEMEM */ #ifdef CONFIG_NODES_SPAN_OTHER_NODES -#define early_pfn_in_nid(pfn, nid) (early_pfn_to_nid(pfn) == (nid)) +bool early_pfn_in_nid(unsigned long pfn, int nid); #else #define early_pfn_in_nid(pfn, nid) (1) #endif diff --git a/mm/page_alloc.c b/mm/page_alloc.c index c5dd74602efc..5c44ed49ca93 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3000,16 +3000,33 @@ int __meminit __early_pfn_to_nid(unsigned long pfn) if (start_pfn <= pfn && pfn < end_pfn) return early_node_map[i].nid; } - - return 0; + /* This is a memory hole */ + return -1; } #endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */ int __meminit early_pfn_to_nid(unsigned long pfn) { - return __early_pfn_to_nid(pfn); + int nid; + + nid = __early_pfn_to_nid(pfn); + if (nid >= 0) + return nid; + /* just returns 0 */ + return 0; } +#ifdef CONFIG_NODES_SPAN_OTHER_NODES +bool __meminit early_pfn_in_nid(unsigned long pfn, int node) +{ + int nid; + + nid = __early_pfn_to_nid(pfn); + if (nid >= 0 && nid != node) + return false; + return true; +} +#endif /* Basic iterator support to walk early_node_map[] */ #define for_each_active_range_index_in_nid(i, nid) \ -- cgit v1.2.3-70-g09d2 From ffa7525c13eb3db0fd19a3e1cffe2ce6f561f5f3 Mon Sep 17 00:00:00 2001 From: Adam Lackorzynski Date: Wed, 18 Feb 2009 14:48:34 -0800 Subject: jsm: additional device support I have a Digi Neo 8 PCI card (114f:00b1) Serial controller: Digi International Digi Neo 8 (rev 05) that works with the jsm driver after using the following patch. Signed-off-by: Adam Lackorzynski Cc: Scott H Kilau Cc: Wendy Xiong Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/jsm/jsm_driver.c | 3 +++ include/linux/pci_ids.h | 1 + 2 files changed, 4 insertions(+) (limited to 'include') diff --git a/drivers/serial/jsm/jsm_driver.c b/drivers/serial/jsm/jsm_driver.c index 92187e28608a..ac79cbe4c2cf 100644 --- a/drivers/serial/jsm/jsm_driver.c +++ b/drivers/serial/jsm/jsm_driver.c @@ -84,6 +84,8 @@ static int jsm_probe_one(struct pci_dev *pdev, const struct pci_device_id *ent) brd->pci_dev = pdev; if (pdev->device == PCIE_DEVICE_ID_NEO_4_IBM) brd->maxports = 4; + else if (pdev->device == PCI_DEVICE_ID_DIGI_NEO_8) + brd->maxports = 8; else brd->maxports = 2; @@ -212,6 +214,7 @@ static struct pci_device_id jsm_pci_tbl[] = { { PCI_DEVICE(PCI_VENDOR_ID_DIGI, PCI_DEVICE_ID_NEO_2RJ45), 0, 0, 2 }, { PCI_DEVICE(PCI_VENDOR_ID_DIGI, PCI_DEVICE_ID_NEO_2RJ45PRI), 0, 0, 3 }, { PCI_DEVICE(PCI_VENDOR_ID_DIGI, PCIE_DEVICE_ID_NEO_4_IBM), 0, 0, 4 }, + { PCI_DEVICE(PCI_VENDOR_ID_DIGI, PCI_DEVICE_ID_DIGI_NEO_8), 0, 0, 5 }, { 0, } }; MODULE_DEVICE_TABLE(pci, jsm_pci_tbl); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 918391b4b109..114b8192eab9 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1445,6 +1445,7 @@ #define PCI_DEVICE_ID_DIGI_DF_M_E 0x0071 #define PCI_DEVICE_ID_DIGI_DF_M_IOM2_A 0x0072 #define PCI_DEVICE_ID_DIGI_DF_M_A 0x0073 +#define PCI_DEVICE_ID_DIGI_NEO_8 0x00B1 #define PCI_DEVICE_ID_NEO_2DB9 0x00C8 #define PCI_DEVICE_ID_NEO_2DB9PRI 0x00C9 #define PCI_DEVICE_ID_NEO_2RJ45 0x00CA -- cgit v1.2.3-70-g09d2 From 97bef7dd05563807539122c488a5dd93ed327722 Mon Sep 17 00:00:00 2001 From: Bernhard Walle Date: Wed, 18 Feb 2009 14:48:40 -0800 Subject: Bernhard has moved Since I don't work for SUSE any more and the bwalle@suse.de address is invalid, correct it in the copyright headers and documentation. Signed-off-by: Bernhard Walle Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/ABI/testing/sysfs-firmware-memmap | 2 +- drivers/firmware/memmap.c | 2 +- include/linux/firmware-map.h | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/Documentation/ABI/testing/sysfs-firmware-memmap b/Documentation/ABI/testing/sysfs-firmware-memmap index 0d99ee6ae02e..eca0d65087dc 100644 --- a/Documentation/ABI/testing/sysfs-firmware-memmap +++ b/Documentation/ABI/testing/sysfs-firmware-memmap @@ -1,6 +1,6 @@ What: /sys/firmware/memmap/ Date: June 2008 -Contact: Bernhard Walle +Contact: Bernhard Walle Description: On all platforms, the firmware provides a memory map which the kernel reads. The resources from that memory map are registered diff --git a/drivers/firmware/memmap.c b/drivers/firmware/memmap.c index 261b9aa3f248..05aa2d406ac6 100644 --- a/drivers/firmware/memmap.c +++ b/drivers/firmware/memmap.c @@ -1,7 +1,7 @@ /* * linux/drivers/firmware/memmap.c * Copyright (C) 2008 SUSE LINUX Products GmbH - * by Bernhard Walle + * by Bernhard Walle * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License v2.0 as published by diff --git a/include/linux/firmware-map.h b/include/linux/firmware-map.h index 6e199c8dfacc..cca686b39123 100644 --- a/include/linux/firmware-map.h +++ b/include/linux/firmware-map.h @@ -1,7 +1,7 @@ /* * include/linux/firmware-map.h: * Copyright (C) 2008 SUSE LINUX Products GmbH - * by Bernhard Walle + * by Bernhard Walle * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License v2.0 as published by -- cgit v1.2.3-70-g09d2 From e4dd61882e2cfe47ea72ecd825671e8e5ae29038 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Wed, 18 Feb 2009 23:31:11 -0800 Subject: vlan: Update skb->mac_header in __vlan_put_tag(). After moving mac addresses in __vlan_put_tag() skb->mac_header needs to be updated. Reported-by: Karl Hiramoto Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- include/linux/if_vlan.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index f8ff918c208f..e1ff5b14310e 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -210,6 +210,7 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, u16 vlan_tci) /* Move the mac addresses to the beginning of the new header. */ memmove(skb->data, skb->data + VLAN_HLEN, 2 * VLAN_ETH_ALEN); + skb->mac_header -= VLAN_HLEN; /* first, the ethernet type */ veth->h_vlan_proto = htons(ETH_P_8021Q); -- cgit v1.2.3-70-g09d2 From ea39f835168f60b01e59d0f348da25d297e7cf94 Mon Sep 17 00:00:00 2001 From: Kristian Høgsberg Date: Thu, 12 Feb 2009 14:37:56 -0500 Subject: drm: Release user fbs in drm_release MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Avoids leaking fbs and associated buffers on release. Signed-off-by: Kristian Høgsberg Tested-by: Tested-by: Chris Wilson Signed-off-by: Eric Anholt Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_crtc.c | 3 +-- drivers/gpu/drm/drm_fops.c | 3 +++ include/drm/drm_crtc.h | 2 +- 3 files changed, 5 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index bfce0992fefb..94a768871734 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -1741,9 +1741,8 @@ out: * RETURNS: * Zero on success, errno on failure. */ -void drm_fb_release(struct file *filp) +void drm_fb_release(struct drm_file *priv) { - struct drm_file *priv = filp->private_data; struct drm_device *dev = priv->minor->dev; struct drm_framebuffer *fb, *tfb; diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c index b06a53715853..6c020fe5431c 100644 --- a/drivers/gpu/drm/drm_fops.c +++ b/drivers/gpu/drm/drm_fops.c @@ -457,6 +457,9 @@ int drm_release(struct inode *inode, struct file *filp) if (dev->driver->driver_features & DRIVER_GEM) drm_gem_release(dev, file_priv); + if (dev->driver->driver_features & DRIVER_MODESET) + drm_fb_release(file_priv); + mutex_lock(&dev->ctxlist_mutex); if (!list_empty(&dev->ctxlist)) { struct drm_ctx_list *pos, *n; diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h index d54de24bf371..5ded1acfb543 100644 --- a/include/drm/drm_crtc.h +++ b/include/drm/drm_crtc.h @@ -609,7 +609,7 @@ extern char *drm_get_dvi_i_subconnector_name(int val); extern char *drm_get_dvi_i_select_name(int val); extern char *drm_get_tv_subconnector_name(int val); extern char *drm_get_tv_select_name(int val); -extern void drm_fb_release(struct file *filp); +extern void drm_fb_release(struct drm_file *file_priv); extern int drm_mode_group_init_legacy_group(struct drm_device *dev, struct drm_mode_group *group); extern struct edid *drm_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter); -- cgit v1.2.3-70-g09d2 From 5c3b82e2b229e78eb32f4ea12d16f3ebeeab3fc7 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Wed, 11 Feb 2009 13:25:09 +0000 Subject: drm: Propagate failure from setting crtc base. Check the error paths within intel_pipe_set_base() to first cleanup and then report back the error. Signed-off-by: Chris Wilson Signed-off-by: Eric Anholt Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_crtc_helper.c | 15 +++++-- drivers/gpu/drm/i915/intel_display.c | 84 +++++++++++++++++++++++------------- include/drm/drm_crtc_helper.h | 10 ++--- 3 files changed, 69 insertions(+), 40 deletions(-) (limited to 'include') diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c index 8ba22c039a11..733028b4d45e 100644 --- a/drivers/gpu/drm/drm_crtc_helper.c +++ b/drivers/gpu/drm/drm_crtc_helper.c @@ -512,8 +512,8 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc, if (drm_mode_equal(&saved_mode, &crtc->mode)) { if (saved_x != crtc->x || saved_y != crtc->y || depth_changed || bpp_changed) { - crtc_funcs->mode_set_base(crtc, crtc->x, crtc->y, - old_fb); + ret = !crtc_funcs->mode_set_base(crtc, crtc->x, crtc->y, + old_fb); goto done; } } @@ -552,7 +552,9 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc, /* Set up the DPLL and any encoders state that needs to adjust or depend * on the DPLL. */ - crtc_funcs->mode_set(crtc, mode, adjusted_mode, x, y, old_fb); + ret = !crtc_funcs->mode_set(crtc, mode, adjusted_mode, x, y, old_fb); + if (!ret) + goto done; list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { @@ -752,6 +754,8 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set) if (!drm_crtc_helper_set_mode(set->crtc, set->mode, set->x, set->y, old_fb)) { + DRM_ERROR("failed to set mode on crtc %p\n", + set->crtc); ret = -EINVAL; goto fail_set_mode; } @@ -765,7 +769,10 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set) old_fb = set->crtc->fb; if (set->crtc->fb != set->fb) set->crtc->fb = set->fb; - crtc_funcs->mode_set_base(set->crtc, set->x, set->y, old_fb); + ret = crtc_funcs->mode_set_base(set->crtc, + set->x, set->y, old_fb); + if (ret != 0) + goto fail_set_mode; } kfree(save_encoders); diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index a440d0db5ccc..ac92799fa8a1 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -343,7 +343,7 @@ intel_wait_for_vblank(struct drm_device *dev) udelay(20000); } -static void +static int intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, struct drm_framebuffer *old_fb) { @@ -361,11 +361,21 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, int dspstride = (pipe == 0) ? DSPASTRIDE : DSPBSTRIDE; int dspcntr_reg = (pipe == 0) ? DSPACNTR : DSPBCNTR; u32 dspcntr, alignment; + int ret; /* no fb bound */ if (!crtc->fb) { DRM_DEBUG("No FB bound\n"); - return; + return 0; + } + + switch (pipe) { + case 0: + case 1: + break; + default: + DRM_ERROR("Can't update pipe %d in SAREA\n", pipe); + return -EINVAL; } intel_fb = to_intel_framebuffer(crtc->fb); @@ -383,20 +393,24 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, case I915_TILING_Y: /* FIXME: Is this true? */ DRM_ERROR("Y tiled not allowed for scan out buffers\n"); - return; + return -EINVAL; default: BUG(); } - if (i915_gem_object_pin(intel_fb->obj, alignment)) - return; - - i915_gem_object_set_to_gtt_domain(intel_fb->obj, 1); - - Start = obj_priv->gtt_offset; - Offset = y * crtc->fb->pitch + x * (crtc->fb->bits_per_pixel / 8); + mutex_lock(&dev->struct_mutex); + ret = i915_gem_object_pin(intel_fb->obj, alignment); + if (ret != 0) { + mutex_unlock(&dev->struct_mutex); + return ret; + } - I915_WRITE(dspstride, crtc->fb->pitch); + ret = i915_gem_object_set_to_gtt_domain(intel_fb->obj, 1); + if (ret != 0) { + i915_gem_object_unpin(intel_fb->obj); + mutex_unlock(&dev->struct_mutex); + return ret; + } dspcntr = I915_READ(dspcntr_reg); /* Mask out pixel format bits in case we change it */ @@ -417,11 +431,17 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, break; default: DRM_ERROR("Unknown color depth\n"); - return; + i915_gem_object_unpin(intel_fb->obj); + mutex_unlock(&dev->struct_mutex); + return -EINVAL; } I915_WRITE(dspcntr_reg, dspcntr); + Start = obj_priv->gtt_offset; + Offset = y * crtc->fb->pitch + x * (crtc->fb->bits_per_pixel / 8); + DRM_DEBUG("Writing base %08lX %08lX %d %d\n", Start, Offset, x, y); + I915_WRITE(dspstride, crtc->fb->pitch); if (IS_I965G(dev)) { I915_WRITE(dspbase, Offset); I915_READ(dspbase); @@ -438,27 +458,24 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, intel_fb = to_intel_framebuffer(old_fb); i915_gem_object_unpin(intel_fb->obj); } + mutex_unlock(&dev->struct_mutex); if (!dev->primary->master) - return; + return 0; master_priv = dev->primary->master->driver_priv; if (!master_priv->sarea_priv) - return; + return 0; - switch (pipe) { - case 0: - master_priv->sarea_priv->pipeA_x = x; - master_priv->sarea_priv->pipeA_y = y; - break; - case 1: + if (pipe) { master_priv->sarea_priv->pipeB_x = x; master_priv->sarea_priv->pipeB_y = y; - break; - default: - DRM_ERROR("Can't update pipe %d in SAREA\n", pipe); - break; + } else { + master_priv->sarea_priv->pipeA_x = x; + master_priv->sarea_priv->pipeA_y = y; } + + return 0; } @@ -706,11 +723,11 @@ static int intel_panel_fitter_pipe (struct drm_device *dev) return 1; } -static void intel_crtc_mode_set(struct drm_crtc *crtc, - struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode, - int x, int y, - struct drm_framebuffer *old_fb) +static int intel_crtc_mode_set(struct drm_crtc *crtc, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode, + int x, int y, + struct drm_framebuffer *old_fb) { struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; @@ -737,6 +754,7 @@ static void intel_crtc_mode_set(struct drm_crtc *crtc, bool is_crt = false, is_lvds = false, is_tv = false; struct drm_mode_config *mode_config = &dev->mode_config; struct drm_connector *connector; + int ret; drm_vblank_pre_modeset(dev, pipe); @@ -777,7 +795,7 @@ static void intel_crtc_mode_set(struct drm_crtc *crtc, ok = intel_find_best_PLL(crtc, adjusted_mode->clock, refclk, &clock); if (!ok) { DRM_ERROR("Couldn't find PLL settings for mode!\n"); - return; + return -EINVAL; } fp = clock.n << 16 | clock.m1 << 8 | clock.m2; @@ -948,9 +966,13 @@ static void intel_crtc_mode_set(struct drm_crtc *crtc, I915_WRITE(dspcntr_reg, dspcntr); /* Flush the plane changes */ - intel_pipe_set_base(crtc, x, y, old_fb); + ret = intel_pipe_set_base(crtc, x, y, old_fb); + if (ret != 0) + return ret; drm_vblank_post_modeset(dev, pipe); + + return 0; } /** Loads the palette/gamma unit for the CRTC with the prepared values */ diff --git a/include/drm/drm_crtc_helper.h b/include/drm/drm_crtc_helper.h index 0c6f0e11b41b..0b0d236c2154 100644 --- a/include/drm/drm_crtc_helper.h +++ b/include/drm/drm_crtc_helper.h @@ -54,13 +54,13 @@ struct drm_crtc_helper_funcs { struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); /* Actually set the mode */ - void (*mode_set)(struct drm_crtc *crtc, struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode, int x, int y, - struct drm_framebuffer *old_fb); + int (*mode_set)(struct drm_crtc *crtc, struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode, int x, int y, + struct drm_framebuffer *old_fb); /* Move the crtc on the current fb to the given position *optional* */ - void (*mode_set_base)(struct drm_crtc *crtc, int x, int y, - struct drm_framebuffer *old_fb); + int (*mode_set_base)(struct drm_crtc *crtc, int x, int y, + struct drm_framebuffer *old_fb); }; struct drm_encoder_helper_funcs { -- cgit v1.2.3-70-g09d2 From ab00b3e5210954cbaff9207db874a9f03197e3ba Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Wed, 11 Feb 2009 14:01:46 -0800 Subject: drm/i915: Keep refs on the object over the lifetime of vmas for GTT mmap. This fixes potential fault at fault time if the object was unreferenced while the mapping still existed. Now, while the mmap_offset only lives for the lifetime of the object, the object also stays alive while a vma exists that needs it. Signed-off-by: Jesse Barnes Signed-off-by: Eric Anholt Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_gem.c | 28 +++++++++++++++++++++++++++ drivers/gpu/drm/i915/i915_drv.c | 2 ++ drivers/gpu/drm/i915/i915_gem.c | 43 ++++++++++++++++++++++++----------------- include/drm/drmP.h | 2 ++ 4 files changed, 57 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index 5dad6b9d0dec..88d3368ffddd 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -463,6 +463,26 @@ drm_gem_object_handle_free(struct kref *kref) } EXPORT_SYMBOL(drm_gem_object_handle_free); +void drm_gem_vm_open(struct vm_area_struct *vma) +{ + struct drm_gem_object *obj = vma->vm_private_data; + + drm_gem_object_reference(obj); +} +EXPORT_SYMBOL(drm_gem_vm_open); + +void drm_gem_vm_close(struct vm_area_struct *vma) +{ + struct drm_gem_object *obj = vma->vm_private_data; + struct drm_device *dev = obj->dev; + + mutex_lock(&dev->struct_mutex); + drm_gem_object_unreference(obj); + mutex_unlock(&dev->struct_mutex); +} +EXPORT_SYMBOL(drm_gem_vm_close); + + /** * drm_gem_mmap - memory map routine for GEM objects * @filp: DRM file pointer @@ -524,6 +544,14 @@ int drm_gem_mmap(struct file *filp, struct vm_area_struct *vma) #endif vma->vm_page_prot = __pgprot(prot); + /* Take a ref for this mapping of the object, so that the fault + * handler can dereference the mmap offset's pointer to the object. + * This reference is cleaned up by the corresponding vm_close + * (which should happen whether the vma was created by this call, or + * by a vm_open due to mremap or partial unmap or whatever). + */ + drm_gem_object_reference(obj); + vma->vm_file = filp; /* Needed for drm_vm_open() */ drm_vm_open_locked(vma); diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index aac12ee31a46..a31cbdbc3c54 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -94,6 +94,8 @@ static int i915_resume(struct drm_device *dev) static struct vm_operations_struct i915_gem_vm_ops = { .fault = i915_gem_fault, + .open = drm_gem_vm_open, + .close = drm_gem_vm_close, }; static struct drm_driver driver = { diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 078858178832..ac534c9a2f81 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -607,8 +607,6 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) case -EAGAIN: return VM_FAULT_OOM; case -EFAULT: - case -EBUSY: - DRM_ERROR("can't insert pfn?? fault or busy...\n"); return VM_FAULT_SIGBUS; default: return VM_FAULT_NOPAGE; @@ -684,6 +682,30 @@ out_free_list: return ret; } +static void +i915_gem_free_mmap_offset(struct drm_gem_object *obj) +{ + struct drm_device *dev = obj->dev; + struct drm_i915_gem_object *obj_priv = obj->driver_private; + struct drm_gem_mm *mm = dev->mm_private; + struct drm_map_list *list; + + list = &obj->map_list; + drm_ht_remove_item(&mm->offset_hash, &list->hash); + + if (list->file_offset_node) { + drm_mm_put_block(list->file_offset_node); + list->file_offset_node = NULL; + } + + if (list->map) { + drm_free(list->map, sizeof(struct drm_map), DRM_MEM_DRIVER); + list->map = NULL; + } + + obj_priv->mmap_offset = 0; +} + /** * i915_gem_get_gtt_alignment - return required GTT alignment for an object * @obj: object to check @@ -2896,9 +2918,6 @@ int i915_gem_init_object(struct drm_gem_object *obj) void i915_gem_free_object(struct drm_gem_object *obj) { struct drm_device *dev = obj->dev; - struct drm_gem_mm *mm = dev->mm_private; - struct drm_map_list *list; - struct drm_map *map; struct drm_i915_gem_object *obj_priv = obj->driver_private; while (obj_priv->pin_count > 0) @@ -2909,19 +2928,7 @@ void i915_gem_free_object(struct drm_gem_object *obj) i915_gem_object_unbind(obj); - list = &obj->map_list; - drm_ht_remove_item(&mm->offset_hash, &list->hash); - - if (list->file_offset_node) { - drm_mm_put_block(list->file_offset_node); - list->file_offset_node = NULL; - } - - map = list->map; - if (map) { - drm_free(map, sizeof(*map), DRM_MEM_DRIVER); - list->map = NULL; - } + i915_gem_free_mmap_offset(obj); drm_free(obj_priv->page_cpu_valid, 1, DRM_MEM_DRIVER); drm_free(obj->driver_private, 1, DRM_MEM_DRIVER); diff --git a/include/drm/drmP.h b/include/drm/drmP.h index 8190b9bcc2d9..e5f4ae989abf 100644 --- a/include/drm/drmP.h +++ b/include/drm/drmP.h @@ -1321,6 +1321,8 @@ void drm_gem_object_free(struct kref *kref); struct drm_gem_object *drm_gem_object_alloc(struct drm_device *dev, size_t size); void drm_gem_object_handle_free(struct kref *kref); +void drm_gem_vm_open(struct vm_area_struct *vma); +void drm_gem_vm_close(struct vm_area_struct *vma); int drm_gem_mmap(struct file *filp, struct vm_area_struct *vma); static inline void -- cgit v1.2.3-70-g09d2 From 3ef0e5ba467366125f04b423f4638baca54a4fc1 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Fri, 20 Feb 2009 15:38:41 -0800 Subject: slab: introduce kzfree() kzfree() is a wrapper for kfree() that additionally zeroes the underlying memory before releasing it to the slab allocator. Currently there is code which memset()s the memory region of an object before releasing it back to the slab allocator to make sure security-sensitive data are really zeroed out after use. These callsites can then just use kzfree() which saves some code, makes users greppable and allows for a stupid destructor that isn't necessarily aware of the actual object size. Signed-off-by: Johannes Weiner Reviewed-by: Pekka Enberg Cc: Matt Mackall Acked-by: Christoph Lameter Cc: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/slab.h | 1 + mm/util.c | 20 ++++++++++++++++++++ 2 files changed, 21 insertions(+) (limited to 'include') diff --git a/include/linux/slab.h b/include/linux/slab.h index f96d13c281e8..24c5602bee99 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -127,6 +127,7 @@ int kmem_ptr_validate(struct kmem_cache *cachep, const void *ptr); void * __must_check __krealloc(const void *, size_t, gfp_t); void * __must_check krealloc(const void *, size_t, gfp_t); void kfree(const void *); +void kzfree(const void *); size_t ksize(const void *); /* diff --git a/mm/util.c b/mm/util.c index cb00b748ce47..37eaccdf3054 100644 --- a/mm/util.c +++ b/mm/util.c @@ -129,6 +129,26 @@ void *krealloc(const void *p, size_t new_size, gfp_t flags) } EXPORT_SYMBOL(krealloc); +/** + * kzfree - like kfree but zero memory + * @p: object to free memory of + * + * The memory of the object @p points to is zeroed before freed. + * If @p is %NULL, kzfree() does nothing. + */ +void kzfree(const void *p) +{ + size_t ks; + void *mem = (void *)p; + + if (unlikely(ZERO_OR_NULL_PTR(mem))) + return; + ks = ksize(mem); + memset(mem, 0, ks); + kfree(mem); +} +EXPORT_SYMBOL(kzfree); + /* * strndup_user - duplicate an existing string from user space * @s: The string to duplicate -- cgit v1.2.3-70-g09d2 From 01b24fee285b098c74318a8be801ef3711ec18d7 Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Fri, 20 Feb 2009 15:38:49 -0800 Subject: spi_bitbang: add more lowlevel function documentation This adds more documentation of the lowlevel API to avoid future bugs. Signed-off-by: Michael Buesch Acked-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/spi/spi_bitbang.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include') diff --git a/include/linux/spi/spi_bitbang.h b/include/linux/spi/spi_bitbang.h index bf8de281b4ed..eed4254bd503 100644 --- a/include/linux/spi/spi_bitbang.h +++ b/include/linux/spi/spi_bitbang.h @@ -83,6 +83,13 @@ extern int spi_bitbang_stop(struct spi_bitbang *spi); * int getmiso(struct spi_device *); * void spidelay(unsigned); * + * setsck()'s is_on parameter is a zero/nonzero boolean. + * + * setmosi()'s is_on parameter is a zero/nonzero boolean. + * + * getmiso() is required to return 0 or 1 only. Any other value is invalid + * and will result in improper operation. + * * A non-inlined routine would call bitbang_txrx_*() routines. The * main loop could easily compile down to a handful of instructions, * especially if the delay is a NOP (to run at peak speed). -- cgit v1.2.3-70-g09d2 From b6adea334c6c89d5e6c94f9196bbf3a279cb53bd Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 20 Feb 2009 15:38:52 -0800 Subject: 8250: fix boot hang with serial console when using with Serial Over Lan port Intel 8257x Ethernet boards have a feature called Serial Over Lan. This feature works by emulating a serial port, and it is detected by kernel as a normal 8250 port. However, this emulation is not perfect, as also noticed on changeset 7500b1f602aad75901774a67a687ee985d85893f. Before this patch, the kernel were trying to check if the serial TX is capable of work using IRQ's. This were done with a code similar this: serial_outp(up, UART_IER, UART_IER_THRI); lsr = serial_in(up, UART_LSR); iir = serial_in(up, UART_IIR); serial_outp(up, UART_IER, 0); if (lsr & UART_LSR_TEMT && iir & UART_IIR_NO_INT) up->bugs |= UART_BUG_TXEN; This works fine for other 8250 ports, but, on 8250-emulated SoL port, the chip is a little lazy to down UART_IIR_NO_INT at UART_IIR register. Due to that, UART_BUG_TXEN is sometimes enabled. However, as TX IRQ keeps working, and the TX polling is now enabled, the driver miss-interprets the IRQ received later, hanging up the machine until a key is pressed at the serial console. This is the 6 version of this patch. Previous versions were trying to introduce a large enough delay between serial_outp and serial_in(up, UART_IIR), but not taking forever. However, the needed delay couldn't be safely determined. At the experimental tests, a delay of 1us solves most of the cases, but still hangs sometimes. Increasing the delay to 5us was better, but still doesn't solve. A very high delay of 50 ms seemed to work every time. However, poking around with delays and pray for it to be enough doesn't seem to be a good approach, even for a quirk. So, instead of playing with random large arbitrary delays, let's just disable UART_BUG_TXEN for all SoL ports. [akpm@linux-foundation.org: fix warnings] Signed-off-by: Mauro Carvalho Chehab Cc: Alan Cox Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/8250.c | 15 +++++++++++++++ drivers/serial/8250_pci.c | 36 ++++++++++++++++++++++++++++++++++++ include/linux/pci_ids.h | 3 +++ include/linux/serial_core.h | 1 + 4 files changed, 55 insertions(+) (limited to 'include') diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c index 0d934bfbdd9b..b4b39811b445 100644 --- a/drivers/serial/8250.c +++ b/drivers/serial/8250.c @@ -2083,6 +2083,20 @@ static int serial8250_startup(struct uart_port *port) serial8250_set_mctrl(&up->port, up->port.mctrl); + /* Serial over Lan (SoL) hack: + Intel 8257x Gigabit ethernet chips have a + 16550 emulation, to be used for Serial Over Lan. + Those chips take a longer time than a normal + serial device to signalize that a transmission + data was queued. Due to that, the above test generally + fails. One solution would be to delay the reading of + iir. However, this is not reliable, since the timeout + is variable. So, let's just don't test if we receive + TX irq. This way, we'll never enable UART_BUG_TXEN. + */ + if (up->port.flags & UPF_NO_TXEN_TEST) + goto dont_test_tx_en; + /* * Do a quick test to see if we receive an * interrupt when we enable the TX irq. @@ -2102,6 +2116,7 @@ static int serial8250_startup(struct uart_port *port) up->bugs &= ~UART_BUG_TXEN; } +dont_test_tx_en: spin_unlock_irqrestore(&up->port.lock, flags); /* diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c index 536d8e510f66..533f82025adf 100644 --- a/drivers/serial/8250_pci.c +++ b/drivers/serial/8250_pci.c @@ -798,6 +798,21 @@ pci_default_setup(struct serial_private *priv, return setup_port(priv, port, bar, offset, board->reg_shift); } +static int skip_tx_en_setup(struct serial_private *priv, + const struct pciserial_board *board, + struct uart_port *port, int idx) +{ + port->flags |= UPF_NO_TXEN_TEST; + printk(KERN_DEBUG "serial8250: skipping TxEn test for device " + "[%04x:%04x] subsystem [%04x:%04x]\n", + priv->dev->vendor, + priv->dev->device, + priv->dev->subsystem_vendor, + priv->dev->subsystem_device); + + return pci_default_setup(priv, board, port, idx); +} + /* This should be in linux/pci_ids.h */ #define PCI_VENDOR_ID_SBSMODULARIO 0x124B #define PCI_SUBVENDOR_ID_SBSMODULARIO 0x124B @@ -864,6 +879,27 @@ static struct pci_serial_quirk pci_serial_quirks[] __refdata = { .init = pci_inteli960ni_init, .setup = pci_default_setup, }, + { + .vendor = PCI_VENDOR_ID_INTEL, + .device = PCI_DEVICE_ID_INTEL_8257X_SOL, + .subvendor = PCI_ANY_ID, + .subdevice = PCI_ANY_ID, + .setup = skip_tx_en_setup, + }, + { + .vendor = PCI_VENDOR_ID_INTEL, + .device = PCI_DEVICE_ID_INTEL_82573L_SOL, + .subvendor = PCI_ANY_ID, + .subdevice = PCI_ANY_ID, + .setup = skip_tx_en_setup, + }, + { + .vendor = PCI_VENDOR_ID_INTEL, + .device = PCI_DEVICE_ID_INTEL_82573E_SOL, + .subvendor = PCI_ANY_ID, + .subdevice = PCI_ANY_ID, + .setup = skip_tx_en_setup, + }, /* * ITE */ diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 114b8192eab9..aca8c458aa8a 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2323,6 +2323,9 @@ #define PCI_DEVICE_ID_INTEL_82378 0x0484 #define PCI_DEVICE_ID_INTEL_I960 0x0960 #define PCI_DEVICE_ID_INTEL_I960RM 0x0962 +#define PCI_DEVICE_ID_INTEL_8257X_SOL 0x1062 +#define PCI_DEVICE_ID_INTEL_82573E_SOL 0x1085 +#define PCI_DEVICE_ID_INTEL_82573L_SOL 0x108F #define PCI_DEVICE_ID_INTEL_82815_MC 0x1130 #define PCI_DEVICE_ID_INTEL_82815_CGC 0x1132 #define PCI_DEVICE_ID_INTEL_82092AA_0 0x1221 diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h index 90bbbf0b1161..df9245c7bd3b 100644 --- a/include/linux/serial_core.h +++ b/include/linux/serial_core.h @@ -296,6 +296,7 @@ struct uart_port { #define UPF_HARDPPS_CD ((__force upf_t) (1 << 11)) #define UPF_LOW_LATENCY ((__force upf_t) (1 << 13)) #define UPF_BUGGY_UART ((__force upf_t) (1 << 14)) +#define UPF_NO_TXEN_TEST ((__force upf_t) (1 << 15)) #define UPF_MAGIC_MULTIPLIER ((__force upf_t) (1 << 16)) #define UPF_CONS_FLOW ((__force upf_t) (1 << 23)) #define UPF_SHARE_IRQ ((__force upf_t) (1 << 24)) -- cgit v1.2.3-70-g09d2 From 216773a787c3c46ef26bf1742c1fdba37d26be45 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Sat, 14 Feb 2009 01:59:06 +0100 Subject: Consolidate driver_probe_done() loops into one place there's a few places that currently loop over driver_probe_done(), and I'm about to add another one. This patch abstracts it into a helper to reduce duplication. Signed-off-by: Arjan van de Ven Signed-off-by: Rafael J. Wysocki Cc: Len Brown Acked-by: Greg KH Signed-off-by: Linus Torvalds --- drivers/base/dd.c | 17 +++++++++++++++++ include/linux/device.h | 2 ++ init/do_mounts.c | 13 +++++++++---- init/do_mounts_md.c | 5 +++-- 4 files changed, 31 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/drivers/base/dd.c b/drivers/base/dd.c index 315bed8d5e7f..135231239103 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -18,9 +18,11 @@ */ #include +#include #include #include #include +#include #include "base.h" #include "power/power.h" @@ -167,6 +169,21 @@ int driver_probe_done(void) return 0; } +/** + * wait_for_device_probe + * Wait for device probing to be completed. + * + * Note: this function polls at 100 msec intervals. + */ +int wait_for_device_probe(void) +{ + /* wait for the known devices to complete their probing */ + while (driver_probe_done() != 0) + msleep(100); + async_synchronize_full(); + return 0; +} + /** * driver_probe_device - attempt to bind device & driver together * @drv: driver to bind a device to diff --git a/include/linux/device.h b/include/linux/device.h index 45e5b1921fbb..47f343c7bdda 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -147,6 +147,8 @@ extern void put_driver(struct device_driver *drv); extern struct device_driver *driver_find(const char *name, struct bus_type *bus); extern int driver_probe_done(void); +extern int wait_for_device_probe(void); + /* sysfs interface for exporting driver attributes */ diff --git a/init/do_mounts.c b/init/do_mounts.c index 708105e163df..8d4ff5afc1d8 100644 --- a/init/do_mounts.c +++ b/init/do_mounts.c @@ -370,10 +370,14 @@ void __init prepare_namespace(void) ssleep(root_delay); } - /* wait for the known devices to complete their probing */ - while (driver_probe_done() != 0) - msleep(100); - async_synchronize_full(); + /* + * wait for the known devices to complete their probing + * + * Note: this is a potential source of long boot delays. + * For example, it is not atypical to wait 5 seconds here + * for the touchpad of a laptop to initialize. + */ + wait_for_device_probe(); md_run_setup(); @@ -399,6 +403,7 @@ void __init prepare_namespace(void) while (driver_probe_done() != 0 || (ROOT_DEV = name_to_dev_t(saved_root_name)) == 0) msleep(100); + async_synchronize_full(); } is_floppy = MAJOR(ROOT_DEV) == FLOPPY_MAJOR; diff --git a/init/do_mounts_md.c b/init/do_mounts_md.c index ff95e3192884..9bdddbcb3d6a 100644 --- a/init/do_mounts_md.c +++ b/init/do_mounts_md.c @@ -281,8 +281,9 @@ static void __init autodetect_raid(void) */ printk(KERN_INFO "md: Waiting for all devices to be available before autodetect\n"); printk(KERN_INFO "md: If you don't use raid, use raid=noautodetect\n"); - while (driver_probe_done() < 0) - msleep(100); + + wait_for_device_probe(); + fd = sys_open("/dev/md0", 0, 0); if (fd >= 0) { sys_ioctl(fd, RAID_AUTORUN, raid_autopart); -- cgit v1.2.3-70-g09d2 From 770824bdc421ff58a64db608294323571c949f4c Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 22 Feb 2009 18:38:50 +0100 Subject: PM: Split up sysdev_[suspend|resume] from device_power_[down|up] Move the sysdev_suspend/resume from the callee to the callers, with no real change in semantics, so that we can rework the disabling of interrupts during suspend/hibernation. This is based on an earlier patch from Linus. Signed-off-by: Rafael J. Wysocki Signed-off-by: Linus Torvalds --- arch/x86/kernel/apm_32.c | 4 ++++ drivers/base/base.h | 2 -- drivers/base/power/main.c | 3 --- drivers/xen/manage.c | 8 ++++++++ include/linux/pm.h | 2 ++ kernel/kexec.c | 7 +++++++ kernel/power/disk.c | 11 +++++++++++ kernel/power/main.c | 8 ++++++-- 8 files changed, 38 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 98807bb095ad..266ec6c18b6c 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -1192,6 +1192,7 @@ static int suspend(int vetoable) device_suspend(PMSG_SUSPEND); local_irq_disable(); device_power_down(PMSG_SUSPEND); + sysdev_suspend(PMSG_SUSPEND); local_irq_enable(); @@ -1208,6 +1209,7 @@ static int suspend(int vetoable) if (err != APM_SUCCESS) apm_error("suspend", err); err = (err == APM_SUCCESS) ? 0 : -EIO; + sysdev_resume(); device_power_up(PMSG_RESUME); local_irq_enable(); device_resume(PMSG_RESUME); @@ -1228,6 +1230,7 @@ static void standby(void) local_irq_disable(); device_power_down(PMSG_SUSPEND); + sysdev_suspend(PMSG_SUSPEND); local_irq_enable(); err = set_system_power_state(APM_STATE_STANDBY); @@ -1235,6 +1238,7 @@ static void standby(void) apm_error("standby", err); local_irq_disable(); + sysdev_resume(); device_power_up(PMSG_RESUME); local_irq_enable(); } diff --git a/drivers/base/base.h b/drivers/base/base.h index 0a5f055dffba..9f50f1b545dc 100644 --- a/drivers/base/base.h +++ b/drivers/base/base.h @@ -88,8 +88,6 @@ extern void driver_detach(struct device_driver *drv); extern int driver_probe_device(struct device_driver *drv, struct device *dev); extern void sysdev_shutdown(void); -extern int sysdev_suspend(pm_message_t state); -extern int sysdev_resume(void); extern char *make_class_name(const char *name, struct kobject *kobj); diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 670c9d6c1407..2d14f4ae6c01 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -333,7 +333,6 @@ static void dpm_power_up(pm_message_t state) */ void device_power_up(pm_message_t state) { - sysdev_resume(); dpm_power_up(state); } EXPORT_SYMBOL_GPL(device_power_up); @@ -577,8 +576,6 @@ int device_power_down(pm_message_t state) } dev->power.status = DPM_OFF_IRQ; } - if (!error) - error = sysdev_suspend(state); if (error) dpm_power_up(resume_event(state)); return error; diff --git a/drivers/xen/manage.c b/drivers/xen/manage.c index 9b91617b9582..56892a142ee2 100644 --- a/drivers/xen/manage.c +++ b/drivers/xen/manage.c @@ -45,6 +45,13 @@ static int xen_suspend(void *data) err); return err; } + err = sysdev_suspend(PMSG_SUSPEND); + if (err) { + printk(KERN_ERR "xen_suspend: sysdev_suspend failed: %d\n", + err); + device_power_up(PMSG_RESUME); + return err; + } xen_mm_pin_all(); gnttab_suspend(); @@ -61,6 +68,7 @@ static int xen_suspend(void *data) gnttab_resume(); xen_mm_unpin_all(); + sysdev_resume(); device_power_up(PMSG_RESUME); if (!*cancelled) { diff --git a/include/linux/pm.h b/include/linux/pm.h index de2e0a8f6728..24ba5f67b3a3 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -381,10 +381,12 @@ struct dev_pm_info { #ifdef CONFIG_PM_SLEEP extern void device_pm_lock(void); +extern int sysdev_resume(void); extern void device_power_up(pm_message_t state); extern void device_resume(pm_message_t state); extern void device_pm_unlock(void); +extern int sysdev_suspend(pm_message_t state); extern int device_power_down(pm_message_t state); extern int device_suspend(pm_message_t state); extern int device_prepare_suspend(pm_message_t state); diff --git a/kernel/kexec.c b/kernel/kexec.c index 8a6d7b08864e..483899578259 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -1465,6 +1465,11 @@ int kernel_kexec(void) error = device_power_down(PMSG_FREEZE); if (error) goto Enable_irqs; + + /* Suspend system devices */ + error = sysdev_suspend(PMSG_FREEZE); + if (error) + goto Power_up_devices; } else #endif { @@ -1477,6 +1482,8 @@ int kernel_kexec(void) #ifdef CONFIG_KEXEC_JUMP if (kexec_image->preserve_context) { + sysdev_resume(); + Power_up_devices: device_power_up(PMSG_RESTORE); Enable_irqs: local_irq_enable(); diff --git a/kernel/power/disk.c b/kernel/power/disk.c index 7b40e94b1d42..4a4a206b1979 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -227,6 +227,12 @@ static int create_image(int platform_mode) "aborting hibernation\n"); goto Enable_irqs; } + sysdev_suspend(PMSG_FREEZE); + if (error) { + printk(KERN_ERR "PM: Some devices failed to power down, " + "aborting hibernation\n"); + goto Power_up_devices; + } if (hibernation_test(TEST_CORE)) goto Power_up; @@ -242,9 +248,11 @@ static int create_image(int platform_mode) if (!in_suspend) platform_leave(platform_mode); Power_up: + sysdev_resume(); /* NOTE: device_power_up() is just a resume() for devices * that suspended with irqs off ... no overall powerup. */ + Power_up_devices: device_power_up(in_suspend ? (error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE); Enable_irqs: @@ -335,6 +343,7 @@ static int resume_target_kernel(void) "aborting resume\n"); goto Enable_irqs; } + sysdev_suspend(PMSG_QUIESCE); /* We'll ignore saved state, but this gets preempt count (etc) right */ save_processor_state(); error = restore_highmem(); @@ -357,6 +366,7 @@ static int resume_target_kernel(void) swsusp_free(); restore_processor_state(); touch_softlockup_watchdog(); + sysdev_resume(); device_power_up(PMSG_RECOVER); Enable_irqs: local_irq_enable(); @@ -440,6 +450,7 @@ int hibernation_platform_enter(void) local_irq_disable(); error = device_power_down(PMSG_HIBERNATE); if (!error) { + sysdev_suspend(PMSG_HIBERNATE); hibernation_ops->enter(); /* We should never get here */ while (1); diff --git a/kernel/power/main.c b/kernel/power/main.c index b4d219016b6c..c9632f841f64 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -298,8 +298,12 @@ static int suspend_enter(suspend_state_t state) goto Done; } - if (!suspend_test(TEST_CORE)) - error = suspend_ops->enter(state); + error = sysdev_suspend(PMSG_SUSPEND); + if (!error) { + if (!suspend_test(TEST_CORE)) + error = suspend_ops->enter(state); + sysdev_resume(); + } device_power_up(PMSG_RESUME); Done: -- cgit v1.2.3-70-g09d2 From cd97f39b7cdf1c8a9c9f52865eec795b7f0c811d Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Tue, 24 Feb 2009 19:19:49 +0100 Subject: i2c-dev: Clarify the unit of ioctl I2C_TIMEOUT The unit in which user-space can set the bus timeout value is jiffies for historical reasons (back when HZ was always 100.) This is however not good because user-space doesn't know how long a jiffy lasts. The timeout value should instead be set in a fixed time unit. Given the original value of HZ, this unit should be 10 ms, for compatibility. Signed-off-by: Jean Delvare Acked-by: Wolfram Sang --- drivers/i2c/i2c-dev.c | 6 +++++- include/linux/i2c-dev.h | 2 +- include/linux/i2c.h | 2 +- 3 files changed, 7 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c index c171988a9f51..7e13d2df9af3 100644 --- a/drivers/i2c/i2c-dev.c +++ b/drivers/i2c/i2c-dev.c @@ -35,6 +35,7 @@ #include #include #include +#include #include static struct i2c_driver i2cdev_driver; @@ -422,7 +423,10 @@ static long i2cdev_ioctl(struct file *file, unsigned int cmd, unsigned long arg) client->adapter->retries = arg; break; case I2C_TIMEOUT: - client->adapter->timeout = arg; + /* For historical reasons, user-space sets the timeout + * value in units of 10 ms. + */ + client->adapter->timeout = msecs_to_jiffies(arg * 10); break; default: /* NOTE: returning a fault code here could cause trouble diff --git a/include/linux/i2c-dev.h b/include/linux/i2c-dev.h index 311315b56b61..fd53bfd26470 100644 --- a/include/linux/i2c-dev.h +++ b/include/linux/i2c-dev.h @@ -33,7 +33,7 @@ */ #define I2C_RETRIES 0x0701 /* number of times a device address should be polled when not acknowledging */ -#define I2C_TIMEOUT 0x0702 /* set timeout in jiffies - call with int */ +#define I2C_TIMEOUT 0x0702 /* set timeout in units of 10 ms */ /* NOTE: Slave address is 7 or 10 bits, but 10-bit addresses * are NOT supported! (due to code brokenness) diff --git a/include/linux/i2c.h b/include/linux/i2c.h index fcfbfea3af72..c86c3b07604c 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -361,7 +361,7 @@ struct i2c_adapter { struct mutex bus_lock; struct mutex clist_lock; - int timeout; + int timeout; /* in jiffies */ int retries; struct device dev; /* the adapter device */ -- cgit v1.2.3-70-g09d2 From fe56cf45f951b3810313584605c1d8a4f20b33a4 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Mon, 23 Feb 2009 15:36:41 -0800 Subject: drm: Fix ordering of bit fields in EDID structure leading huge vsync values. Signed-off-by: Jesse Barnes Reviewed-by: Eric Anholt Signed-off-by: Dave Airlie --- include/drm/drm_edid.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/drm/drm_edid.h b/include/drm/drm_edid.h index c707c15f5164..ff8d27af4786 100644 --- a/include/drm/drm_edid.h +++ b/include/drm/drm_edid.h @@ -58,10 +58,10 @@ struct detailed_pixel_timing { u8 hsync_pulse_width_lo; u8 vsync_pulse_width_lo:4; u8 vsync_offset_lo:4; - u8 hsync_pulse_width_hi:2; - u8 hsync_offset_hi:2; u8 vsync_pulse_width_hi:2; u8 vsync_offset_hi:2; + u8 hsync_pulse_width_hi:2; + u8 hsync_offset_hi:2; u8 width_mm_lo; u8 height_mm_lo; u8 height_mm_hi:4; -- cgit v1.2.3-70-g09d2 From 7bec756c74b1a5079d5074144bb77a6b3e7d7783 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Mon, 23 Feb 2009 16:09:34 -0800 Subject: drm: disable encoders before re-routing them MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In some cases we may receive a mode config that has a different CRTC<->encoder map that the current configuration. In that case, we need to disable any re-routed encoders before setting the mode, otherwise they may not pick up the new CRTC (if the output types are incompatible for example). Tested-by: Kristian Høgsberg Signed-off-by: Jesse Barnes Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_crtc_helper.c | 76 ++++++++++++++++++++++++++++++++++++--- include/drm/drm_crtc_helper.h | 1 + 2 files changed, 73 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c index 733028b4d45e..1c3a8c557140 100644 --- a/drivers/gpu/drm/drm_crtc_helper.c +++ b/drivers/gpu/drm/drm_crtc_helper.c @@ -452,6 +452,59 @@ static void drm_setup_crtcs(struct drm_device *dev) kfree(modes); kfree(enabled); } + +/** + * drm_encoder_crtc_ok - can a given crtc drive a given encoder? + * @encoder: encoder to test + * @crtc: crtc to test + * + * Return false if @encoder can't be driven by @crtc, true otherwise. + */ +static bool drm_encoder_crtc_ok(struct drm_encoder *encoder, + struct drm_crtc *crtc) +{ + struct drm_device *dev; + struct drm_crtc *tmp; + int crtc_mask = 1; + + WARN(!crtc, "checking null crtc?"); + + dev = crtc->dev; + + list_for_each_entry(tmp, &dev->mode_config.crtc_list, head) { + if (tmp == crtc) + break; + crtc_mask <<= 1; + } + + if (encoder->possible_crtcs & crtc_mask) + return true; + return false; +} + +/* + * Check the CRTC we're going to map each output to vs. its current + * CRTC. If they don't match, we have to disable the output and the CRTC + * since the driver will have to re-route things. + */ +static void +drm_crtc_prepare_encoders(struct drm_device *dev) +{ + struct drm_encoder_helper_funcs *encoder_funcs; + struct drm_encoder *encoder; + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { + encoder_funcs = encoder->helper_private; + /* Disable unused encoders */ + if (encoder->crtc == NULL) + (*encoder_funcs->dpms)(encoder, DRM_MODE_DPMS_OFF); + /* Disable encoders whose CRTC is about to change */ + if (encoder_funcs->get_crtc && + encoder->crtc != (*encoder_funcs->get_crtc)(encoder)) + (*encoder_funcs->dpms)(encoder, DRM_MODE_DPMS_OFF); + } +} + /** * drm_crtc_set_mode - set a mode * @crtc: CRTC to program @@ -547,6 +600,8 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc, encoder_funcs->prepare(encoder); } + drm_crtc_prepare_encoders(dev); + crtc_funcs->prepare(crtc); /* Set up the DPLL and any encoders state that needs to adjust or depend @@ -617,7 +672,7 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set) struct drm_device *dev; struct drm_crtc **save_crtcs, *new_crtc; struct drm_encoder **save_encoders, *new_encoder; - struct drm_framebuffer *old_fb; + struct drm_framebuffer *old_fb = NULL; bool save_enabled; bool mode_changed = false; bool fb_changed = false; @@ -668,9 +723,10 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set) * and then just flip_or_move it */ if (set->crtc->fb != set->fb) { /* If we have no fb then treat it as a full mode set */ - if (set->crtc->fb == NULL) + if (set->crtc->fb == NULL) { + DRM_DEBUG("crtc has no fb, full mode set\n"); mode_changed = true; - else if ((set->fb->bits_per_pixel != + } else if ((set->fb->bits_per_pixel != set->crtc->fb->bits_per_pixel) || set->fb->depth != set->crtc->fb->depth) fb_changed = true; @@ -682,7 +738,7 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set) fb_changed = true; if (set->mode && !drm_mode_equal(set->mode, &set->crtc->mode)) { - DRM_DEBUG("modes are different\n"); + DRM_DEBUG("modes are different, full mode set\n"); drm_mode_debug_printmodeline(&set->crtc->mode); drm_mode_debug_printmodeline(set->mode); mode_changed = true; @@ -708,6 +764,7 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set) } if (new_encoder != connector->encoder) { + DRM_DEBUG("encoder changed, full mode switch\n"); mode_changed = true; connector->encoder = new_encoder; } @@ -734,10 +791,20 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set) if (set->connectors[ro] == connector) new_crtc = set->crtc; } + + /* Make sure the new CRTC will work with the encoder */ + if (new_crtc && + !drm_encoder_crtc_ok(connector->encoder, new_crtc)) { + ret = -EINVAL; + goto fail_set_mode; + } if (new_crtc != connector->encoder->crtc) { + DRM_DEBUG("crtc changed, full mode switch\n"); mode_changed = true; connector->encoder->crtc = new_crtc; } + DRM_DEBUG("setting connector %d crtc to %p\n", + connector->base.id, new_crtc); } /* mode_set_base is not a required function */ @@ -781,6 +848,7 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set) fail_set_mode: set->crtc->enabled = save_enabled; + set->crtc->fb = old_fb; count = 0; list_for_each_entry(connector, &dev->mode_config.connector_list, head) { if (!connector->encoder) diff --git a/include/drm/drm_crtc_helper.h b/include/drm/drm_crtc_helper.h index 0b0d236c2154..c7d4b2e606a5 100644 --- a/include/drm/drm_crtc_helper.h +++ b/include/drm/drm_crtc_helper.h @@ -76,6 +76,7 @@ struct drm_encoder_helper_funcs { void (*mode_set)(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); + struct drm_crtc *(*get_crtc)(struct drm_encoder *encoder); /* detect for DAC style encoders */ enum drm_connector_status (*detect)(struct drm_encoder *encoder, struct drm_connector *connector); -- cgit v1.2.3-70-g09d2 From 4ab0d47d0ab311eb181532c1ecb6d02905685071 Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Tue, 24 Feb 2009 17:35:12 -0800 Subject: gpu/drm, x86, PAT: io_mapping_create_wc and resource_size_t io_mapping_create_wc should take a resource_size_t parameter in place of unsigned long. With unsigned long, there will be no way to map greater than 4GB address in i386/32 bit. On x86, greater than 4GB addresses cannot be mapped on i386 without PAE. Return error for such a case. Patch also adds a structure for io_mapping, that saves the base, size and type on HAVE_ATOMIC_IOMAP archs, that can be used to verify the offset on io_mapping_map calls. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Suresh Siddha Cc: Dave Airlie Cc: Jesse Barnes Cc: Eric Anholt Cc: Keith Packard Signed-off-by: Ingo Molnar --- arch/x86/include/asm/iomap.h | 3 +++ arch/x86/mm/iomap_32.c | 18 +++++++++++++++++ include/linux/io-mapping.h | 46 +++++++++++++++++++++++++++++++++----------- 3 files changed, 56 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/arch/x86/include/asm/iomap.h b/arch/x86/include/asm/iomap.h index c1f06289b14b..86af26091d6c 100644 --- a/arch/x86/include/asm/iomap.h +++ b/arch/x86/include/asm/iomap.h @@ -23,6 +23,9 @@ #include #include +int +is_io_mapping_possible(resource_size_t base, unsigned long size); + void * iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot); diff --git a/arch/x86/mm/iomap_32.c b/arch/x86/mm/iomap_32.c index ca53224fc56c..6c2b1af16926 100644 --- a/arch/x86/mm/iomap_32.c +++ b/arch/x86/mm/iomap_32.c @@ -20,6 +20,24 @@ #include #include +#ifdef CONFIG_X86_PAE +int +is_io_mapping_possible(resource_size_t base, unsigned long size) +{ + return 1; +} +#else +int +is_io_mapping_possible(resource_size_t base, unsigned long size) +{ + /* There is no way to map greater than 1 << 32 address without PAE */ + if (base + size > 0x100000000ULL) + return 0; + + return 1; +} +#endif + /* Map 'pfn' using fixed map 'type' and protections 'prot' */ void * diff --git a/include/linux/io-mapping.h b/include/linux/io-mapping.h index 82df31726a54..cbc2f0cd631b 100644 --- a/include/linux/io-mapping.h +++ b/include/linux/io-mapping.h @@ -30,11 +30,14 @@ * See Documentation/io_mapping.txt */ -/* this struct isn't actually defined anywhere */ -struct io_mapping; - #ifdef CONFIG_HAVE_ATOMIC_IOMAP +struct io_mapping { + resource_size_t base; + unsigned long size; + pgprot_t prot; +}; + /* * For small address space machines, mapping large objects * into the kernel virtual space isn't practical. Where @@ -43,23 +46,40 @@ struct io_mapping; */ static inline struct io_mapping * -io_mapping_create_wc(unsigned long base, unsigned long size) +io_mapping_create_wc(resource_size_t base, unsigned long size) { - return (struct io_mapping *) base; + struct io_mapping *iomap; + + if (!is_io_mapping_possible(base, size)) + return NULL; + + iomap = kmalloc(sizeof(*iomap), GFP_KERNEL); + if (!iomap) + return NULL; + + iomap->base = base; + iomap->size = size; + iomap->prot = pgprot_writecombine(__pgprot(__PAGE_KERNEL)); + return iomap; } static inline void io_mapping_free(struct io_mapping *mapping) { + kfree(mapping); } /* Atomic map/unmap */ static inline void * io_mapping_map_atomic_wc(struct io_mapping *mapping, unsigned long offset) { - offset += (unsigned long) mapping; - return iomap_atomic_prot_pfn(offset >> PAGE_SHIFT, KM_USER0, - __pgprot(__PAGE_KERNEL_WC)); + resource_size_t phys_addr; + unsigned long pfn; + + BUG_ON(offset >= mapping->size); + phys_addr = mapping->base + offset; + pfn = (unsigned long) (phys_addr >> PAGE_SHIFT); + return iomap_atomic_prot_pfn(pfn, KM_USER0, mapping->prot); } static inline void @@ -71,8 +91,9 @@ io_mapping_unmap_atomic(void *vaddr) static inline void * io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) { - offset += (unsigned long) mapping; - return ioremap_wc(offset, PAGE_SIZE); + BUG_ON(offset >= mapping->size); + resource_size_t phys_addr = mapping->base + offset; + return ioremap_wc(phys_addr, PAGE_SIZE); } static inline void @@ -83,9 +104,12 @@ io_mapping_unmap(void *vaddr) #else +/* this struct isn't actually defined anywhere */ +struct io_mapping; + /* Create the io_mapping object*/ static inline struct io_mapping * -io_mapping_create_wc(unsigned long base, unsigned long size) +io_mapping_create_wc(resource_size_t base, unsigned long size) { return (struct io_mapping *) ioremap_wc(base, size); } -- cgit v1.2.3-70-g09d2 From 8fed43684174b68f04d01d1210fd00536af790df Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Wed, 25 Feb 2009 20:28:24 +0100 Subject: ide: fix refcounting in device drivers During host driver module removal del_gendisk() results in a final put on drive->gendev and freeing the drive by drive_release_dev(). Convert device drivers from using struct kref to use struct device so device driver's object holds reference on ->gendev and prevents drive from prematurely going away. Also fix ->remove methods to not erroneously drop reference on a host driver by using only put_device() instead of ide*_put(). Reported-by: Stanislaw Gruszka Tested-by: Stanislaw Gruszka Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ide-cd.c | 27 ++++++++++++++++++--------- drivers/ide/ide-cd.h | 2 +- drivers/ide/ide-gd.c | 26 +++++++++++++++++--------- drivers/ide/ide-gd.h | 2 +- drivers/ide/ide-tape.c | 29 +++++++++++++++++++---------- include/linux/ide.h | 2 +- 6 files changed, 57 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c index 690475b834de..ddfbea41d296 100644 --- a/drivers/ide/ide-cd.c +++ b/drivers/ide/ide-cd.c @@ -55,7 +55,7 @@ static DEFINE_MUTEX(idecd_ref_mutex); -static void ide_cd_release(struct kref *); +static void ide_cd_release(struct device *); static struct cdrom_info *ide_cd_get(struct gendisk *disk) { @@ -67,7 +67,7 @@ static struct cdrom_info *ide_cd_get(struct gendisk *disk) if (ide_device_get(cd->drive)) cd = NULL; else - kref_get(&cd->kref); + get_device(&cd->dev); } mutex_unlock(&idecd_ref_mutex); @@ -79,7 +79,7 @@ static void ide_cd_put(struct cdrom_info *cd) ide_drive_t *drive = cd->drive; mutex_lock(&idecd_ref_mutex); - kref_put(&cd->kref, ide_cd_release); + put_device(&cd->dev); ide_device_put(drive); mutex_unlock(&idecd_ref_mutex); } @@ -1798,15 +1798,17 @@ static void ide_cd_remove(ide_drive_t *drive) ide_debug_log(IDE_DBG_FUNC, "Call %s\n", __func__); ide_proc_unregister_driver(drive, info->driver); - + device_del(&info->dev); del_gendisk(info->disk); - ide_cd_put(info); + mutex_lock(&idecd_ref_mutex); + put_device(&info->dev); + mutex_unlock(&idecd_ref_mutex); } -static void ide_cd_release(struct kref *kref) +static void ide_cd_release(struct device *dev) { - struct cdrom_info *info = to_ide_drv(kref, cdrom_info); + struct cdrom_info *info = to_ide_drv(dev, cdrom_info); struct cdrom_device_info *devinfo = &info->devinfo; ide_drive_t *drive = info->drive; struct gendisk *g = info->disk; @@ -2005,7 +2007,12 @@ static int ide_cd_probe(ide_drive_t *drive) ide_init_disk(g, drive); - kref_init(&info->kref); + info->dev.parent = &drive->gendev; + info->dev.release = ide_cd_release; + dev_set_name(&info->dev, dev_name(&drive->gendev)); + + if (device_register(&info->dev)) + goto out_free_disk; info->drive = drive; info->driver = &ide_cdrom_driver; @@ -2019,7 +2026,7 @@ static int ide_cd_probe(ide_drive_t *drive) g->driverfs_dev = &drive->gendev; g->flags = GENHD_FL_CD | GENHD_FL_REMOVABLE; if (ide_cdrom_setup(drive)) { - ide_cd_release(&info->kref); + put_device(&info->dev); goto failed; } @@ -2029,6 +2036,8 @@ static int ide_cd_probe(ide_drive_t *drive) add_disk(g); return 0; +out_free_disk: + put_disk(g); out_free_cd: kfree(info); failed: diff --git a/drivers/ide/ide-cd.h b/drivers/ide/ide-cd.h index ac40d6cb90a2..c878bfcf1116 100644 --- a/drivers/ide/ide-cd.h +++ b/drivers/ide/ide-cd.h @@ -80,7 +80,7 @@ struct cdrom_info { ide_drive_t *drive; struct ide_driver *driver; struct gendisk *disk; - struct kref kref; + struct device dev; /* Buffer for table of contents. NULL if we haven't allocated a TOC buffer for this device yet. */ diff --git a/drivers/ide/ide-gd.c b/drivers/ide/ide-gd.c index 7857b209c6df..047109419902 100644 --- a/drivers/ide/ide-gd.c +++ b/drivers/ide/ide-gd.c @@ -25,7 +25,7 @@ module_param(debug_mask, ulong, 0644); static DEFINE_MUTEX(ide_disk_ref_mutex); -static void ide_disk_release(struct kref *); +static void ide_disk_release(struct device *); static struct ide_disk_obj *ide_disk_get(struct gendisk *disk) { @@ -37,7 +37,7 @@ static struct ide_disk_obj *ide_disk_get(struct gendisk *disk) if (ide_device_get(idkp->drive)) idkp = NULL; else - kref_get(&idkp->kref); + get_device(&idkp->dev); } mutex_unlock(&ide_disk_ref_mutex); return idkp; @@ -48,7 +48,7 @@ static void ide_disk_put(struct ide_disk_obj *idkp) ide_drive_t *drive = idkp->drive; mutex_lock(&ide_disk_ref_mutex); - kref_put(&idkp->kref, ide_disk_release); + put_device(&idkp->dev); ide_device_put(drive); mutex_unlock(&ide_disk_ref_mutex); } @@ -66,17 +66,18 @@ static void ide_gd_remove(ide_drive_t *drive) struct gendisk *g = idkp->disk; ide_proc_unregister_driver(drive, idkp->driver); - + device_del(&idkp->dev); del_gendisk(g); - drive->disk_ops->flush(drive); - ide_disk_put(idkp); + mutex_lock(&ide_disk_ref_mutex); + put_device(&idkp->dev); + mutex_unlock(&ide_disk_ref_mutex); } -static void ide_disk_release(struct kref *kref) +static void ide_disk_release(struct device *dev) { - struct ide_disk_obj *idkp = to_ide_drv(kref, ide_disk_obj); + struct ide_disk_obj *idkp = to_ide_drv(dev, ide_disk_obj); ide_drive_t *drive = idkp->drive; struct gendisk *g = idkp->disk; @@ -348,7 +349,12 @@ static int ide_gd_probe(ide_drive_t *drive) ide_init_disk(g, drive); - kref_init(&idkp->kref); + idkp->dev.parent = &drive->gendev; + idkp->dev.release = ide_disk_release; + dev_set_name(&idkp->dev, dev_name(&drive->gendev)); + + if (device_register(&idkp->dev)) + goto out_free_disk; idkp->drive = drive; idkp->driver = &ide_gd_driver; @@ -373,6 +379,8 @@ static int ide_gd_probe(ide_drive_t *drive) add_disk(g); return 0; +out_free_disk: + put_disk(g); out_free_idkp: kfree(idkp); failed: diff --git a/drivers/ide/ide-gd.h b/drivers/ide/ide-gd.h index a86779f0756b..b604bdd318a1 100644 --- a/drivers/ide/ide-gd.h +++ b/drivers/ide/ide-gd.h @@ -17,7 +17,7 @@ struct ide_disk_obj { ide_drive_t *drive; struct ide_driver *driver; struct gendisk *disk; - struct kref kref; + struct device dev; unsigned int openers; /* protected by BKL for now */ /* Last failed packet command */ diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c index d7ecd3c79757..bb450a7608c2 100644 --- a/drivers/ide/ide-tape.c +++ b/drivers/ide/ide-tape.c @@ -169,7 +169,7 @@ typedef struct ide_tape_obj { ide_drive_t *drive; struct ide_driver *driver; struct gendisk *disk; - struct kref kref; + struct device dev; /* * failed_pc points to the last failed packet command, or contains @@ -267,7 +267,7 @@ static DEFINE_MUTEX(idetape_ref_mutex); static struct class *idetape_sysfs_class; -static void ide_tape_release(struct kref *); +static void ide_tape_release(struct device *); static struct ide_tape_obj *ide_tape_get(struct gendisk *disk) { @@ -279,7 +279,7 @@ static struct ide_tape_obj *ide_tape_get(struct gendisk *disk) if (ide_device_get(tape->drive)) tape = NULL; else - kref_get(&tape->kref); + get_device(&tape->dev); } mutex_unlock(&idetape_ref_mutex); return tape; @@ -290,7 +290,7 @@ static void ide_tape_put(struct ide_tape_obj *tape) ide_drive_t *drive = tape->drive; mutex_lock(&idetape_ref_mutex); - kref_put(&tape->kref, ide_tape_release); + put_device(&tape->dev); ide_device_put(drive); mutex_unlock(&idetape_ref_mutex); } @@ -308,7 +308,7 @@ static struct ide_tape_obj *ide_tape_chrdev_get(unsigned int i) mutex_lock(&idetape_ref_mutex); tape = idetape_devs[i]; if (tape) - kref_get(&tape->kref); + get_device(&tape->dev); mutex_unlock(&idetape_ref_mutex); return tape; } @@ -2256,15 +2256,17 @@ static void ide_tape_remove(ide_drive_t *drive) idetape_tape_t *tape = drive->driver_data; ide_proc_unregister_driver(drive, tape->driver); - + device_del(&tape->dev); ide_unregister_region(tape->disk); - ide_tape_put(tape); + mutex_lock(&idetape_ref_mutex); + put_device(&tape->dev); + mutex_unlock(&idetape_ref_mutex); } -static void ide_tape_release(struct kref *kref) +static void ide_tape_release(struct device *dev) { - struct ide_tape_obj *tape = to_ide_drv(kref, ide_tape_obj); + struct ide_tape_obj *tape = to_ide_drv(dev, ide_tape_obj); ide_drive_t *drive = tape->drive; struct gendisk *g = tape->disk; @@ -2407,7 +2409,12 @@ static int ide_tape_probe(ide_drive_t *drive) ide_init_disk(g, drive); - kref_init(&tape->kref); + tape->dev.parent = &drive->gendev; + tape->dev.release = ide_tape_release; + dev_set_name(&tape->dev, dev_name(&drive->gendev)); + + if (device_register(&tape->dev)) + goto out_free_disk; tape->drive = drive; tape->driver = &idetape_driver; @@ -2436,6 +2443,8 @@ static int ide_tape_probe(ide_drive_t *drive) return 0; +out_free_disk: + put_disk(g); out_free_tape: kfree(tape); failed: diff --git a/include/linux/ide.h b/include/linux/ide.h index 194da5a4b0d6..fe235b65207e 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -663,7 +663,7 @@ typedef struct ide_drive_s ide_drive_t; #define to_ide_device(dev) container_of(dev, ide_drive_t, gendev) #define to_ide_drv(obj, cont_type) \ - container_of(obj, struct cont_type, kref) + container_of(obj, struct cont_type, dev) #define ide_drv_g(disk, cont_type) \ container_of((disk)->private_data, struct cont_type, driver) -- cgit v1.2.3-70-g09d2 From a682604838763981613e42015cd0e39f2989d6bb Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 25 Feb 2009 18:03:42 -0800 Subject: rcu: Teach RCU that idle task is not quiscent state at boot This patch fixes a bug located by Vegard Nossum with the aid of kmemcheck, updated based on review comments from Nick Piggin, Ingo Molnar, and Andrew Morton. And cleans up the variable-name and function-name language. ;-) The boot CPU runs in the context of its idle thread during boot-up. During this time, idle_cpu(0) will always return nonzero, which will fool Classic and Hierarchical RCU into deciding that a large chunk of the boot-up sequence is a big long quiescent state. This in turn causes RCU to prematurely end grace periods during this time. This patch changes the rcutree.c and rcuclassic.c rcu_check_callbacks() function to ignore the idle task as a quiescent state until the system has started up the scheduler in rest_init(), introducing a new non-API function rcu_idle_now_means_idle() to inform RCU of this transition. RCU maintains an internal rcu_idle_cpu_truthful variable to track this state, which is then used by rcu_check_callback() to determine if it should believe idle_cpu(). Because this patch has the effect of disallowing RCU grace periods during long stretches of the boot-up sequence, this patch also introduces Josh Triplett's UP-only optimization that makes synchronize_rcu() be a no-op if num_online_cpus() returns 1. This allows boot-time code that calls synchronize_rcu() to proceed normally. Note, however, that RCU callbacks registered by call_rcu() will likely queue up until later in the boot sequence. Although rcuclassic and rcutree can also use this same optimization after boot completes, rcupreempt must restrict its use of this optimization to the portion of the boot sequence before the scheduler starts up, given that an rcupreempt RCU read-side critical section may be preeempted. In addition, this patch takes Nick Piggin's suggestion to make the system_state global variable be __read_mostly. Changes since v4: o Changes the name of the introduced function and variable to be less emotional. ;-) Changes since v3: o WARN_ON(nr_context_switches() > 0) to verify that RCU switches out of boot-time mode before the first context switch, as suggested by Nick Piggin. Changes since v2: o Created rcu_blocking_is_gp() internal-to-RCU API that determines whether a call to synchronize_rcu() is itself a grace period. o The definition of rcu_blocking_is_gp() for rcuclassic and rcutree checks to see if but a single CPU is online. o The definition of rcu_blocking_is_gp() for rcupreempt checks to see both if but a single CPU is online and if the system is still in early boot. This allows rcupreempt to again work correctly if running on a single CPU after booting is complete. o Added check to rcupreempt's synchronize_sched() for there being but one online CPU. Tested all three variants both SMP and !SMP, booted fine, passed a short rcutorture test on both x86 and Power. Located-by: Vegard Nossum Tested-by: Vegard Nossum Tested-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Signed-off-by: Ingo Molnar --- include/linux/rcuclassic.h | 6 ++++++ include/linux/rcupdate.h | 4 ++++ include/linux/rcupreempt.h | 15 +++++++++++++++ include/linux/rcutree.h | 6 ++++++ init/main.c | 3 ++- kernel/rcuclassic.c | 4 ++-- kernel/rcupdate.c | 12 ++++++++++++ kernel/rcupreempt.c | 3 +++ kernel/rcutree.c | 4 ++-- 9 files changed, 52 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/rcuclassic.h b/include/linux/rcuclassic.h index f3f697df1d71..80044a4f3ab9 100644 --- a/include/linux/rcuclassic.h +++ b/include/linux/rcuclassic.h @@ -181,4 +181,10 @@ extern long rcu_batches_completed_bh(void); #define rcu_enter_nohz() do { } while (0) #define rcu_exit_nohz() do { } while (0) +/* A context switch is a grace period for rcuclassic. */ +static inline int rcu_blocking_is_gp(void) +{ + return num_online_cpus() == 1; +} + #endif /* __LINUX_RCUCLASSIC_H */ diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 921340a7b71c..528343e6da51 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -52,6 +52,9 @@ struct rcu_head { void (*func)(struct rcu_head *head); }; +/* Internal to kernel, but needed by rcupreempt.h. */ +extern int rcu_scheduler_active; + #if defined(CONFIG_CLASSIC_RCU) #include #elif defined(CONFIG_TREE_RCU) @@ -265,6 +268,7 @@ extern void rcu_barrier_sched(void); /* Internal to kernel */ extern void rcu_init(void); +extern void rcu_scheduler_starting(void); extern int rcu_needs_cpu(int cpu); #endif /* __LINUX_RCUPDATE_H */ diff --git a/include/linux/rcupreempt.h b/include/linux/rcupreempt.h index 3e05c09b54a2..74304b4538d8 100644 --- a/include/linux/rcupreempt.h +++ b/include/linux/rcupreempt.h @@ -142,4 +142,19 @@ static inline void rcu_exit_nohz(void) #define rcu_exit_nohz() do { } while (0) #endif /* CONFIG_NO_HZ */ +/* + * A context switch is a grace period for rcupreempt synchronize_rcu() + * only during early boot, before the scheduler has been initialized. + * So, how the heck do we get a context switch? Well, if the caller + * invokes synchronize_rcu(), they are willing to accept a context + * switch, so we simply pretend that one happened. + * + * After boot, there might be a blocked or preempted task in an RCU + * read-side critical section, so we cannot then take the fastpath. + */ +static inline int rcu_blocking_is_gp(void) +{ + return num_online_cpus() == 1 && !rcu_scheduler_active; +} + #endif /* __LINUX_RCUPREEMPT_H */ diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h index d4368b7975c3..a722fb67bb2d 100644 --- a/include/linux/rcutree.h +++ b/include/linux/rcutree.h @@ -326,4 +326,10 @@ static inline void rcu_exit_nohz(void) } #endif /* CONFIG_NO_HZ */ +/* A context switch is a grace period for rcutree. */ +static inline int rcu_blocking_is_gp(void) +{ + return num_online_cpus() == 1; +} + #endif /* __LINUX_RCUTREE_H */ diff --git a/init/main.c b/init/main.c index 844209453c02..83697e160b3a 100644 --- a/init/main.c +++ b/init/main.c @@ -97,7 +97,7 @@ static inline void mark_rodata_ro(void) { } extern void tc_init(void); #endif -enum system_states system_state; +enum system_states system_state __read_mostly; EXPORT_SYMBOL(system_state); /* @@ -463,6 +463,7 @@ static noinline void __init_refok rest_init(void) * at least once to get things moving: */ init_idle_bootup_task(current); + rcu_scheduler_starting(); preempt_enable_no_resched(); schedule(); preempt_disable(); diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c index bd5a9003497c..654c640a6b9c 100644 --- a/kernel/rcuclassic.c +++ b/kernel/rcuclassic.c @@ -679,8 +679,8 @@ int rcu_needs_cpu(int cpu) void rcu_check_callbacks(int cpu, int user) { if (user || - (idle_cpu(cpu) && !in_softirq() && - hardirq_count() <= (1 << HARDIRQ_SHIFT))) { + (idle_cpu(cpu) && rcu_scheduler_active && + !in_softirq() && hardirq_count() <= (1 << HARDIRQ_SHIFT))) { /* * Get here if this CPU took its interrupt from user diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index d92a76a881aa..cae8a059cf47 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -44,6 +44,7 @@ #include #include #include +#include enum rcu_barrier { RCU_BARRIER_STD, @@ -55,6 +56,7 @@ static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL}; static atomic_t rcu_barrier_cpu_count; static DEFINE_MUTEX(rcu_barrier_mutex); static struct completion rcu_barrier_completion; +int rcu_scheduler_active __read_mostly; /* * Awaken the corresponding synchronize_rcu() instance now that a @@ -80,6 +82,10 @@ void wakeme_after_rcu(struct rcu_head *head) void synchronize_rcu(void) { struct rcu_synchronize rcu; + + if (rcu_blocking_is_gp()) + return; + init_completion(&rcu.completion); /* Will wake me after RCU finished. */ call_rcu(&rcu.head, wakeme_after_rcu); @@ -175,3 +181,9 @@ void __init rcu_init(void) __rcu_init(); } +void rcu_scheduler_starting(void) +{ + WARN_ON(num_online_cpus() != 1); + WARN_ON(nr_context_switches() > 0); + rcu_scheduler_active = 1; +} diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c index 33cfc50781f9..5d59e850fb71 100644 --- a/kernel/rcupreempt.c +++ b/kernel/rcupreempt.c @@ -1181,6 +1181,9 @@ void __synchronize_sched(void) { struct rcu_synchronize rcu; + if (num_online_cpus() == 1) + return; /* blocking is gp if only one CPU! */ + init_completion(&rcu.completion); /* Will wake me after RCU finished. */ call_rcu_sched(&rcu.head, wakeme_after_rcu); diff --git a/kernel/rcutree.c b/kernel/rcutree.c index b2fd602a6f6f..97ce31579ec0 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -948,8 +948,8 @@ static void rcu_do_batch(struct rcu_data *rdp) void rcu_check_callbacks(int cpu, int user) { if (user || - (idle_cpu(cpu) && !in_softirq() && - hardirq_count() <= (1 << HARDIRQ_SHIFT))) { + (idle_cpu(cpu) && rcu_scheduler_active && + !in_softirq() && hardirq_count() <= (1 << HARDIRQ_SHIFT))) { /* * Get here if this CPU took its interrupt from user -- cgit v1.2.3-70-g09d2 From 1e42807918d17e8c93bf14fbb74be84b141334c1 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 23 Feb 2009 09:03:10 +0100 Subject: block: reduce stack footprint of blk_recount_segments() blk_recalc_rq_segments() requires a request structure passed in, which we don't have from blk_recount_segments(). So the latter allocates one on the stack, using > 400 bytes of stack for that. This can cause us to spill over one page of stack from ext4 at least: 0) 4560 400 blk_recount_segments+0x43/0x62 1) 4160 32 bio_phys_segments+0x1c/0x24 2) 4128 32 blk_rq_bio_prep+0x2a/0xf9 3) 4096 32 init_request_from_bio+0xf9/0xfe 4) 4064 112 __make_request+0x33c/0x3f6 5) 3952 144 generic_make_request+0x2d1/0x321 6) 3808 64 submit_bio+0xb9/0xc3 7) 3744 48 submit_bh+0xea/0x10e 8) 3696 368 ext4_mb_init_cache+0x257/0xa6a [ext4] 9) 3328 288 ext4_mb_regular_allocator+0x421/0xcd9 [ext4] 10) 3040 160 ext4_mb_new_blocks+0x211/0x4b4 [ext4] 11) 2880 336 ext4_ext_get_blocks+0xb61/0xd45 [ext4] 12) 2544 96 ext4_get_blocks_wrap+0xf2/0x200 [ext4] 13) 2448 80 ext4_da_get_block_write+0x6e/0x16b [ext4] 14) 2368 352 mpage_da_map_blocks+0x7e/0x4b3 [ext4] 15) 2016 352 ext4_da_writepages+0x2ce/0x43c [ext4] 16) 1664 32 do_writepages+0x2d/0x3c 17) 1632 144 __writeback_single_inode+0x162/0x2cd 18) 1488 96 generic_sync_sb_inodes+0x1e3/0x32b 19) 1392 16 sync_sb_inodes+0xe/0x10 20) 1376 48 writeback_inodes+0x69/0xb3 21) 1328 208 balance_dirty_pages_ratelimited_nr+0x187/0x2f9 22) 1120 224 generic_file_buffered_write+0x1d4/0x2c4 23) 896 176 __generic_file_aio_write_nolock+0x35f/0x393 24) 720 80 generic_file_aio_write+0x6c/0xc8 25) 640 80 ext4_file_write+0xa9/0x137 [ext4] 26) 560 320 do_sync_write+0xf0/0x137 27) 240 48 vfs_write+0xb3/0x13c 28) 192 64 sys_write+0x4c/0x74 29) 128 128 system_call_fastpath+0x16/0x1b Split the segment counting out into a __blk_recalc_rq_segments() helper to avoid allocating an onstack request just for checking the physical segment count. Signed-off-by: Jens Axboe --- block/blk-merge.c | 94 ++++++++++++++++++++++++++++---------------------- include/linux/blkdev.h | 2 ++ 2 files changed, 55 insertions(+), 41 deletions(-) (limited to 'include') diff --git a/block/blk-merge.c b/block/blk-merge.c index b92f5b0866b0..a104593e70c3 100644 --- a/block/blk-merge.c +++ b/block/blk-merge.c @@ -38,72 +38,84 @@ void blk_recalc_rq_sectors(struct request *rq, int nsect) } } -void blk_recalc_rq_segments(struct request *rq) +static unsigned int __blk_recalc_rq_segments(struct request_queue *q, + struct bio *bio, + unsigned int *seg_size_ptr) { - int nr_phys_segs; unsigned int phys_size; struct bio_vec *bv, *bvprv = NULL; - int seg_size; - int cluster; - struct req_iterator iter; - int high, highprv = 1; - struct request_queue *q = rq->q; + int cluster, i, high, highprv = 1; + unsigned int seg_size, nr_phys_segs; + struct bio *fbio; - if (!rq->bio) - return; + if (!bio) + return 0; + fbio = bio; cluster = test_bit(QUEUE_FLAG_CLUSTER, &q->queue_flags); seg_size = 0; phys_size = nr_phys_segs = 0; - rq_for_each_segment(bv, rq, iter) { - /* - * the trick here is making sure that a high page is never - * considered part of another segment, since that might - * change with the bounce page. - */ - high = page_to_pfn(bv->bv_page) > q->bounce_pfn; - if (high || highprv) - goto new_segment; - if (cluster) { - if (seg_size + bv->bv_len > q->max_segment_size) - goto new_segment; - if (!BIOVEC_PHYS_MERGEABLE(bvprv, bv)) - goto new_segment; - if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bv)) + for_each_bio(bio) { + bio_for_each_segment(bv, bio, i) { + /* + * the trick here is making sure that a high page is + * never considered part of another segment, since that + * might change with the bounce page. + */ + high = page_to_pfn(bv->bv_page) > q->bounce_pfn; + if (high || highprv) goto new_segment; + if (cluster) { + if (seg_size + bv->bv_len > q->max_segment_size) + goto new_segment; + if (!BIOVEC_PHYS_MERGEABLE(bvprv, bv)) + goto new_segment; + if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bv)) + goto new_segment; + + seg_size += bv->bv_len; + bvprv = bv; + continue; + } +new_segment: + if (nr_phys_segs == 1 && seg_size > + fbio->bi_seg_front_size) + fbio->bi_seg_front_size = seg_size; - seg_size += bv->bv_len; + nr_phys_segs++; bvprv = bv; - continue; + seg_size = bv->bv_len; + highprv = high; } -new_segment: - if (nr_phys_segs == 1 && seg_size > rq->bio->bi_seg_front_size) - rq->bio->bi_seg_front_size = seg_size; - - nr_phys_segs++; - bvprv = bv; - seg_size = bv->bv_len; - highprv = high; } - if (nr_phys_segs == 1 && seg_size > rq->bio->bi_seg_front_size) + if (seg_size_ptr) + *seg_size_ptr = seg_size; + + return nr_phys_segs; +} + +void blk_recalc_rq_segments(struct request *rq) +{ + unsigned int seg_size = 0, phys_segs; + + phys_segs = __blk_recalc_rq_segments(rq->q, rq->bio, &seg_size); + + if (phys_segs == 1 && seg_size > rq->bio->bi_seg_front_size) rq->bio->bi_seg_front_size = seg_size; if (seg_size > rq->biotail->bi_seg_back_size) rq->biotail->bi_seg_back_size = seg_size; - rq->nr_phys_segments = nr_phys_segs; + rq->nr_phys_segments = phys_segs; } void blk_recount_segments(struct request_queue *q, struct bio *bio) { - struct request rq; struct bio *nxt = bio->bi_next; - rq.q = q; - rq.bio = rq.biotail = bio; + bio->bi_next = NULL; - blk_recalc_rq_segments(&rq); + bio->bi_phys_segments = __blk_recalc_rq_segments(q, bio, NULL); bio->bi_next = nxt; - bio->bi_phys_segments = rq.nr_phys_segments; bio->bi_flags |= (1 << BIO_SEG_VALID); } EXPORT_SYMBOL(blk_recount_segments); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index dcaa0fd84b02..465d6babc847 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -708,6 +708,8 @@ struct req_iterator { }; /* This should not be used directly - use rq_for_each_segment */ +#define for_each_bio(_bio) \ + for (; _bio; _bio = _bio->bi_next) #define __rq_for_each_bio(_bio, rq) \ if ((rq->bio)) \ for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next) -- cgit v1.2.3-70-g09d2 From 54e991242850edc8c53f71fa5aa3ba7a93ce38f5 Mon Sep 17 00:00:00 2001 From: Dhaval Giani Date: Fri, 27 Feb 2009 15:13:54 +0530 Subject: sched: don't allow setuid to succeed if the user does not have rt bandwidth Impact: fix hung task with certain (non-default) rt-limit settings Corey Hickey reported that on using setuid to change the uid of a rt process, the process would be unkillable and not be running. This is because there was no rt runtime for that user group. Add in a check to see if a user can attach an rt task to its task group. On failure, return EINVAL, which is also returned in CONFIG_CGROUP_SCHED. Reported-by: Corey Hickey Signed-off-by: Dhaval Giani Acked-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- include/linux/sched.h | 4 ++++ kernel/sched.c | 13 +++++++++++-- kernel/sys.c | 31 ++++++++++++++++++++----------- kernel/user.c | 18 ++++++++++++++++++ 4 files changed, 53 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 8981e52c714f..8c216e057c94 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2291,9 +2291,13 @@ extern long sched_group_rt_runtime(struct task_group *tg); extern int sched_group_set_rt_period(struct task_group *tg, long rt_period_us); extern long sched_group_rt_period(struct task_group *tg); +extern int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk); #endif #endif +extern int task_can_switch_user(struct user_struct *up, + struct task_struct *tsk); + #ifdef CONFIG_TASK_XACCT static inline void add_rchar(struct task_struct *tsk, ssize_t amt) { diff --git a/kernel/sched.c b/kernel/sched.c index c3baa9653d1d..8e2558c2ba67 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -9224,6 +9224,16 @@ static int sched_rt_global_constraints(void) return ret; } + +int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk) +{ + /* Don't accept realtime tasks when there is no way for them to run */ + if (rt_task(tsk) && tg->rt_bandwidth.rt_runtime == 0) + return 0; + + return 1; +} + #else /* !CONFIG_RT_GROUP_SCHED */ static int sched_rt_global_constraints(void) { @@ -9317,8 +9327,7 @@ cpu_cgroup_can_attach(struct cgroup_subsys *ss, struct cgroup *cgrp, struct task_struct *tsk) { #ifdef CONFIG_RT_GROUP_SCHED - /* Don't accept realtime tasks when there is no way for them to run */ - if (rt_task(tsk) && cgroup_tg(cgrp)->rt_bandwidth.rt_runtime == 0) + if (!sched_rt_can_attach(cgroup_tg(cgrp), tsk)) return -EINVAL; #else /* We don't support RT-tasks being in separate groups */ diff --git a/kernel/sys.c b/kernel/sys.c index f145c415bc16..37f458e6882a 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -559,7 +559,7 @@ error: abort_creds(new); return retval; } - + /* * change the user struct in a credentials set to match the new UID */ @@ -571,6 +571,11 @@ static int set_user(struct cred *new) if (!new_user) return -EAGAIN; + if (!task_can_switch_user(new_user, current)) { + free_uid(new_user); + return -EINVAL; + } + if (atomic_read(&new_user->processes) >= current->signal->rlim[RLIMIT_NPROC].rlim_cur && new_user != INIT_USER) { @@ -631,10 +636,11 @@ SYSCALL_DEFINE2(setreuid, uid_t, ruid, uid_t, euid) goto error; } - retval = -EAGAIN; - if (new->uid != old->uid && set_user(new) < 0) - goto error; - + if (new->uid != old->uid) { + retval = set_user(new); + if (retval < 0) + goto error; + } if (ruid != (uid_t) -1 || (euid != (uid_t) -1 && euid != old->uid)) new->suid = new->euid; @@ -680,9 +686,10 @@ SYSCALL_DEFINE1(setuid, uid_t, uid) retval = -EPERM; if (capable(CAP_SETUID)) { new->suid = new->uid = uid; - if (uid != old->uid && set_user(new) < 0) { - retval = -EAGAIN; - goto error; + if (uid != old->uid) { + retval = set_user(new); + if (retval < 0) + goto error; } } else if (uid != old->uid && uid != new->suid) { goto error; @@ -734,11 +741,13 @@ SYSCALL_DEFINE3(setresuid, uid_t, ruid, uid_t, euid, uid_t, suid) goto error; } - retval = -EAGAIN; if (ruid != (uid_t) -1) { new->uid = ruid; - if (ruid != old->uid && set_user(new) < 0) - goto error; + if (ruid != old->uid) { + retval = set_user(new); + if (retval < 0) + goto error; + } } if (euid != (uid_t) -1) new->euid = euid; diff --git a/kernel/user.c b/kernel/user.c index 3551ac742395..6a9b696128c8 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -362,6 +362,24 @@ static void free_user(struct user_struct *up, unsigned long flags) #endif +#if defined(CONFIG_RT_GROUP_SCHED) && defined(CONFIG_USER_SCHED) +/* + * We need to check if a setuid can take place. This function should be called + * before successfully completing the setuid. + */ +int task_can_switch_user(struct user_struct *up, struct task_struct *tsk) +{ + + return sched_rt_can_attach(up->tg, tsk); + +} +#else +int task_can_switch_user(struct user_struct *up, struct task_struct *tsk) +{ + return 1; +} +#endif + /* * Locate the user_struct for the passed UID. If found, take a ref on it. The * caller must undo that ref with free_uid(). -- cgit v1.2.3-70-g09d2 From 5170836679185357dc1b7660bad13287b39e1e33 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 27 Feb 2009 14:03:03 -0800 Subject: Fix recursive lock in free_uid()/free_user_ns() free_uid() and free_user_ns() are corecursive when CONFIG_USER_SCHED=n, but free_user_ns() is called from free_uid() by way of uid_hash_remove(), which requires uidhash_lock to be held. free_user_ns() then calls free_uid() to complete the destruction. Fix this by deferring the destruction of the user_namespace. Signed-off-by: David Howells Acked-by: Serge Hallyn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/user_namespace.h | 1 + kernel/user_namespace.c | 21 +++++++++++++++++---- 2 files changed, 18 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/user_namespace.h b/include/linux/user_namespace.h index 315bcd375224..cc4f45361dbb 100644 --- a/include/linux/user_namespace.h +++ b/include/linux/user_namespace.h @@ -13,6 +13,7 @@ struct user_namespace { struct kref kref; struct hlist_head uidhash_table[UIDHASH_SZ]; struct user_struct *creator; + struct work_struct destroyer; }; extern struct user_namespace init_user_ns; diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c index 79084311ee57..076c7c8215b0 100644 --- a/kernel/user_namespace.c +++ b/kernel/user_namespace.c @@ -60,12 +60,25 @@ int create_user_ns(struct cred *new) return 0; } -void free_user_ns(struct kref *kref) +/* + * Deferred destructor for a user namespace. This is required because + * free_user_ns() may be called with uidhash_lock held, but we need to call + * back to free_uid() which will want to take the lock again. + */ +static void free_user_ns_work(struct work_struct *work) { - struct user_namespace *ns; - - ns = container_of(kref, struct user_namespace, kref); + struct user_namespace *ns = + container_of(work, struct user_namespace, destroyer); free_uid(ns->creator); kfree(ns); } + +void free_user_ns(struct kref *kref) +{ + struct user_namespace *ns = + container_of(kref, struct user_namespace, kref); + + INIT_WORK(&ns->destroyer, free_user_ns_work); + schedule_work(&ns->destroyer); +} EXPORT_SYMBOL(free_user_ns); -- cgit v1.2.3-70-g09d2 From 5c2522218059ca1f4174a568923b988aad3ddfda Mon Sep 17 00:00:00 2001 From: Chris Leech Date: Fri, 27 Feb 2009 10:01:36 +0000 Subject: net headers: cleanup dcbnl.h 1) add an include for 2) change dcbmsg.dcb_family from unsigned char to __u8 to be more consistent with use of kernel types Signed-off-by: Chris Leech Acked-by: Sam Ravnborg Signed-off-by: David S. Miller --- include/linux/dcbnl.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/dcbnl.h b/include/linux/dcbnl.h index b0ef274e0031..7d2e10006188 100644 --- a/include/linux/dcbnl.h +++ b/include/linux/dcbnl.h @@ -20,10 +20,12 @@ #ifndef __LINUX_DCBNL_H__ #define __LINUX_DCBNL_H__ +#include + #define DCB_PROTO_VERSION 1 struct dcbmsg { - unsigned char dcb_family; + __u8 dcb_family; __u8 cmd; __u16 dcb_pad; }; -- cgit v1.2.3-70-g09d2 From 709ab3261e3ed789c0bb31c6ab53c9eccb276522 Mon Sep 17 00:00:00 2001 From: Chris Leech Date: Fri, 27 Feb 2009 10:01:42 +0000 Subject: net headers: export dcbnl.h The DCB netlink interface is required for building the userspace tools available at e1000.sourceforge.net Signed-off-by: Chris Leech Signed-off-by: David S. Miller --- include/linux/Kbuild | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/Kbuild b/include/linux/Kbuild index b97cdc516a8f..106c3ba50844 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -52,6 +52,7 @@ header-y += const.h header-y += cgroupstats.h header-y += cramfs_fs.h header-y += cycx_cfm.h +header-y += dcbnl.h header-y += dlmconstants.h header-y += dlm_device.h header-y += dlm_netlink.h -- cgit v1.2.3-70-g09d2 From 5ce04e3de8c36ba37c56e94e3c4dc7973c7f546c Mon Sep 17 00:00:00 2001 From: "Pallipadi, Venkatesh" Date: Sun, 1 Mar 2009 08:53:27 -0800 Subject: fix warning in io_mapping_map_wc() Signed-off-by: Venkatesh Pallipadi Signed-off-by: Ingo Molnar --- include/linux/io-mapping.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/io-mapping.h b/include/linux/io-mapping.h index cbc2f0cd631b..0adb0f91568c 100644 --- a/include/linux/io-mapping.h +++ b/include/linux/io-mapping.h @@ -91,8 +91,11 @@ io_mapping_unmap_atomic(void *vaddr) static inline void * io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) { + resource_size_t phys_addr; + BUG_ON(offset >= mapping->size); - resource_size_t phys_addr = mapping->base + offset; + phys_addr = mapping->base + offset; + return ioremap_wc(phys_addr, PAGE_SIZE); } -- cgit v1.2.3-70-g09d2 From 17edde520927070a6bf14a6a75027c0b843443e5 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sun, 22 Feb 2009 00:11:09 -0800 Subject: netns: Remove net_alive It turns out that net_alive is unnecessary, and the original problem that led to it being added was simply that the icmp code thought it was a network device and wound up being unable to handle packets while there were still packets in the network namespace. Now that icmp and tcp have been fixed to properly register themselves this problem is no longer present and we have a stronger guarantee that packets will not arrive in a network namespace then that provided by net_alive in netif_receive_skb. So remove net_alive allowing packet reception run a little faster. Additionally document the strong reason why network namespace cleanup is safe so that if something happens again someone else will have a chance of figuring it out. Signed-off-by: Eric W. Biederman Signed-off-by: David S. Miller --- include/net/net_namespace.h | 27 +++++++++++++++++---------- net/core/dev.c | 6 ------ net/core/net_namespace.c | 3 --- 3 files changed, 17 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index 6fc13d905c5f..ded434b032a4 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -109,11 +109,6 @@ extern struct list_head net_namespace_list; #ifdef CONFIG_NET_NS extern void __put_net(struct net *net); -static inline int net_alive(struct net *net) -{ - return net && atomic_read(&net->count); -} - static inline struct net *get_net(struct net *net) { atomic_inc(&net->count); @@ -145,11 +140,6 @@ int net_eq(const struct net *net1, const struct net *net2) } #else -static inline int net_alive(struct net *net) -{ - return 1; -} - static inline struct net *get_net(struct net *net) { return net; @@ -234,6 +224,23 @@ struct pernet_operations { void (*exit)(struct net *net); }; +/* + * Use these carefully. If you implement a network device and it + * needs per network namespace operations use device pernet operations, + * otherwise use pernet subsys operations. + * + * This is critically important. Most of the network code cleanup + * runs with the assumption that dev_remove_pack has been called so no + * new packets will arrive during and after the cleanup functions have + * been called. dev_remove_pack is not per namespace so instead the + * guarantee of no more packets arriving in a network namespace is + * provided by ensuring that all network devices and all sockets have + * left the network namespace before the cleanup methods are called. + * + * For the longest time the ipv4 icmp code was registered as a pernet + * device which caused kernel oops, and panics during network + * namespace cleanup. So please don't get this wrong. + */ extern int register_pernet_subsys(struct pernet_operations *); extern void unregister_pernet_subsys(struct pernet_operations *); extern int register_pernet_gen_subsys(int *id, struct pernet_operations *); diff --git a/net/core/dev.c b/net/core/dev.c index 72b0d26fd46d..9e4afe650e7a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2267,12 +2267,6 @@ int netif_receive_skb(struct sk_buff *skb) rcu_read_lock(); - /* Don't receive packets in an exiting network namespace */ - if (!net_alive(dev_net(skb->dev))) { - kfree_skb(skb); - goto out; - } - #ifdef CONFIG_NET_CLS_ACT if (skb->tc_verd & TC_NCLS) { skb->tc_verd = CLR_TC_NCLS(skb->tc_verd); diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 2adb1a7d361f..e3bebd36f053 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -157,9 +157,6 @@ static void cleanup_net(struct work_struct *work) struct pernet_operations *ops; struct net *net; - /* Be very certain incoming network packets will not find us */ - rcu_barrier(); - net = container_of(work, struct net, work); mutex_lock(&net_mutex); -- cgit v1.2.3-70-g09d2 From 9d40bbda599def1e1d155d7f7dca14fe8744bd2b Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 4 Mar 2009 23:46:25 -0800 Subject: vlan: Fix vlan-in-vlan crashes. As analyzed by Patrick McHardy, vlan needs to reset it's netdev_ops pointer in it's ->init() function but this leaves the compat method pointers stale. Add a netdev_resync_ops() and call it from the vlan code. Any other driver which changes ->netdev_ops after register_netdevice() will need to call this new function after doing so too. With help from Patrick McHardy. Tested-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 + net/8021q/vlan_dev.c | 3 ++- net/core/dev.c | 56 ++++++++++++++++++++++++++++------------------- 3 files changed, 37 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ec54785d34f9..659366734f3f 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1079,6 +1079,7 @@ extern void synchronize_net(void); extern int register_netdevice_notifier(struct notifier_block *nb); extern int unregister_netdevice_notifier(struct notifier_block *nb); extern int init_dummy_netdev(struct net_device *dev); +extern void netdev_resync_ops(struct net_device *dev); extern int call_netdevice_notifiers(unsigned long val, struct net_device *dev); extern struct net_device *dev_get_by_index(struct net *net, int ifindex); diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 4a19acd3a32b..1b34135cf990 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -553,7 +553,7 @@ static int vlan_dev_neigh_setup(struct net_device *dev, struct neigh_parms *pa) int err = 0; if (netif_device_present(real_dev) && ops->ndo_neigh_setup) - err = ops->ndo_neigh_setup(dev, pa); + err = ops->ndo_neigh_setup(real_dev, pa); return err; } @@ -639,6 +639,7 @@ static int vlan_dev_init(struct net_device *dev) dev->hard_header_len = real_dev->hard_header_len + VLAN_HLEN; dev->netdev_ops = &vlan_netdev_ops; } + netdev_resync_ops(dev); if (is_vlan_dev(real_dev)) subclass = 1; diff --git a/net/core/dev.c b/net/core/dev.c index 2dd484ed3dbb..f1129706ce7b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4282,6 +4282,39 @@ unsigned long netdev_fix_features(unsigned long features, const char *name) } EXPORT_SYMBOL(netdev_fix_features); +/* Some devices need to (re-)set their netdev_ops inside + * ->init() or similar. If that happens, we have to setup + * the compat pointers again. + */ +void netdev_resync_ops(struct net_device *dev) +{ +#ifdef CONFIG_COMPAT_NET_DEV_OPS + const struct net_device_ops *ops = dev->netdev_ops; + + dev->init = ops->ndo_init; + dev->uninit = ops->ndo_uninit; + dev->open = ops->ndo_open; + dev->change_rx_flags = ops->ndo_change_rx_flags; + dev->set_rx_mode = ops->ndo_set_rx_mode; + dev->set_multicast_list = ops->ndo_set_multicast_list; + dev->set_mac_address = ops->ndo_set_mac_address; + dev->validate_addr = ops->ndo_validate_addr; + dev->do_ioctl = ops->ndo_do_ioctl; + dev->set_config = ops->ndo_set_config; + dev->change_mtu = ops->ndo_change_mtu; + dev->neigh_setup = ops->ndo_neigh_setup; + dev->tx_timeout = ops->ndo_tx_timeout; + dev->get_stats = ops->ndo_get_stats; + dev->vlan_rx_register = ops->ndo_vlan_rx_register; + dev->vlan_rx_add_vid = ops->ndo_vlan_rx_add_vid; + dev->vlan_rx_kill_vid = ops->ndo_vlan_rx_kill_vid; +#ifdef CONFIG_NET_POLL_CONTROLLER + dev->poll_controller = ops->ndo_poll_controller; +#endif +#endif +} +EXPORT_SYMBOL(netdev_resync_ops); + /** * register_netdevice - register a network device * @dev: device to register @@ -4326,28 +4359,7 @@ int register_netdevice(struct net_device *dev) * This is temporary until all network devices are converted. */ if (dev->netdev_ops) { - const struct net_device_ops *ops = dev->netdev_ops; - - dev->init = ops->ndo_init; - dev->uninit = ops->ndo_uninit; - dev->open = ops->ndo_open; - dev->change_rx_flags = ops->ndo_change_rx_flags; - dev->set_rx_mode = ops->ndo_set_rx_mode; - dev->set_multicast_list = ops->ndo_set_multicast_list; - dev->set_mac_address = ops->ndo_set_mac_address; - dev->validate_addr = ops->ndo_validate_addr; - dev->do_ioctl = ops->ndo_do_ioctl; - dev->set_config = ops->ndo_set_config; - dev->change_mtu = ops->ndo_change_mtu; - dev->neigh_setup = ops->ndo_neigh_setup; - dev->tx_timeout = ops->ndo_tx_timeout; - dev->get_stats = ops->ndo_get_stats; - dev->vlan_rx_register = ops->ndo_vlan_rx_register; - dev->vlan_rx_add_vid = ops->ndo_vlan_rx_add_vid; - dev->vlan_rx_kill_vid = ops->ndo_vlan_rx_kill_vid; -#ifdef CONFIG_NET_POLL_CONTROLLER - dev->poll_controller = ops->ndo_poll_controller; -#endif + netdev_resync_ops(dev); } else { char drivername[64]; pr_info("%s (%s): not using net_device_ops yet\n", -- cgit v1.2.3-70-g09d2 From e7d3ef13d52a126438f687a1a32da65ff926ed57 Mon Sep 17 00:00:00 2001 From: Stuart Hayes Date: Wed, 4 Mar 2009 11:59:46 -0800 Subject: libata: change drive ready wait after hard reset to 5s This fixes problems during resume with drives that take longer than 1s to be ready. The ATA-6 spec appears to allow 5 seconds for a drive to be ready. On one affected system, this patch changes "PM: resume devices took..." message from 17 seconds to 4 seconds, and gets rid of a lot of ugly timeout/error messages. Without this patch, the libata code moves on after 1s, tries to send a soft reset (which the drive doesn't see because it isn't ready) which also times out, then an IDENTIFY command is sent to the drive which times out, and finally the error handler will try to send another hard reset which will finally get things working. Signed-off-by: Stuart Hayes Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- include/linux/libata.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/libata.h b/include/linux/libata.h index 5d87bc09a1f5..dd818c7decd7 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -275,7 +275,7 @@ enum { * advised to wait only for the following duration before * doing SRST. */ - ATA_TMOUT_PMP_SRST_WAIT = 1000, + ATA_TMOUT_PMP_SRST_WAIT = 5000, /* ATA bus states */ BUS_UNKNOWN = 0, -- cgit v1.2.3-70-g09d2 From 5825627c9463581fd9e70f8285685889ae5bb9bb Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Fri, 27 Feb 2009 17:35:43 +0900 Subject: libata: fix dma_unmap_sg misuse libata passes the returned value of dma_map_sg() to dma_unmap_sg(),which is the misuse of dma_unmap_sg(). DMA-mapping.txt says: To unmap a scatterlist, just call: pci_unmap_sg(pdev, sglist, nents, direction); Again, make sure DMA activity has already finished. PLEASE NOTE: The 'nents' argument to the pci_unmap_sg call must be the _same_ one you passed into the pci_map_sg call, it should _NOT_ be the 'count' value _returned_ from the pci_map_sg call. Signed-off-by: FUJITA Tomonori Acked-by: Bartlomiej Zolnierkiewicz Acked-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 4 ++-- include/linux/libata.h | 1 + 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 9fbf0595f3d4..5e324cea3019 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4612,7 +4612,7 @@ void ata_sg_clean(struct ata_queued_cmd *qc) VPRINTK("unmapping %u sg elements\n", qc->n_elem); if (qc->n_elem) - dma_unmap_sg(ap->dev, sg, qc->n_elem, dir); + dma_unmap_sg(ap->dev, sg, qc->orig_n_elem, dir); qc->flags &= ~ATA_QCFLAG_DMAMAP; qc->sg = NULL; @@ -4727,7 +4727,7 @@ static int ata_sg_setup(struct ata_queued_cmd *qc) return -1; DPRINTK("%d sg elements mapped\n", n_elem); - + qc->orig_n_elem = qc->n_elem; qc->n_elem = n_elem; qc->flags |= ATA_QCFLAG_DMAMAP; diff --git a/include/linux/libata.h b/include/linux/libata.h index dd818c7decd7..fbf064e13ad5 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -530,6 +530,7 @@ struct ata_queued_cmd { unsigned long flags; /* ATA_QCFLAG_xxx */ unsigned int tag; unsigned int n_elem; + unsigned int orig_n_elem; int dma_dir; -- cgit v1.2.3-70-g09d2 From 84bda12af31f930e4200c5244aa111de2485d7b0 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 2 Mar 2009 18:53:26 +0900 Subject: libata: align ap->sector_buf ap->sector_buf is used as DMA target and should at least be aligned on cacheline. This caused problems on some embedded machines. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- include/linux/libata.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/libata.h b/include/linux/libata.h index fbf064e13ad5..dc18b87ed722 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -751,7 +751,8 @@ struct ata_port { acpi_handle acpi_handle; struct ata_acpi_gtm __acpi_init_gtm; /* use ata_acpi_init_gtm() */ #endif - u8 sector_buf[ATA_SECT_SIZE]; /* owned by EH */ + /* owned by EH */ + u8 sector_buf[ATA_SECT_SIZE] ____cacheline_aligned; }; /* The following initializer overrides a method to NULL whether one of -- cgit v1.2.3-70-g09d2 From 849d7130001ab740a5a4778a561049841fdd77c9 Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Thu, 5 Mar 2009 16:10:57 +0100 Subject: ide: allow to wrap interrupt handler Signed-off-by: Stanislaw Gruszka Cc: Andrew Victor [bart: minor checkpatch.pl / CodingStyle fixups] Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ide-io.c | 1 + drivers/ide/ide-probe.c | 7 ++++++- include/linux/ide.h | 1 + 3 files changed, 8 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c index 9ff90cb1dbf1..a9a6c208288a 100644 --- a/drivers/ide/ide-io.c +++ b/drivers/ide/ide-io.c @@ -1162,6 +1162,7 @@ out_early: return irq_ret; } +EXPORT_SYMBOL_GPL(ide_intr); /** * ide_do_drive_cmd - issue IDE special command diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c index ce0818a993f6..ee8e3e7cad51 100644 --- a/drivers/ide/ide-probe.c +++ b/drivers/ide/ide-probe.c @@ -950,6 +950,7 @@ static int ide_port_setup_devices(ide_hwif_t *hwif) static int init_irq (ide_hwif_t *hwif) { struct ide_io_ports *io_ports = &hwif->io_ports; + irq_handler_t irq_handler; int sa = 0; mutex_lock(&ide_cfg_mtx); @@ -959,6 +960,10 @@ static int init_irq (ide_hwif_t *hwif) hwif->timer.function = &ide_timer_expiry; hwif->timer.data = (unsigned long)hwif; + irq_handler = hwif->host->irq_handler; + if (irq_handler == NULL) + irq_handler = ide_intr; + #if defined(__mc68000__) sa = IRQF_SHARED; #endif /* __mc68000__ */ @@ -969,7 +974,7 @@ static int init_irq (ide_hwif_t *hwif) if (io_ports->ctl_addr) hwif->tp_ops->set_irq(hwif, 1); - if (request_irq(hwif->irq, &ide_intr, sa, hwif->name, hwif)) + if (request_irq(hwif->irq, irq_handler, sa, hwif->name, hwif)) goto out_up; if (!hwif->rqsize) { diff --git a/include/linux/ide.h b/include/linux/ide.h index fe235b65207e..e0cedfe9fad4 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -866,6 +866,7 @@ struct ide_host { unsigned int n_ports; struct device *dev[2]; unsigned int (*init_chipset)(struct pci_dev *); + irq_handler_t irq_handler; unsigned long host_flags; void *host_priv; ide_hwif_t *cur_port; /* for hosts requiring serialization */ -- cgit v1.2.3-70-g09d2 From ebcad5aaea26da3cb2ca90b7f31a67a027eb60db Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Thu, 5 Mar 2009 16:10:59 +0100 Subject: remove stale comment from HDIO_GET_IDENTITY returns 256 words currently. Noticed by Norman Diamond. Signed-off-by: Bartlomiej Zolnierkiewicz --- include/linux/hdreg.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include') diff --git a/include/linux/hdreg.h b/include/linux/hdreg.h index c37e9241fae7..ed21bd3dbd25 100644 --- a/include/linux/hdreg.h +++ b/include/linux/hdreg.h @@ -511,7 +511,6 @@ struct hd_driveid { unsigned short words69_70[2]; /* reserved words 69-70 * future command overlap and queuing */ - /* HDIO_GET_IDENTITY currently returns only words 0 through 70 */ unsigned short words71_74[4]; /* reserved words 71-74 * for IDENTIFY PACKET DEVICE command */ -- cgit v1.2.3-70-g09d2 From d42ad15b759d05a87f22b484af63987eff38ea88 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Thu, 5 Mar 2009 17:20:55 +0100 Subject: ata: add CFA specific identify data words Declare CFA specific identify data words 162 and 163 for future use. Signed-off-by: Sergei Shtylyov Signed-off-by: Jeff Garzik [bart: update patch summary/description] Signed-off-by: Bartlomiej Zolnierkiewicz --- include/linux/ata.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/ata.h b/include/linux/ata.h index 08a86d5cdf1b..9a061accd8b8 100644 --- a/include/linux/ata.h +++ b/include/linux/ata.h @@ -89,6 +89,8 @@ enum { ATA_ID_DLF = 128, ATA_ID_CSFO = 129, ATA_ID_CFA_POWER = 160, + ATA_ID_CFA_KEY_MGMT = 162, + ATA_ID_CFA_MODES = 163, ATA_ID_ROT_SPEED = 217, ATA_ID_PIO4 = (1 << 1), -- cgit v1.2.3-70-g09d2 From 1f6ff364ceda516f88351a8ab640e656beed0b26 Mon Sep 17 00:00:00 2001 From: Abhijeet Joglekar Date: Fri, 27 Feb 2009 10:54:35 -0800 Subject: [SCSI] libfc: Pass lport in exch_mgr_reset fc_exch_mgr structure is private to fc_exch.c. To export exch_mgr_reset to transport, transport needs access to the exch manager. Change exch_mgr_reset to use lport param which is the shared structure between libFC and transport. Alternatively, fc_exch_mgr definition can be moved to libfc.h so that lport can be accessed from mp*. Signed-off-by: Abhijeet Joglekar Signed-off-by: Robert Love Signed-off-by: James Bottomley --- drivers/scsi/libfc/fc_exch.c | 3 ++- drivers/scsi/libfc/fc_lport.c | 4 ++-- drivers/scsi/libfc/fc_rport.c | 4 ++-- include/scsi/libfc.h | 4 ++-- 4 files changed, 8 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/drivers/scsi/libfc/fc_exch.c b/drivers/scsi/libfc/fc_exch.c index 66db08a5f27f..a09416fd843c 100644 --- a/drivers/scsi/libfc/fc_exch.c +++ b/drivers/scsi/libfc/fc_exch.c @@ -1480,10 +1480,11 @@ static void fc_exch_reset(struct fc_exch *ep) * If sid is non-zero, reset only exchanges we source from that FID. * If did is non-zero, reset only exchanges destined to that FID. */ -void fc_exch_mgr_reset(struct fc_exch_mgr *mp, u32 sid, u32 did) +void fc_exch_mgr_reset(struct fc_lport *lp, u32 sid, u32 did) { struct fc_exch *ep; struct fc_exch *next; + struct fc_exch_mgr *mp = lp->emp; spin_lock_bh(&mp->em_lock); restart: diff --git a/drivers/scsi/libfc/fc_lport.c b/drivers/scsi/libfc/fc_lport.c index 0b9bdb1fb807..5db223ce3b25 100644 --- a/drivers/scsi/libfc/fc_lport.c +++ b/drivers/scsi/libfc/fc_lport.c @@ -663,7 +663,7 @@ int fc_lport_destroy(struct fc_lport *lport) { lport->tt.frame_send = fc_frame_drop; lport->tt.fcp_abort_io(lport); - lport->tt.exch_mgr_reset(lport->emp, 0, 0); + lport->tt.exch_mgr_reset(lport, 0, 0); return 0; } EXPORT_SYMBOL(fc_lport_destroy); @@ -973,7 +973,7 @@ static void fc_lport_enter_reset(struct fc_lport *lport) lport->tt.disc_stop(lport); - lport->tt.exch_mgr_reset(lport->emp, 0, 0); + lport->tt.exch_mgr_reset(lport, 0, 0); fc_host_fabric_name(lport->host) = 0; fc_host_port_id(lport->host) = 0; diff --git a/drivers/scsi/libfc/fc_rport.c b/drivers/scsi/libfc/fc_rport.c index e780d8caf70e..dec7bae0e56d 100644 --- a/drivers/scsi/libfc/fc_rport.c +++ b/drivers/scsi/libfc/fc_rport.c @@ -1285,7 +1285,7 @@ void fc_rport_terminate_io(struct fc_rport *rport) struct fc_rport_libfc_priv *rdata = rport->dd_data; struct fc_lport *lport = rdata->local_port; - lport->tt.exch_mgr_reset(lport->emp, 0, rport->port_id); - lport->tt.exch_mgr_reset(lport->emp, rport->port_id, 0); + lport->tt.exch_mgr_reset(lport, 0, rport->port_id); + lport->tt.exch_mgr_reset(lport, rport->port_id, 0); } EXPORT_SYMBOL(fc_rport_terminate_io); diff --git a/include/scsi/libfc.h b/include/scsi/libfc.h index 9f2876397dda..042f4ade73d7 100644 --- a/include/scsi/libfc.h +++ b/include/scsi/libfc.h @@ -472,7 +472,7 @@ struct libfc_function_template { * If s_id is non-zero, reset only exchanges originating from that FID. * If d_id is non-zero, reset only exchanges sending to that FID. */ - void (*exch_mgr_reset)(struct fc_exch_mgr *, + void (*exch_mgr_reset)(struct fc_lport *, u32 s_id, u32 d_id); void (*rport_flush_queue)(void); @@ -916,7 +916,7 @@ struct fc_seq *fc_seq_start_next(struct fc_seq *sp); * If s_id is non-zero, reset only exchanges originating from that FID. * If d_id is non-zero, reset only exchanges sending to that FID. */ -void fc_exch_mgr_reset(struct fc_exch_mgr *, u32 s_id, u32 d_id); +void fc_exch_mgr_reset(struct fc_lport *, u32 s_id, u32 d_id); /* * Functions for fc_functions_template -- cgit v1.2.3-70-g09d2 From bc0e17f691085315ae9303eb5b0883fe16dfe6b1 Mon Sep 17 00:00:00 2001 From: Vasu Dev Date: Fri, 27 Feb 2009 10:54:57 -0800 Subject: [SCSI] libfc, fcoe: fixed locking issues with lport->lp_mutex around lport->link_status The fcoe_xmit could call fc_pause in case the pending skb queue len is larger than FCOE_MAX_QUEUE_DEPTH, the fc_pause was trying to grab lport->lp_muex to change lport->link_status and that had these issues :- 1. The fcoe_xmit was getting called with bh disabled, thus causing "BUG: scheduling while atomic" when grabbing lport->lp_muex with bh disabled. 2. fc_linkup and fc_linkdown function calls lport_enter function with lport->lp_mutex held and these enter function in turn calls fcoe_xmit to send lport related FC frame, e.g. fc_linkup => fc_lport_enter_flogi to send flogi req. In this case grabbing the same lport->lp_mutex again in fc_puase from fcoe_xmit would cause deadlock. The lport->lp_mutex was used for setting FC_PAUSE in fcoe_xmit path but FC_PAUSE bit was not used anywhere beside just setting and clear this bit in lport->link_status, instead used a separate field qfull in fc_lport to eliminate need for lport->lp_mutex to track pending queue full condition and in turn avoid above described two locking issues. Also added check for lp->qfull in fc_fcp_lport_queue_ready to trigger SCSI_MLQUEUE_HOST_BUSY when lp->qfull is set to prevent more scsi-ml cmds while lp->qfull is set. This patch eliminated FC_LINK_UP and FC_PAUSE and instead used dedicated fields in fc_lport for this, this simplified all related conditional code. Also removed fc_pause and fc_unpause functions and instead used newly added lport->qfull directly in fcoe. Signed-off-by: Vasu Dev Signed-off-by: Robert Love Signed-off-by: James Bottomley --- drivers/scsi/fcoe/fcoe_sw.c | 6 +++--- drivers/scsi/fcoe/libfcoe.c | 41 +++++++++++++++++------------------------ drivers/scsi/libfc/fc_fcp.c | 4 ++-- drivers/scsi/libfc/fc_lport.c | 36 ++++++------------------------------ include/scsi/libfc.h | 12 ++---------- 5 files changed, 30 insertions(+), 69 deletions(-) (limited to 'include') diff --git a/drivers/scsi/fcoe/fcoe_sw.c b/drivers/scsi/fcoe/fcoe_sw.c index dc4cd5e25760..cf83675a0fb9 100644 --- a/drivers/scsi/fcoe/fcoe_sw.c +++ b/drivers/scsi/fcoe/fcoe_sw.c @@ -116,7 +116,8 @@ static int fcoe_sw_lport_config(struct fc_lport *lp) { int i = 0; - lp->link_status = 0; + lp->link_up = 0; + lp->qfull = 0; lp->max_retry_count = 3; lp->e_d_tov = 2 * 1000; /* FC-FS default */ lp->r_a_tov = 2 * 2 * 1000; @@ -181,9 +182,8 @@ static int fcoe_sw_netdev_config(struct fc_lport *lp, struct net_device *netdev) if (fc_set_mfs(lp, mfs)) return -EINVAL; - lp->link_status = ~FC_PAUSE & ~FC_LINK_UP; if (!fcoe_link_ok(lp)) - lp->link_status |= FC_LINK_UP; + lp->link_up = 1; /* offload features support */ if (fc->real_dev->features & NETIF_F_SG) diff --git a/drivers/scsi/fcoe/libfcoe.c b/drivers/scsi/fcoe/libfcoe.c index e419f486cdb3..296071043f55 100644 --- a/drivers/scsi/fcoe/libfcoe.c +++ b/drivers/scsi/fcoe/libfcoe.c @@ -504,7 +504,7 @@ int fcoe_xmit(struct fc_lport *lp, struct fc_frame *fp) if (rc) { fcoe_insert_wait_queue(lp, skb); if (fc->fcoe_pending_queue.qlen > FCOE_MAX_QUEUE_DEPTH) - fc_pause(lp); + lp->qfull = 1; } return 0; @@ -718,7 +718,7 @@ static void fcoe_recv_flogi(struct fcoe_softc *fc, struct fc_frame *fp, u8 *sa) * fcoe_watchdog - fcoe timer callback * @vp: * - * This checks the pending queue length for fcoe and put fcoe to be paused state + * This checks the pending queue length for fcoe and set lport qfull * if the FCOE_MAX_QUEUE_DEPTH is reached. This is done for all fc_lport on the * fcoe_hostlist. * @@ -728,17 +728,17 @@ void fcoe_watchdog(ulong vp) { struct fc_lport *lp; struct fcoe_softc *fc; - int paused = 0; + int qfilled = 0; read_lock(&fcoe_hostlist_lock); list_for_each_entry(fc, &fcoe_hostlist, list) { lp = fc->lp; if (lp) { if (fc->fcoe_pending_queue.qlen > FCOE_MAX_QUEUE_DEPTH) - paused = 1; + qfilled = 1; if (fcoe_check_wait_queue(lp) < FCOE_MAX_QUEUE_DEPTH) { - if (paused) - fc_unpause(lp); + if (qfilled) + lp->qfull = 0; } } } @@ -767,8 +767,7 @@ void fcoe_watchdog(ulong vp) **/ static int fcoe_check_wait_queue(struct fc_lport *lp) { - int rc, unpause = 0; - int paused = 0; + int rc; struct sk_buff *skb; struct fcoe_softc *fc; @@ -776,10 +775,10 @@ static int fcoe_check_wait_queue(struct fc_lport *lp) spin_lock_bh(&fc->fcoe_pending_queue.lock); /* - * is this interface paused? + * if interface pending queue full then set qfull in lport. */ if (fc->fcoe_pending_queue.qlen > FCOE_MAX_QUEUE_DEPTH) - paused = 1; + lp->qfull = 1; if (fc->fcoe_pending_queue.qlen) { while ((skb = __skb_dequeue(&fc->fcoe_pending_queue)) != NULL) { spin_unlock_bh(&fc->fcoe_pending_queue.lock); @@ -791,11 +790,9 @@ static int fcoe_check_wait_queue(struct fc_lport *lp) spin_lock_bh(&fc->fcoe_pending_queue.lock); } if (fc->fcoe_pending_queue.qlen < FCOE_MAX_QUEUE_DEPTH) - unpause = 1; + lp->qfull = 0; } spin_unlock_bh(&fc->fcoe_pending_queue.lock); - if ((unpause) && (paused)) - fc_unpause(lp); return fc->fcoe_pending_queue.qlen; } @@ -873,7 +870,7 @@ static int fcoe_device_notification(struct notifier_block *notifier, struct net_device *real_dev = ptr; struct fcoe_softc *fc; struct fcoe_dev_stats *stats; - u16 new_status; + u32 new_link_up; u32 mfs; int rc = NOTIFY_OK; @@ -890,17 +887,15 @@ static int fcoe_device_notification(struct notifier_block *notifier, goto out; } - new_status = lp->link_status; + new_link_up = lp->link_up; switch (event) { case NETDEV_DOWN: case NETDEV_GOING_DOWN: - new_status &= ~FC_LINK_UP; + new_link_up = 0; break; case NETDEV_UP: case NETDEV_CHANGE: - new_status &= ~FC_LINK_UP; - if (!fcoe_link_ok(lp)) - new_status |= FC_LINK_UP; + new_link_up = !fcoe_link_ok(lp); break; case NETDEV_CHANGEMTU: mfs = fc->real_dev->mtu - @@ -908,17 +903,15 @@ static int fcoe_device_notification(struct notifier_block *notifier, sizeof(struct fcoe_crc_eof)); if (mfs >= FC_MIN_MAX_FRAME) fc_set_mfs(lp, mfs); - new_status &= ~FC_LINK_UP; - if (!fcoe_link_ok(lp)) - new_status |= FC_LINK_UP; + new_link_up = !fcoe_link_ok(lp); break; case NETDEV_REGISTER: break; default: FC_DBG("unknown event %ld call", event); } - if (lp->link_status != new_status) { - if ((new_status & FC_LINK_UP) == FC_LINK_UP) + if (lp->link_up != new_link_up) { + if (new_link_up) fc_linkup(lp); else { stats = lp->dev_stats[smp_processor_id()]; diff --git a/drivers/scsi/libfc/fc_fcp.c b/drivers/scsi/libfc/fc_fcp.c index 404e63ff46b8..f440aaca39c2 100644 --- a/drivers/scsi/libfc/fc_fcp.c +++ b/drivers/scsi/libfc/fc_fcp.c @@ -1621,7 +1621,7 @@ out: static inline int fc_fcp_lport_queue_ready(struct fc_lport *lp) { /* lock ? */ - return (lp->state == LPORT_ST_READY) && (lp->link_status & FC_LINK_UP); + return (lp->state == LPORT_ST_READY) && lp->link_up && !lp->qfull; } /** @@ -1890,7 +1890,7 @@ int fc_eh_abort(struct scsi_cmnd *sc_cmd) lp = shost_priv(sc_cmd->device->host); if (lp->state != LPORT_ST_READY) return rc; - else if (!(lp->link_status & FC_LINK_UP)) + else if (!lp->link_up) return rc; spin_lock_irqsave(lp->host->host_lock, flags); diff --git a/drivers/scsi/libfc/fc_lport.c b/drivers/scsi/libfc/fc_lport.c index 5db223ce3b25..a6ab692f5f51 100644 --- a/drivers/scsi/libfc/fc_lport.c +++ b/drivers/scsi/libfc/fc_lport.c @@ -250,7 +250,7 @@ void fc_get_host_port_state(struct Scsi_Host *shost) { struct fc_lport *lp = shost_priv(shost); - if ((lp->link_status & FC_LINK_UP) == FC_LINK_UP) + if (lp->link_up) fc_host_port_state(shost) = FC_PORTSTATE_ONLINE; else fc_host_port_state(shost) = FC_PORTSTATE_OFFLINE; @@ -577,8 +577,8 @@ void fc_linkup(struct fc_lport *lport) fc_host_port_id(lport->host)); mutex_lock(&lport->lp_mutex); - if ((lport->link_status & FC_LINK_UP) != FC_LINK_UP) { - lport->link_status |= FC_LINK_UP; + if (!lport->link_up) { + lport->link_up = 1; if (lport->state == LPORT_ST_RESET) fc_lport_enter_flogi(lport); @@ -597,8 +597,8 @@ void fc_linkdown(struct fc_lport *lport) FC_DEBUG_LPORT("Link is down for port (%6x)\n", fc_host_port_id(lport->host)); - if ((lport->link_status & FC_LINK_UP) == FC_LINK_UP) { - lport->link_status &= ~(FC_LINK_UP); + if (lport->link_up) { + lport->link_up = 0; fc_lport_enter_reset(lport); lport->tt.fcp_cleanup(lport); } @@ -606,30 +606,6 @@ void fc_linkdown(struct fc_lport *lport) } EXPORT_SYMBOL(fc_linkdown); -/** - * fc_pause - Pause the flow of frames - * @lport: The lport to be paused - */ -void fc_pause(struct fc_lport *lport) -{ - mutex_lock(&lport->lp_mutex); - lport->link_status |= FC_PAUSE; - mutex_unlock(&lport->lp_mutex); -} -EXPORT_SYMBOL(fc_pause); - -/** - * fc_unpause - Unpause the flow of frames - * @lport: The lport to be unpaused - */ -void fc_unpause(struct fc_lport *lport) -{ - mutex_lock(&lport->lp_mutex); - lport->link_status &= ~(FC_PAUSE); - mutex_unlock(&lport->lp_mutex); -} -EXPORT_SYMBOL(fc_unpause); - /** * fc_fabric_logoff - Logout of the fabric * @lport: fc_lport pointer to logoff the fabric @@ -977,7 +953,7 @@ static void fc_lport_enter_reset(struct fc_lport *lport) fc_host_fabric_name(lport->host) = 0; fc_host_port_id(lport->host) = 0; - if ((lport->link_status & FC_LINK_UP) == FC_LINK_UP) + if (lport->link_up) fc_lport_enter_flogi(lport); } diff --git a/include/scsi/libfc.h b/include/scsi/libfc.h index 042f4ade73d7..b9e6c1cd8914 100644 --- a/include/scsi/libfc.h +++ b/include/scsi/libfc.h @@ -68,9 +68,6 @@ /* * FC HBA status */ -#define FC_PAUSE (1 << 1) -#define FC_LINK_UP (1 << 0) - enum fc_lport_state { LPORT_ST_NONE = 0, LPORT_ST_FLOGI, @@ -603,7 +600,8 @@ struct fc_lport { /* Operational Information */ struct libfc_function_template tt; - u16 link_status; + u8 link_up; + u8 qfull; enum fc_lport_state state; unsigned long boot_time; @@ -703,12 +701,6 @@ void fc_linkup(struct fc_lport *); */ void fc_linkdown(struct fc_lport *); -/* - * Pause and unpause traffic. - */ -void fc_pause(struct fc_lport *); -void fc_unpause(struct fc_lport *); - /* * Configure the local port. */ -- cgit v1.2.3-70-g09d2 From 6755db1cd4587084be85f860b7aa7c0cc9d776dc Mon Sep 17 00:00:00 2001 From: Chris Leech Date: Fri, 27 Feb 2009 10:55:02 -0800 Subject: [SCSI] libfc: rport retry on LS_RJT from certain ELS This allows any rport ELS to retry on LS_RJT. The rport error handling would only retry on resource allocation failures and exchange timeouts. I have a target that will occasionally reject PLOGI when we do a quick LOGO/PLOGI. When a critical ELS was rejected, libfc would fail silently leaving the rport in a dead state. The retry count and delay are managed by fc_rport_error_retry. If the retry count is exceeded fc_rport_error will be called. When retrying is not the correct course of action, fc_rport_error can be called directly. Signed-off-by: Chris Leech Signed-off-by: Robert Love Signed-off-by: James Bottomley --- drivers/scsi/libfc/fc_exch.c | 2 - drivers/scsi/libfc/fc_rport.c | 111 ++++++++++++++++++++++++------------------ include/scsi/fc/fc_fs.h | 5 ++ 3 files changed, 69 insertions(+), 49 deletions(-) (limited to 'include') diff --git a/drivers/scsi/libfc/fc_exch.c b/drivers/scsi/libfc/fc_exch.c index dd269e5f953e..8c4018956d4c 100644 --- a/drivers/scsi/libfc/fc_exch.c +++ b/drivers/scsi/libfc/fc_exch.c @@ -32,8 +32,6 @@ #include #include -#define FC_DEF_R_A_TOV (10 * 1000) /* resource allocation timeout */ - /* * fc_exch_debug can be set in debugger or at compile time to get more logs. */ diff --git a/drivers/scsi/libfc/fc_rport.c b/drivers/scsi/libfc/fc_rport.c index 717575934152..600a8fffa940 100644 --- a/drivers/scsi/libfc/fc_rport.c +++ b/drivers/scsi/libfc/fc_rport.c @@ -81,6 +81,7 @@ static void fc_rport_recv_logo_req(struct fc_rport *, struct fc_seq *, struct fc_frame *); static void fc_rport_timeout(struct work_struct *); static void fc_rport_error(struct fc_rport *, struct fc_frame *); +static void fc_rport_error_retry(struct fc_rport *, struct fc_frame *); static void fc_rport_work(struct work_struct *); static const char *fc_rport_state_names[] = { @@ -410,57 +411,73 @@ static void fc_rport_timeout(struct work_struct *work) } /** - * fc_rport_error - Handler for any errors + * fc_rport_error - Error handler, called once retries have been exhausted * @rport: The fc_rport object * @fp: The frame pointer * - * If the error was caused by a resource allocation failure - * then wait for half a second and retry, otherwise retry - * immediately. - * * Locking Note: The rport lock is expected to be held before * calling this routine */ static void fc_rport_error(struct fc_rport *rport, struct fc_frame *fp) { struct fc_rport_libfc_priv *rdata = rport->dd_data; - unsigned long delay = 0; FC_DEBUG_RPORT("Error %ld in state %s, retries %d\n", PTR_ERR(fp), fc_rport_state(rport), rdata->retries); - if (!fp || PTR_ERR(fp) == -FC_EX_TIMEOUT) { - /* - * Memory allocation failure, or the exchange timed out. - * Retry after delay - */ - if (rdata->retries < rdata->local_port->max_retry_count) { - rdata->retries++; - if (!fp) - delay = msecs_to_jiffies(500); - get_device(&rport->dev); - schedule_delayed_work(&rdata->retry_work, delay); - } else { - switch (rdata->rp_state) { - case RPORT_ST_PLOGI: - case RPORT_ST_PRLI: - case RPORT_ST_LOGO: - rdata->event = RPORT_EV_FAILED; - queue_work(rport_event_queue, - &rdata->event_work); - break; - case RPORT_ST_RTV: - fc_rport_enter_ready(rport); - break; - case RPORT_ST_NONE: - case RPORT_ST_READY: - case RPORT_ST_INIT: - break; - } - } + switch (rdata->rp_state) { + case RPORT_ST_PLOGI: + case RPORT_ST_PRLI: + case RPORT_ST_LOGO: + rdata->event = RPORT_EV_FAILED; + queue_work(rport_event_queue, + &rdata->event_work); + break; + case RPORT_ST_RTV: + fc_rport_enter_ready(rport); + break; + case RPORT_ST_NONE: + case RPORT_ST_READY: + case RPORT_ST_INIT: + break; } } +/** + * fc_rport_error_retry - Error handler when retries are desired + * @rport: The fc_rport object + * @fp: The frame pointer + * + * If the error was an exchange timeout retry immediately, + * otherwise wait for E_D_TOV. + * + * Locking Note: The rport lock is expected to be held before + * calling this routine + */ +static void fc_rport_error_retry(struct fc_rport *rport, struct fc_frame *fp) +{ + struct fc_rport_libfc_priv *rdata = rport->dd_data; + unsigned long delay = FC_DEF_E_D_TOV; + + /* make sure this isn't an FC_EX_CLOSED error, never retry those */ + if (PTR_ERR(fp) == -FC_EX_CLOSED) + return fc_rport_error(rport, fp); + + if (rdata->retries < rdata->local_port->max_retry_count) { + FC_DEBUG_RPORT("Error %ld in state %s, retrying\n", + PTR_ERR(fp), fc_rport_state(rport)); + rdata->retries++; + /* no additional delay on exchange timeouts */ + if (PTR_ERR(fp) == -FC_EX_TIMEOUT) + delay = 0; + get_device(&rport->dev); + schedule_delayed_work(&rdata->retry_work, delay); + return; + } + + return fc_rport_error(rport, fp); +} + /** * fc_rport_plogi_recv_resp - Handle incoming ELS PLOGI response * @sp: current sequence in the PLOGI exchange @@ -495,7 +512,7 @@ static void fc_rport_plogi_resp(struct fc_seq *sp, struct fc_frame *fp, } if (IS_ERR(fp)) { - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); goto err; } @@ -527,7 +544,7 @@ static void fc_rport_plogi_resp(struct fc_seq *sp, struct fc_frame *fp, else fc_rport_enter_prli(rport); } else - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); out: fc_frame_free(fp); @@ -557,14 +574,14 @@ static void fc_rport_enter_plogi(struct fc_rport *rport) rport->maxframe_size = FC_MIN_MAX_PAYLOAD; fp = fc_frame_alloc(lport, sizeof(struct fc_els_flogi)); if (!fp) { - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); return; } rdata->e_d_tov = lport->e_d_tov; if (!lport->tt.elsct_send(lport, rport, fp, ELS_PLOGI, fc_rport_plogi_resp, rport, lport->e_d_tov)) - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); else get_device(&rport->dev); } @@ -604,7 +621,7 @@ static void fc_rport_prli_resp(struct fc_seq *sp, struct fc_frame *fp, } if (IS_ERR(fp)) { - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); goto err; } @@ -662,7 +679,7 @@ static void fc_rport_logo_resp(struct fc_seq *sp, struct fc_frame *fp, rport->port_id); if (IS_ERR(fp)) { - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); goto err; } @@ -712,13 +729,13 @@ static void fc_rport_enter_prli(struct fc_rport *rport) fp = fc_frame_alloc(lport, sizeof(*pp)); if (!fp) { - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); return; } if (!lport->tt.elsct_send(lport, rport, fp, ELS_PRLI, fc_rport_prli_resp, rport, lport->e_d_tov)) - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); else get_device(&rport->dev); } @@ -809,13 +826,13 @@ static void fc_rport_enter_rtv(struct fc_rport *rport) fp = fc_frame_alloc(lport, sizeof(struct fc_els_rtv)); if (!fp) { - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); return; } if (!lport->tt.elsct_send(lport, rport, fp, ELS_RTV, fc_rport_rtv_resp, rport, lport->e_d_tov)) - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); else get_device(&rport->dev); } @@ -840,13 +857,13 @@ static void fc_rport_enter_logo(struct fc_rport *rport) fp = fc_frame_alloc(lport, sizeof(struct fc_els_logo)); if (!fp) { - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); return; } if (!lport->tt.elsct_send(lport, rport, fp, ELS_LOGO, fc_rport_logo_resp, rport, lport->e_d_tov)) - fc_rport_error(rport, fp); + fc_rport_error_retry(rport, fp); else get_device(&rport->dev); } diff --git a/include/scsi/fc/fc_fs.h b/include/scsi/fc/fc_fs.h index 3e4801d2bdbb..1b7af3a64c7c 100644 --- a/include/scsi/fc/fc_fs.h +++ b/include/scsi/fc/fc_fs.h @@ -337,4 +337,9 @@ enum fc_pf_rjt_reason { FC_RJT_VENDOR = 0xff, /* vendor specific reject */ }; +/* default timeout values */ + +#define FC_DEF_E_D_TOV 2000UL +#define FC_DEF_R_A_TOV 10000UL + #endif /* _FC_FS_H_ */ -- cgit v1.2.3-70-g09d2 From 5101ff99f59aefb72e0c96e82aa32048ac9f8425 Mon Sep 17 00:00:00 2001 From: Robert Love Date: Fri, 27 Feb 2009 10:55:18 -0800 Subject: [SCSI] libfc: Don't violate transport template for rogue port creation Signed-off-by: Robert Love Signed-off-by: James Bottomley --- drivers/scsi/libfc/fc_disc.c | 6 +++--- drivers/scsi/libfc/fc_lport.c | 4 ++-- drivers/scsi/libfc/fc_rport.c | 3 +++ include/scsi/libfc.h | 5 +++++ 4 files changed, 13 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/drivers/scsi/libfc/fc_disc.c b/drivers/scsi/libfc/fc_disc.c index dd1564c9e04a..15e3f840c049 100644 --- a/drivers/scsi/libfc/fc_disc.c +++ b/drivers/scsi/libfc/fc_disc.c @@ -430,7 +430,7 @@ static int fc_disc_new_target(struct fc_disc *disc, dp.ids.port_name = ids->port_name; dp.ids.node_name = ids->node_name; dp.ids.roles = ids->roles; - rport = fc_rport_rogue_create(&dp); + rport = lport->tt.rport_create(&dp); } if (!rport) error = -ENOMEM; @@ -617,7 +617,7 @@ static int fc_disc_gpn_ft_parse(struct fc_disc *disc, void *buf, size_t len) if ((dp.ids.port_id != fc_host_port_id(lport->host)) && (dp.ids.port_name != lport->wwpn)) { - rport = fc_rport_rogue_create(&dp); + rport = lport->tt.rport_create(&dp); if (rport) { rdata = rport->dd_data; rdata->ops = &fc_disc_rport_ops; @@ -769,7 +769,7 @@ static void fc_disc_single(struct fc_disc *disc, struct fc_disc_port *dp) if (rport) fc_disc_del_target(disc, rport); - new_rport = fc_rport_rogue_create(dp); + new_rport = lport->tt.rport_create(dp); if (new_rport) { rdata = new_rport->dd_data; rdata->ops = &fc_disc_rport_ops; diff --git a/drivers/scsi/libfc/fc_lport.c b/drivers/scsi/libfc/fc_lport.c index e6ea4f119530..07335ae2947c 100644 --- a/drivers/scsi/libfc/fc_lport.c +++ b/drivers/scsi/libfc/fc_lport.c @@ -232,7 +232,7 @@ static void fc_lport_ptp_setup(struct fc_lport *lport, lport->ptp_rp = NULL; } - lport->ptp_rp = fc_rport_rogue_create(&dp); + lport->ptp_rp = lport->tt.rport_create(&dp); lport->tt.rport_login(lport->ptp_rp); @@ -1282,7 +1282,7 @@ static void fc_lport_enter_dns(struct fc_lport *lport) fc_lport_state_enter(lport, LPORT_ST_DNS); - rport = fc_rport_rogue_create(&dp); + rport = lport->tt.rport_create(&dp); if (!rport) goto err; diff --git a/drivers/scsi/libfc/fc_rport.c b/drivers/scsi/libfc/fc_rport.c index 600a8fffa940..81b3ca188789 100644 --- a/drivers/scsi/libfc/fc_rport.c +++ b/drivers/scsi/libfc/fc_rport.c @@ -1271,6 +1271,9 @@ static void fc_rport_flush_queue(void) int fc_rport_init(struct fc_lport *lport) { + if (!lport->tt.rport_create) + lport->tt.rport_create = fc_rport_rogue_create; + if (!lport->tt.rport_login) lport->tt.rport_login = fc_rport_login; diff --git a/include/scsi/libfc.h b/include/scsi/libfc.h index b9e6c1cd8914..37df48e13b97 100644 --- a/include/scsi/libfc.h +++ b/include/scsi/libfc.h @@ -489,6 +489,11 @@ struct libfc_function_template { * Remote Port interfaces */ + /* + * Create a remote port + */ + struct fc_rport *(*rport_create)(struct fc_disc_port *); + /* * Initiates the RP state machine. It is called from the LP module. * This function will issue the following commands to the N_Port -- cgit v1.2.3-70-g09d2 From 0ae4d4ae47d2ccbcad813b0d6d8fe12590c7d648 Mon Sep 17 00:00:00 2001 From: Robert Love Date: Fri, 27 Feb 2009 10:55:39 -0800 Subject: [SCSI] libfc: Cleanup libfc_function_template comments Made the comments more like the comments for struct scsi_host_template. Signed-off-by: Robert Love Signed-off-by: James Bottomley --- include/scsi/libfc.h | 107 +++++++++++++++++++++++++++++++-------------------- 1 file changed, 66 insertions(+), 41 deletions(-) (limited to 'include') diff --git a/include/scsi/libfc.h b/include/scsi/libfc.h index 37df48e13b97..282829cdf352 100644 --- a/include/scsi/libfc.h +++ b/include/scsi/libfc.h @@ -336,31 +336,17 @@ struct fc_exch { struct libfc_function_template { - /** - * Mandatory Fields - * - * These handlers must be implemented by the LLD. - */ - /* * Interface to send a FC frame - */ - int (*frame_send)(struct fc_lport *lp, struct fc_frame *fp); - - /** - * Optional Fields * - * The LLD may choose to implement any of the following handlers. - * If LLD doesn't specify hander and leaves its pointer NULL then - * the default libfc function will be used for that handler. - */ - - /** - * ELS/CT interfaces + * STATUS: REQUIRED */ + int (*frame_send)(struct fc_lport *lp, struct fc_frame *fp); /* - * elsct_send - sends ELS/CT frame + * Interface to send ELS/CT frames + * + * STATUS: OPTIONAL */ struct fc_seq *(*elsct_send)(struct fc_lport *lport, struct fc_rport *rport, @@ -370,9 +356,6 @@ struct libfc_function_template { struct fc_frame *fp, void *arg), void *arg, u32 timer_msec); - /** - * Exhance Manager interfaces - */ /* * Send the FC frame payload using a new exchange and sequence. @@ -404,6 +387,8 @@ struct libfc_function_template { * timer_msec argument is specified. The timer is canceled when * it fires or when the exchange is done. The exchange timeout handler * is registered by EM layer. + * + * STATUS: OPTIONAL */ struct fc_seq *(*exch_seq_send)(struct fc_lport *lp, struct fc_frame *fp, @@ -415,14 +400,18 @@ struct libfc_function_template { void *arg, unsigned int timer_msec); /* - * send a frame using existing sequence and exchange. + * Send a frame using an existing sequence and exchange. + * + * STATUS: OPTIONAL */ int (*seq_send)(struct fc_lport *lp, struct fc_seq *sp, struct fc_frame *fp); /* - * Send ELS response using mainly infomation - * in exchange and sequence in EM layer. + * Send an ELS response using infomation from a previous + * exchange and sequence. + * + * STATUS: OPTIONAL */ void (*seq_els_rsp_send)(struct fc_seq *sp, enum fc_els_cmd els_cmd, struct fc_seq_els_data *els_data); @@ -434,6 +423,8 @@ struct libfc_function_template { * A timer_msec can be specified for abort timeout, if non-zero * timer_msec value is specified then exchange resp handler * will be called with timeout error if no response to abort. + * + * STATUS: OPTIONAL */ int (*seq_exch_abort)(const struct fc_seq *req_sp, unsigned int timer_msec); @@ -441,6 +432,8 @@ struct libfc_function_template { /* * Indicate that an exchange/sequence tuple is complete and the memory * allocated for the related objects may be freed. + * + * STATUS: OPTIONAL */ void (*exch_done)(struct fc_seq *sp); @@ -448,6 +441,8 @@ struct libfc_function_template { * Assigns a EM and a free XID for an new exchange and then * allocates a new exchange and sequence pair. * The fp can be used to determine free XID. + * + * STATUS: OPTIONAL */ struct fc_exch *(*exch_get)(struct fc_lport *lp, struct fc_frame *fp); @@ -455,12 +450,16 @@ struct libfc_function_template { * Release previously assigned XID by exch_get API. * The LLD may implement this if XID is assigned by LLD * in exch_get(). + * + * STATUS: OPTIONAL */ void (*exch_put)(struct fc_lport *lp, struct fc_exch_mgr *mp, u16 ex_id); /* * Start a new sequence on the same exchange/sequence tuple. + * + * STATUS: OPTIONAL */ struct fc_seq *(*seq_start_next)(struct fc_seq *sp); @@ -468,26 +467,33 @@ struct libfc_function_template { * Reset an exchange manager, completing all sequences and exchanges. * If s_id is non-zero, reset only exchanges originating from that FID. * If d_id is non-zero, reset only exchanges sending to that FID. + * + * STATUS: OPTIONAL */ void (*exch_mgr_reset)(struct fc_lport *, u32 s_id, u32 d_id); - void (*rport_flush_queue)(void); - /** - * Local Port interfaces + /* + * Flush the rport work queue. Generally used before shutdown. + * + * STATUS: OPTIONAL */ + void (*rport_flush_queue)(void); /* - * Receive a frame to a local port. + * Receive a frame for a local port. + * + * STATUS: OPTIONAL */ void (*lport_recv)(struct fc_lport *lp, struct fc_seq *sp, struct fc_frame *fp); - int (*lport_reset)(struct fc_lport *); - - /** - * Remote Port interfaces + /* + * Reset the local port. + * + * STATUS: OPTIONAL */ + int (*lport_reset)(struct fc_lport *); /* * Create a remote port @@ -502,26 +508,33 @@ struct libfc_function_template { * - PLOGI * - PRLI * - RTV + * + * STATUS: OPTIONAL */ int (*rport_login)(struct fc_rport *rport); /* * Logoff, and remove the rport from the transport if * it had been added. This will send a LOGO to the target. + * + * STATUS: OPTIONAL */ int (*rport_logoff)(struct fc_rport *rport); /* * Recieve a request from a remote port. + * + * STATUS: OPTIONAL */ void (*rport_recv_req)(struct fc_seq *, struct fc_frame *, struct fc_rport *); - struct fc_rport *(*rport_lookup)(const struct fc_lport *, u32); - - /** - * FCP interfaces + /* + * lookup an rport by it's port ID. + * + * STATUS: OPTIONAL */ + struct fc_rport *(*rport_lookup)(const struct fc_lport *, u32); /* * Send a fcp cmd from fsp pkt. @@ -529,30 +542,38 @@ struct libfc_function_template { * * The resp handler is called when FCP_RSP received. * + * STATUS: OPTIONAL */ int (*fcp_cmd_send)(struct fc_lport *lp, struct fc_fcp_pkt *fsp, void (*resp)(struct fc_seq *, struct fc_frame *fp, void *arg)); /* - * Used at least durring linkdown and reset + * Cleanup the FCP layer, used durring link down and reset + * + * STATUS: OPTIONAL */ void (*fcp_cleanup)(struct fc_lport *lp); /* * Abort all I/O on a local port + * + * STATUS: OPTIONAL */ void (*fcp_abort_io)(struct fc_lport *lp); - /** - * Discovery interfaces + /* + * Receive a request for the discovery layer. + * + * STATUS: OPTIONAL */ - void (*disc_recv_req)(struct fc_seq *, struct fc_frame *, struct fc_lport *); /* * Start discovery for a local port. + * + * STATUS: OPTIONAL */ void (*disc_start)(void (*disc_callback)(struct fc_lport *, enum fc_disc_event), @@ -561,6 +582,8 @@ struct libfc_function_template { /* * Stop discovery for a given lport. This will remove * all discovered rports + * + * STATUS: OPTIONAL */ void (*disc_stop) (struct fc_lport *); @@ -568,6 +591,8 @@ struct libfc_function_template { * Stop discovery for a given lport. This will block * until all discovered rports are deleted from the * FC transport class + * + * STATUS: OPTIONAL */ void (*disc_stop_final) (struct fc_lport *); }; -- cgit v1.2.3-70-g09d2 From ab96ddec7213004b632d24dc2cdcd2df5f16f50b Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Sat, 7 Mar 2009 13:39:22 -0800 Subject: Input: serio - fix protocol number for TouchIT213 Protocol 0x37 has been reserved for iNexio devices and Sahara was supposed to get 0x38. Reported-by: Claudio Nieder Signed-off-by: Dmitry Torokhov --- include/linux/serio.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/serio.h b/include/linux/serio.h index 1bcb357a01a1..e0417e4d3f15 100644 --- a/include/linux/serio.h +++ b/include/linux/serio.h @@ -212,7 +212,7 @@ static inline void serio_unpin_driver(struct serio *serio) #define SERIO_FUJITSU 0x35 #define SERIO_ZHENHUA 0x36 #define SERIO_INEXIO 0x37 -#define SERIO_TOUCHIT213 0x37 +#define SERIO_TOUCHIT213 0x38 #define SERIO_W8001 0x39 #endif -- cgit v1.2.3-70-g09d2 From 129f8ae9b1b5be94517da76009ea956e89104ce8 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Mon, 9 Mar 2009 15:07:33 -0400 Subject: Revert "[CPUFREQ] Disable sysfs ui for p4-clockmod." This reverts commit e088e4c9cdb618675874becb91b2fd581ee707e6. Removing the sysfs interface for p4-clockmod was flagged as a regression in bug 12826. Course of action: - Find out the remaining causes of overheating, and fix them if possible. ACPI should be doing the right thing automatically. If it isn't, we need to fix that. - mark p4-clockmod ui as deprecated - try again with the removal in six months. It's not really feasible to printk about the deprecation, because it needs to happen at all the sysfs entry points, which means adding a lot of strcmp("p4-clockmod".. calls to the core, which.. bleuch. Signed-off-by: Dave Jones --- arch/x86/kernel/cpu/cpufreq/p4-clockmod.c | 1 - drivers/cpufreq/cpufreq.c | 51 +++++++++++-------------------- include/linux/cpufreq.h | 1 - 3 files changed, 18 insertions(+), 35 deletions(-) (limited to 'include') diff --git a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c index b585e04cbc9e..3178c3acd97e 100644 --- a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c +++ b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c @@ -277,7 +277,6 @@ static struct cpufreq_driver p4clockmod_driver = { .name = "p4-clockmod", .owner = THIS_MODULE, .attr = p4clockmod_attr, - .hide_interface = 1, }; diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index b55cb67435bd..d6daf3c507d3 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -754,11 +754,6 @@ static struct kobj_type ktype_cpufreq = { .release = cpufreq_sysfs_release, }; -static struct kobj_type ktype_empty_cpufreq = { - .sysfs_ops = &sysfs_ops, - .release = cpufreq_sysfs_release, -}; - /** * cpufreq_add_dev - add a CPU device @@ -892,36 +887,26 @@ static int cpufreq_add_dev(struct sys_device *sys_dev) memcpy(&new_policy, policy, sizeof(struct cpufreq_policy)); /* prepare interface data */ - if (!cpufreq_driver->hide_interface) { - ret = kobject_init_and_add(&policy->kobj, &ktype_cpufreq, - &sys_dev->kobj, "cpufreq"); + ret = kobject_init_and_add(&policy->kobj, &ktype_cpufreq, &sys_dev->kobj, + "cpufreq"); + if (ret) + goto err_out_driver_exit; + + /* set up files for this cpu device */ + drv_attr = cpufreq_driver->attr; + while ((drv_attr) && (*drv_attr)) { + ret = sysfs_create_file(&policy->kobj, &((*drv_attr)->attr)); if (ret) goto err_out_driver_exit; - - /* set up files for this cpu device */ - drv_attr = cpufreq_driver->attr; - while ((drv_attr) && (*drv_attr)) { - ret = sysfs_create_file(&policy->kobj, - &((*drv_attr)->attr)); - if (ret) - goto err_out_driver_exit; - drv_attr++; - } - if (cpufreq_driver->get) { - ret = sysfs_create_file(&policy->kobj, - &cpuinfo_cur_freq.attr); - if (ret) - goto err_out_driver_exit; - } - if (cpufreq_driver->target) { - ret = sysfs_create_file(&policy->kobj, - &scaling_cur_freq.attr); - if (ret) - goto err_out_driver_exit; - } - } else { - ret = kobject_init_and_add(&policy->kobj, &ktype_empty_cpufreq, - &sys_dev->kobj, "cpufreq"); + drv_attr++; + } + if (cpufreq_driver->get) { + ret = sysfs_create_file(&policy->kobj, &cpuinfo_cur_freq.attr); + if (ret) + goto err_out_driver_exit; + } + if (cpufreq_driver->target) { + ret = sysfs_create_file(&policy->kobj, &scaling_cur_freq.attr); if (ret) goto err_out_driver_exit; } diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 384b38d3e8e2..161042746afc 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -234,7 +234,6 @@ struct cpufreq_driver { int (*suspend) (struct cpufreq_policy *policy, pm_message_t pmsg); int (*resume) (struct cpufreq_policy *policy); struct freq_attr **attr; - bool hide_interface; }; /* flags */ -- cgit v1.2.3-70-g09d2 From 34f42a070fc98f5dc07e9fa2338b7b8d1dc347eb Mon Sep 17 00:00:00 2001 From: Robert Love Date: Fri, 27 Feb 2009 10:55:45 -0800 Subject: [SCSI] libfc, fcoe: Fix kerneldoc comments 1) Added '()' for function names in kerneldoc comments 2) Changed comment bookends from '**/' to '*/'. The comment on the the mailing list was that '**/' "is consistently unconventional. Not wrong, just odd." The Documentation/kernel-doc-nano-HOWTO.txt states that kerneldoc comment blocks should end with '**/' but most (if not all) instance I found under drivers/scsi/ were only using the '*/' so I converted to that style. 3) Removed incorrect linebreaks in kerneldoc comments where found 4) Removed a few unnecessary blank comment lines in kerneldoc comment blocks Signed-off-by: Robert Love Signed-off-by: James Bottomley --- drivers/scsi/fcoe/fc_transport_fcoe.c | 77 ++++++++-------- drivers/scsi/fcoe/fcoe_sw.c | 42 ++++----- drivers/scsi/fcoe/libfcoe.c | 162 +++++++++++++++++----------------- drivers/scsi/libfc/fc_disc.c | 45 +++++----- drivers/scsi/libfc/fc_fcp.c | 46 +++++----- drivers/scsi/libfc/fc_lport.c | 77 ++++++++-------- drivers/scsi/libfc/fc_rport.c | 46 +++++----- include/scsi/libfc.h | 14 +-- 8 files changed, 252 insertions(+), 257 deletions(-) (limited to 'include') diff --git a/drivers/scsi/fcoe/fc_transport_fcoe.c b/drivers/scsi/fcoe/fc_transport_fcoe.c index bf7fe6fc0820..0eea4c434c31 100644 --- a/drivers/scsi/fcoe/fc_transport_fcoe.c +++ b/drivers/scsi/fcoe/fc_transport_fcoe.c @@ -33,19 +33,19 @@ static LIST_HEAD(fcoe_transports); static DEFINE_MUTEX(fcoe_transports_lock); /** - * fcoe_transport_default - returns ptr to the default transport fcoe_sw - **/ + * fcoe_transport_default() - Returns ptr to the default transport fcoe_sw + */ struct fcoe_transport *fcoe_transport_default(void) { return &fcoe_sw_transport; } /** - * fcoe_transport_to_pcidev - get the pci dev from a netdev + * fcoe_transport_to_pcidev() - get the pci dev from a netdev * @netdev: the netdev that pci dev will be retrived from * * Returns: NULL or the corrsponding pci_dev - **/ + */ struct pci_dev *fcoe_transport_pcidev(const struct net_device *netdev) { if (!netdev->dev.parent) @@ -54,16 +54,14 @@ struct pci_dev *fcoe_transport_pcidev(const struct net_device *netdev) } /** - * fcoe_transport_device_lookup - find out netdev is managed by the - * transport - * assign a transport to a device + * fcoe_transport_device_lookup() - Lookup a transport * @netdev: the netdev the transport to be attached to * * This will look for existing offload driver, if not found, it falls back to * the default sw hba (fcoe_sw) as its fcoe transport. * * Returns: 0 for success - **/ + */ static struct fcoe_transport_internal *fcoe_transport_device_lookup( struct fcoe_transport *t, struct net_device *netdev) { @@ -81,14 +79,14 @@ static struct fcoe_transport_internal *fcoe_transport_device_lookup( return NULL; } /** - * fcoe_transport_device_add - assign a transport to a device + * fcoe_transport_device_add() - Assign a transport to a device * @netdev: the netdev the transport to be attached to * * This will look for existing offload driver, if not found, it falls back to * the default sw hba (fcoe_sw) as its fcoe transport. * * Returns: 0 for success - **/ + */ static int fcoe_transport_device_add(struct fcoe_transport *t, struct net_device *netdev) { @@ -123,14 +121,14 @@ static int fcoe_transport_device_add(struct fcoe_transport *t, } /** - * fcoe_transport_device_remove - remove a device from its transport + * fcoe_transport_device_remove() - Remove a device from its transport * @netdev: the netdev the transport to be attached to * - * this removes the device from the transport so the given transport will + * This removes the device from the transport so the given transport will * not manage this device any more * * Returns: 0 for success - **/ + */ static int fcoe_transport_device_remove(struct fcoe_transport *t, struct net_device *netdev) { @@ -155,13 +153,13 @@ static int fcoe_transport_device_remove(struct fcoe_transport *t, } /** - * fcoe_transport_device_remove_all - remove all from transport devlist + * fcoe_transport_device_remove_all() - Remove all from transport devlist * - * this removes the device from the transport so the given transport will + * This removes the device from the transport so the given transport will * not manage this device any more * * Returns: 0 for success - **/ + */ static void fcoe_transport_device_remove_all(struct fcoe_transport *t) { struct fcoe_transport_internal *ti, *tmp; @@ -175,16 +173,16 @@ static void fcoe_transport_device_remove_all(struct fcoe_transport *t) } /** - * fcoe_transport_match - use the bus device match function to match the hw - * @t: the fcoe transport - * @netdev: + * fcoe_transport_match() - Use the bus device match function to match the hw + * @t: The fcoe transport to check + * @netdev: The netdev to match against * - * This function is used to check if the givne transport wants to manage the + * This function is used to check if the given transport wants to manage the * input netdev. if the transports implements the match function, it will be * called, o.w. we just compare the pci vendor and device id. * * Returns: true for match up - **/ + */ static bool fcoe_transport_match(struct fcoe_transport *t, struct net_device *netdev) { @@ -210,15 +208,15 @@ static bool fcoe_transport_match(struct fcoe_transport *t, } /** - * fcoe_transport_lookup - check if the transport is already registered + * fcoe_transport_lookup() - Check if the transport is already registered * @t: the transport to be looked up * * This compares the parent device (pci) vendor and device id * * Returns: NULL if not found * - * TODO - return default sw transport if no other transport is found - **/ + * TODO: return default sw transport if no other transport is found + */ static struct fcoe_transport *fcoe_transport_lookup( struct net_device *netdev) { @@ -239,11 +237,11 @@ static struct fcoe_transport *fcoe_transport_lookup( } /** - * fcoe_transport_register - adds a fcoe transport to the fcoe transports list + * fcoe_transport_register() - Adds a fcoe transport to the fcoe transports list * @t: ptr to the fcoe transport to be added * * Returns: 0 for success - **/ + */ int fcoe_transport_register(struct fcoe_transport *t) { struct fcoe_transport *tt; @@ -269,11 +267,11 @@ int fcoe_transport_register(struct fcoe_transport *t) EXPORT_SYMBOL_GPL(fcoe_transport_register); /** - * fcoe_transport_unregister - remove the tranport fro the fcoe transports list + * fcoe_transport_unregister() - Remove the tranport fro the fcoe transports list * @t: ptr to the fcoe transport to be removed * * Returns: 0 for success - **/ + */ int fcoe_transport_unregister(struct fcoe_transport *t) { struct fcoe_transport *tt, *tmp; @@ -294,8 +292,8 @@ int fcoe_transport_unregister(struct fcoe_transport *t) } EXPORT_SYMBOL_GPL(fcoe_transport_unregister); -/* - * fcoe_load_transport_driver - load an offload driver by alias name +/** + * fcoe_load_transport_driver() - Load an offload driver by alias name * @netdev: the target net device * * Requests for an offload driver module as the fcoe transport, if fails, it @@ -307,7 +305,7 @@ EXPORT_SYMBOL_GPL(fcoe_transport_unregister); * 3. pure hw fcoe hba may not have netdev * * Returns: 0 for success - **/ + */ int fcoe_load_transport_driver(struct net_device *netdev) { struct pci_dev *pci; @@ -335,14 +333,14 @@ int fcoe_load_transport_driver(struct net_device *netdev) EXPORT_SYMBOL_GPL(fcoe_load_transport_driver); /** - * fcoe_transport_attach - load transport to fcoe + * fcoe_transport_attach() - Load transport to fcoe * @netdev: the netdev the transport to be attached to * * This will look for existing offload driver, if not found, it falls back to * the default sw hba (fcoe_sw) as its fcoe transport. * * Returns: 0 for success - **/ + */ int fcoe_transport_attach(struct net_device *netdev) { struct fcoe_transport *t; @@ -373,11 +371,11 @@ int fcoe_transport_attach(struct net_device *netdev) EXPORT_SYMBOL_GPL(fcoe_transport_attach); /** - * fcoe_transport_release - unload transport from fcoe + * fcoe_transport_release() - Unload transport from fcoe * @netdev: the net device on which fcoe is to be released * * Returns: 0 for success - **/ + */ int fcoe_transport_release(struct net_device *netdev) { struct fcoe_transport *t; @@ -410,12 +408,12 @@ int fcoe_transport_release(struct net_device *netdev) EXPORT_SYMBOL_GPL(fcoe_transport_release); /** - * fcoe_transport_init - initializes fcoe transport layer + * fcoe_transport_init() - Initializes fcoe transport layer * * This prepares for the fcoe transport layer * * Returns: none - **/ + */ int __init fcoe_transport_init(void) { INIT_LIST_HEAD(&fcoe_transports); @@ -424,12 +422,13 @@ int __init fcoe_transport_init(void) } /** - * fcoe_transport_exit - cleans up the fcoe transport layer + * fcoe_transport_exit() - Cleans up the fcoe transport layer + * * This cleans up the fcoe transport layer. removing any transport on the list, * note that the transport destroy func is not called here. * * Returns: none - **/ + */ int __exit fcoe_transport_exit(void) { struct fcoe_transport *t, *tmp; diff --git a/drivers/scsi/fcoe/fcoe_sw.c b/drivers/scsi/fcoe/fcoe_sw.c index cf83675a0fb9..007d1fbae808 100644 --- a/drivers/scsi/fcoe/fcoe_sw.c +++ b/drivers/scsi/fcoe/fcoe_sw.c @@ -104,13 +104,12 @@ static struct scsi_host_template fcoe_sw_shost_template = { .max_sectors = 0xffff, }; -/* - * fcoe_sw_lport_config - sets up the fc_lport +/** + * fcoe_sw_lport_config() - sets up the fc_lport * @lp: ptr to the fc_lport * @shost: ptr to the parent scsi host * * Returns: 0 for success - * */ static int fcoe_sw_lport_config(struct fc_lport *lp) { @@ -137,16 +136,14 @@ static int fcoe_sw_lport_config(struct fc_lport *lp) return 0; } -/* - * fcoe_sw_netdev_config - sets up fcoe_softc for lport and network - * related properties +/** + * fcoe_sw_netdev_config() - Set up netdev for SW FCoE * @lp : ptr to the fc_lport * @netdev : ptr to the associated netdevice struct * * Must be called after fcoe_sw_lport_config() as it will use lport mutex * * Returns : 0 for success - * */ static int fcoe_sw_netdev_config(struct fc_lport *lp, struct net_device *netdev) { @@ -224,16 +221,15 @@ static int fcoe_sw_netdev_config(struct fc_lport *lp, struct net_device *netdev) return 0; } -/* - * fcoe_sw_shost_config - sets up fc_lport->host +/** + * fcoe_sw_shost_config() - Sets up fc_lport->host * @lp : ptr to the fc_lport * @shost : ptr to the associated scsi host * @dev : device associated to scsi host * - * Must be called after fcoe_sw_lport_config) and fcoe_sw_netdev_config() + * Must be called after fcoe_sw_lport_config() and fcoe_sw_netdev_config() * * Returns : 0 for success - * */ static int fcoe_sw_shost_config(struct fc_lport *lp, struct Scsi_Host *shost, struct device *dev) @@ -261,8 +257,8 @@ static int fcoe_sw_shost_config(struct fc_lport *lp, struct Scsi_Host *shost, return 0; } -/* - * fcoe_sw_em_config - allocates em for this lport +/** + * fcoe_sw_em_config() - allocates em for this lport * @lp: the port that em is to allocated for * * Returns : 0 on success @@ -279,8 +275,8 @@ static inline int fcoe_sw_em_config(struct fc_lport *lp) return 0; } -/* - * fcoe_sw_destroy - FCoE software HBA tear-down function +/** + * fcoe_sw_destroy() - FCoE software HBA tear-down function * @netdev: ptr to the associated net_device * * Returns: 0 if link is OK for use by FCoE. @@ -353,8 +349,8 @@ static struct libfc_function_template fcoe_sw_libfc_fcn_templ = { .frame_send = fcoe_xmit, }; -/* - * fcoe_sw_create - this function creates the fcoe interface +/** + * fcoe_sw_create() - this function creates the fcoe interface * @netdev: pointer the associated netdevice * * Creates fc_lport struct and scsi_host for lport, configures lport @@ -440,8 +436,8 @@ out_host_put: return rc; } -/* - * fcoe_sw_match - the fcoe sw transport match function +/** + * fcoe_sw_match() - The FCoE SW transport match function * * Returns : false always */ @@ -461,8 +457,8 @@ struct fcoe_transport fcoe_sw_transport = { .device = 0xffff, }; -/* - * fcoe_sw_init - registers fcoe_sw_transport +/** + * fcoe_sw_init() - Registers fcoe_sw_transport * * Returns : 0 on success */ @@ -480,8 +476,8 @@ int __init fcoe_sw_init(void) return 0; } -/* - * fcoe_sw_exit - unregisters fcoe_sw_transport +/** + * fcoe_sw_exit() - Unregisters fcoe_sw_transport * * Returns : 0 on success */ diff --git a/drivers/scsi/fcoe/libfcoe.c b/drivers/scsi/fcoe/libfcoe.c index 296071043f55..5e68652985b2 100644 --- a/drivers/scsi/fcoe/libfcoe.c +++ b/drivers/scsi/fcoe/libfcoe.c @@ -91,13 +91,13 @@ static struct notifier_block fcoe_cpu_notifier = { }; /** - * fcoe_create_percpu_data - creates the associated cpu data + * fcoe_create_percpu_data() - creates the associated cpu data * @cpu: index for the cpu where fcoe cpu data will be created * * create percpu stats block, from cpu add notifier * * Returns: none - **/ + */ static void fcoe_create_percpu_data(int cpu) { struct fc_lport *lp; @@ -115,13 +115,13 @@ static void fcoe_create_percpu_data(int cpu) } /** - * fcoe_destroy_percpu_data - destroys the associated cpu data + * fcoe_destroy_percpu_data() - destroys the associated cpu data * @cpu: index for the cpu where fcoe cpu data will destroyed * * destroy percpu stats block called by cpu add/remove notifier * * Retuns: none - **/ + */ static void fcoe_destroy_percpu_data(int cpu) { struct fc_lport *lp; @@ -137,7 +137,7 @@ static void fcoe_destroy_percpu_data(int cpu) } /** - * fcoe_cpu_callback - fcoe cpu hotplug event callback + * fcoe_cpu_callback() - fcoe cpu hotplug event callback * @nfb: callback data block * @action: event triggering the callback * @hcpu: index for the cpu of this event @@ -145,7 +145,7 @@ static void fcoe_destroy_percpu_data(int cpu) * this creates or destroys per cpu data for fcoe * * Returns NOTIFY_OK always. - **/ + */ static int fcoe_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { @@ -166,7 +166,7 @@ static int fcoe_cpu_callback(struct notifier_block *nfb, unsigned long action, #endif /* CONFIG_HOTPLUG_CPU */ /** - * fcoe_rcv - this is the fcoe receive function called by NET_RX_SOFTIRQ + * fcoe_rcv() - this is the fcoe receive function called by NET_RX_SOFTIRQ * @skb: the receive skb * @dev: associated net device * @ptype: context @@ -175,7 +175,7 @@ static int fcoe_cpu_callback(struct notifier_block *nfb, unsigned long action, * this function will receive the packet and build fc frame and pass it up * * Returns: 0 for success - **/ + */ int fcoe_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *ptype, struct net_device *olddev) { @@ -265,11 +265,11 @@ err2: EXPORT_SYMBOL_GPL(fcoe_rcv); /** - * fcoe_start_io - pass to netdev to start xmit for fcoe + * fcoe_start_io() - pass to netdev to start xmit for fcoe * @skb: the skb to be xmitted * * Returns: 0 for success - **/ + */ static inline int fcoe_start_io(struct sk_buff *skb) { int rc; @@ -283,12 +283,12 @@ static inline int fcoe_start_io(struct sk_buff *skb) } /** - * fcoe_get_paged_crc_eof - in case we need alloc a page for crc_eof + * fcoe_get_paged_crc_eof() - in case we need alloc a page for crc_eof * @skb: the skb to be xmitted * @tlen: total len * * Returns: 0 for success - **/ + */ static int fcoe_get_paged_crc_eof(struct sk_buff *skb, int tlen) { struct fcoe_percpu_s *fps; @@ -326,13 +326,12 @@ static int fcoe_get_paged_crc_eof(struct sk_buff *skb, int tlen) } /** - * fcoe_fc_crc - calculates FC CRC in this fcoe skb + * fcoe_fc_crc() - calculates FC CRC in this fcoe skb * @fp: the fc_frame containg data to be checksummed * * This uses crc32() to calculate the crc for fc frame * Return : 32 bit crc - * - **/ + */ u32 fcoe_fc_crc(struct fc_frame *fp) { struct sk_buff *skb = fp_skb(fp); @@ -363,13 +362,12 @@ u32 fcoe_fc_crc(struct fc_frame *fp) EXPORT_SYMBOL_GPL(fcoe_fc_crc); /** - * fcoe_xmit - FCoE frame transmit function + * fcoe_xmit() - FCoE frame transmit function * @lp: the associated local port * @fp: the fc_frame to be transmitted * * Return : 0 for success - * - **/ + */ int fcoe_xmit(struct fc_lport *lp, struct fc_frame *fp) { int wlen, rc = 0; @@ -511,12 +509,11 @@ int fcoe_xmit(struct fc_lport *lp, struct fc_frame *fp) } EXPORT_SYMBOL_GPL(fcoe_xmit); -/* - * fcoe_percpu_receive_thread - recv thread per cpu +/** + * fcoe_percpu_receive_thread() - recv thread per cpu * @arg: ptr to the fcoe per cpu struct * * Return: 0 for success - * */ int fcoe_percpu_receive_thread(void *arg) { @@ -658,7 +655,7 @@ int fcoe_percpu_receive_thread(void *arg) } /** - * fcoe_recv_flogi - flogi receive function + * fcoe_recv_flogi() - flogi receive function * @fc: associated fcoe_softc * @fp: the recieved frame * @sa: the source address of this flogi @@ -667,7 +664,7 @@ int fcoe_percpu_receive_thread(void *arg) * mac address for the initiator, eitehr OUI based or GW based. * * Returns: none - **/ + */ static void fcoe_recv_flogi(struct fcoe_softc *fc, struct fc_frame *fp, u8 *sa) { struct fc_frame_header *fh; @@ -715,7 +712,7 @@ static void fcoe_recv_flogi(struct fcoe_softc *fc, struct fc_frame *fp, u8 *sa) } /** - * fcoe_watchdog - fcoe timer callback + * fcoe_watchdog() - fcoe timer callback * @vp: * * This checks the pending queue length for fcoe and set lport qfull @@ -723,7 +720,7 @@ static void fcoe_recv_flogi(struct fcoe_softc *fc, struct fc_frame *fp, u8 *sa) * fcoe_hostlist. * * Returns: 0 for success - **/ + */ void fcoe_watchdog(ulong vp) { struct fc_lport *lp; @@ -750,7 +747,7 @@ void fcoe_watchdog(ulong vp) /** - * fcoe_check_wait_queue - put the skb into fcoe pending xmit queue + * fcoe_check_wait_queue() - put the skb into fcoe pending xmit queue * @lp: the fc_port for this skb * @skb: the associated skb to be xmitted * @@ -764,7 +761,7 @@ void fcoe_watchdog(ulong vp) * by the next skb transmit. * * Returns: 0 for success - **/ + */ static int fcoe_check_wait_queue(struct fc_lport *lp) { int rc; @@ -797,12 +794,12 @@ static int fcoe_check_wait_queue(struct fc_lport *lp) } /** - * fcoe_insert_wait_queue_head - puts skb to fcoe pending queue head + * fcoe_insert_wait_queue_head() - puts skb to fcoe pending queue head * @lp: the fc_port for this skb * @skb: the associated skb to be xmitted * * Returns: none - **/ + */ static void fcoe_insert_wait_queue_head(struct fc_lport *lp, struct sk_buff *skb) { @@ -815,12 +812,12 @@ static void fcoe_insert_wait_queue_head(struct fc_lport *lp, } /** - * fcoe_insert_wait_queue - put the skb into fcoe pending queue tail + * fcoe_insert_wait_queue() - put the skb into fcoe pending queue tail * @lp: the fc_port for this skb * @skb: the associated skb to be xmitted * * Returns: none - **/ + */ static void fcoe_insert_wait_queue(struct fc_lport *lp, struct sk_buff *skb) { @@ -833,10 +830,9 @@ static void fcoe_insert_wait_queue(struct fc_lport *lp, } /** - * fcoe_dev_setup - setup link change notification interface - * - **/ -static void fcoe_dev_setup(void) + * fcoe_dev_setup() - setup link change notification interface + */ +static void fcoe_dev_setup() { /* * here setup a interface specific wd time to @@ -846,15 +842,15 @@ static void fcoe_dev_setup(void) } /** - * fcoe_dev_setup - cleanup link change notification interface - **/ + * fcoe_dev_setup() - cleanup link change notification interface + */ static void fcoe_dev_cleanup(void) { unregister_netdevice_notifier(&fcoe_notifier); } /** - * fcoe_device_notification - netdev event notification callback + * fcoe_device_notification() - netdev event notification callback * @notifier: context of the notification * @event: type of event * @ptr: fixed array for output parsed ifname @@ -862,7 +858,7 @@ static void fcoe_dev_cleanup(void) * This function is called by the ethernet driver in case of link change event * * Returns: 0 for success - **/ + */ static int fcoe_device_notification(struct notifier_block *notifier, ulong event, void *ptr) { @@ -926,12 +922,12 @@ out: } /** - * fcoe_if_to_netdev - parse a name buffer to get netdev + * fcoe_if_to_netdev() - parse a name buffer to get netdev * @ifname: fixed array for output parsed ifname * @buffer: incoming buffer to be copied * * Returns: NULL or ptr to netdeive - **/ + */ static struct net_device *fcoe_if_to_netdev(const char *buffer) { char *cp; @@ -948,11 +944,11 @@ static struct net_device *fcoe_if_to_netdev(const char *buffer) } /** - * fcoe_netdev_to_module_owner - finds out the nic drive moddule of the netdev + * fcoe_netdev_to_module_owner() - finds out the nic drive moddule of the netdev * @netdev: the target netdev * * Returns: ptr to the struct module, NULL for failure - **/ + */ static struct module *fcoe_netdev_to_module_owner( const struct net_device *netdev) { @@ -972,12 +968,14 @@ static struct module *fcoe_netdev_to_module_owner( } /** - * fcoe_ethdrv_get - holds the nic driver module by try_module_get() for - * the corresponding netdev. + * fcoe_ethdrv_get() - Hold the Ethernet driver * @netdev: the target netdev * + * Holds the Ethernet driver module by try_module_get() for + * the corresponding netdev. + * * Returns: 0 for succsss - **/ + */ static int fcoe_ethdrv_get(const struct net_device *netdev) { struct module *owner; @@ -992,12 +990,14 @@ static int fcoe_ethdrv_get(const struct net_device *netdev) } /** - * fcoe_ethdrv_get - releases the nic driver module by module_put for - * the corresponding netdev. + * fcoe_ethdrv_put() - Release the Ethernet driver * @netdev: the target netdev * + * Releases the Ethernet driver module by module_put for + * the corresponding netdev. + * * Returns: 0 for succsss - **/ + */ static int fcoe_ethdrv_put(const struct net_device *netdev) { struct module *owner; @@ -1013,12 +1013,12 @@ static int fcoe_ethdrv_put(const struct net_device *netdev) } /** - * fcoe_destroy- handles the destroy from sysfs + * fcoe_destroy() - handles the destroy from sysfs * @buffer: expcted to be a eth if name * @kp: associated kernel param * * Returns: 0 for success - **/ + */ static int fcoe_destroy(const char *buffer, struct kernel_param *kp) { int rc; @@ -1051,12 +1051,12 @@ out_nodev: } /** - * fcoe_create - handles the create call from sysfs + * fcoe_create() - Handles the create call from sysfs * @buffer: expcted to be a eth if name * @kp: associated kernel param * * Returns: 0 for success - **/ + */ static int fcoe_create(const char *buffer, struct kernel_param *kp) { int rc; @@ -1097,8 +1097,8 @@ module_param_call(destroy, fcoe_destroy, NULL, NULL, S_IWUSR); __MODULE_PARM_TYPE(destroy, "string"); MODULE_PARM_DESC(destroy, "Destroy fcoe port"); -/* - * fcoe_link_ok - check if link is ok for the fc_lport +/** + * fcoe_link_ok() - Check if link is ok for the fc_lport * @lp: ptr to the fc_lport * * Any permanently-disqualifying conditions have been previously checked. @@ -1142,9 +1142,8 @@ int fcoe_link_ok(struct fc_lport *lp) } EXPORT_SYMBOL_GPL(fcoe_link_ok); -/* - * fcoe_percpu_clean - frees skb of the corresponding lport from the per - * cpu queue. +/** + * fcoe_percpu_clean() - Clear the pending skbs for an lport * @lp: the fc_lport */ void fcoe_percpu_clean(struct fc_lport *lp) @@ -1178,11 +1177,11 @@ void fcoe_percpu_clean(struct fc_lport *lp) EXPORT_SYMBOL_GPL(fcoe_percpu_clean); /** - * fcoe_clean_pending_queue - dequeue skb and free it + * fcoe_clean_pending_queue() - Dequeue a skb and free it * @lp: the corresponding fc_lport * * Returns: none - **/ + */ void fcoe_clean_pending_queue(struct fc_lport *lp) { struct fcoe_softc *fc = lport_priv(lp); @@ -1199,12 +1198,12 @@ void fcoe_clean_pending_queue(struct fc_lport *lp) EXPORT_SYMBOL_GPL(fcoe_clean_pending_queue); /** - * libfc_host_alloc - allocate a Scsi_Host with room for the fc_lport + * libfc_host_alloc() - Allocate a Scsi_Host with room for the fc_lport * @sht: ptr to the scsi host templ * @priv_size: size of private data after fc_lport * * Returns: ptr to Scsi_Host - * TODO - to libfc? + * TODO: to libfc? */ static inline struct Scsi_Host *libfc_host_alloc( struct scsi_host_template *sht, int priv_size) @@ -1213,7 +1212,7 @@ static inline struct Scsi_Host *libfc_host_alloc( } /** - * fcoe_host_alloc - allocate a Scsi_Host with room for the fcoe_softc + * fcoe_host_alloc() - Allocate a Scsi_Host with room for the fcoe_softc * @sht: ptr to the scsi host templ * @priv_size: size of private data after fc_lport * @@ -1225,8 +1224,8 @@ struct Scsi_Host *fcoe_host_alloc(struct scsi_host_template *sht, int priv_size) } EXPORT_SYMBOL_GPL(fcoe_host_alloc); -/* - * fcoe_reset - resets the fcoe +/** + * fcoe_reset() - Resets the fcoe * @shost: shost the reset is from * * Returns: always 0 @@ -1239,8 +1238,8 @@ int fcoe_reset(struct Scsi_Host *shost) } EXPORT_SYMBOL_GPL(fcoe_reset); -/* - * fcoe_wwn_from_mac - converts 48-bit IEEE MAC address to 64-bit FC WWN. +/** + * fcoe_wwn_from_mac() - Converts 48-bit IEEE MAC address to 64-bit FC WWN. * @mac: mac address * @scheme: check port * @port: port indicator for converting @@ -1279,8 +1278,9 @@ u64 fcoe_wwn_from_mac(unsigned char mac[MAX_ADDR_LEN], return wwn; } EXPORT_SYMBOL_GPL(fcoe_wwn_from_mac); -/* - * fcoe_hostlist_lookup_softc - find the corresponding lport by a given device + +/** + * fcoe_hostlist_lookup_softc() - find the corresponding lport by a given device * @device: this is currently ptr to net_device * * Returns: NULL or the located fcoe_softc @@ -1301,8 +1301,8 @@ static struct fcoe_softc *fcoe_hostlist_lookup_softc( return NULL; } -/* - * fcoe_hostlist_lookup - find the corresponding lport by netdev +/** + * fcoe_hostlist_lookup() - Find the corresponding lport by netdev * @netdev: ptr to net_device * * Returns: 0 for success @@ -1317,8 +1317,8 @@ struct fc_lport *fcoe_hostlist_lookup(const struct net_device *netdev) } EXPORT_SYMBOL_GPL(fcoe_hostlist_lookup); -/* - * fcoe_hostlist_add - add a lport to lports list +/** + * fcoe_hostlist_add() - Add a lport to lports list * @lp: ptr to the fc_lport to badded * * Returns: 0 for success @@ -1338,8 +1338,8 @@ int fcoe_hostlist_add(const struct fc_lport *lp) } EXPORT_SYMBOL_GPL(fcoe_hostlist_add); -/* - * fcoe_hostlist_remove - remove a lport from lports list +/** + * fcoe_hostlist_remove() - remove a lport from lports list * @lp: ptr to the fc_lport to badded * * Returns: 0 for success @@ -1359,12 +1359,12 @@ int fcoe_hostlist_remove(const struct fc_lport *lp) EXPORT_SYMBOL_GPL(fcoe_hostlist_remove); /** - * fcoe_libfc_config - sets up libfc related properties for lport + * fcoe_libfc_config() - sets up libfc related properties for lport * @lp: ptr to the fc_lport * @tt: libfc function template * * Returns : 0 for success - **/ + */ int fcoe_libfc_config(struct fc_lport *lp, struct libfc_function_template *tt) { /* Set the function pointers set by the LLDD */ @@ -1382,14 +1382,14 @@ int fcoe_libfc_config(struct fc_lport *lp, struct libfc_function_template *tt) EXPORT_SYMBOL_GPL(fcoe_libfc_config); /** - * fcoe_init - fcoe module loading initialization + * fcoe_init() - fcoe module loading initialization * * Initialization routine * 1. Will create fc transport software structure * 2. initialize the link list of port information structure * * Returns 0 on success, negative on failure - **/ + */ static int __init fcoe_init(void) { int cpu; @@ -1452,10 +1452,10 @@ static int __init fcoe_init(void) module_init(fcoe_init); /** - * fcoe_exit - fcoe module unloading cleanup + * fcoe_exit() - fcoe module unloading cleanup * * Returns 0 on success, negative on failure - **/ + */ static void __exit fcoe_exit(void) { u32 idx; diff --git a/drivers/scsi/libfc/fc_disc.c b/drivers/scsi/libfc/fc_disc.c index cfbce893941a..e57556ea5b48 100644 --- a/drivers/scsi/libfc/fc_disc.c +++ b/drivers/scsi/libfc/fc_disc.c @@ -64,7 +64,7 @@ static void fc_disc_single(struct fc_disc *, struct fc_disc_port *); static void fc_disc_restart(struct fc_disc *); /** - * fc_disc_lookup_rport - lookup a remote port by port_id + * fc_disc_lookup_rport() - lookup a remote port by port_id * @lport: Fibre Channel host port instance * @port_id: remote port port_id to match */ @@ -92,7 +92,7 @@ struct fc_rport *fc_disc_lookup_rport(const struct fc_lport *lport, } /** - * fc_disc_stop_rports - delete all the remote ports associated with the lport + * fc_disc_stop_rports() - delete all the remote ports associated with the lport * @disc: The discovery job to stop rports on * * Locking Note: This function expects that the lport mutex is locked before @@ -117,7 +117,7 @@ void fc_disc_stop_rports(struct fc_disc *disc) } /** - * fc_disc_rport_callback - Event handler for rport events + * fc_disc_rport_callback() - Event handler for rport events * @lport: The lport which is receiving the event * @rport: The rport which the event has occured on * @event: The event that occured @@ -151,7 +151,7 @@ static void fc_disc_rport_callback(struct fc_lport *lport, } /** - * fc_disc_recv_rscn_req - Handle Registered State Change Notification (RSCN) + * fc_disc_recv_rscn_req() - Handle Registered State Change Notification (RSCN) * @sp: Current sequence of the RSCN exchange * @fp: RSCN Frame * @lport: Fibre Channel host port instance @@ -265,7 +265,7 @@ reject: } /** - * fc_disc_recv_req - Handle incoming requests + * fc_disc_recv_req() - Handle incoming requests * @sp: Current sequence of the request exchange * @fp: The frame * @lport: The FC local port @@ -294,7 +294,7 @@ static void fc_disc_recv_req(struct fc_seq *sp, struct fc_frame *fp, } /** - * fc_disc_restart - Restart discovery + * fc_disc_restart() - Restart discovery * @lport: FC discovery context * * Locking Note: This function expects that the disc mutex @@ -322,7 +322,7 @@ static void fc_disc_restart(struct fc_disc *disc) } /** - * fc_disc_start - Fibre Channel Target discovery + * fc_disc_start() - Fibre Channel Target discovery * @lport: FC local port * * Returns non-zero if discovery cannot be started. @@ -383,7 +383,7 @@ static struct fc_rport_operations fc_disc_rport_ops = { }; /** - * fc_disc_new_target - Handle new target found by discovery + * fc_disc_new_target() - Handle new target found by discovery * @lport: FC local port * @rport: The previous FC remote port (NULL if new remote port) * @ids: Identifiers for the new FC remote port @@ -446,7 +446,7 @@ static int fc_disc_new_target(struct fc_disc *disc, } /** - * fc_disc_del_target - Delete a target + * fc_disc_del_target() - Delete a target * @disc: FC discovery context * @rport: The remote port to be removed */ @@ -459,7 +459,7 @@ static void fc_disc_del_target(struct fc_disc *disc, struct fc_rport *rport) } /** - * fc_disc_done - Discovery has been completed + * fc_disc_done() - Discovery has been completed * @disc: FC discovery context */ static void fc_disc_done(struct fc_disc *disc) @@ -479,7 +479,7 @@ static void fc_disc_done(struct fc_disc *disc) } /** - * fc_disc_error - Handle error on dNS request + * fc_disc_error() - Handle error on dNS request * @disc: FC discovery context * @fp: The frame pointer */ @@ -519,7 +519,7 @@ static void fc_disc_error(struct fc_disc *disc, struct fc_frame *fp) } /** - * fc_disc_gpn_ft_req - Send Get Port Names by FC-4 type (GPN_FT) request + * fc_disc_gpn_ft_req() - Send Get Port Names by FC-4 type (GPN_FT) request * @lport: FC discovery context * * Locking Note: This function expects that the disc_mutex is locked @@ -553,7 +553,7 @@ err: } /** - * fc_disc_gpn_ft_parse - Parse the list of IDs and names resulting from a request + * fc_disc_gpn_ft_parse() - Parse the list of IDs and names resulting from a request * @lport: Fibre Channel host port instance * @buf: GPN_FT response buffer * @len: size of response buffer @@ -658,7 +658,10 @@ static int fc_disc_gpn_ft_parse(struct fc_disc *disc, void *buf, size_t len) return error; } -/* +/** + * fc_disc_timeout() - Retry handler for the disc component + * @work: Structure holding disc obj that needs retry discovery + * * Handle retry of memory allocation for remote ports. */ static void fc_disc_timeout(struct work_struct *work) @@ -673,7 +676,7 @@ static void fc_disc_timeout(struct work_struct *work) } /** - * fc_disc_gpn_ft_resp - Handle a response frame from Get Port Names (GPN_FT) + * fc_disc_gpn_ft_resp() - Handle a response frame from Get Port Names (GPN_FT) * @sp: Current sequence of GPN_FT exchange * @fp: response frame * @lp_arg: Fibre Channel host port instance @@ -712,9 +715,7 @@ static void fc_disc_gpn_ft_resp(struct fc_seq *sp, struct fc_frame *fp, fr_len(fp)); } else if (ntohs(cp->ct_cmd) == FC_FS_ACC) { - /* - * Accepted. Parse response. - */ + /* Accepted, parse the response. */ buf = cp + 1; len -= sizeof(*cp); } else if (ntohs(cp->ct_cmd) == FC_FS_RJT) { @@ -746,7 +747,7 @@ static void fc_disc_gpn_ft_resp(struct fc_seq *sp, struct fc_frame *fp, } /** - * fc_disc_single - Discover the directory information for a single target + * fc_disc_single() - Discover the directory information for a single target * @lport: FC local port * @dp: The port to rediscover * @@ -782,7 +783,7 @@ out: } /** - * fc_disc_stop - Stop discovery for a given lport + * fc_disc_stop() - Stop discovery for a given lport * @lport: The lport that discovery should stop for */ void fc_disc_stop(struct fc_lport *lport) @@ -796,7 +797,7 @@ void fc_disc_stop(struct fc_lport *lport) } /** - * fc_disc_stop_final - Stop discovery for a given lport + * fc_disc_stop_final() - Stop discovery for a given lport * @lport: The lport that discovery should stop for * * This function will block until discovery has been @@ -809,7 +810,7 @@ void fc_disc_stop_final(struct fc_lport *lport) } /** - * fc_disc_init - Initialize the discovery block + * fc_disc_init() - Initialize the discovery block * @lport: FC local port */ int fc_disc_init(struct fc_lport *lport) diff --git a/drivers/scsi/libfc/fc_fcp.c b/drivers/scsi/libfc/fc_fcp.c index ecc72612c385..2a631d7dbcec 100644 --- a/drivers/scsi/libfc/fc_fcp.c +++ b/drivers/scsi/libfc/fc_fcp.c @@ -161,7 +161,7 @@ static struct fc_fcp_pkt *fc_fcp_pkt_alloc(struct fc_lport *lp, gfp_t gfp) } /** - * fc_fcp_pkt_release - release hold on scsi_pkt packet + * fc_fcp_pkt_release() - release hold on scsi_pkt packet * @fsp: fcp packet struct * * This is used by upper layer scsi driver. @@ -183,8 +183,7 @@ static void fc_fcp_pkt_hold(struct fc_fcp_pkt *fsp) } /** - * fc_fcp_pkt_destory - release hold on scsi_pkt packet - * + * fc_fcp_pkt_destory() - release hold on scsi_pkt packet * @seq: exchange sequence * @fsp: fcp packet struct * @@ -199,7 +198,7 @@ static void fc_fcp_pkt_destroy(struct fc_seq *seq, void *fsp) } /** - * fc_fcp_lock_pkt - lock a packet and get a ref to it. + * fc_fcp_lock_pkt() - lock a packet and get a ref to it. * @fsp: fcp packet * * We should only return error if we return a command to scsi-ml before @@ -291,9 +290,7 @@ static void fc_fcp_recv_data(struct fc_fcp_pkt *fsp, struct fc_frame *fp) buf = fc_frame_payload_get(fp, 0); if (offset + len > fsp->data_len) { - /* - * this should never happen - */ + /* this should never happen */ if ((fr_flags(fp) & FCPHF_CRC_UNCHECKED) && fc_frame_crc_check(fp)) goto crc_err; @@ -387,8 +384,8 @@ crc_err: fc_fcp_complete_locked(fsp); } -/* - * fc_fcp_send_data - Send SCSI data to target. +/** + * fc_fcp_send_data() - Send SCSI data to target. * @fsp: ptr to fc_fcp_pkt * @sp: ptr to this sequence * @offset: starting offset for this data request @@ -610,8 +607,8 @@ static void fc_fcp_abts_resp(struct fc_fcp_pkt *fsp, struct fc_frame *fp) } } -/* - * fc_fcp_reduce_can_queue - drop can_queue +/** + * fc_fcp_reduce_can_queue() - drop can_queue * @lp: lport to drop queueing for * * If we are getting memory allocation failures, then we may @@ -642,9 +639,11 @@ done: spin_unlock_irqrestore(lp->host->host_lock, flags); } -/* - * exch mgr calls this routine to process scsi - * exchanges. +/** + * fc_fcp_recv() - Reveive FCP frames + * @seq: The sequence the frame is on + * @fp: The FC frame + * @arg: The related FCP packet * * Return : None * Context : called from Soft IRQ context @@ -832,7 +831,7 @@ err: } /** - * fc_fcp_complete_locked - complete processing of a fcp packet + * fc_fcp_complete_locked() - complete processing of a fcp packet * @fsp: fcp packet * * This function may sleep if a timer is pending. The packet lock must be @@ -900,7 +899,7 @@ static void fc_fcp_cleanup_cmd(struct fc_fcp_pkt *fsp, int error) } /** - * fc_fcp_cleanup_each_cmd - run fn on each active command + * fc_fcp_cleanup_each_cmd() - Cleanup active commads * @lp: logical port * @id: target id * @lun: lun @@ -952,7 +951,7 @@ static void fc_fcp_abort_io(struct fc_lport *lp) } /** - * fc_fcp_pkt_send - send a fcp packet to the lower level. + * fc_fcp_pkt_send() - send a fcp packet to the lower level. * @lp: fc lport * @fsp: fc packet. * @@ -1727,7 +1726,7 @@ out: EXPORT_SYMBOL(fc_queuecommand); /** - * fc_io_compl - Handle responses for completed commands + * fc_io_compl() - Handle responses for completed commands * @fsp: scsi packet * * Translates a error to a Linux SCSI error. @@ -1857,7 +1856,7 @@ static void fc_io_compl(struct fc_fcp_pkt *fsp) } /** - * fc_fcp_complete - complete processing of a fcp packet + * fc_fcp_complete() - complete processing of a fcp packet * @fsp: fcp packet * * This function may sleep if a fsp timer is pending. @@ -1874,9 +1873,10 @@ void fc_fcp_complete(struct fc_fcp_pkt *fsp) EXPORT_SYMBOL(fc_fcp_complete); /** - * fc_eh_abort - Abort a command...from scsi host template + * fc_eh_abort() - Abort a command * @sc_cmd: scsi command to abort * + * From scsi host template. * send ABTS to the target device and wait for the response * sc_cmd is the pointer to the command to be aborted. */ @@ -1920,7 +1920,7 @@ release_pkt: EXPORT_SYMBOL(fc_eh_abort); /** - * fc_eh_device_reset: Reset a single LUN + * fc_eh_device_reset() Reset a single LUN * @sc_cmd: scsi command * * Set from scsi host template to send tm cmd to the target and wait for the @@ -1973,7 +1973,7 @@ out: EXPORT_SYMBOL(fc_eh_device_reset); /** - * fc_eh_host_reset - The reset function will reset the ports on the host. + * fc_eh_host_reset() - The reset function will reset the ports on the host. * @sc_cmd: scsi command */ int fc_eh_host_reset(struct scsi_cmnd *sc_cmd) @@ -1999,7 +1999,7 @@ int fc_eh_host_reset(struct scsi_cmnd *sc_cmd) EXPORT_SYMBOL(fc_eh_host_reset); /** - * fc_slave_alloc - configure queue depth + * fc_slave_alloc() - configure queue depth * @sdev: scsi device * * Configures queue depth based on host's cmd_per_len. If not set diff --git a/drivers/scsi/libfc/fc_lport.c b/drivers/scsi/libfc/fc_lport.c index c00de2244c78..2ae50a1188e6 100644 --- a/drivers/scsi/libfc/fc_lport.c +++ b/drivers/scsi/libfc/fc_lport.c @@ -139,7 +139,7 @@ static int fc_frame_drop(struct fc_lport *lport, struct fc_frame *fp) } /** - * fc_lport_rport_callback - Event handler for rport events + * fc_lport_rport_callback() - Event handler for rport events * @lport: The lport which is receiving the event * @rport: The rport which the event has occured on * @event: The event that occured @@ -195,7 +195,7 @@ static void fc_lport_rport_callback(struct fc_lport *lport, } /** - * fc_lport_state - Return a string which represents the lport's state + * fc_lport_state() - Return a string which represents the lport's state * @lport: The lport whose state is to converted to a string */ static const char *fc_lport_state(struct fc_lport *lport) @@ -209,7 +209,7 @@ static const char *fc_lport_state(struct fc_lport *lport) } /** - * fc_lport_ptp_setup - Create an rport for point-to-point mode + * fc_lport_ptp_setup() - Create an rport for point-to-point mode * @lport: The lport to attach the ptp rport to * @fid: The FID of the ptp rport * @remote_wwpn: The WWPN of the ptp rport @@ -351,7 +351,7 @@ static void fc_lport_add_fc4_type(struct fc_lport *lport, enum fc_fh_type type) } /** - * fc_lport_recv_rlir_req - Handle received Registered Link Incident Report. + * fc_lport_recv_rlir_req() - Handle received Registered Link Incident Report. * @lport: Fibre Channel local port recieving the RLIR * @sp: current sequence in the RLIR exchange * @fp: RLIR request frame @@ -370,7 +370,7 @@ static void fc_lport_recv_rlir_req(struct fc_seq *sp, struct fc_frame *fp, } /** - * fc_lport_recv_echo_req - Handle received ECHO request + * fc_lport_recv_echo_req() - Handle received ECHO request * @lport: Fibre Channel local port recieving the ECHO * @sp: current sequence in the ECHO exchange * @fp: ECHO request frame @@ -412,7 +412,7 @@ static void fc_lport_recv_echo_req(struct fc_seq *sp, struct fc_frame *in_fp, } /** - * fc_lport_recv_echo_req - Handle received Request Node ID data request + * fc_lport_recv_echo_req() - Handle received Request Node ID data request * @lport: Fibre Channel local port recieving the RNID * @sp: current sequence in the RNID exchange * @fp: RNID request frame @@ -479,7 +479,7 @@ static void fc_lport_recv_rnid_req(struct fc_seq *sp, struct fc_frame *in_fp, } /** - * fc_lport_recv_adisc_req - Handle received Address Discovery Request + * fc_lport_recv_adisc_req() - Handle received Address Discovery Request * @lport: Fibre Channel local port recieving the ADISC * @sp: current sequence in the ADISC exchange * @fp: ADISC request frame @@ -529,7 +529,7 @@ static void fc_lport_recv_adisc_req(struct fc_seq *sp, struct fc_frame *in_fp, } /** - * fc_lport_recv_logo_req - Handle received fabric LOGO request + * fc_lport_recv_logo_req() - Handle received fabric LOGO request * @lport: Fibre Channel local port recieving the LOGO * @sp: current sequence in the LOGO exchange * @fp: LOGO request frame @@ -546,7 +546,7 @@ static void fc_lport_recv_logo_req(struct fc_seq *sp, struct fc_frame *fp, } /** - * fc_fabric_login - Start the lport state machine + * fc_fabric_login() - Start the lport state machine * @lport: The lport that should log into the fabric * * Locking Note: This function should not be called @@ -568,7 +568,7 @@ int fc_fabric_login(struct fc_lport *lport) EXPORT_SYMBOL(fc_fabric_login); /** - * fc_linkup - Handler for transport linkup events + * fc_linkup() - Handler for transport linkup events * @lport: The lport whose link is up */ void fc_linkup(struct fc_lport *lport) @@ -588,7 +588,7 @@ void fc_linkup(struct fc_lport *lport) EXPORT_SYMBOL(fc_linkup); /** - * fc_linkdown - Handler for transport linkdown events + * fc_linkdown() - Handler for transport linkdown events * @lport: The lport whose link is down */ void fc_linkdown(struct fc_lport *lport) @@ -607,12 +607,12 @@ void fc_linkdown(struct fc_lport *lport) EXPORT_SYMBOL(fc_linkdown); /** - * fc_fabric_logoff - Logout of the fabric + * fc_fabric_logoff() - Logout of the fabric * @lport: fc_lport pointer to logoff the fabric * * Return value: * 0 for success, -1 for failure - **/ + */ int fc_fabric_logoff(struct fc_lport *lport) { lport->tt.disc_stop_final(lport); @@ -625,7 +625,7 @@ int fc_fabric_logoff(struct fc_lport *lport) EXPORT_SYMBOL(fc_fabric_logoff); /** - * fc_lport_destroy - unregister a fc_lport + * fc_lport_destroy() - unregister a fc_lport * @lport: fc_lport pointer to unregister * * Return value: @@ -635,7 +635,7 @@ EXPORT_SYMBOL(fc_fabric_logoff); * clean-up all the allocated memory * and free up other system resources. * - **/ + */ int fc_lport_destroy(struct fc_lport *lport) { lport->tt.frame_send = fc_frame_drop; @@ -646,15 +646,14 @@ int fc_lport_destroy(struct fc_lport *lport) EXPORT_SYMBOL(fc_lport_destroy); /** - * fc_set_mfs - sets up the mfs for the corresponding fc_lport + * fc_set_mfs() - sets up the mfs for the corresponding fc_lport * @lport: fc_lport pointer to unregister * @mfs: the new mfs for fc_lport * * Set mfs for the given fc_lport to the new mfs. * * Return: 0 for success - * - **/ + */ int fc_set_mfs(struct fc_lport *lport, u32 mfs) { unsigned int old_mfs; @@ -683,7 +682,7 @@ int fc_set_mfs(struct fc_lport *lport, u32 mfs) EXPORT_SYMBOL(fc_set_mfs); /** - * fc_lport_disc_callback - Callback for discovery events + * fc_lport_disc_callback() - Callback for discovery events * @lport: FC local port * @event: The discovery event */ @@ -708,7 +707,7 @@ void fc_lport_disc_callback(struct fc_lport *lport, enum fc_disc_event event) } /** - * fc_rport_enter_ready - Enter the ready state and start discovery + * fc_rport_enter_ready() - Enter the ready state and start discovery * @lport: Fibre Channel local port that is ready * * Locking Note: The lport lock is expected to be held before calling @@ -725,7 +724,7 @@ static void fc_lport_enter_ready(struct fc_lport *lport) } /** - * fc_lport_recv_flogi_req - Receive a FLOGI request + * fc_lport_recv_flogi_req() - Receive a FLOGI request * @sp_in: The sequence the FLOGI is on * @rx_fp: The frame the FLOGI is in * @lport: The lport that recieved the request @@ -815,7 +814,7 @@ out: } /** - * fc_lport_recv_req - The generic lport request handler + * fc_lport_recv_req() - The generic lport request handler * @lport: The lport that received the request * @sp: The sequence the request is on * @fp: The frame the request is in @@ -911,7 +910,7 @@ static void fc_lport_recv_req(struct fc_lport *lport, struct fc_seq *sp, } /** - * fc_lport_reset - Reset an lport + * fc_lport_reset() - Reset an lport * @lport: The lport which should be reset * * Locking Note: This functions should not be called with the @@ -928,7 +927,7 @@ int fc_lport_reset(struct fc_lport *lport) EXPORT_SYMBOL(fc_lport_reset); /** - * fc_rport_enter_reset - Reset the local port + * fc_rport_enter_reset() - Reset the local port * @lport: Fibre Channel local port to be reset * * Locking Note: The lport lock is expected to be held before calling @@ -960,7 +959,7 @@ static void fc_lport_enter_reset(struct fc_lport *lport) } /** - * fc_lport_error - Handler for any errors + * fc_lport_error() - Handler for any errors * @lport: The fc_lport object * @fp: The frame pointer * @@ -1007,8 +1006,8 @@ static void fc_lport_error(struct fc_lport *lport, struct fc_frame *fp) } /** - * fc_lport_rft_id_resp - Handle response to Register Fibre - * Channel Types by ID (RPN_ID) request + * fc_lport_rft_id_resp() - Handle response to Register Fibre + * Channel Types by ID (RPN_ID) request * @sp: current sequence in RPN_ID exchange * @fp: response frame * @lp_arg: Fibre Channel host port instance @@ -1059,8 +1058,8 @@ err: } /** - * fc_lport_rpn_id_resp - Handle response to Register Port - * Name by ID (RPN_ID) request + * fc_lport_rpn_id_resp() - Handle response to Register Port + * Name by ID (RPN_ID) request * @sp: current sequence in RPN_ID exchange * @fp: response frame * @lp_arg: Fibre Channel host port instance @@ -1111,7 +1110,7 @@ err: } /** - * fc_lport_scr_resp - Handle response to State Change Register (SCR) request + * fc_lport_scr_resp() - Handle response to State Change Register (SCR) request * @sp: current sequence in SCR exchange * @fp: response frame * @lp_arg: Fibre Channel lport port instance that sent the registration request @@ -1157,7 +1156,7 @@ err: } /** - * fc_lport_enter_scr - Send a State Change Register (SCR) request + * fc_lport_enter_scr() - Send a State Change Register (SCR) request * @lport: Fibre Channel local port to register for state changes * * Locking Note: The lport lock is expected to be held before calling @@ -1184,7 +1183,7 @@ static void fc_lport_enter_scr(struct fc_lport *lport) } /** - * fc_lport_enter_rft_id - Register FC4-types with the name server + * fc_lport_enter_rft_id() - Register FC4-types with the name server * @lport: Fibre Channel local port to register * * Locking Note: The lport lock is expected to be held before calling @@ -1226,7 +1225,7 @@ static void fc_lport_enter_rft_id(struct fc_lport *lport) } /** - * fc_rport_enter_rft_id - Register port name with the name server + * fc_rport_enter_rft_id() - Register port name with the name server * @lport: Fibre Channel local port to register * * Locking Note: The lport lock is expected to be held before calling @@ -1259,7 +1258,7 @@ static struct fc_rport_operations fc_lport_rport_ops = { }; /** - * fc_rport_enter_dns - Create a rport to the name server + * fc_rport_enter_dns() - Create a rport to the name server * @lport: Fibre Channel local port requesting a rport for the name server * * Locking Note: The lport lock is expected to be held before calling @@ -1296,7 +1295,7 @@ err: } /** - * fc_lport_timeout - Handler for the retry_work timer. + * fc_lport_timeout() - Handler for the retry_work timer. * @work: The work struct of the fc_lport */ static void fc_lport_timeout(struct work_struct *work) @@ -1337,7 +1336,7 @@ static void fc_lport_timeout(struct work_struct *work) } /** - * fc_lport_logo_resp - Handle response to LOGO request + * fc_lport_logo_resp() - Handle response to LOGO request * @sp: current sequence in LOGO exchange * @fp: response frame * @lp_arg: Fibre Channel lport port instance that sent the LOGO request @@ -1383,7 +1382,7 @@ err: } /** - * fc_rport_enter_logo - Logout of the fabric + * fc_rport_enter_logo() - Logout of the fabric * @lport: Fibre Channel local port to be logged out * * Locking Note: The lport lock is expected to be held before calling @@ -1415,7 +1414,7 @@ static void fc_lport_enter_logo(struct fc_lport *lport) } /** - * fc_lport_flogi_resp - Handle response to FLOGI request + * fc_lport_flogi_resp() - Handle response to FLOGI request * @sp: current sequence in FLOGI exchange * @fp: response frame * @lp_arg: Fibre Channel lport port instance that sent the FLOGI request @@ -1510,7 +1509,7 @@ err: } /** - * fc_rport_enter_flogi - Send a FLOGI request to the fabric manager + * fc_rport_enter_flogi() - Send a FLOGI request to the fabric manager * @lport: Fibre Channel local port to be logged in to the fabric * * Locking Note: The lport lock is expected to be held before calling diff --git a/drivers/scsi/libfc/fc_rport.c b/drivers/scsi/libfc/fc_rport.c index 4f23a9bb15a1..6f07de15c491 100644 --- a/drivers/scsi/libfc/fc_rport.c +++ b/drivers/scsi/libfc/fc_rport.c @@ -146,7 +146,7 @@ struct fc_rport *fc_rport_rogue_create(struct fc_disc_port *dp) } /** - * fc_rport_state - return a string for the state the rport is in + * fc_rport_state() - return a string for the state the rport is in * @rport: The rport whose state we want to get a string for */ static const char *fc_rport_state(struct fc_rport *rport) @@ -161,7 +161,7 @@ static const char *fc_rport_state(struct fc_rport *rport) } /** - * fc_set_rport_loss_tmo - Set the remote port loss timeout in seconds. + * fc_set_rport_loss_tmo() - Set the remote port loss timeout in seconds. * @rport: Pointer to Fibre Channel remote port structure * @timeout: timeout in seconds */ @@ -175,7 +175,7 @@ void fc_set_rport_loss_tmo(struct fc_rport *rport, u32 timeout) EXPORT_SYMBOL(fc_set_rport_loss_tmo); /** - * fc_plogi_get_maxframe - Get max payload from the common service parameters + * fc_plogi_get_maxframe() - Get max payload from the common service parameters * @flp: FLOGI payload structure * @maxval: upper limit, may be less than what is in the service parameters */ @@ -198,7 +198,7 @@ fc_plogi_get_maxframe(struct fc_els_flogi *flp, unsigned int maxval) } /** - * fc_rport_state_enter - Change the rport's state + * fc_rport_state_enter() - Change the rport's state * @rport: The rport whose state should change * @new: The new state of the rport * @@ -292,7 +292,7 @@ static void fc_rport_work(struct work_struct *work) } /** - * fc_rport_login - Start the remote port login state machine + * fc_rport_login() - Start the remote port login state machine * @rport: Fibre Channel remote port * * Locking Note: Called without the rport lock held. This @@ -315,7 +315,7 @@ int fc_rport_login(struct fc_rport *rport) } /** - * fc_rport_logoff - Logoff and remove an rport + * fc_rport_logoff() - Logoff and remove an rport * @rport: Fibre Channel remote port to be removed * * Locking Note: Called without the rport lock held. This @@ -353,7 +353,7 @@ int fc_rport_logoff(struct fc_rport *rport) } /** - * fc_rport_enter_ready - The rport is ready + * fc_rport_enter_ready() - The rport is ready * @rport: Fibre Channel remote port that is ready * * Locking Note: The rport lock is expected to be held before calling @@ -372,7 +372,7 @@ static void fc_rport_enter_ready(struct fc_rport *rport) } /** - * fc_rport_timeout - Handler for the retry_work timer. + * fc_rport_timeout() - Handler for the retry_work timer. * @work: The work struct of the fc_rport_libfc_priv * * Locking Note: Called without the rport lock held. This @@ -411,7 +411,7 @@ static void fc_rport_timeout(struct work_struct *work) } /** - * fc_rport_error - Error handler, called once retries have been exhausted + * fc_rport_error() - Error handler, called once retries have been exhausted * @rport: The fc_rport object * @fp: The frame pointer * @@ -444,7 +444,7 @@ static void fc_rport_error(struct fc_rport *rport, struct fc_frame *fp) } /** - * fc_rport_error_retry - Error handler when retries are desired + * fc_rport_error_retry() - Error handler when retries are desired * @rport: The fc_rport object * @fp: The frame pointer * @@ -479,7 +479,7 @@ static void fc_rport_error_retry(struct fc_rport *rport, struct fc_frame *fp) } /** - * fc_rport_plogi_recv_resp - Handle incoming ELS PLOGI response + * fc_rport_plogi_recv_resp() - Handle incoming ELS PLOGI response * @sp: current sequence in the PLOGI exchange * @fp: response frame * @rp_arg: Fibre Channel remote port @@ -554,7 +554,7 @@ err: } /** - * fc_rport_enter_plogi - Send Port Login (PLOGI) request to peer + * fc_rport_enter_plogi() - Send Port Login (PLOGI) request to peer * @rport: Fibre Channel remote port to send PLOGI to * * Locking Note: The rport lock is expected to be held before calling @@ -587,7 +587,7 @@ static void fc_rport_enter_plogi(struct fc_rport *rport) } /** - * fc_rport_prli_resp - Process Login (PRLI) response handler + * fc_rport_prli_resp() - Process Login (PRLI) response handler * @sp: current sequence in the PRLI exchange * @fp: response frame * @rp_arg: Fibre Channel remote port @@ -657,7 +657,7 @@ err: } /** - * fc_rport_logo_resp - Logout (LOGO) response handler + * fc_rport_logo_resp() - Logout (LOGO) response handler * @sp: current sequence in the LOGO exchange * @fp: response frame * @rp_arg: Fibre Channel remote port @@ -706,7 +706,7 @@ err: } /** - * fc_rport_enter_prli - Send Process Login (PRLI) request to peer + * fc_rport_enter_prli() - Send Process Login (PRLI) request to peer * @rport: Fibre Channel remote port to send PRLI to * * Locking Note: The rport lock is expected to be held before calling @@ -741,7 +741,7 @@ static void fc_rport_enter_prli(struct fc_rport *rport) } /** - * fc_rport_els_rtv_resp - Request Timeout Value response handler + * fc_rport_els_rtv_resp() - Request Timeout Value response handler * @sp: current sequence in the RTV exchange * @fp: response frame * @rp_arg: Fibre Channel remote port @@ -807,7 +807,7 @@ err: } /** - * fc_rport_enter_rtv - Send Request Timeout Value (RTV) request to peer + * fc_rport_enter_rtv() - Send Request Timeout Value (RTV) request to peer * @rport: Fibre Channel remote port to send RTV to * * Locking Note: The rport lock is expected to be held before calling @@ -838,7 +838,7 @@ static void fc_rport_enter_rtv(struct fc_rport *rport) } /** - * fc_rport_enter_logo - Send Logout (LOGO) request to peer + * fc_rport_enter_logo() - Send Logout (LOGO) request to peer * @rport: Fibre Channel remote port to send LOGO to * * Locking Note: The rport lock is expected to be held before calling @@ -870,7 +870,7 @@ static void fc_rport_enter_logo(struct fc_rport *rport) /** - * fc_rport_recv_req - Receive a request from a rport + * fc_rport_recv_req() - Receive a request from a rport * @sp: current sequence in the PLOGI exchange * @fp: response frame * @rp_arg: Fibre Channel remote port @@ -931,7 +931,7 @@ void fc_rport_recv_req(struct fc_seq *sp, struct fc_frame *fp, } /** - * fc_rport_recv_plogi_req - Handle incoming Port Login (PLOGI) request + * fc_rport_recv_plogi_req() - Handle incoming Port Login (PLOGI) request * @rport: Fibre Channel remote port that initiated PLOGI * @sp: current sequence in the PLOGI exchange * @fp: PLOGI request frame @@ -1053,7 +1053,7 @@ static void fc_rport_recv_plogi_req(struct fc_rport *rport, } /** - * fc_rport_recv_prli_req - Handle incoming Process Login (PRLI) request + * fc_rport_recv_prli_req() - Handle incoming Process Login (PRLI) request * @rport: Fibre Channel remote port that initiated PRLI * @sp: current sequence in the PRLI exchange * @fp: PRLI request frame @@ -1204,7 +1204,7 @@ static void fc_rport_recv_prli_req(struct fc_rport *rport, } /** - * fc_rport_recv_prlo_req - Handle incoming Process Logout (PRLO) request + * fc_rport_recv_prlo_req() - Handle incoming Process Logout (PRLO) request * @rport: Fibre Channel remote port that initiated PRLO * @sp: current sequence in the PRLO exchange * @fp: PRLO request frame @@ -1235,7 +1235,7 @@ static void fc_rport_recv_prlo_req(struct fc_rport *rport, struct fc_seq *sp, } /** - * fc_rport_recv_logo_req - Handle incoming Logout (LOGO) request + * fc_rport_recv_logo_req() - Handle incoming Logout (LOGO) request * @rport: Fibre Channel remote port that initiated LOGO * @sp: current sequence in the LOGO exchange * @fp: LOGO request frame diff --git a/include/scsi/libfc.h b/include/scsi/libfc.h index 282829cdf352..a2e126b86e3e 100644 --- a/include/scsi/libfc.h +++ b/include/scsi/libfc.h @@ -665,7 +665,7 @@ struct fc_lport { struct delayed_work disc_work; }; -/** +/* * FC_LPORT HELPER FUNCTIONS *****************************/ static inline void *lport_priv(const struct fc_lport *lp) @@ -697,7 +697,7 @@ static inline void fc_lport_state_enter(struct fc_lport *lp, } -/** +/* * LOCAL PORT LAYER *****************************/ int fc_lport_init(struct fc_lport *lp); @@ -747,19 +747,19 @@ int fc_lport_reset(struct fc_lport *); int fc_set_mfs(struct fc_lport *lp, u32 mfs); -/** +/* * REMOTE PORT LAYER *****************************/ int fc_rport_init(struct fc_lport *lp); void fc_rport_terminate_io(struct fc_rport *rp); -/** +/* * DISCOVERY LAYER *****************************/ int fc_disc_init(struct fc_lport *lp); -/** +/* * SCSI LAYER *****************************/ /* @@ -820,7 +820,7 @@ int fc_change_queue_type(struct scsi_device *sdev, int tag_type); */ void fc_fcp_destroy(struct fc_lport *); -/** +/* * ELS/CT interface *****************************/ /* @@ -829,7 +829,7 @@ void fc_fcp_destroy(struct fc_lport *); int fc_elsct_init(struct fc_lport *lp); -/** +/* * EXCHANGE MANAGER LAYER *****************************/ /* -- cgit v1.2.3-70-g09d2 From fc47ff6b1b27fb736f255ed8cd490356e0cd228f Mon Sep 17 00:00:00 2001 From: Robert Love Date: Fri, 27 Feb 2009 10:55:55 -0800 Subject: [SCSI] libfc, fcoe: Remove unnecessary cast by removing inline wrapper Comment from "Andrew Morton " > +{ > + return (struct fcoe_softc *)lport_priv(lp); unneeded/undesirable cast of void*. There are probably zillions of instances of this - there always are. This whole inline function was unnecessary. The FCoE layer knows that it's data structure is stored in the lport private data, it can just access it from lport_priv(). Signed-off-by: Robert Love Signed-off-by: James Bottomley --- drivers/scsi/fcoe/fcoe_sw.c | 2 +- drivers/scsi/fcoe/libfcoe.c | 12 ++++++------ include/scsi/libfcoe.h | 8 +------- 3 files changed, 8 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/drivers/scsi/fcoe/fcoe_sw.c b/drivers/scsi/fcoe/fcoe_sw.c index 007d1fbae808..f667dce1bbba 100644 --- a/drivers/scsi/fcoe/fcoe_sw.c +++ b/drivers/scsi/fcoe/fcoe_sw.c @@ -297,7 +297,7 @@ static int fcoe_sw_destroy(struct net_device *netdev) if (!lp) return -ENODEV; - fc = fcoe_softc(lp); + fc = lport_priv(lp); /* Logout of the fabric */ fc_fabric_logoff(lp); diff --git a/drivers/scsi/fcoe/libfcoe.c b/drivers/scsi/fcoe/libfcoe.c index d006d6576817..02f044a87159 100644 --- a/drivers/scsi/fcoe/libfcoe.c +++ b/drivers/scsi/fcoe/libfcoe.c @@ -387,7 +387,7 @@ int fcoe_xmit(struct fc_lport *lp, struct fc_frame *fp) WARN_ON((fr_len(fp) % sizeof(u32)) != 0); - fc = fcoe_softc(lp); + fc = lport_priv(lp); /* * if it is a flogi then we need to learn gw-addr * and my own fcid @@ -768,7 +768,7 @@ static int fcoe_check_wait_queue(struct fc_lport *lp) struct sk_buff *skb; struct fcoe_softc *fc; - fc = fcoe_softc(lp); + fc = lport_priv(lp); spin_lock_bh(&fc->fcoe_pending_queue.lock); /* @@ -805,7 +805,7 @@ static void fcoe_insert_wait_queue_head(struct fc_lport *lp, { struct fcoe_softc *fc; - fc = fcoe_softc(lp); + fc = lport_priv(lp); spin_lock_bh(&fc->fcoe_pending_queue.lock); __skb_queue_head(&fc->fcoe_pending_queue, skb); spin_unlock_bh(&fc->fcoe_pending_queue.lock); @@ -823,7 +823,7 @@ static void fcoe_insert_wait_queue(struct fc_lport *lp, { struct fcoe_softc *fc; - fc = fcoe_softc(lp); + fc = lport_priv(lp); spin_lock_bh(&fc->fcoe_pending_queue.lock); __skb_queue_tail(&fc->fcoe_pending_queue, skb); spin_unlock_bh(&fc->fcoe_pending_queue.lock); @@ -1113,7 +1113,7 @@ MODULE_PARM_DESC(destroy, "Destroy fcoe port"); */ int fcoe_link_ok(struct fc_lport *lp) { - struct fcoe_softc *fc = fcoe_softc(lp); + struct fcoe_softc *fc = lport_priv(lp); struct net_device *dev = fc->real_dev; struct ethtool_cmd ecmd = { ETHTOOL_GSET }; int rc = 0; @@ -1329,7 +1329,7 @@ int fcoe_hostlist_add(const struct fc_lport *lp) fc = fcoe_hostlist_lookup_softc(fcoe_netdev(lp)); if (!fc) { - fc = fcoe_softc(lp); + fc = lport_priv(lp); write_lock_bh(&fcoe_hostlist_lock); list_add_tail(&fc->list, &fcoe_hostlist); write_unlock_bh(&fcoe_hostlist_lock); diff --git a/include/scsi/libfcoe.h b/include/scsi/libfcoe.h index 89fdbb9a6a1b..f43d3833a7a8 100644 --- a/include/scsi/libfcoe.h +++ b/include/scsi/libfcoe.h @@ -58,16 +58,10 @@ struct fcoe_softc { u8 address_mode; }; -static inline struct fcoe_softc *fcoe_softc( - const struct fc_lport *lp) -{ - return (struct fcoe_softc *)lport_priv(lp); -} - static inline struct net_device *fcoe_netdev( const struct fc_lport *lp) { - return fcoe_softc(lp)->real_dev; + return ((struct fcoe_softc *)lport_priv(lp))->real_dev; } static inline struct fcoe_hdr *skb_fcoe_header(const struct sk_buff *skb) -- cgit v1.2.3-70-g09d2 From 74846bf85e42b4ba851d3f0dd67aa1df245895f9 Mon Sep 17 00:00:00 2001 From: Yi Zou Date: Fri, 27 Feb 2009 10:56:16 -0800 Subject: [SCSI] fcoe: ETH_P_8021Q is already in if_ether and fcoe is not using it anyway Signed-off-by: Yi Zou Signed-off-by: Robert Love Signed-off-by: James Bottomley --- include/scsi/fc/fc_fcoe.h | 4 ---- 1 file changed, 4 deletions(-) (limited to 'include') diff --git a/include/scsi/fc/fc_fcoe.h b/include/scsi/fc/fc_fcoe.h index 57aaa8f0d613..f271d9cc0fc2 100644 --- a/include/scsi/fc/fc_fcoe.h +++ b/include/scsi/fc/fc_fcoe.h @@ -31,10 +31,6 @@ #define ETH_P_FCOE 0x8906 /* FCOE ether type */ #endif -#ifndef ETH_P_8021Q -#define ETH_P_8021Q 0x8100 -#endif - /* * FC_FCOE_OUI hasn't been standardized yet. XXX TBD. */ -- cgit v1.2.3-70-g09d2 From c826a3145736e3baabebccfd0aecfbb6dae059f2 Mon Sep 17 00:00:00 2001 From: Vasu Dev Date: Fri, 27 Feb 2009 10:56:27 -0800 Subject: [SCSI] fcoe: Out of order tx frames was causing several check condition SCSI status frames followed by these errors in log. [sdp] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK [sdp] Sense Key : Aborted Command [current] [sdp] Add. Sense: Data phase error This was causing some test apps to exit due to write failure under heavy load. This was due to a race around adding and removing tx frame skb in fcoe_pending_queue, Chris Leech helped me to find that brief unlocking period when pulling skb from fcoe_pending_queue in various contexts (fcoe_watchdog and fcoe_xmit) and then adding skb back into fcoe_pending_queue up on a failed fcoe_start_io could change skb/tx frame order in fcoe_pending_queue. Thanks Chris. This patch allows only single context to pull skb from fcoe_pending_queue at any time to prevent above described ordering issue/race by use of fcoe_pending_queue_active flag. This patch simplified fcoe_watchdog with modified fcoe_check_wait_queue by use of FCOE_LOW_QUEUE_DEPTH instead previously used several conditionals to clear and set lp->qfull. I think FCOE_MAX_QUEUE_DEPTH with FCOE_LOW_QUEUE_DEPTH will work better in re/setting lp->qfull and these could be fine tuned for performance. Signed-off-by: Vasu Dev Signed-off-by: Robert Love Signed-off-by: James Bottomley --- drivers/scsi/fcoe/fcoe_sw.c | 1 + drivers/scsi/fcoe/libfcoe.c | 45 +++++++++++++++++++++------------------------ include/scsi/libfcoe.h | 1 + 3 files changed, 23 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/drivers/scsi/fcoe/fcoe_sw.c b/drivers/scsi/fcoe/fcoe_sw.c index 37d359db1648..da210eba1941 100644 --- a/drivers/scsi/fcoe/fcoe_sw.c +++ b/drivers/scsi/fcoe/fcoe_sw.c @@ -188,6 +188,7 @@ static int fcoe_sw_netdev_config(struct fc_lport *lp, struct net_device *netdev) skb_queue_head_init(&fc->fcoe_pending_queue); + fc->fcoe_pending_queue_active = 0; /* setup Source Mac Address */ memcpy(fc->ctl_src_addr, fc->real_dev->dev_addr, diff --git a/drivers/scsi/fcoe/libfcoe.c b/drivers/scsi/fcoe/libfcoe.c index 7265e0937995..14dd8a0402b6 100644 --- a/drivers/scsi/fcoe/libfcoe.c +++ b/drivers/scsi/fcoe/libfcoe.c @@ -49,6 +49,7 @@ static int debug_fcoe; #define FCOE_MAX_QUEUE_DEPTH 256 +#define FCOE_LOW_QUEUE_DEPTH 32 /* destination address mode */ #define FCOE_GW_ADDR_MODE 0x00 @@ -723,21 +724,12 @@ static void fcoe_recv_flogi(struct fcoe_softc *fc, struct fc_frame *fp, u8 *sa) */ void fcoe_watchdog(ulong vp) { - struct fc_lport *lp; struct fcoe_softc *fc; - int qfilled = 0; read_lock(&fcoe_hostlist_lock); list_for_each_entry(fc, &fcoe_hostlist, list) { - lp = fc->lp; - if (lp) { - if (fc->fcoe_pending_queue.qlen > FCOE_MAX_QUEUE_DEPTH) - qfilled = 1; - if (fcoe_check_wait_queue(lp) < FCOE_MAX_QUEUE_DEPTH) { - if (qfilled) - lp->qfull = 0; - } - } + if (fc->lp) + fcoe_check_wait_queue(fc->lp); } read_unlock(&fcoe_hostlist_lock); @@ -753,8 +745,8 @@ void fcoe_watchdog(ulong vp) * * This empties the wait_queue, dequeue the head of the wait_queue queue * and calls fcoe_start_io() for each packet, if all skb have been - * transmitted, return 0 if a error occurs, then restore wait_queue and - * try again later. + * transmitted, return qlen or -1 if a error occurs, then restore + * wait_queue and try again later. * * The wait_queue is used when the skb transmit fails. skb will go * in the wait_queue which will be emptied by the time function OR @@ -764,33 +756,38 @@ void fcoe_watchdog(ulong vp) */ static int fcoe_check_wait_queue(struct fc_lport *lp) { - int rc; struct sk_buff *skb; struct fcoe_softc *fc; + int rc = -1; fc = lport_priv(lp); spin_lock_bh(&fc->fcoe_pending_queue.lock); - /* - * if interface pending queue full then set qfull in lport. - */ - if (fc->fcoe_pending_queue.qlen > FCOE_MAX_QUEUE_DEPTH) - lp->qfull = 1; + if (fc->fcoe_pending_queue_active) + goto out; + fc->fcoe_pending_queue_active = 1; if (fc->fcoe_pending_queue.qlen) { while ((skb = __skb_dequeue(&fc->fcoe_pending_queue)) != NULL) { spin_unlock_bh(&fc->fcoe_pending_queue.lock); rc = fcoe_start_io(skb); - if (rc) { + if (rc) fcoe_insert_wait_queue_head(lp, skb); - return rc; - } spin_lock_bh(&fc->fcoe_pending_queue.lock); + if (rc) + break; } - if (fc->fcoe_pending_queue.qlen < FCOE_MAX_QUEUE_DEPTH) + /* + * if interface pending queue is below FCOE_LOW_QUEUE_DEPTH + * then clear qfull flag. + */ + if (fc->fcoe_pending_queue.qlen < FCOE_LOW_QUEUE_DEPTH) lp->qfull = 0; } + fc->fcoe_pending_queue_active = 0; + rc = fc->fcoe_pending_queue.qlen; +out: spin_unlock_bh(&fc->fcoe_pending_queue.lock); - return fc->fcoe_pending_queue.qlen; + return rc; } /** diff --git a/include/scsi/libfcoe.h b/include/scsi/libfcoe.h index f43d3833a7a8..941818f29f59 100644 --- a/include/scsi/libfcoe.h +++ b/include/scsi/libfcoe.h @@ -46,6 +46,7 @@ struct fcoe_softc { struct net_device *phys_dev; /* device with ethtool_ops */ struct packet_type fcoe_packet_type; struct sk_buff_head fcoe_pending_queue; + u8 fcoe_pending_queue_active; u8 dest_addr[ETH_ALEN]; u8 ctl_src_addr[ETH_ALEN]; -- cgit v1.2.3-70-g09d2 From ae46141ff08f1965b17c531b571953c39ce8b9e2 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 10 Mar 2009 20:33:18 -0400 Subject: NFSv3: Fix posix ACL code Fix a memory leak due to allocation in the XDR layer. In cases where the RPC call needs to be retransmitted, we end up allocating new pages without clearing the old ones. Fix this by moving the allocation into nfs3_proc_setacls(). Also fix an issue discovered by Kevin Rudd, whereby the amount of memory reserved for the acls in the xdr_buf->head was miscalculated, and causing corruption. Signed-off-by: Trond Myklebust --- fs/nfs/nfs3acl.c | 27 +++++++++++++++++++++------ fs/nfs/nfs3xdr.c | 34 +++++++++++++--------------------- include/linux/nfs_xdr.h | 2 ++ include/linux/nfsacl.h | 3 +++ 4 files changed, 39 insertions(+), 27 deletions(-) (limited to 'include') diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index cef62557c87d..6bbf0e6daad2 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -292,7 +292,7 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, { struct nfs_server *server = NFS_SERVER(inode); struct nfs_fattr fattr; - struct page *pages[NFSACL_MAXPAGES] = { }; + struct page *pages[NFSACL_MAXPAGES]; struct nfs3_setaclargs args = { .inode = inode, .mask = NFS_ACL, @@ -303,7 +303,7 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, .rpc_argp = &args, .rpc_resp = &fattr, }; - int status, count; + int status; status = -EOPNOTSUPP; if (!nfs_server_capable(inode, NFS_CAP_ACLS)) @@ -319,6 +319,20 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, if (S_ISDIR(inode->i_mode)) { args.mask |= NFS_DFACL; args.acl_default = dfacl; + args.len = nfsacl_size(acl, dfacl); + } else + args.len = nfsacl_size(acl, NULL); + + if (args.len > NFS_ACL_INLINE_BUFSIZE) { + unsigned int npages = 1 + ((args.len - 1) >> PAGE_SHIFT); + + status = -ENOMEM; + do { + args.pages[args.npages] = alloc_page(GFP_KERNEL); + if (args.pages[args.npages] == NULL) + goto out_freepages; + args.npages++; + } while (args.npages < npages); } dprintk("NFS call setacl\n"); @@ -329,10 +343,6 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, nfs_zap_acl_cache(inode); dprintk("NFS reply setacl: %d\n", status); - /* pages may have been allocated at the xdr layer. */ - for (count = 0; count < NFSACL_MAXPAGES && args.pages[count]; count++) - __free_page(args.pages[count]); - switch (status) { case 0: status = nfs_refresh_inode(inode, &fattr); @@ -346,6 +356,11 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, case -ENOTSUPP: status = -EOPNOTSUPP; } +out_freepages: + while (args.npages != 0) { + args.npages--; + __free_page(args.pages[args.npages]); + } out: return status; } diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index 11cdddec1432..6cdeacffde46 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -82,8 +82,10 @@ #define NFS3_commitres_sz (1+NFS3_wcc_data_sz+2) #define ACL3_getaclargs_sz (NFS3_fh_sz+1) -#define ACL3_setaclargs_sz (NFS3_fh_sz+1+2*(2+5*3)) -#define ACL3_getaclres_sz (1+NFS3_post_op_attr_sz+1+2*(2+5*3)) +#define ACL3_setaclargs_sz (NFS3_fh_sz+1+ \ + XDR_QUADLEN(NFS_ACL_INLINE_BUFSIZE)) +#define ACL3_getaclres_sz (1+NFS3_post_op_attr_sz+1+ \ + XDR_QUADLEN(NFS_ACL_INLINE_BUFSIZE)) #define ACL3_setaclres_sz (1+NFS3_post_op_attr_sz) /* @@ -703,28 +705,18 @@ nfs3_xdr_setaclargs(struct rpc_rqst *req, __be32 *p, struct nfs3_setaclargs *args) { struct xdr_buf *buf = &req->rq_snd_buf; - unsigned int base, len_in_head, len = nfsacl_size( - (args->mask & NFS_ACL) ? args->acl_access : NULL, - (args->mask & NFS_DFACL) ? args->acl_default : NULL); - int count, err; + unsigned int base; + int err; p = xdr_encode_fhandle(p, NFS_FH(args->inode)); *p++ = htonl(args->mask); - base = (char *)p - (char *)buf->head->iov_base; - /* put as much of the acls into head as possible. */ - len_in_head = min_t(unsigned int, buf->head->iov_len - base, len); - len -= len_in_head; - req->rq_slen = xdr_adjust_iovec(req->rq_svec, p + (len_in_head >> 2)); - - for (count = 0; (count << PAGE_SHIFT) < len; count++) { - args->pages[count] = alloc_page(GFP_KERNEL); - if (!args->pages[count]) { - while (count) - __free_page(args->pages[--count]); - return -ENOMEM; - } - } - xdr_encode_pages(buf, args->pages, 0, len); + req->rq_slen = xdr_adjust_iovec(req->rq_svec, p); + base = req->rq_slen; + + if (args->npages != 0) + xdr_encode_pages(buf, args->pages, 0, args->len); + else + req->rq_slen += args->len; err = nfsacl_encode(buf, base, args->inode, (args->mask & NFS_ACL) ? diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index a550b528319f..2e5f00066afd 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -406,6 +406,8 @@ struct nfs3_setaclargs { int mask; struct posix_acl * acl_access; struct posix_acl * acl_default; + size_t len; + unsigned int npages; struct page ** pages; }; diff --git a/include/linux/nfsacl.h b/include/linux/nfsacl.h index 54487a99beb8..43011b69297c 100644 --- a/include/linux/nfsacl.h +++ b/include/linux/nfsacl.h @@ -37,6 +37,9 @@ #define NFSACL_MAXPAGES ((2*(8+12*NFS_ACL_MAX_ENTRIES) + PAGE_SIZE-1) \ >> PAGE_SHIFT) +#define NFS_ACL_MAX_ENTRIES_INLINE (5) +#define NFS_ACL_INLINE_BUFSIZE ((2*(2+3*NFS_ACL_MAX_ENTRIES_INLINE)) << 2) + static inline unsigned int nfsacl_size(struct posix_acl *acl_access, struct posix_acl *acl_default) { -- cgit v1.2.3-70-g09d2 From 78851e1aa4c3b796d5f0bb11b445016726302b44 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 10 Mar 2009 20:33:19 -0400 Subject: NLM: Shrink the IPv4-only version of nlm_cmp_addr() Clean up/micro-optimatization: Make the AF_INET-only version of nlm_cmp_addr() smaller. This matches the style of nlm_privileged_requester(), and makes the AF_INET-only version of nlm_cmp_addr() nearly the same size as it was before IPv6 support. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/lockd/lockd.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include') diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h index aa6fe7026de7..51855dfd8adb 100644 --- a/include/linux/lockd/lockd.h +++ b/include/linux/lockd/lockd.h @@ -346,6 +346,7 @@ static inline int __nlm_cmp_addr4(const struct sockaddr *sap1, return sin1->sin_addr.s_addr == sin2->sin_addr.s_addr; } +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) static inline int __nlm_cmp_addr6(const struct sockaddr *sap1, const struct sockaddr *sap2) { @@ -353,6 +354,13 @@ static inline int __nlm_cmp_addr6(const struct sockaddr *sap1, const struct sockaddr_in6 *sin2 = (const struct sockaddr_in6 *)sap2; return ipv6_addr_equal(&sin1->sin6_addr, &sin2->sin6_addr); } +#else /* !(CONFIG_IPV6 || CONFIG_IPV6_MODULE) */ +static inline int __nlm_cmp_addr6(const struct sockaddr *sap1, + const struct sockaddr *sap2) +{ + return 0; +} +#endif /* !(CONFIG_IPV6 || CONFIG_IPV6_MODULE) */ /* * Compare two host addresses -- cgit v1.2.3-70-g09d2 From 260cf8a2cbe022131a4850ea6175ed716dc2663c Mon Sep 17 00:00:00 2001 From: Pantelis Koukousoulas Date: Tue, 10 Mar 2009 13:16:14 +0200 Subject: drm: fix EDID parser problem with positive/negative hsync/vsync Comparing the layouts of struct detail_pixel_timing with x.org's struct detailed_timings and how those are handled, it appears that the hsync_positive and vsync_positive fields are backwards. This patch fixes https://bugs.freedesktop.org/show_bug.cgi?id=20019 for me. It was tested on 2 monitors, LG FLATRON L225WS 22" and a YAKUMO 17" for which more details are unknown. Signed-off-by: Pantelis Koukousoulas Signed-off-by: Dave Airlie --- include/drm/drm_edid.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/drm/drm_edid.h b/include/drm/drm_edid.h index ff8d27af4786..a11cc9d32591 100644 --- a/include/drm/drm_edid.h +++ b/include/drm/drm_edid.h @@ -69,8 +69,8 @@ struct detailed_pixel_timing { u8 hborder; u8 vborder; u8 unknown0:1; - u8 vsync_positive:1; u8 hsync_positive:1; + u8 vsync_positive:1; u8 separate_sync:2; u8 stereo:1; u8 unknown6:1; -- cgit v1.2.3-70-g09d2 From 76e6eee03353f01bfca707d4dbb1f10a4ee27dc0 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 12 Mar 2009 14:35:43 -0600 Subject: cpumask: tsk_cpumask for accessing the struct task_struct's cpus_allowed. This allows us to change the representation (to a dangling bitmap or cpumask_var_t) without breaking all the callers: they can use tsk_cpumask() now and won't see a difference as the changes roll into linux-next. Signed-off-by: Rusty Russell --- include/linux/sched.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 8c216e057c94..011db2f4c94c 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1419,6 +1419,9 @@ struct task_struct { #endif }; +/* Future-safe accessor for struct task_struct's cpus_allowed. */ +#define tsk_cpumask(tsk) (&(tsk)->cpus_allowed) + /* * Priority of a process goes from 0..MAX_PRIO-1, valid RT * priority is 0..MAX_RT_PRIO-1, and SCHED_NORMAL/SCHED_BATCH -- cgit v1.2.3-70-g09d2 From 45e575ab9bfada5a5ef1b6174f8e749b1ecf0864 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 12 Mar 2009 14:35:44 -0600 Subject: cpumask: mm_cpumask for accessing the struct mm_struct's cpu_vm_mask. This allows us to change the representation (to a dangling bitmap or cpumask_var_t) without breaking all the callers: they can use mm_cpumask() now and won't see a difference as the changes roll into linux-next. Signed-off-by: Rusty Russell --- include/linux/mm_types.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include') diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 92915e81443f..d84feb7bdbf0 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -276,4 +276,7 @@ struct mm_struct { #endif }; +/* Future-safe accessor for struct mm_struct's cpu_vm_mask. */ +#define mm_cpumask(mm) (&(mm)->cpu_vm_mask) + #endif /* _LINUX_MM_TYPES_H */ -- cgit v1.2.3-70-g09d2 From 446c92b2901bedb3725d29b4e73def8aba623ffc Mon Sep 17 00:00:00 2001 From: Uwe Kleine-König Date: Thu, 12 Mar 2009 18:03:16 +0100 Subject: [ARM] 5421/1: ftrace: fix crash due to tracing of __naked functions MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This is a fix for the following crash observed in 2.6.29-rc3: http://lkml.org/lkml/2009/1/29/150 On ARM it doesn't make sense to trace a naked function because then mcount is called without stack and frame pointer being set up and there is no chance to restore the lr register to the value before mcount was called. Reported-by: Matthias Kaehlcke Tested-by: Matthias Kaehlcke Cc: Abhishek Sagar Cc: Steven Rostedt Cc: Ingo Molnar Signed-off-by: Uwe Kleine-König Signed-off-by: Russell King --- arch/arm/kernel/fiq.c | 4 ++-- arch/arm/mm/copypage-feroceon.c | 2 +- arch/arm/mm/copypage-v3.c | 2 +- arch/arm/mm/copypage-v4mc.c | 2 +- arch/arm/mm/copypage-v4wb.c | 2 +- arch/arm/mm/copypage-v4wt.c | 2 +- arch/arm/mm/copypage-xsc3.c | 2 +- arch/arm/mm/copypage-xscale.c | 2 +- include/linux/compiler-gcc.h | 10 +++++++++- 9 files changed, 18 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/arch/arm/kernel/fiq.c b/arch/arm/kernel/fiq.c index 36f81d967979..6ff7919613d7 100644 --- a/arch/arm/kernel/fiq.c +++ b/arch/arm/kernel/fiq.c @@ -88,7 +88,7 @@ void set_fiq_handler(void *start, unsigned int length) * disable irqs for the duration. Note - these functions are almost * entirely coded in assembly. */ -void __attribute__((naked)) set_fiq_regs(struct pt_regs *regs) +void __naked set_fiq_regs(struct pt_regs *regs) { register unsigned long tmp; asm volatile ( @@ -106,7 +106,7 @@ void __attribute__((naked)) set_fiq_regs(struct pt_regs *regs) : "r" (®s->ARM_r8), "I" (PSR_I_BIT | PSR_F_BIT | FIQ_MODE)); } -void __attribute__((naked)) get_fiq_regs(struct pt_regs *regs) +void __naked get_fiq_regs(struct pt_regs *regs) { register unsigned long tmp; asm volatile ( diff --git a/arch/arm/mm/copypage-feroceon.c b/arch/arm/mm/copypage-feroceon.c index c3ba6a94da0c..70997d5bee2d 100644 --- a/arch/arm/mm/copypage-feroceon.c +++ b/arch/arm/mm/copypage-feroceon.c @@ -13,7 +13,7 @@ #include #include -static void __attribute__((naked)) +static void __naked feroceon_copy_user_page(void *kto, const void *kfrom) { asm("\ diff --git a/arch/arm/mm/copypage-v3.c b/arch/arm/mm/copypage-v3.c index 70ed96c8af8e..de9c06854ad7 100644 --- a/arch/arm/mm/copypage-v3.c +++ b/arch/arm/mm/copypage-v3.c @@ -15,7 +15,7 @@ * * FIXME: do we need to handle cache stuff... */ -static void __attribute__((naked)) +static void __naked v3_copy_user_page(void *kto, const void *kfrom) { asm("\n\ diff --git a/arch/arm/mm/copypage-v4mc.c b/arch/arm/mm/copypage-v4mc.c index 1601698b9800..7370a7142b04 100644 --- a/arch/arm/mm/copypage-v4mc.c +++ b/arch/arm/mm/copypage-v4mc.c @@ -44,7 +44,7 @@ static DEFINE_SPINLOCK(minicache_lock); * instruction. If your processor does not supply this, you have to write your * own copy_user_highpage that does the right thing. */ -static void __attribute__((naked)) +static void __naked mc_copy_user_page(void *from, void *to) { asm volatile( diff --git a/arch/arm/mm/copypage-v4wb.c b/arch/arm/mm/copypage-v4wb.c index 3ec93dab7656..9ab098414227 100644 --- a/arch/arm/mm/copypage-v4wb.c +++ b/arch/arm/mm/copypage-v4wb.c @@ -22,7 +22,7 @@ * instruction. If your processor does not supply this, you have to write your * own copy_user_highpage that does the right thing. */ -static void __attribute__((naked)) +static void __naked v4wb_copy_user_page(void *kto, const void *kfrom) { asm("\ diff --git a/arch/arm/mm/copypage-v4wt.c b/arch/arm/mm/copypage-v4wt.c index 0f1188efae45..300efafd6643 100644 --- a/arch/arm/mm/copypage-v4wt.c +++ b/arch/arm/mm/copypage-v4wt.c @@ -20,7 +20,7 @@ * dirty data in the cache. However, we do have to ensure that * subsequent reads are up to date. */ -static void __attribute__((naked)) +static void __naked v4wt_copy_user_page(void *kto, const void *kfrom) { asm("\ diff --git a/arch/arm/mm/copypage-xsc3.c b/arch/arm/mm/copypage-xsc3.c index 39a994542cad..bc4525f5ab23 100644 --- a/arch/arm/mm/copypage-xsc3.c +++ b/arch/arm/mm/copypage-xsc3.c @@ -29,7 +29,7 @@ * if we eventually end up using our copied page. * */ -static void __attribute__((naked)) +static void __naked xsc3_mc_copy_user_page(void *kto, const void *kfrom) { asm("\ diff --git a/arch/arm/mm/copypage-xscale.c b/arch/arm/mm/copypage-xscale.c index d18f2397ee2d..76824d3e966a 100644 --- a/arch/arm/mm/copypage-xscale.c +++ b/arch/arm/mm/copypage-xscale.c @@ -42,7 +42,7 @@ static DEFINE_SPINLOCK(minicache_lock); * Dcache aliasing issue. The writes will be forwarded to the write buffer, * and merged as appropriate. */ -static void __attribute__((naked)) +static void __naked mc_copy_user_page(void *from, void *to) { /* diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h index 1514d534deeb..a3ed7cb8ca34 100644 --- a/include/linux/compiler-gcc.h +++ b/include/linux/compiler-gcc.h @@ -52,7 +52,15 @@ #define __deprecated __attribute__((deprecated)) #define __packed __attribute__((packed)) #define __weak __attribute__((weak)) -#define __naked __attribute__((naked)) + +/* + * it doesn't make sense on ARM (currently the only user of __naked) to trace + * naked functions because then mcount is called without stack and frame pointer + * being set up and there is no chance to restore the lr register to the value + * before mcount was called. + */ +#define __naked __attribute__((naked)) notrace + #define __noreturn __attribute__((noreturn)) /* -- cgit v1.2.3-70-g09d2 From 5d82720a7f41f0c877e026c7d17e3bf20ccdbae0 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Fri, 13 Mar 2009 21:16:13 +0100 Subject: ide: save the returned value of dma_map_sg dma_map_sg could return a value different to 'nents' argument of dma_map_sg so the ide stack needs to save it for the later usage (e.g. for_each_sg). The ide stack also needs to save the original sg_nents value for pci_unmap_sg. Signed-off-by: FUJITA Tomonori [bart: backport to Linus' tree] Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ide-dma.c | 12 +++++++++--- include/linux/ide.h | 1 + 2 files changed, 10 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c index 72ebab0bc755..059c90bb5ad2 100644 --- a/drivers/ide/ide-dma.c +++ b/drivers/ide/ide-dma.c @@ -128,6 +128,7 @@ int ide_build_sglist(ide_drive_t *drive, struct request *rq) { ide_hwif_t *hwif = drive->hwif; struct scatterlist *sg = hwif->sg_table; + int i; ide_map_sg(drive, rq); @@ -136,8 +137,13 @@ int ide_build_sglist(ide_drive_t *drive, struct request *rq) else hwif->sg_dma_direction = DMA_TO_DEVICE; - return dma_map_sg(hwif->dev, sg, hwif->sg_nents, - hwif->sg_dma_direction); + i = dma_map_sg(hwif->dev, sg, hwif->sg_nents, hwif->sg_dma_direction); + if (i) { + hwif->orig_sg_nents = hwif->sg_nents; + hwif->sg_nents = i; + } + + return i; } EXPORT_SYMBOL_GPL(ide_build_sglist); @@ -156,7 +162,7 @@ void ide_destroy_dmatable(ide_drive_t *drive) { ide_hwif_t *hwif = drive->hwif; - dma_unmap_sg(hwif->dev, hwif->sg_table, hwif->sg_nents, + dma_unmap_sg(hwif->dev, hwif->sg_table, hwif->orig_sg_nents, hwif->sg_dma_direction); } EXPORT_SYMBOL_GPL(ide_destroy_dmatable); diff --git a/include/linux/ide.h b/include/linux/ide.h index e0cedfe9fad4..25087aead657 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -797,6 +797,7 @@ typedef struct hwif_s { struct scatterlist *sg_table; int sg_max_nents; /* Maximum number of entries in it */ int sg_nents; /* Current number of entries in it */ + int orig_sg_nents; int sg_dma_direction; /* dma transfer direction */ /* data phase of the active command (currently only valid for PIO/DMA) */ -- cgit v1.2.3-70-g09d2 From 87092698c665e0a358caf9825ae13114343027e8 Mon Sep 17 00:00:00 2001 From: un'ichi Nomura Date: Mon, 9 Mar 2009 10:40:52 +0100 Subject: block: Add gfp_mask parameter to bio_integrity_clone() Stricter gfp_mask might be required for clone allocation. For example, request-based dm may clone bio in interrupt context so it has to use GFP_ATOMIC. Signed-off-by: Kiyoshi Ueda Signed-off-by: Jun'ichi Nomura Acked-by: Martin K. Petersen Cc: Alasdair G Kergon Signed-off-by: Jens Axboe --- fs/bio-integrity.c | 5 +++-- fs/bio.c | 2 +- include/linux/bio.h | 4 ++-- 3 files changed, 6 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c index 549b0144da11..fe2b1aa2464e 100644 --- a/fs/bio-integrity.c +++ b/fs/bio-integrity.c @@ -685,19 +685,20 @@ EXPORT_SYMBOL(bio_integrity_split); * bio_integrity_clone - Callback for cloning bios with integrity metadata * @bio: New bio * @bio_src: Original bio + * @gfp_mask: Memory allocation mask * @bs: bio_set to allocate bip from * * Description: Called to allocate a bip when cloning a bio */ int bio_integrity_clone(struct bio *bio, struct bio *bio_src, - struct bio_set *bs) + gfp_t gfp_mask, struct bio_set *bs) { struct bio_integrity_payload *bip_src = bio_src->bi_integrity; struct bio_integrity_payload *bip; BUG_ON(bip_src == NULL); - bip = bio_integrity_alloc_bioset(bio, GFP_NOIO, bip_src->bip_vcnt, bs); + bip = bio_integrity_alloc_bioset(bio, gfp_mask, bip_src->bip_vcnt, bs); if (bip == NULL) return -EIO; diff --git a/fs/bio.c b/fs/bio.c index 124b95c4d582..cf747378b977 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -463,7 +463,7 @@ struct bio *bio_clone(struct bio *bio, gfp_t gfp_mask) if (bio_integrity(bio)) { int ret; - ret = bio_integrity_clone(b, bio, fs_bio_set); + ret = bio_integrity_clone(b, bio, gfp_mask, fs_bio_set); if (ret < 0) return NULL; diff --git a/include/linux/bio.h b/include/linux/bio.h index 1b16108a5417..d8bd43bfdcf5 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -531,7 +531,7 @@ extern void bio_integrity_endio(struct bio *, int); extern void bio_integrity_advance(struct bio *, unsigned int); extern void bio_integrity_trim(struct bio *, unsigned int, unsigned int); extern void bio_integrity_split(struct bio *, struct bio_pair *, int); -extern int bio_integrity_clone(struct bio *, struct bio *, struct bio_set *); +extern int bio_integrity_clone(struct bio *, struct bio *, gfp_t, struct bio_set *); extern int bioset_integrity_create(struct bio_set *, int); extern void bioset_integrity_free(struct bio_set *); extern void bio_integrity_init_slab(void); @@ -542,7 +542,7 @@ extern void bio_integrity_init_slab(void); #define bioset_integrity_create(a, b) (0) #define bio_integrity_prep(a) (0) #define bio_integrity_enabled(a) (0) -#define bio_integrity_clone(a, b, c) (0) +#define bio_integrity_clone(a, b, c,d ) (0) #define bioset_integrity_free(a) do { } while (0) #define bio_integrity_free(a, b) do { } while (0) #define bio_integrity_endio(a, b) do { } while (0) -- cgit v1.2.3-70-g09d2 From b1e93a68ca41e7e73766f95ba32ca05cf9052e15 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 16 Mar 2009 15:06:42 +0100 Subject: netfilter: conntrack: don't deliver events for racy packets This patch skips the delivery of conntrack events if the packet was drop due to a race condition in the conntrack insertion. Signed-off-by: Pablo Neira Ayuso Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_core.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index c25068e38516..5a449b44ba33 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -62,7 +62,8 @@ static inline int nf_conntrack_confirm(struct sk_buff *skb) if (ct && ct != &nf_conntrack_untracked) { if (!nf_ct_is_confirmed(ct) && !nf_ct_is_dying(ct)) ret = __nf_conntrack_confirm(skb); - nf_ct_deliver_cached_events(ct); + if (likely(ret == NF_ACCEPT)) + nf_ct_deliver_cached_events(ct); } return ret; } -- cgit v1.2.3-70-g09d2 From 76a67ec6fb79ff3570dcb5342142c16098299911 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Mon, 16 Mar 2009 18:34:20 -0400 Subject: nfsd: nfsd should drop CAP_MKNOD for non-root Since creating a device node is normally an operation requiring special privilege, Igor Zhbanov points out that it is surprising (to say the least) that a client can, for example, create a device node on a filesystem exported with root_squash. So, make sure CAP_MKNOD is among the capabilities dropped when an nfsd thread handles a request from a non-root user. Reported-by: Igor Zhbanov Cc: stable@kernel.org Signed-off-by: J. Bruce Fields --- include/linux/capability.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/capability.h b/include/linux/capability.h index 1b9872556131..4864a43b2b45 100644 --- a/include/linux/capability.h +++ b/include/linux/capability.h @@ -393,8 +393,10 @@ struct cpu_vfs_cap_data { # define CAP_FULL_SET ((kernel_cap_t){{ ~0, ~0 }}) # define CAP_INIT_EFF_SET ((kernel_cap_t){{ ~CAP_TO_MASK(CAP_SETPCAP), ~0 }}) # define CAP_FS_SET ((kernel_cap_t){{ CAP_FS_MASK_B0, CAP_FS_MASK_B1 } }) -# define CAP_NFSD_SET ((kernel_cap_t){{ CAP_FS_MASK_B0|CAP_TO_MASK(CAP_SYS_RESOURCE), \ - CAP_FS_MASK_B1 } }) +# define CAP_NFSD_SET ((kernel_cap_t){{ CAP_FS_MASK_B0 \ + | CAP_TO_MASK(CAP_SYS_RESOURCE) \ + | CAP_TO_MASK(CAP_MKNOD), \ + CAP_FS_MASK_B1 } }) #endif /* _KERNEL_CAPABILITY_U32S != 2 */ -- cgit v1.2.3-70-g09d2 From e2fc4d19292ef2eb208f76976ddc3320cc5839b6 Mon Sep 17 00:00:00 2001 From: Maciej Sosnowski Date: Sat, 21 Mar 2009 13:31:23 -0700 Subject: dca: add missing copyright/license headers In two dca files copyright and license headers are missing. This patch adds them there. Signed-off-by: Maciej Sosnowski Signed-off-by: David S. Miller --- drivers/dca/dca-sysfs.c | 21 +++++++++++++++++++++ include/linux/dca.h | 20 ++++++++++++++++++++ 2 files changed, 41 insertions(+) (limited to 'include') diff --git a/drivers/dca/dca-sysfs.c b/drivers/dca/dca-sysfs.c index bb538b9690e0..ee916c9857ee 100644 --- a/drivers/dca/dca-sysfs.c +++ b/drivers/dca/dca-sysfs.c @@ -1,3 +1,24 @@ +/* + * Copyright(c) 2007 - 2009 Intel Corporation. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 + * Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * The full GNU General Public License is included in this distribution in the + * file called COPYING. + */ + #include #include #include diff --git a/include/linux/dca.h b/include/linux/dca.h index b00a753eda53..9c20c7e87d0a 100644 --- a/include/linux/dca.h +++ b/include/linux/dca.h @@ -1,3 +1,23 @@ +/* + * Copyright(c) 2007 - 2009 Intel Corporation. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 + * Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * The full GNU General Public License is included in this distribution in the + * file called COPYING. + */ #ifndef DCA_H #define DCA_H /* DCA Provider API */ -- cgit v1.2.3-70-g09d2