From patchwork Wed Oct 27 16:18:10 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Matthew Auld X-Patchwork-Id: 12587573 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id E857DC433F5 for ; Wed, 27 Oct 2021 16:19:52 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id AC59460E76 for ; Wed, 27 Oct 2021 16:19:52 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org AC59460E76 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id E5A5A6E5A9; Wed, 27 Oct 2021 16:19:51 +0000 (UTC) Received: from mga06.intel.com (mga06.intel.com [134.134.136.31]) by gabe.freedesktop.org (Postfix) with ESMTPS id DD8606E5A9; Wed, 27 Oct 2021 16:19:50 +0000 (UTC) X-IronPort-AV: E=McAfee;i="6200,9189,10150"; a="291038180" X-IronPort-AV: E=Sophos;i="5.87,187,1631602800"; d="scan'208";a="291038180" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 27 Oct 2021 09:19:47 -0700 X-IronPort-AV: E=Sophos;i="5.87,187,1631602800"; d="scan'208";a="597417902" Received: from shetherx-mobl.ger.corp.intel.com (HELO mwauld-desk1.intel.com) ([10.213.218.37]) by orsmga004-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 27 Oct 2021 09:19:45 -0700 From: Matthew Auld To: intel-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org, =?utf-8?q?Thomas_Hellstr=C3=B6m?= Date: Wed, 27 Oct 2021 17:18:10 +0100 Message-Id: <20211027161813.3094681-1-matthew.auld@intel.com> X-Mailer: git-send-email 2.26.3 MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH v2 1/4] drm/i915/clflush: fixup handling of cache_dirty X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" In theory if clflush_work_create() somehow fails here, and we don't yet have mm.pages populated then we end up resetting cache_dirty, which is likely wrong, since that will potentially skip the flush-on-acquire, if it was needed. It looks like intel_user_framebuffer_dirty() can arrive here before the pages are populated. v2(Thomas): - Move setting cache_dirty out of the async portion, also add a comment for why that should still be safe. Signed-off-by: Matthew Auld Cc: Thomas Hellström --- drivers/gpu/drm/i915/gem/i915_gem_clflush.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c index f0435c6feb68..47586a8a1b73 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c @@ -109,12 +109,20 @@ bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, I915_FENCE_GFP); dma_resv_add_excl_fence(obj->base.resv, &clflush->base.dma); dma_fence_work_commit(&clflush->base); + /* + * We must have successfully populated the pages(since we are + * holding a pin on the pages as per the flush worker) to reach + * this point, which must mean we have already done the required + * flush-on-acquire, hence resetting cache_dirty here should be + * safe. + */ + obj->cache_dirty = false; } else if (obj->mm.pages) { __do_clflush(obj); + obj->cache_dirty = false; } else { GEM_BUG_ON(obj->write_domain != I915_GEM_DOMAIN_CPU); } - obj->cache_dirty = false; return true; } From patchwork Wed Oct 27 16:18:11 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Matthew Auld X-Patchwork-Id: 12587575 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id EBDF2C433EF for ; Wed, 27 Oct 2021 16:19:58 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id B83F9603E8 for ; Wed, 27 Oct 2021 16:19:58 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org B83F9603E8 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 5CCDF6E8B6; Wed, 27 Oct 2021 16:19:54 +0000 (UTC) Received: from mga06.intel.com (mga06.intel.com [134.134.136.31]) by gabe.freedesktop.org (Postfix) with ESMTPS id 88EE26E7D0; Wed, 27 Oct 2021 16:19:52 +0000 (UTC) X-IronPort-AV: E=McAfee;i="6200,9189,10150"; a="291038189" X-IronPort-AV: E=Sophos;i="5.87,187,1631602800"; d="scan'208";a="291038189" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 27 Oct 2021 09:19:48 -0700 X-IronPort-AV: E=Sophos;i="5.87,187,1631602800"; d="scan'208";a="597417905" Received: from shetherx-mobl.ger.corp.intel.com (HELO mwauld-desk1.intel.com) ([10.213.218.37]) by orsmga004-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 27 Oct 2021 09:19:47 -0700 From: Matthew Auld To: intel-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org, =?utf-8?q?Thomas_Hellstr=C3=B6m?= Date: Wed, 27 Oct 2021 17:18:11 +0100 Message-Id: <20211027161813.3094681-2-matthew.auld@intel.com> X-Mailer: git-send-email 2.26.3 In-Reply-To: <20211027161813.3094681-1-matthew.auld@intel.com> References: <20211027161813.3094681-1-matthew.auld@intel.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH v2 2/4] drm/i915/clflush: disallow on discrete X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" We seem to have an unfortunate issue where we arrive from: i915_gem_object_flush_if_display+0x86/0xd0 [i915] intel_user_framebuffer_dirty+0x1a/0x50 [i915] drm_mode_dirtyfb_ioctl+0xfb/0x1b0 which can be before the pages are populated(and pinned for display), and so i915_gem_object_has_struct_page() might still return true, as per the ttm backend. We could re-order the later get_pages() call here, but since on discrete everything should already be coherent, with the exception of the display engine, and even there display surfaces must be allocated in device local-memory anyway, so there should in theory be no conceivable reason to ever call i915_gem_clflush_object() on discrete. References: https://gitlab.freedesktop.org/drm/intel/-/issues/4320 Signed-off-by: Matthew Auld Cc: Thomas Hellström Reviewed-by: Thomas Hellström --- drivers/gpu/drm/i915/gem/i915_gem_clflush.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c index 47586a8a1b73..aa91252c9324 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c @@ -69,6 +69,8 @@ static struct clflush *clflush_work_create(struct drm_i915_gem_object *obj) bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, unsigned int flags) { + + struct drm_i915_private *i915 = to_i915(obj->base.dev); struct clflush *clflush; assert_object_held(obj); @@ -80,7 +82,7 @@ bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, * anything not backed by physical memory we consider to be always * coherent and not need clflushing. */ - if (!i915_gem_object_has_struct_page(obj)) { + if (!i915_gem_object_has_struct_page(obj) || IS_DGFX(i915)) { obj->cache_dirty = false; return false; } @@ -105,7 +107,7 @@ bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, if (clflush) { i915_sw_fence_await_reservation(&clflush->base.chain, obj->base.resv, NULL, true, - i915_fence_timeout(to_i915(obj->base.dev)), + i915_fence_timeout(i915), I915_FENCE_GFP); dma_resv_add_excl_fence(obj->base.resv, &clflush->base.dma); dma_fence_work_commit(&clflush->base); From patchwork Wed Oct 27 16:18:12 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Matthew Auld X-Patchwork-Id: 12587577 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 34C2EC433F5 for ; Wed, 27 Oct 2021 16:20:01 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 024D860E76 for ; Wed, 27 Oct 2021 16:20:00 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 024D860E76 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 323066E7D0; Wed, 27 Oct 2021 16:19:56 +0000 (UTC) Received: from mga06.intel.com (mga06.intel.com [134.134.136.31]) by gabe.freedesktop.org (Postfix) with ESMTPS id 25C826E7D0; Wed, 27 Oct 2021 16:19:54 +0000 (UTC) X-IronPort-AV: E=McAfee;i="6200,9189,10150"; a="291038211" X-IronPort-AV: E=Sophos;i="5.87,187,1631602800"; d="scan'208";a="291038211" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 27 Oct 2021 09:19:49 -0700 X-IronPort-AV: E=Sophos;i="5.87,187,1631602800"; d="scan'208";a="597417913" Received: from shetherx-mobl.ger.corp.intel.com (HELO mwauld-desk1.intel.com) ([10.213.218.37]) by orsmga004-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 27 Oct 2021 09:19:48 -0700 From: Matthew Auld To: intel-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org, =?utf-8?q?Thomas_Hellstr=C3=B6m?= Date: Wed, 27 Oct 2021 17:18:12 +0100 Message-Id: <20211027161813.3094681-3-matthew.auld@intel.com> X-Mailer: git-send-email 2.26.3 In-Reply-To: <20211027161813.3094681-1-matthew.auld@intel.com> References: <20211027161813.3094681-1-matthew.auld@intel.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH v2 3/4] drm/i915: move cpu_write_needs_clflush X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" Move it next to its partner in crime; gpu_write_needs_clflush. For better readability lets keep gpu vs cpu at least in the same file. Signed-off-by: Matthew Auld Cc: Thomas Hellström Reviewed-by: Thomas Hellström --- drivers/gpu/drm/i915/gem/i915_gem_domain.c | 12 ++++++++++++ drivers/gpu/drm/i915/gem/i915_gem_object.h | 15 ++------------- drivers/gpu/drm/i915/i915_gem.c | 2 +- 3 files changed, 15 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c index b684a62bf3b0..d30d5a699788 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c @@ -22,6 +22,18 @@ static bool gpu_write_needs_clflush(struct drm_i915_gem_object *obj) obj->cache_level == I915_CACHE_WT); } +bool i915_gem_cpu_write_needs_clflush(struct drm_i915_gem_object *obj) +{ + if (obj->cache_dirty) + return false; + + if (!(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE)) + return true; + + /* Currently in use by HW (display engine)? Keep flushed. */ + return i915_gem_object_is_framebuffer(obj); +} + static void flush_write_domain(struct drm_i915_gem_object *obj, unsigned int flush_domains) { diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h index a5479ac7a4ad..de0c8145cc4c 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h @@ -522,6 +522,7 @@ void i915_gem_object_set_cache_coherency(struct drm_i915_gem_object *obj, bool i915_gem_object_can_bypass_llc(struct drm_i915_gem_object *obj); void i915_gem_object_flush_if_display(struct drm_i915_gem_object *obj); void i915_gem_object_flush_if_display_locked(struct drm_i915_gem_object *obj); +bool i915_gem_cpu_write_needs_clflush(struct drm_i915_gem_object *obj); int __must_check i915_gem_object_set_to_wc_domain(struct drm_i915_gem_object *obj, bool write); @@ -542,23 +543,11 @@ void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj); void __i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj); void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj); -static inline bool cpu_write_needs_clflush(struct drm_i915_gem_object *obj) -{ - if (obj->cache_dirty) - return false; - - if (!(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE)) - return true; - - /* Currently in use by HW (display engine)? Keep flushed. */ - return i915_gem_object_is_framebuffer(obj); -} - static inline void __start_cpu_write(struct drm_i915_gem_object *obj) { obj->read_domains = I915_GEM_DOMAIN_CPU; obj->write_domain = I915_GEM_DOMAIN_CPU; - if (cpu_write_needs_clflush(obj)) + if (i915_gem_cpu_write_needs_clflush(obj)) obj->cache_dirty = true; } diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 981e383d1a5d..d0e642c82064 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -764,7 +764,7 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data, * perspective, requiring manual detiling by the client. */ if (!i915_gem_object_has_struct_page(obj) || - cpu_write_needs_clflush(obj)) + i915_gem_cpu_write_needs_clflush(obj)) /* Note that the gtt paths might fail with non-page-backed user * pointers (e.g. gtt mappings when moving data between * textures). Fallback to the shmem path in that case. From patchwork Wed Oct 27 16:18:13 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Matthew Auld X-Patchwork-Id: 12587579 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id F3DB0C433FE for ; Wed, 27 Oct 2021 16:20:12 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id B9D6D603E8 for ; Wed, 27 Oct 2021 16:20:12 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org B9D6D603E8 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 3A9B86E8C9; Wed, 27 Oct 2021 16:20:03 +0000 (UTC) Received: from mga06.intel.com (mga06.intel.com [134.134.136.31]) by gabe.freedesktop.org (Postfix) with ESMTPS id 362FF6E7D0; Wed, 27 Oct 2021 16:19:55 +0000 (UTC) X-IronPort-AV: E=McAfee;i="6200,9189,10150"; a="291038231" X-IronPort-AV: E=Sophos;i="5.87,187,1631602800"; d="scan'208";a="291038231" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 27 Oct 2021 09:19:51 -0700 X-IronPort-AV: E=Sophos;i="5.87,187,1631602800"; d="scan'208";a="597417917" Received: from shetherx-mobl.ger.corp.intel.com (HELO mwauld-desk1.intel.com) ([10.213.218.37]) by orsmga004-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 27 Oct 2021 09:19:49 -0700 From: Matthew Auld To: intel-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org, =?utf-8?q?Thomas_Hellstr=C3=B6m?= Date: Wed, 27 Oct 2021 17:18:13 +0100 Message-Id: <20211027161813.3094681-4-matthew.auld@intel.com> X-Mailer: git-send-email 2.26.3 In-Reply-To: <20211027161813.3094681-1-matthew.auld@intel.com> References: <20211027161813.3094681-1-matthew.auld@intel.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH v2 4/4] drm/i915: stop setting cache_dirty on discrete X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" Should not be needed. Even with non-coherent display, we should be using device local-memory there, and not system memory. v2: also add a warning in i915_gem_clflush_object Signed-off-by: Matthew Auld Cc: Thomas Hellström Reviewed-by: Thomas Hellström #v1 --- drivers/gpu/drm/i915/gem/i915_gem_clflush.c | 7 ++++++- drivers/gpu/drm/i915/gem/i915_gem_domain.c | 10 ++++++++++ drivers/gpu/drm/i915/gem/i915_gem_object.c | 7 +++++-- drivers/gpu/drm/i915/gem/i915_gem_pages.c | 1 + 4 files changed, 22 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c index aa91252c9324..539b2116b2d5 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c @@ -75,6 +75,11 @@ bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, assert_object_held(obj); + if (IS_DGFX(i915)) { + WARN_ON_ONCE(obj->cache_dirty); + return false; + } + /* * Stolen memory is always coherent with the GPU as it is explicitly * marked as wc by the system, or the system is cache-coherent. @@ -82,7 +87,7 @@ bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, * anything not backed by physical memory we consider to be always * coherent and not need clflushing. */ - if (!i915_gem_object_has_struct_page(obj) || IS_DGFX(i915)) { + if (!i915_gem_object_has_struct_page(obj)) { obj->cache_dirty = false; return false; } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c index d30d5a699788..26532c07d467 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c @@ -18,18 +18,28 @@ static bool gpu_write_needs_clflush(struct drm_i915_gem_object *obj) { + struct drm_i915_private *i915 = to_i915(obj->base.dev); + + if (IS_DGFX(i915)) + return false; + return !(obj->cache_level == I915_CACHE_NONE || obj->cache_level == I915_CACHE_WT); } bool i915_gem_cpu_write_needs_clflush(struct drm_i915_gem_object *obj) { + struct drm_i915_private *i915 = to_i915(obj->base.dev); + if (obj->cache_dirty) return false; if (!(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE)) return true; + if (IS_DGFX(i915)) + return false; + /* Currently in use by HW (display engine)? Keep flushed. */ return i915_gem_object_is_framebuffer(obj); } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index 55b0f1df3192..591ee3cb7275 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -114,18 +114,21 @@ void __i915_gem_object_fini(struct drm_i915_gem_object *obj) void i915_gem_object_set_cache_coherency(struct drm_i915_gem_object *obj, unsigned int cache_level) { + struct drm_i915_private *i915 = to_i915(obj->base.dev); + obj->cache_level = cache_level; if (cache_level != I915_CACHE_NONE) obj->cache_coherent = (I915_BO_CACHE_COHERENT_FOR_READ | I915_BO_CACHE_COHERENT_FOR_WRITE); - else if (HAS_LLC(to_i915(obj->base.dev))) + else if (HAS_LLC(i915)) obj->cache_coherent = I915_BO_CACHE_COHERENT_FOR_READ; else obj->cache_coherent = 0; obj->cache_dirty = - !(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE); + !(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE) && + !IS_DGFX(i915); } bool i915_gem_object_can_bypass_llc(struct drm_i915_gem_object *obj) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c index 308e22a80af4..c4f684b7cc51 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c @@ -26,6 +26,7 @@ void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj, /* Make the pages coherent with the GPU (flushing any swapin). */ if (obj->cache_dirty) { + WARN_ON_ONCE(IS_DGFX(i915)); obj->write_domain = 0; if (i915_gem_object_has_struct_page(obj)) drm_clflush_sg(pages);