Merge tag 'amd-drm-next-6.2-2022-11-18' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.2-2022-11-18:

amdgpu:
- SR-IOV fixes
- Clean up DC checks
- DCN 3.2.x fixes
- DCN 3.1.x fixes
- Don't enable degamma on asics which don't support it
- IP discovery fixes
- BACO fixes
- Fix vbios allocation handling when vkms is enabled
- Drop buggy tdr advanced mode GPU reset handling
- Fix the build when DCN is not set in kconfig
- MST DSC fixes
- Userptr fixes
- FRU and RAS EEPROM fixes
- VCN 4.x RAS support
- Aldrebaran CU occupancy reporting fix
- PSP ring cleanup

amdkfd:
- Memory limit fix
- Enable cooperative launch on gfx 10.3

amd-drm-next-6.2-2022-11-11:

amdgpu:
- SMU 13.x updates
- GPUVM TLB race fix
- DCN 3.1.4 updates
- DCN 3.2.x updates
- PSR fixes
- Kerneldoc fix
- Vega10 fan fix
- GPUVM locking fixes in error pathes
- BACO fix for Beige Goby
- EEPROM I2C address cleanup
- GFXOFF fix
- Fix DC memory leak in error pathes
- Flexible array updates
- Mtype fix for GPUVM PTEs
- Move Kconfig into amdgpu directory
- SR-IOV updates
- Fix possible memory leak in CS IOCTL error path

amdkfd:
- Fix possible memory overrun
- CRIU fixes

radeon:
- ACPI ref count fix
- HDA audio notifier support
- Move Kconfig into radeon directory

UAPI:
- Add new GEM_CREATE flags to help to transition more KFD functionality to the DRM UAPI.
  These are used internally in the driver to align location based memory coherency
  requirements from memory allocated in the KFD with how we manage GPUVM PTEs.  They
  are currently blocked in the GEM_CREATE IOCTL as we don't have a user right now.
  They are just used internally in the kernel driver for now for existing KFD memory
  allocations. So a change to the UAPI header, but no functional change in the UAPI.

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221118170807.6505-1-alexander.deucher@amd.com
Signed-off-by: Dave Airlie <airlied@redhat.com>

8 files changed, 82 insertions, 15 deletions

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c
index a88a71460521..5947c2cb0f30 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c
@@ -72,6 +72,23 @@ static void dcn32_init_crb(struct hubbub *hubbub)
 	REG_UPDATE(DCHUBBUB_DEBUG_CTRL_0, DET_DEPTH, 0x47F);
 }
 
+void hubbub32_set_request_limit(struct hubbub *hubbub, int memory_channel_count, int words_per_channel)
+{
+	struct dcn20_hubbub *hubbub2 = TO_DCN20_HUBBUB(hubbub);
+
+	uint32_t request_limit = 3 * memory_channel_count * words_per_channel / 4;
+
+	ASSERT((request_limit & (~0xFFF)) == 0); //field is only 24 bits long
+	ASSERT(request_limit > 0); //field is only 24 bits long
+
+	if (request_limit > 0xFFF)
+		request_limit = 0xFFF;
+
+	if (request_limit > 0)
+		REG_UPDATE(SDPIF_REQUEST_RATE_LIMIT, SDPIF_REQUEST_RATE_LIMIT, request_limit);
+}
+
+
 void dcn32_program_det_size(struct hubbub *hubbub, int hubp_inst, unsigned int det_buffer_size_in_kbyte)
 {
 	struct dcn20_hubbub *hubbub2 = TO_DCN20_HUBBUB(hubbub);
@@ -949,6 +966,7 @@ static const struct hubbub_funcs hubbub32_funcs = {
 	.init_crb = dcn32_init_crb,
 	.hubbub_read_state = hubbub2_read_state,
 	.force_usr_retraining_allow = hubbub32_force_usr_retraining_allow,
+	.set_request_limit = hubbub32_set_request_limit
 };
 
 void hubbub32_construct(struct dcn20_hubbub *hubbub2,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.h
index cda94e0e31bf..786f9ce07f92 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.h
@@ -82,7 +82,8 @@
 	SR(DCN_VM_FAULT_ADDR_MSB),\
 	SR(DCN_VM_FAULT_ADDR_LSB),\
 	SR(DCN_VM_FAULT_CNTL),\
-	SR(DCN_VM_FAULT_STATUS)
+	SR(DCN_VM_FAULT_STATUS),\
+	SR(SDPIF_REQUEST_RATE_LIMIT)
 
 #define HUBBUB_MASK_SH_LIST_DCN32(mask_sh)\
 	HUBBUB_SF(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, mask_sh), \
@@ -159,7 +160,8 @@
 	HUBBUB_SF(DCN_VM_FAULT_STATUS, DCN_VM_ERROR_VMID, mask_sh), \
 	HUBBUB_SF(DCN_VM_FAULT_STATUS, DCN_VM_ERROR_TABLE_LEVEL, mask_sh), \
 	HUBBUB_SF(DCN_VM_FAULT_STATUS, DCN_VM_ERROR_PIPE, mask_sh), \
-	HUBBUB_SF(DCN_VM_FAULT_STATUS, DCN_VM_ERROR_INTERRUPT_STATUS, mask_sh)
+	HUBBUB_SF(DCN_VM_FAULT_STATUS, DCN_VM_ERROR_INTERRUPT_STATUS, mask_sh),\
+	HUBBUB_SF(SDPIF_REQUEST_RATE_LIMIT, SDPIF_REQUEST_RATE_LIMIT, mask_sh)
 
 bool hubbub32_program_urgent_watermarks(
 		struct hubbub *hubbub,
@@ -200,4 +202,6 @@ void hubbub32_construct(struct dcn20_hubbub *hubbub2,
 	int pixel_chunk_size_kb,
 	int config_return_buffer_size_kb);
 
+void hubbub32_set_request_limit(struct hubbub *hubbub, int umc_count, int words_per_umc);
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index 5e0018efe055..763311ffb967 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -283,8 +283,7 @@ static uint32_t dcn32_calculate_cab_allocation(struct dc *dc, struct dc_state *c
 			using the max for calculation */
 
 		if (hubp->curs_attr.width > 0) {
-				// Round cursor width to next multiple of 64
-				cursor_size = (((hubp->curs_attr.width + 63) / 64) * 64) * hubp->curs_attr.height;
+				cursor_size = hubp->curs_attr.pitch * hubp->curs_attr.height;
 
 				switch (pipe->stream->cursor_attributes.color_format) {
 				case CURSOR_MODE_MONO:
@@ -309,9 +308,9 @@ static uint32_t dcn32_calculate_cab_allocation(struct dc *dc, struct dc_state *c
 						cursor_size > 16384) {
 					/* cursor_num_mblk = CEILING(num_cursors*cursor_width*cursor_width*cursor_Bpe/mblk_bytes, 1)
 					 */
-					cache_lines_used += (((hubp->curs_attr.width * hubp->curs_attr.height * cursor_bpp +
-										DCN3_2_MALL_MBLK_SIZE_BYTES - 1) / DCN3_2_MALL_MBLK_SIZE_BYTES) *
-										DCN3_2_MALL_MBLK_SIZE_BYTES) / dc->caps.cache_line_size + 2;
+					cache_lines_used += (((cursor_size + DCN3_2_MALL_MBLK_SIZE_BYTES - 1) /
+							DCN3_2_MALL_MBLK_SIZE_BYTES) * DCN3_2_MALL_MBLK_SIZE_BYTES) /
+							dc->caps.cache_line_size + 2;
 				}
 				break;
 			}
@@ -727,10 +726,7 @@ void dcn32_update_mall_sel(struct dc *dc, struct dc_state *context)
 		struct hubp *hubp = pipe->plane_res.hubp;
 
 		if (pipe->stream && pipe->plane_state && hubp && hubp->funcs->hubp_update_mall_sel) {
-			//Round cursor width up to next multiple of 64
-			int cursor_width = ((hubp->curs_attr.width + 63) / 64) * 64;
-			int cursor_height = hubp->curs_attr.height;
-			int cursor_size = cursor_width * cursor_height;
+			int cursor_size = hubp->curs_attr.pitch * hubp->curs_attr.height;
 
 			switch (hubp->curs_attr.color_format) {
 			case CURSOR_MODE_MONO:
@@ -984,6 +980,9 @@ void dcn32_init_hw(struct dc *dc)
 	if (dc->res_pool->hubbub->funcs->init_crb)
 		dc->res_pool->hubbub->funcs->init_crb(dc->res_pool->hubbub);
 
+	if (dc->res_pool->hubbub->funcs->set_request_limit && dc->config.sdpif_request_limit_words_per_umc > 0)
+		dc->res_pool->hubbub->funcs->set_request_limit(dc->res_pool->hubbub, dc->ctx->dc_bios->vram_info.num_chans, dc->config.sdpif_request_limit_words_per_umc);
+
 	// Get DMCUB capabilities
 	if (dc->ctx->dmub_srv) {
 		dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv->dmub);
@@ -1367,6 +1366,33 @@ void dcn32_update_phantom_vp_position(struct dc *dc,
 	}
 }
 
+/* Treat the phantom pipe as if it needs to be fully enabled.
+ * If the pipe was previously in use but not phantom, it would
+ * have been disabled earlier in the sequence so we need to run
+ * the full enable sequence.
+ */
+void dcn32_apply_update_flags_for_phantom(struct pipe_ctx *phantom_pipe)
+{
+	phantom_pipe->update_flags.raw = 0;
+	if (phantom_pipe->stream && phantom_pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) {
+		if (phantom_pipe->stream && phantom_pipe->plane_state) {
+			phantom_pipe->update_flags.bits.enable = 1;
+			phantom_pipe->update_flags.bits.mpcc = 1;
+			phantom_pipe->update_flags.bits.dppclk = 1;
+			phantom_pipe->update_flags.bits.hubp_interdependent = 1;
+			phantom_pipe->update_flags.bits.hubp_rq_dlg_ttu = 1;
+			phantom_pipe->update_flags.bits.gamut_remap = 1;
+			phantom_pipe->update_flags.bits.scaler = 1;
+			phantom_pipe->update_flags.bits.viewport = 1;
+			phantom_pipe->update_flags.bits.det_size = 1;
+			if (!phantom_pipe->top_pipe && !phantom_pipe->prev_odm_pipe) {
+				phantom_pipe->update_flags.bits.odm = 1;
+				phantom_pipe->update_flags.bits.global_sync = 1;
+			}
+		}
+	}
+}
+
 bool dcn32_dsc_pg_status(
 		struct dce_hwseq *hws,
 		unsigned int dsc_inst)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
index ac3657a5b9ea..7de36529cf99 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
@@ -92,6 +92,8 @@ void dcn32_update_phantom_vp_position(struct dc *dc,
 		struct dc_state *context,
 		struct pipe_ctx *phantom_pipe);
 
+void dcn32_apply_update_flags_for_phantom(struct pipe_ctx *phantom_pipe);
+
 bool dcn32_dsc_pg_status(
 		struct dce_hwseq *hws,
 		unsigned int dsc_inst);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
index 45a949ba6f3f..dc4649458567 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
@@ -110,6 +110,7 @@ static const struct hw_sequencer_funcs dcn32_funcs = {
 	.update_visual_confirm_color = dcn20_update_visual_confirm_color,
 	.update_phantom_vp_position = dcn32_update_phantom_vp_position,
 	.update_dsc_pg = dcn32_update_dsc_pg,
+	.apply_update_flags_for_phantom = dcn32_apply_update_flags_for_phantom,
 };
 
 static const struct hwseq_private_funcs dcn32_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
index 2b33eeb213e2..2ee798965bc2 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
@@ -167,6 +167,13 @@ static void optc32_phantom_crtc_post_enable(struct timing_generator *optc)
 	REG_WAIT(OTG_CLOCK_CONTROL, OTG_BUSY, 0, 1, 100000);
 }
 
+static void optc32_disable_phantom_otg(struct timing_generator *optc)
+{
+	struct optc *optc1 = DCN10TG_FROM_TG(optc);
+
+	REG_UPDATE(OTG_CONTROL, OTG_MASTER_EN, 0);
+}
+
 static void optc32_set_odm_bypass(struct timing_generator *optc,
 		const struct dc_crtc_timing *dc_crtc_timing)
 {
@@ -260,6 +267,7 @@ static struct timing_generator_funcs dcn32_tg_funcs = {
 		.enable_crtc = optc32_enable_crtc,
 		.disable_crtc = optc32_disable_crtc,
 		.phantom_crtc_post_enable = optc32_phantom_crtc_post_enable,
+		.disable_phantom_crtc = optc32_disable_phantom_otg,
 		/* used by enable_timing_synchronization. Not need for FPGA */
 		.is_counter_moving = optc1_is_counter_moving,
 		.get_position = optc1_get_position,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 4ba9a8662185..cdeff6de725d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -106,8 +106,6 @@ enum dcn32_clk_src_array_id {
  */
 
 /* DCN */
-/* TODO awful hack. fixup dcn20_dwb.h */
-#undef BASE_INNER
 #define BASE_INNER(seg) ctx->dcn_reg_offsets[seg]
 
 #define BASE(seg) BASE_INNER(seg)
@@ -167,6 +165,9 @@ enum dcn32_clk_src_array_id {
 	REG_STRUCT.reg_name[id] = BASE(reg ## block ## id ## _ ## temp_name ## _BASE_IDX) + \
 		reg ## block ## id ## _ ## temp_name
 
+#define SF_DWB2(reg_name, block, id, field_name, post_fix)	\
+	.field_name = reg_name ## __ ## field_name ## post_fix
+
 #define DCCG_SRII(reg_name, block, id)\
 	REG_STRUCT.block ## _ ## reg_name[id] = BASE(reg ## block ## id ## _ ## reg_name ## _BASE_IDX) + \
 		reg ## block ## id ## _ ## reg_name
@@ -724,6 +725,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.enable_dp_dig_pixel_rate_div_policy = 1,
 	.allow_sw_cursor_fallback = false,
 	.alloc_extra_way_for_cursor = true,
+	.min_prefetch_in_strobe_ns = 60000, // 60us
 };
 
 static const struct dc_debug_options debug_defaults_diags = {
@@ -829,6 +831,7 @@ static struct clock_source *dcn32_clock_source_create(
 		return &clk_src->base;
 	}
 
+	kfree(clk_src);
 	BREAK_TO_DEBUGGER();
 	return NULL;
 }
@@ -1678,7 +1681,7 @@ static void dcn32_enable_phantom_plane(struct dc *dc,
 
 		/* Shadow pipe has small viewport. */
 		phantom_plane->clip_rect.y = 0;
-		phantom_plane->clip_rect.height = phantom_stream->timing.v_addressable;
+		phantom_plane->clip_rect.height = phantom_stream->src.height;
 
 		phantom_plane->is_phantom = true;
 
@@ -2116,6 +2119,7 @@ static bool dcn32_resource_construct(
 	dc->caps.cache_num_ways = 16;
 	dc->caps.max_cab_allocation_bytes = 67108864; // 64MB = 1024 * 1024 * 64
 	dc->caps.subvp_fw_processing_delay_us = 15;
+	dc->caps.subvp_drr_max_vblank_margin_us = 40;
 	dc->caps.subvp_prefetch_end_to_mall_start_us = 15;
 	dc->caps.subvp_swath_height_margin_lines = 16;
 	dc->caps.subvp_pstate_allow_width_us = 20;
@@ -2411,6 +2415,9 @@ static bool dcn32_resource_construct(
 		pool->base.oem_device = NULL;
 	}
 
+	if (ASICREV_IS_GC_11_0_3(dc->ctx->asic_id.hw_internal_rev) && (dc->config.sdpif_request_limit_words_per_umc == 0))
+		dc->config.sdpif_request_limit_words_per_umc = 16;
+
 	DC_FP_END();
 
 	return true;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
index f76120e67c16..f6bc9bd5da31 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
@@ -1244,7 +1244,8 @@ void dcn32_restore_mall_state(struct dc *dc,
       SR(DCHUBBUB_ARB_FCLK_PSTATE_CHANGE_WATERMARK_C),                         \
       SR(DCHUBBUB_ARB_FCLK_PSTATE_CHANGE_WATERMARK_D),                         \
       SR(DCN_VM_FAULT_ADDR_MSB), SR(DCN_VM_FAULT_ADDR_LSB),                    \
-      SR(DCN_VM_FAULT_CNTL), SR(DCN_VM_FAULT_STATUS)                           \
+      SR(DCN_VM_FAULT_CNTL), SR(DCN_VM_FAULT_STATUS),                          \
+      SR(SDPIF_REQUEST_RATE_LIMIT)                                             \
   )
 
 /* DCCG */