From 7b5bdae7740eb6a3d09f9cd4e4b07362a15b86b3 Mon Sep 17 00:00:00 2001 From: Umesh Nerlige Ramappa Date: Fri, 28 Jun 2024 17:56:43 -0700 Subject: i915/perf: Remove code to update PWR_CLK_STATE for gen12 PWR_CLK_STATE only needs to be modified up until gen11. For gen12 this code is not applicable. Remove code to update context image with PWR_CLK_STATE for gen12. Fixes: 00a7f0d7155c ("drm/i915/tgl: Add perf support on TGL") Signed-off-by: Umesh Nerlige Ramappa Reviewed-by: Ashutosh Dixit Link: https://patchwork.freedesktop.org/patch/msgid/20240629005643.3050678-1-umesh.nerlige.ramappa@intel.com --- drivers/gpu/drm/i915/i915_perf.c | 33 --------------------------------- 1 file changed, 33 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index 0b1cd4c7a525..025a79fe5920 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -2748,26 +2748,6 @@ oa_configure_all_contexts(struct i915_perf_stream *stream, return 0; } -static int -gen12_configure_all_contexts(struct i915_perf_stream *stream, - const struct i915_oa_config *oa_config, - struct i915_active *active) -{ - struct flex regs[] = { - { - GEN8_R_PWR_CLK_STATE(RENDER_RING_BASE), - CTX_R_PWR_CLK_STATE, - }, - }; - - if (stream->engine->class != RENDER_CLASS) - return 0; - - return oa_configure_all_contexts(stream, - regs, ARRAY_SIZE(regs), - active); -} - static int lrc_configure_all_contexts(struct i915_perf_stream *stream, const struct i915_oa_config *oa_config, @@ -2874,7 +2854,6 @@ gen12_enable_metric_set(struct i915_perf_stream *stream, { struct drm_i915_private *i915 = stream->perf->i915; struct intel_uncore *uncore = stream->uncore; - struct i915_oa_config *oa_config = stream->oa_config; bool periodic = stream->periodic; u32 period_exponent = stream->period_exponent; u32 sqcnt1; @@ -2918,15 +2897,6 @@ gen12_enable_metric_set(struct i915_perf_stream *stream, intel_uncore_rmw(uncore, GEN12_SQCNT1, 0, sqcnt1); - /* - * Update all contexts prior writing the mux configurations as we need - * to make sure all slices/subslices are ON before writing to NOA - * registers. - */ - ret = gen12_configure_all_contexts(stream, oa_config, active); - if (ret) - return ret; - /* * For Gen12, performance counters are context * saved/restored. Only enable it for the context that @@ -2980,9 +2950,6 @@ static void gen12_disable_metric_set(struct i915_perf_stream *stream) _MASKED_BIT_DISABLE(GEN12_DISABLE_DOP_GATING)); } - /* Reset all contexts' slices/subslices configurations. */ - gen12_configure_all_contexts(stream, NULL, NULL); - /* disable the context save/restore or OAR counters */ if (stream->ctx) gen12_configure_oar_context(stream, NULL); -- cgit v1.2.3 From 7df0be6e6280c6fca01d039864bb123e5e36604b Mon Sep 17 00:00:00 2001 From: Nitin Gote Date: Thu, 11 Jul 2024 22:02:08 +0530 Subject: drm/i915/gt: Do not consider preemption during execlists_dequeue for gen8 We're seeing a GPU hang issue on a CHV platform, which was caused by commit bac24f59f454 ("drm/i915/execlists: Enable coarse preemption boundaries for Gen8"). The Gen8 platform only supports timeslicing and doesn't have a preemption mechanism, as its engines do not have a preemption timer. Commit 751f82b353a6 ("drm/i915/gt: Only disable preemption on Gen8 render engines") addressed this issue only for render engines. This patch extends that fix by ensuring that preemption is not considered for all engines on Gen8 platforms. v4: - Use the correct Fixes tag (Rodrigo Vivi) - Reworded commit log (Andi Shyti) v3: - Inside need_preempt(), condition of can_preempt() is not required as simplified can_preempt() is enough. (Chris Wilson) v2: Simplify can_preempt() function (Tvrtko Ursulin) Fixes: 751f82b353a6 ("drm/i915/gt: Only disable preemption on gen8 render engines") Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/11396 Suggested-by: Andi Shyti Signed-off-by: Nitin Gote Cc: Chris Wilson CC: # v5.12+ Reviewed-by: Jonathan Cavitt Reviewed-by: Andi Shyti Signed-off-by: Andi Shyti Link: https://patchwork.freedesktop.org/patch/msgid/20240711163208.1355736-1-nitin.r.gote@intel.com --- drivers/gpu/drm/i915/gt/intel_execlists_submission.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c index 21829439e686..72090f52fb85 100644 --- a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c +++ b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c @@ -3315,11 +3315,7 @@ static void remove_from_engine(struct i915_request *rq) static bool can_preempt(struct intel_engine_cs *engine) { - if (GRAPHICS_VER(engine->i915) > 8) - return true; - - /* GPGPU on bdw requires extra w/a; not implemented */ - return engine->class != RENDER_CLASS; + return GRAPHICS_VER(engine->i915) > 8; } static void kick_execlists(const struct i915_request *rq, int prio) -- cgit v1.2.3 From d75dec1fcbcb05b021c08b62551649567ab8955c Mon Sep 17 00:00:00 2001 From: Jonathan Cavitt Date: Fri, 12 Jul 2024 14:41:56 -0700 Subject: drm/i915: Allow NULL memory region Prevent a NULL pointer access in intel_memory_regions_hw_probe. Fixes: 05da7d9f717b ("drm/i915/gem: Downgrade stolen lmem setup warning") Reported-by: Dan Carpenter Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/11704 Signed-off-by: Jonathan Cavitt Reviewed-by: Nirmoy Das Reviewed-by: Andi Shyti Signed-off-by: Andi Shyti Link: https://patchwork.freedesktop.org/patch/msgid/20240712214156.3969584-1-jonathan.cavitt@intel.com --- drivers/gpu/drm/i915/intel_memory_region.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/intel_memory_region.c b/drivers/gpu/drm/i915/intel_memory_region.c index 172dfa7c3588..d40ee1b42110 100644 --- a/drivers/gpu/drm/i915/intel_memory_region.c +++ b/drivers/gpu/drm/i915/intel_memory_region.c @@ -368,8 +368,10 @@ int intel_memory_regions_hw_probe(struct drm_i915_private *i915) goto out_cleanup; } - mem->id = i; - i915->mm.regions[i] = mem; + if (mem) { /* Skip on non-fatal errors */ + mem->id = i; + i915->mm.regions[i] = mem; + } } for (i = 0; i < ARRAY_SIZE(i915->mm.regions); i++) { -- cgit v1.2.3 From 104bcfae57d80c484185c012cc0ee8d8d5889a7e Mon Sep 17 00:00:00 2001 From: John Harrison Date: Fri, 21 Jun 2024 17:46:35 -0700 Subject: drm/i915/arl: Enable Wa_14019159160 for ARL The context switch out workaround also applies to ARL. Signed-off-by: John Harrison Reviewed-by: Vinay Belgaumkar Signed-off-by: Daniele Ceraolo Spurio Link: https://patchwork.freedesktop.org/patch/msgid/20240622004636.662081-2-John.C.Harrison@Intel.com --- drivers/gpu/drm/i915/gt/uc/intel_guc.c | 2 +- drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc.c index 5e60a34692af..097fc6bd1285 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.c @@ -296,7 +296,7 @@ static u32 guc_ctl_wa_flags(struct intel_guc *guc) /* Wa_16019325821 */ /* Wa_14019159160 */ - if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 71))) + if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74))) flags |= GUC_WA_RCS_CCS_SWITCHOUT; /* diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c index 7995f059f30d..f1fe5f905453 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c @@ -850,7 +850,7 @@ static void guc_waklv_init(struct intel_guc *guc) remain = guc_ads_waklv_size(guc); /* Wa_14019159160 */ - if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 71))) + if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74))) guc_waklv_enable_simple(guc, GUC_WORKAROUND_KLV_SERIALIZED_RA_MODE, &offset, &remain); -- cgit v1.2.3 From e4a0251d36fc6c190c545aa95702eec0efe05a4f Mon Sep 17 00:00:00 2001 From: John Harrison Date: Fri, 21 Jun 2024 17:46:36 -0700 Subject: drm/i915/guc: Extend w/a 14019159160 There is a new part to an existing workaround, so enable that piece as well. v2: Extend even further. v3: Drop DG2 as there are CI failures still to resolve. Also re-order the parameters to a function to reduce excessive line wrapping. Signed-off-by: John Harrison Reviewed-by: Vinay Belgaumkar Signed-off-by: Daniele Ceraolo Spurio Link: https://patchwork.freedesktop.org/patch/msgid/20240622004636.662081-3-John.C.Harrison@Intel.com --- drivers/gpu/drm/i915/gt/uc/abi/guc_klvs_abi.h | 1 + drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c | 18 +++++++++--------- 2 files changed, 10 insertions(+), 9 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gt/uc/abi/guc_klvs_abi.h b/drivers/gpu/drm/i915/gt/uc/abi/guc_klvs_abi.h index 37ff539a6963..0c709e6c15be 100644 --- a/drivers/gpu/drm/i915/gt/uc/abi/guc_klvs_abi.h +++ b/drivers/gpu/drm/i915/gt/uc/abi/guc_klvs_abi.h @@ -107,6 +107,7 @@ enum { enum { GUC_WORKAROUND_KLV_SERIALIZED_RA_MODE = 0x9001, GUC_WORKAROUND_KLV_BLOCK_INTERRUPTS_WHEN_MGSR_BLOCKED = 0x9002, + GUC_WORKAROUND_KLV_AVOID_GFX_CLEAR_WHILE_ACTIVE = 0x9006, }; #endif /* _ABI_GUC_KLVS_ABI_H */ diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c index f1fe5f905453..46fabbfc775e 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c @@ -815,8 +815,7 @@ engine_instance_list: return PAGE_ALIGN(total_size); } -static void guc_waklv_enable_simple(struct intel_guc *guc, - u32 klv_id, u32 *offset, u32 *remain) +static void guc_waklv_enable_simple(struct intel_guc *guc, u32 *offset, u32 *remain, u32 klv_id) { u32 size; u32 klv_entry[] = { @@ -850,19 +849,20 @@ static void guc_waklv_init(struct intel_guc *guc) remain = guc_ads_waklv_size(guc); /* Wa_14019159160 */ - if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74))) - guc_waklv_enable_simple(guc, - GUC_WORKAROUND_KLV_SERIALIZED_RA_MODE, - &offset, &remain); + if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74))) { + guc_waklv_enable_simple(guc, &offset, &remain, + GUC_WORKAROUND_KLV_SERIALIZED_RA_MODE); + guc_waklv_enable_simple(guc, &offset, &remain, + GUC_WORKAROUND_KLV_AVOID_GFX_CLEAR_WHILE_ACTIVE); + } /* Wa_16021333562 */ if ((GUC_FIRMWARE_VER(guc) >= MAKE_GUC_VER(70, 21, 1)) && (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74)) || IS_MEDIA_GT_IP_RANGE(gt, IP_VER(13, 0), IP_VER(13, 0)) || IS_DG2(gt->i915))) - guc_waklv_enable_simple(guc, - GUC_WORKAROUND_KLV_BLOCK_INTERRUPTS_WHEN_MGSR_BLOCKED, - &offset, &remain); + guc_waklv_enable_simple(guc, &offset, &remain, + GUC_WORKAROUND_KLV_BLOCK_INTERRUPTS_WHEN_MGSR_BLOCKED); size = guc_ads_waklv_size(guc) - remain; if (!size) -- cgit v1.2.3 From 843f10ce6539cd5e4989415e18cb809f4cff8b91 Mon Sep 17 00:00:00 2001 From: Nitin Gote Date: Wed, 31 Jul 2024 21:26:14 +0530 Subject: drm/i915/gt: Add Wa_14019789679 Wa_14019789679 implementation for MTL, ARL and DG2. v2: Corrected condition v3: - Fix indentation (Jani Nikula) - dword size should be 0x1 and initialize dword to 0 instead of MI_NOOP (Tejas) - Use IS_GFX_GT_IP_RANGE() (Tejas) v4: - 3DSTATE_MESH_CONTROL instruction is 3 dwords long Align with dword size. (Roper, Matthew D) - Add RCS engine check. (Tejas) Bspec: 47083 Signed-off-by: Nitin Gote Reviewed-by: Tejas Upadhyay Reviewed-by: Matt Roper Reviewed-by: Andi Shyti Signed-off-by: Andi Shyti Link: https://patchwork.freedesktop.org/patch/msgid/20240731155614.3460645-1-nitin.r.gote@intel.com --- drivers/gpu/drm/i915/gt/intel_gpu_commands.h | 1 + drivers/gpu/drm/i915/gt/intel_workarounds.c | 16 +++++++++++++++- 2 files changed, 16 insertions(+), 1 deletion(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gt/intel_gpu_commands.h b/drivers/gpu/drm/i915/gt/intel_gpu_commands.h index 2bd8d98d2110..5394bc7d4daf 100644 --- a/drivers/gpu/drm/i915/gt/intel_gpu_commands.h +++ b/drivers/gpu/drm/i915/gt/intel_gpu_commands.h @@ -220,6 +220,7 @@ #define GFX_OP_DESTBUFFER_INFO ((0x3<<29)|(0x1d<<24)|(0x8e<<16)|1) #define GFX_OP_DRAWRECT_INFO ((0x3<<29)|(0x1d<<24)|(0x80<<16)|(0x3)) #define GFX_OP_DRAWRECT_INFO_I965 ((0x7900<<16)|0x2) +#define CMD_3DSTATE_MESH_CONTROL ((0x3 << 29) | (0x3 << 27) | (0x0 << 24) | (0x77 << 16) | (0x3)) #define XY_CTRL_SURF_INSTR_SIZE 5 #define MI_FLUSH_DW_SIZE 3 diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index 09a287c1aedd..bfe6d8fc820f 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -974,7 +974,12 @@ int intel_engine_emit_ctx_wa(struct i915_request *rq) if (ret) return ret; - cs = intel_ring_begin(rq, (wal->count * 2 + 2)); + if ((IS_GFX_GT_IP_RANGE(rq->engine->gt, IP_VER(12, 70), IP_VER(12, 74)) || + IS_DG2(rq->i915)) && rq->engine->class == RENDER_CLASS) + cs = intel_ring_begin(rq, (wal->count * 2 + 6)); + else + cs = intel_ring_begin(rq, (wal->count * 2 + 2)); + if (IS_ERR(cs)) return PTR_ERR(cs); @@ -1004,6 +1009,15 @@ int intel_engine_emit_ctx_wa(struct i915_request *rq) } *cs++ = MI_NOOP; + /* Wa_14019789679 */ + if ((IS_GFX_GT_IP_RANGE(rq->engine->gt, IP_VER(12, 70), IP_VER(12, 74)) || + IS_DG2(rq->i915)) && rq->engine->class == RENDER_CLASS) { + *cs++ = CMD_3DSTATE_MESH_CONTROL; + *cs++ = 0; + *cs++ = 0; + *cs++ = MI_NOOP; + } + intel_uncore_forcewake_put__locked(uncore, fw); spin_unlock(&uncore->lock); intel_gt_mcr_unlock(wal->gt, flags); -- cgit v1.2.3 From fceff12e52985e49c464f402e11b2f97bce3cc24 Mon Sep 17 00:00:00 2001 From: Krzysztof Niemiec Date: Thu, 1 Aug 2024 17:40:48 +0200 Subject: drm/i915/gt: Empty uabi engines list during intel_engines_release() While the uabi_engines_llist is populated in intel_engines_init() during driver load, the corresponding function intel_engines_release() does not correctly get rid of it. This can lead to a UAF if, after failed initialization (for example when gt is set wedged on init), we try to access the engines. Suggested-by: Chris Wilson Signed-off-by: Krzysztof Niemiec Reviewed-by: Andi Shyti Signed-off-by: Andi Shyti Link: https://patchwork.freedesktop.org/patch/msgid/20240801154047.115176-2-krzysztof.niemiec@intel.com --- drivers/gpu/drm/i915/gt/intel_engine_cs.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c index 3b740ca25000..4d30a86016f2 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c @@ -693,6 +693,8 @@ void intel_engines_release(struct intel_gt *gt) memset(&engine->reset, 0, sizeof(engine->reset)); } + + llist_del_all(>->i915->uabi_engines_llist); } void intel_engine_free_request_pool(struct intel_engine_cs *engine) -- cgit v1.2.3 From 60a2066c50058086510c91f404eb582029650970 Mon Sep 17 00:00:00 2001 From: Andi Shyti Date: Fri, 2 Aug 2024 10:38:49 +0200 Subject: drm/i915/gem: Adjust vma offset for framebuffer mmap offset When mapping a framebuffer object, the virtual memory area (VMA) offset ('vm_pgoff') should be adjusted by the start of the 'vma_node' associated with the object. This ensures that the VMA offset is correctly aligned with the corresponding offset within the GGTT aperture. Increment vm_pgoff by the start of the vma_node with the offset= provided by the user. Suggested-by: Chris Wilson Signed-off-by: Andi Shyti Reviewed-by: Jonathan Cavitt Reviewed-by: Rodrigo Vivi Cc: # v4.9+ [Joonas: Add Cc: stable] Signed-off-by: Joonas Lahtinen Link: https://patchwork.freedesktop.org/patch/msgid/20240802083850.103694-2-andi.shyti@linux.intel.com --- drivers/gpu/drm/i915/gem/i915_gem_mman.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index a2195e28b625..ce10dd259812 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -1084,6 +1084,8 @@ int i915_gem_fb_mmap(struct drm_i915_gem_object *obj, struct vm_area_struct *vma mmo = mmap_offset_attach(obj, mmap_type, NULL); if (IS_ERR(mmo)) return PTR_ERR(mmo); + + vma->vm_pgoff += drm_vma_node_start(&mmo->vma_node); } /* -- cgit v1.2.3 From 97b6784753da06d9d40232328efc5c5367e53417 Mon Sep 17 00:00:00 2001 From: Andi Shyti Date: Fri, 2 Aug 2024 10:38:50 +0200 Subject: drm/i915/gem: Fix Virtual Memory mapping boundaries calculation Calculating the size of the mapped area as the lesser value between the requested size and the actual size does not consider the partial mapping offset. This can cause page fault access. Fix the calculation of the starting and ending addresses, the total size is now deduced from the difference between the end and start addresses. Additionally, the calculations have been rewritten in a clearer and more understandable form. Fixes: c58305af1835 ("drm/i915: Use remap_io_mapping() to prefault all PTE in a single pass") Reported-by: Jann Horn Co-developed-by: Chris Wilson Signed-off-by: Chris Wilson Signed-off-by: Andi Shyti Cc: Joonas Lahtinen Cc: Matthew Auld Cc: Rodrigo Vivi Cc: # v4.9+ Reviewed-by: Jann Horn Reviewed-by: Jonathan Cavitt [Joonas: Add Requires: tag] Requires: 60a2066c5005 ("drm/i915/gem: Adjust vma offset for framebuffer mmap offset") Signed-off-by: Joonas Lahtinen Link: https://patchwork.freedesktop.org/patch/msgid/20240802083850.103694-3-andi.shyti@linux.intel.com --- drivers/gpu/drm/i915/gem/i915_gem_mman.c | 53 ++++++++++++++++++++++++++++---- 1 file changed, 47 insertions(+), 6 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index ce10dd259812..cac6d4184506 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -290,6 +290,41 @@ out: return i915_error_to_vmf_fault(err); } +static void set_address_limits(struct vm_area_struct *area, + struct i915_vma *vma, + unsigned long obj_offset, + unsigned long *start_vaddr, + unsigned long *end_vaddr) +{ + unsigned long vm_start, vm_end, vma_size; /* user's memory parameters */ + long start, end; /* memory boundaries */ + + /* + * Let's move into the ">> PAGE_SHIFT" + * domain to be sure not to lose bits + */ + vm_start = area->vm_start >> PAGE_SHIFT; + vm_end = area->vm_end >> PAGE_SHIFT; + vma_size = vma->size >> PAGE_SHIFT; + + /* + * Calculate the memory boundaries by considering the offset + * provided by the user during memory mapping and the offset + * provided for the partial mapping. + */ + start = vm_start; + start -= obj_offset; + start += vma->gtt_view.partial.offset; + end = start + vma_size; + + start = max_t(long, start, vm_start); + end = min_t(long, end, vm_end); + + /* Let's move back into the "<< PAGE_SHIFT" domain */ + *start_vaddr = (unsigned long)start << PAGE_SHIFT; + *end_vaddr = (unsigned long)end << PAGE_SHIFT; +} + static vm_fault_t vm_fault_gtt(struct vm_fault *vmf) { #define MIN_CHUNK_PAGES (SZ_1M >> PAGE_SHIFT) @@ -302,14 +337,18 @@ static vm_fault_t vm_fault_gtt(struct vm_fault *vmf) struct i915_ggtt *ggtt = to_gt(i915)->ggtt; bool write = area->vm_flags & VM_WRITE; struct i915_gem_ww_ctx ww; + unsigned long obj_offset; + unsigned long start, end; /* memory boundaries */ intel_wakeref_t wakeref; struct i915_vma *vma; pgoff_t page_offset; + unsigned long pfn; int srcu; int ret; - /* We don't use vmf->pgoff since that has the fake offset */ + obj_offset = area->vm_pgoff - drm_vma_node_start(&mmo->vma_node); page_offset = (vmf->address - area->vm_start) >> PAGE_SHIFT; + page_offset += obj_offset; trace_i915_gem_object_fault(obj, page_offset, true, write); @@ -402,12 +441,14 @@ retry: if (ret) goto err_unpin; + set_address_limits(area, vma, obj_offset, &start, &end); + + pfn = (ggtt->gmadr.start + i915_ggtt_offset(vma)) >> PAGE_SHIFT; + pfn += (start - area->vm_start) >> PAGE_SHIFT; + pfn += obj_offset - vma->gtt_view.partial.offset; + /* Finally, remap it using the new GTT offset */ - ret = remap_io_mapping(area, - area->vm_start + (vma->gtt_view.partial.offset << PAGE_SHIFT), - (ggtt->gmadr.start + i915_ggtt_offset(vma)) >> PAGE_SHIFT, - min_t(u64, vma->size, area->vm_end - area->vm_start), - &ggtt->iomap); + ret = remap_io_mapping(area, start, pfn, end - start, &ggtt->iomap); if (ret) goto err_fence; -- cgit v1.2.3 From 54bf0af90844fbf18f5be3272eda69198dfdb622 Mon Sep 17 00:00:00 2001 From: David Gow Date: Sun, 4 Aug 2024 17:18:47 +0800 Subject: drm/i915: Allow evicting to use the requested placement In commit a78a8da51b36 ("drm/ttm: replace busy placement with flags v6"), the old system of having a separate placement list (for placements which should be used without eviction) and a 'busy' placement list (for placements which should be attempted if eviction is required) was replaced with a new one where placements could be marked 'FALLBACK' (to be attempted if eviction is required) or 'DESIRED' (to be attempted first, but not if eviction is required). i915 had always included the requested placement in the list of 'busy' placements: i.e., the placement could be used either if eviction is required or not. But when the new system was put in place, the requested (first) placement was marked 'DESIRED', so would never be used if eviction became necessary. While a bug in the original commit prevented this flag from working, when this was fixed in 4a0e7b3c ("drm/i915: fix applying placement flag"), it caused long hangs on DG2 systems with small BAR. Don't mark the requested placement DESIRED (or FALLBACK), allowing it to be used in both situations. This matches the old behaviour, and resolves the hangs. Thanks to Justin Brewer for bisecting the issue. Fixes: a78a8da51b36 ("drm/ttm: replace busy placement with flags v6") Fixes: 4a0e7b3c3753 ("drm/i915: fix applying placement flag") Link: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/11255 Signed-off-by: David Gow Reviewed-by: Jonathan Cavitt Reviewed-by: Andi Shyti Signed-off-by: Andi Shyti Link: https://patchwork.freedesktop.org/patch/msgid/20240804091851.122186-2-david@davidgow.net --- drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 1 - 1 file changed, 1 deletion(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c index e6f177183c0f..fb848fd8ba15 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c @@ -165,7 +165,6 @@ i915_ttm_placement_from_obj(const struct drm_i915_gem_object *obj, i915_ttm_place_from_region(num_allowed ? obj->mm.placements[0] : obj->mm.region, &places[0], obj->bo_offset, obj->base.size, flags); - places[0].flags |= TTM_PL_FLAG_DESIRED; /* Cache this on object? */ for (i = 0; i < num_allowed; ++i) { -- cgit v1.2.3 From 92653f2a572505adaf7f13f695c1907e71a1dc84 Mon Sep 17 00:00:00 2001 From: David Gow Date: Sun, 4 Aug 2024 17:18:48 +0800 Subject: drm/i915: Attempt to get pages without eviction first In commit a78a8da51b36 ("drm/ttm: replace busy placement with flags v6"), __i915_ttm_get_pages was updated to use flags instead of the separate 'busy' placement list. However, the behaviour was subtly changed. Originally, the function would attempt to use the preferred placement without eviction, and give an opportunity to restart the operation before falling back to allowing eviction. This was unintentionally changed, as the preferred placement was not given the TTM_PL_FLAG_DESIRED flag, and so eviction could be triggered in that first pass. This caused thrashing, and a significant performance regression on DG2 systems with small BAR. For example, Minecraft and Team Fortress 2 would drop to single-digit framerates. Restore the original behaviour by marking the initial placement as desired on that first attempt. Also, rework this to use a separate struct ttm_palcement, as the individual placements are marked 'const', so hot-patching the flags is even more dodgy than before. Thanks to Justin Brewer for bisecting this. Fixes: a78a8da51b36 ("drm/ttm: replace busy placement with flags v6") Link: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/11255 Signed-off-by: David Gow Reviewed-by: Jonathan Cavitt Reviewed-by: Andi Shyti Signed-off-by: Andi Shyti Link: https://patchwork.freedesktop.org/patch/msgid/20240804091851.122186-3-david@davidgow.net --- drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c index fb848fd8ba15..5c72462d1f57 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c @@ -778,13 +778,16 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj, .interruptible = true, .no_wait_gpu = false, }; - int real_num_busy; + struct ttm_placement initial_placement; + struct ttm_place initial_place; int ret; /* First try only the requested placement. No eviction. */ - real_num_busy = placement->num_placement; - placement->num_placement = 1; - ret = ttm_bo_validate(bo, placement, &ctx); + initial_placement.num_placement = 1; + memcpy(&initial_place, placement->placement, sizeof(struct ttm_place)); + initial_place.flags |= TTM_PL_FLAG_DESIRED; + initial_placement.placement = &initial_place; + ret = ttm_bo_validate(bo, &initial_placement, &ctx); if (ret) { ret = i915_ttm_err_to_gem(ret); /* @@ -799,7 +802,6 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj, * If the initial attempt fails, allow all accepted placements, * evicting if necessary. */ - placement->num_placement = real_num_busy; ret = ttm_bo_validate(bo, placement, &ctx); if (ret) return i915_ttm_err_to_gem(ret); -- cgit v1.2.3 From 0829b5bcdd3ba077c408e71e22ac2bfa85bf2c95 Mon Sep 17 00:00:00 2001 From: Tvrtko Ursulin Date: Tue, 21 May 2024 11:12:01 +0100 Subject: drm/i915: 2 GiB of relocations ought to be enough for anybody* Kernel test robot reports i915 can hit a warn in kvmalloc_node which has a purpose of dissalowing crazy size kernel allocations. This was added in 7661809d493b ("mm: don't allow oversized kvmalloc() calls"): /* Don't even allow crazy sizes */ if (WARN_ON_ONCE(size > INT_MAX)) return NULL; This would be kind of okay since i915 at one point dropped the need for making a shadow copy of the relocation list, but then it got re-added in fd1500fcd442 ("Revert "drm/i915/gem: Drop relocation slowpath".") a year after Linus added the above warning. It is plausible that the issue was not seen until now because to trigger gem_exec_reloc test requires a combination of an relatively older generation hardware but with at least 8GiB of RAM installed. Probably even more depending on runtime checks. Lets cap what we allow userspace to pass in using the matching limit. There should be no issue for real userspace since we are talking about "crazy" number of relocations which have no practical purpose. *) Well IGT tests might get upset but they can be easily adjusted. Signed-off-by: Tvrtko Ursulin Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-lkp/202405151008.6ddd1aaf-oliver.sang@intel.com Cc: Kees Cook Cc: Kent Overstreet Cc: Joonas Lahtinen Cc: Rodrigo Vivi Reviewed-by: Kees Cook Reviewed-by: Joonas Lahtinen Signed-off-by: Tvrtko Ursulin Link: https://patchwork.freedesktop.org/patch/msgid/20240521101201.18978-1-tursulin@igalia.com --- drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index 090724fa766c..8aff06933f54 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -1533,7 +1533,7 @@ static int eb_relocate_vma(struct i915_execbuffer *eb, struct eb_vma *ev) u64_to_user_ptr(entry->relocs_ptr); unsigned long remain = entry->relocation_count; - if (unlikely(remain > N_RELOC(ULONG_MAX))) + if (unlikely(remain > N_RELOC(INT_MAX))) return -EINVAL; /* @@ -1641,7 +1641,7 @@ static int check_relocations(const struct drm_i915_gem_exec_object2 *entry) if (size == 0) return 0; - if (size > N_RELOC(ULONG_MAX)) + if (size > N_RELOC(INT_MAX)) return -EINVAL; addr = u64_to_user_ptr(entry->relocs_ptr); -- cgit v1.2.3 From cecab0bb737fb74813733f2472761fc3a93857b1 Mon Sep 17 00:00:00 2001 From: Andi Shyti Date: Wed, 7 Aug 2024 11:45:53 +0100 Subject: drm/i915/gem: Improve pfn calculation readability in vm_fault_gtt() By moving the pfn calculation to the set_address_limits() function we improve code readability. This way, set_address_limits() is responsible for calculating all memory mapping paramenters: "start", "end" and "pfn". This suggestion from Jonathan was made during the review of commit 8bdd9ef7e9b1 ("drm/i915/gem: Fix Virtual Memory mapping boundaries calculation"), which I liked, but it got lost on the way. Suggested-by: Jonathan Cavitt Signed-off-by: Andi Shyti Reviewed-by: Krzysztof Niemiec Reviewed-by: Jonathan Cavitt Link: https://patchwork.freedesktop.org/patch/msgid/20240807104553.481763-1-andi.shyti@linux.intel.com --- drivers/gpu/drm/i915/gem/i915_gem_mman.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index cac6d4184506..e9b2424156f0 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -293,8 +293,10 @@ out: static void set_address_limits(struct vm_area_struct *area, struct i915_vma *vma, unsigned long obj_offset, + resource_size_t gmadr_start, unsigned long *start_vaddr, - unsigned long *end_vaddr) + unsigned long *end_vaddr, + unsigned long *pfn) { unsigned long vm_start, vm_end, vma_size; /* user's memory parameters */ long start, end; /* memory boundaries */ @@ -323,6 +325,10 @@ static void set_address_limits(struct vm_area_struct *area, /* Let's move back into the "<< PAGE_SHIFT" domain */ *start_vaddr = (unsigned long)start << PAGE_SHIFT; *end_vaddr = (unsigned long)end << PAGE_SHIFT; + + *pfn = (gmadr_start + i915_ggtt_offset(vma)) >> PAGE_SHIFT; + *pfn += (*start_vaddr - area->vm_start) >> PAGE_SHIFT; + *pfn += obj_offset - vma->gtt_view.partial.offset; } static vm_fault_t vm_fault_gtt(struct vm_fault *vmf) @@ -441,11 +447,13 @@ retry: if (ret) goto err_unpin; - set_address_limits(area, vma, obj_offset, &start, &end); - - pfn = (ggtt->gmadr.start + i915_ggtt_offset(vma)) >> PAGE_SHIFT; - pfn += (start - area->vm_start) >> PAGE_SHIFT; - pfn += obj_offset - vma->gtt_view.partial.offset; + /* + * Dump all the necessary parameters in this function to perform the + * arithmetic calculation for the virtual address start and end and + * the PFN (Page Frame Number). + */ + set_address_limits(area, vma, obj_offset, ggtt->gmadr.start, + &start, &end, &pfn); /* Finally, remap it using the new GTT offset */ ret = remap_io_mapping(area, start, pfn, end - start, &ggtt->iomap); -- cgit v1.2.3 From b7b930d104c38a545d862896759863d649be8252 Mon Sep 17 00:00:00 2001 From: Andi Shyti Date: Wed, 7 Aug 2024 14:05:15 +0100 Subject: drm/i915: Replace double blank with single blank after comma in gem/ and gt/ Do not use double blanks, ", " in function parameters where it's not required by any alignment purpose. Replase it with a single blank, ", ". Signed-off-by: Andi Shyti Reviewed-by: Jani Nikula Link: https://patchwork.freedesktop.org/patch/msgid/20240807130516.491053-2-andi.shyti@linux.intel.com --- drivers/gpu/drm/i915/gem/i915_gem_object_types.h | 2 +- drivers/gpu/drm/i915/gt/selftest_migrate.c | 2 +- drivers/gpu/drm/i915/gt/uc/intel_uc.c | 2 +- drivers/gpu/drm/i915/selftests/intel_memory_region.c | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h index 1495b6074492..68413c05c812 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h @@ -535,7 +535,7 @@ struct drm_i915_gem_object { * I915_CACHE_NONE. The only exception is userptr objects, where we * instead force I915_CACHE_LLC, but we also don't allow userspace to * ever change the @cache_level for such objects. Another special case - * is dma-buf, which doesn't rely on @cache_dirty, but there we + * is dma-buf, which doesn't rely on @cache_dirty, but there we * always do a forced flush when acquiring the pages, if there is a * chance that the pages can be read directly from main memory with * the GPU. diff --git a/drivers/gpu/drm/i915/gt/selftest_migrate.c b/drivers/gpu/drm/i915/gt/selftest_migrate.c index 3eff364ccf3a..ca460cee4f8b 100644 --- a/drivers/gpu/drm/i915/gt/selftest_migrate.c +++ b/drivers/gpu/drm/i915/gt/selftest_migrate.c @@ -336,7 +336,7 @@ static int clear(struct intel_migrate *migrate, if (vaddr[x] != val) { pr_err("%ps failed, (%u != %u), offset: %zu\n", - fn, vaddr[x], val, x * sizeof(u32)); + fn, vaddr[x], val, x * sizeof(u32)); igt_hexdump(vaddr + i * 1024, 4096); err = -EINVAL; } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c index 7a63abf8f644..5b8080ec5315 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c @@ -99,7 +99,7 @@ static void __confirm_options(struct intel_uc *uc) } if (!intel_uc_supports_guc(uc)) - gt_info(gt, "Incompatible option enable_guc=%d - %s\n", + gt_info(gt, "Incompatible option enable_guc=%d - %s\n", i915->params.enable_guc, "GuC is not supported!"); if (i915->params.enable_guc & ENABLE_GUC_SUBMISSION && diff --git a/drivers/gpu/drm/i915/selftests/intel_memory_region.c b/drivers/gpu/drm/i915/selftests/intel_memory_region.c index ae6070b5bf07..f08f6674911e 100644 --- a/drivers/gpu/drm/i915/selftests/intel_memory_region.c +++ b/drivers/gpu/drm/i915/selftests/intel_memory_region.c @@ -517,7 +517,7 @@ static int igt_mock_max_segment(void *arg) if (!IS_ALIGNED(daddr, ps)) { pr_err("%s: Created an unaligned scatterlist entry, addr=%pa, ps=%u\n", - __func__, &daddr, ps); + __func__, &daddr, ps); err = -EINVAL; goto out_close; } -- cgit v1.2.3 From a857add73e61277fb8f688234c9d383130313ea7 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Wed, 7 Aug 2024 10:10:14 +0100 Subject: drm/i915/gt: Mark the GT as dead when mmio is unreliable After we detect that mmio is returning all 0xff, we believe that the GPU has dropped off the pci bus and is dead. Mark the device as wedged such that we can propagate the failure back to userspace and wait for recovery. Signed-off-by: Chris Wilson Signed-off-by: Andi Shyti Reviewed-by: Jonathan Cavitt Link: https://patchwork.freedesktop.org/patch/msgid/20240807091014.469992-1-andi.shyti@linux.intel.com --- drivers/gpu/drm/i915/gt/intel_gt.h | 6 ++++++ drivers/gpu/drm/i915/gt/intel_gt_types.h | 2 ++ drivers/gpu/drm/i915/gt/intel_reset.c | 12 +++++++++++- drivers/gpu/drm/i915/intel_uncore.c | 7 +++++-- 4 files changed, 24 insertions(+), 3 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gt/intel_gt.h b/drivers/gpu/drm/i915/gt/intel_gt.h index b5e114d284ad..b73555889d50 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.h +++ b/drivers/gpu/drm/i915/gt/intel_gt.h @@ -208,4 +208,10 @@ enum i915_map_type intel_gt_coherent_map_type(struct intel_gt *gt, void intel_gt_bind_context_set_ready(struct intel_gt *gt); void intel_gt_bind_context_set_unready(struct intel_gt *gt); bool intel_gt_is_bind_context_ready(struct intel_gt *gt); + +static inline void intel_gt_set_wedged_async(struct intel_gt *gt) +{ + queue_work(system_highpri_wq, >->wedge); +} + #endif /* __INTEL_GT_H__ */ diff --git a/drivers/gpu/drm/i915/gt/intel_gt_types.h b/drivers/gpu/drm/i915/gt/intel_gt_types.h index cfdd2ad5e954..bcee084b1f27 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_types.h +++ b/drivers/gpu/drm/i915/gt/intel_gt_types.h @@ -292,6 +292,8 @@ struct intel_gt { struct gt_defaults defaults; struct kobject *sysfs_defaults; + struct work_struct wedge; + struct i915_perf_gt perf; /** link: &ggtt.gt_list */ diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c index 6161f7a3ff70..76810ebb5e83 100644 --- a/drivers/gpu/drm/i915/gt/intel_reset.c +++ b/drivers/gpu/drm/i915/gt/intel_reset.c @@ -1013,6 +1013,15 @@ static void __intel_gt_set_wedged(struct intel_gt *gt) GT_TRACE(gt, "end\n"); } +static void set_wedged_work(struct work_struct *w) +{ + struct intel_gt *gt = container_of(w, struct intel_gt, wedge); + intel_wakeref_t wf; + + with_intel_runtime_pm(gt->uncore->rpm, wf) + __intel_gt_set_wedged(gt); +} + void intel_gt_set_wedged(struct intel_gt *gt) { intel_wakeref_t wakeref; @@ -1614,6 +1623,7 @@ void intel_gt_init_reset(struct intel_gt *gt) init_waitqueue_head(>->reset.queue); mutex_init(>->reset.mutex); init_srcu_struct(>->reset.backoff_srcu); + INIT_WORK(>->wedge, set_wedged_work); /* * While undesirable to wait inside the shrinker, complain anyway. @@ -1640,7 +1650,7 @@ static void intel_wedge_me(struct work_struct *work) struct intel_wedge_me *w = container_of(work, typeof(*w), work.work); gt_err(w->gt, "%s timed out, cancelling all in-flight rendering.\n", w->name); - intel_gt_set_wedged(w->gt); + set_wedged_work(&w->gt->wedge); } void __intel_init_wedge(struct intel_wedge_me *w, diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index 2eba289d88ad..6aa179a3e92a 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -24,6 +24,7 @@ #include #include +#include "gt/intel_gt.h" #include "gt/intel_engine_regs.h" #include "gt/intel_gt_regs.h" @@ -180,14 +181,16 @@ fw_domain_wait_ack_clear(const struct intel_uncore_forcewake_domain *d) if (!wait_ack_clear(d, FORCEWAKE_KERNEL)) return; - if (fw_ack(d) == ~0) + if (fw_ack(d) == ~0) { drm_err(&d->uncore->i915->drm, "%s: MMIO unreliable (forcewake register returns 0xFFFFFFFF)!\n", intel_uncore_forcewake_domain_to_str(d->id)); - else + intel_gt_set_wedged_async(d->uncore->gt); + } else { drm_err(&d->uncore->i915->drm, "%s: timed out waiting for forcewake ack to clear.\n", intel_uncore_forcewake_domain_to_str(d->id)); + } add_taint_for_CI(d->uncore->i915, TAINT_WARN); /* CI now unreliable */ } -- cgit v1.2.3 From 437ad4534a15ccf538ddb440cb29d3f445251605 Mon Sep 17 00:00:00 2001 From: Jesus Narvaez Date: Thu, 8 Aug 2024 13:49:43 -0700 Subject: drm/i915/guc: Change GEM_WARN_ON to guc_err to prevent taints in CI This warning was supposed to catch a harmless issue, but changing to guc_error should prevent kernel taints in CI runs. Signed-off-by: Jesus Narvaez Reviewed-by: Jonathan Cavitt Signed-off-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20240808204943.911727-1-jesus.narvaez@intel.com --- drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c index 9400d0eb682b..c3a5d9e1288e 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c @@ -2014,11 +2014,12 @@ void intel_guc_submission_reset_finish(struct intel_guc *guc) /* * Technically possible for either of these values to be non-zero here, - * but very unlikely + harmless. Regardless let's add a warn so we can + * but very unlikely + harmless. Regardless let's add an error so we can * see in CI if this happens frequently / a precursor to taking down the * machine. */ - GEM_WARN_ON(atomic_read(&guc->outstanding_submission_g2h)); + if (atomic_read(&guc->outstanding_submission_g2h)) + guc_err(guc, "Unexpected outstanding GuC to Host in reset finish\n"); atomic_set(&guc->outstanding_submission_g2h, 0); intel_guc_global_policies_update(guc); -- cgit v1.2.3 From 0523374e303051c94457df9170f12a752d8b09cf Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Tue, 13 Aug 2024 17:06:18 +0300 Subject: drm/i915/gt: remove stray declaration of intel_gt_release_all() When intel_gt_release_all() was removed from the code in commit e89950553385 ("drm/i915: do not clean GT table on error path"), its declaration in the header file remained. Remove it. Signed-off-by: Luca Coelho Reviewed-by: Jonathan Cavitt Reviewed-by: Andi Shyti Signed-off-by: Andi Shyti Link: https://patchwork.freedesktop.org/patch/msgid/20240813140618.387553-1-luciano.coelho@intel.com --- drivers/gpu/drm/i915/gt/intel_gt.h | 1 - 1 file changed, 1 deletion(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gt/intel_gt.h b/drivers/gpu/drm/i915/gt/intel_gt.h index b73555889d50..998ca029b73a 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.h +++ b/drivers/gpu/drm/i915/gt/intel_gt.h @@ -174,7 +174,6 @@ static inline bool intel_gt_is_wedged(const struct intel_gt *gt) int intel_gt_probe_all(struct drm_i915_private *i915); int intel_gt_tiles_init(struct drm_i915_private *i915); -void intel_gt_release_all(struct drm_i915_private *i915); #define for_each_gt(gt__, i915__, id__) \ for ((id__) = 0; \ -- cgit v1.2.3 From 609d8b1c422cd8d1f93bf526fb236f2e07687e7c Mon Sep 17 00:00:00 2001 From: Andi Shyti Date: Wed, 7 Aug 2024 11:05:20 +0100 Subject: drm/i915/gem: Do not look for the exact address in node In preparation for the upcoming partial memory mapping feature, we want to make sure that when looking for a node we consider also the offset and not just the starting address of the virtual memory node. Signed-off-by: Andi Shyti Reviewed-by: Nirmoy Das Link: https://patchwork.freedesktop.org/patch/msgid/20240807100521.478266-2-andi.shyti@linux.intel.com --- drivers/gpu/drm/i915/gem/i915_gem_mman.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index e9b2424156f0..99fde0a05632 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -1079,9 +1079,9 @@ int i915_gem_mmap(struct file *filp, struct vm_area_struct *vma) rcu_read_lock(); drm_vma_offset_lock_lookup(dev->vma_offset_manager); - node = drm_vma_offset_exact_lookup_locked(dev->vma_offset_manager, - vma->vm_pgoff, - vma_pages(vma)); + node = drm_vma_offset_lookup_locked(dev->vma_offset_manager, + vma->vm_pgoff, + vma_pages(vma)); if (node && drm_vma_node_is_allowed(node, priv)) { /* * Skip 0-refcnted objects as it is in the process of being -- cgit v1.2.3 From 255fc1703e42321b5afdedc8259ad03c7cc533ec Mon Sep 17 00:00:00 2001 From: Andi Shyti Date: Wed, 7 Aug 2024 11:05:21 +0100 Subject: drm/i915/gem: Calculate object page offset for partial memory mapping To enable partial memory mapping of GPU virtual memory, it's necessary to introduce an offset to the object's memory (obj->mm.pages) scatterlist. This adjustment compensates for instances when userspace mappings do not start from the beginning of the object. Based on a patch by Chris Wilson. Signed-off-by: Andi Shyti Cc: Chris Wilson Cc: Lionel Landwerlin Reviewed-by: Nirmoy Das Link: https://patchwork.freedesktop.org/patch/msgid/20240807100521.478266-3-andi.shyti@linux.intel.com --- drivers/gpu/drm/i915/gem/i915_gem_mman.c | 4 +++- drivers/gpu/drm/i915/i915_mm.c | 12 +++++++++++- drivers/gpu/drm/i915/i915_mm.h | 3 ++- 3 files changed, 16 insertions(+), 3 deletions(-) (limited to 'drivers/gpu') diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index 99fde0a05632..21274aa9bddd 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -252,6 +252,7 @@ static vm_fault_t vm_fault_cpu(struct vm_fault *vmf) struct vm_area_struct *area = vmf->vma; struct i915_mmap_offset *mmo = area->vm_private_data; struct drm_i915_gem_object *obj = mmo->obj; + unsigned long obj_offset; resource_size_t iomap; int err; @@ -273,10 +274,11 @@ static vm_fault_t vm_fault_cpu(struct vm_fault *vmf) iomap -= obj->mm.region->region.start; } + obj_offset = area->vm_pgoff - drm_vma_node_start(&mmo->vma_node); /* PTEs are revoked in obj->ops->put_pages() */ err = remap_io_sg(area, area->vm_start, area->vm_end - area->vm_start, - obj->mm.pages->sgl, iomap); + obj->mm.pages->sgl, obj_offset, iomap); if (area->vm_flags & VM_WRITE) { GEM_BUG_ON(!i915_gem_object_has_pinned_pages(obj)); diff --git a/drivers/gpu/drm/i915/i915_mm.c b/drivers/gpu/drm/i915/i915_mm.c index 7998bc74ab49..f5c97a620962 100644 --- a/drivers/gpu/drm/i915/i915_mm.c +++ b/drivers/gpu/drm/i915/i915_mm.c @@ -122,13 +122,15 @@ int remap_io_mapping(struct vm_area_struct *vma, * @addr: target user address to start at * @size: size of map area * @sgl: Start sg entry + * @offset: offset from the start of the page * @iobase: Use stored dma address offset by this address or pfn if -1 * * Note: this is only safe if the mm semaphore is held when called. */ int remap_io_sg(struct vm_area_struct *vma, unsigned long addr, unsigned long size, - struct scatterlist *sgl, resource_size_t iobase) + struct scatterlist *sgl, unsigned long offset, + resource_size_t iobase) { struct remap_pfn r = { .mm = vma->vm_mm, @@ -141,6 +143,14 @@ int remap_io_sg(struct vm_area_struct *vma, /* We rely on prevalidation of the io-mapping to skip track_pfn(). */ GEM_BUG_ON((vma->vm_flags & EXPECTED_FLAGS) != EXPECTED_FLAGS); + while (offset >= sg_dma_len(r.sgt.sgp) >> PAGE_SHIFT) { + offset -= sg_dma_len(r.sgt.sgp) >> PAGE_SHIFT; + r.sgt = __sgt_iter(__sg_next(r.sgt.sgp), use_dma(iobase)); + if (!r.sgt.sgp) + return -EINVAL; + } + r.sgt.curr = offset << PAGE_SHIFT; + if (!use_dma(iobase)) flush_cache_range(vma, addr, size); diff --git a/drivers/gpu/drm/i915/i915_mm.h b/drivers/gpu/drm/i915/i915_mm.h index 04c8974d822b..69f9351b1a1c 100644 --- a/drivers/gpu/drm/i915/i915_mm.h +++ b/drivers/gpu/drm/i915/i915_mm.h @@ -30,6 +30,7 @@ int remap_io_mapping(struct vm_area_struct *vma, int remap_io_sg(struct vm_area_struct *vma, unsigned long addr, unsigned long size, - struct scatterlist *sgl, resource_size_t iobase); + struct scatterlist *sgl, unsigned long offset, + resource_size_t iobase); #endif /* __I915_MM_H__ */ -- cgit v1.2.3