From patchwork Tue Dec 29 15:49:01 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?q?Micha=C5=82_Winiarski?= X-Patchwork-Id: 7930221 Return-Path: X-Original-To: patchwork-intel-gfx@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 40879BEEE5 for ; Tue, 29 Dec 2015 15:49:13 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 35EC0201DD for ; Tue, 29 Dec 2015 15:49:12 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) by mail.kernel.org (Postfix) with ESMTP id 1E8AE201BB for ; Tue, 29 Dec 2015 15:49:11 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 7835189A61; Tue, 29 Dec 2015 07:49:10 -0800 (PST) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga14.intel.com (mga14.intel.com [192.55.52.115]) by gabe.freedesktop.org (Postfix) with ESMTP id 08F5A89A61 for ; Tue, 29 Dec 2015 07:49:09 -0800 (PST) Received: from orsmga002.jf.intel.com ([10.7.209.21]) by fmsmga103.fm.intel.com with ESMTP; 29 Dec 2015 07:49:08 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.20,495,1444719600"; d="scan'208";a="880480229" Received: from irsmsx103.ger.corp.intel.com ([163.33.3.157]) by orsmga002.jf.intel.com with ESMTP; 29 Dec 2015 07:49:07 -0800 Received: from mwiniars-desk1.ger.corp.intel.com (172.28.173.39) by IRSMSX103.ger.corp.intel.com (163.33.3.157) with Microsoft SMTP Server id 14.3.248.2; Tue, 29 Dec 2015 15:49:06 +0000 From: =?UTF-8?q?Micha=C5=82=20Winiarski?= To: Date: Tue, 29 Dec 2015 16:49:01 +0100 Message-ID: <1451404141-30354-1-git-send-email-michal.winiarski@intel.com> X-Mailer: git-send-email 2.5.0 In-Reply-To: <1450787830-14443-1-git-send-email-michal.winiarski@intel.com> References: <1450787830-14443-1-git-send-email-michal.winiarski@intel.com> MIME-Version: 1.0 X-Originating-IP: [172.28.173.39] Subject: [Intel-gfx] [PATCH v7] drm/i915: Avoid writing relocs with addresses in non-canonical form X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.18 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" X-Spam-Status: No, score=-4.2 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_MED, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP According to PRM, some parts of HW require the addresses to be in a canonical form, where bits [63:48] == [47]. Let's convert addresses to canonical form prior to relocating and return converted offsets to userspace. We also need to make sure that userspace is using addresses in canonical form in case of softpin. v2: Whitespace fixup, gen8_canonical_addr description (Chris, Ville) v3: Rebase on top of softpin, fix a hole in relocate_entry, s/expect/require (Chris) v4: Handle softpin in validate_exec_list (Chris) v5: Convert back to canonical form at copy_to_user time (Chris) v6: Don't use struct exec_object2 in place of exec_object v7: Use sign_extend64 for converting to canonical form (Joonas), reject non-canonical and non-page-aligned offset for softpin (Chris) Cc: Chris Wilson Cc: Michel Thierry Cc: Ville Syrjälä Signed-off-by: Micha? Winiarski --- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 45 +++++++++++++++++++++++++++--- 1 file changed, 41 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 5d01ea6..dd081fd 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -249,6 +249,25 @@ static inline int use_cpu_reloc(struct drm_i915_gem_object *obj) obj->cache_level != I915_CACHE_NONE); } +/* Used to convert any address to canonical form. + * Starting from gen8, some commands (e.g. STATE_BASE_ADDRESS, + * MI_LOAD_REGISTER_MEM and others, see Broadwell PRM Vol2a) require the + * addresses to be in a canonical form: + * "GraphicsAddress[63:48] are ignored by the HW and assumed to be in correct + * canonical form [63:48] == [47]." + */ +static inline uint64_t gen8_canonical_addr(uint64_t address) +{ + return sign_extend64(address, 47); +} + +static inline uint64_t +relocation_target(struct drm_i915_gem_relocation_entry *reloc, + uint64_t target_offset) +{ + return gen8_canonical_addr((int)reloc->delta + target_offset); +} + static int relocate_entry_cpu(struct drm_i915_gem_object *obj, struct drm_i915_gem_relocation_entry *reloc, @@ -256,7 +275,7 @@ relocate_entry_cpu(struct drm_i915_gem_object *obj, { struct drm_device *dev = obj->base.dev; uint32_t page_offset = offset_in_page(reloc->offset); - uint64_t delta = reloc->delta + target_offset; + uint64_t delta = relocation_target(reloc, target_offset); char *vaddr; int ret; @@ -292,7 +311,7 @@ relocate_entry_gtt(struct drm_i915_gem_object *obj, { struct drm_device *dev = obj->base.dev; struct drm_i915_private *dev_priv = dev->dev_private; - uint64_t delta = reloc->delta + target_offset; + uint64_t delta = relocation_target(reloc, target_offset); uint64_t offset; void __iomem *reloc_page; int ret; @@ -347,7 +366,7 @@ relocate_entry_clflush(struct drm_i915_gem_object *obj, { struct drm_device *dev = obj->base.dev; uint32_t page_offset = offset_in_page(reloc->offset); - uint64_t delta = (int)reloc->delta + target_offset; + uint64_t delta = relocation_target(reloc, target_offset); char *vaddr; int ret; @@ -395,7 +414,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj, target_i915_obj = target_vma->obj; target_obj = &target_vma->obj->base; - target_offset = target_vma->node.start; + target_offset = gen8_canonical_addr(target_vma->node.start); /* Sandybridge PPGTT errata: We need a global gtt mapping for MI and * pipe_control writes because the gpu doesn't properly redirect them @@ -994,6 +1013,20 @@ validate_exec_list(struct drm_device *dev, if (exec[i].flags & invalid_flags) return -EINVAL; + /* Offset can be used as input (EXEC_OBJECT_PINNED), reject + * any non-page-aligned or non-canonial addresses. + */ + if (exec[i].flags & EXEC_OBJECT_PINNED && + (exec[i].offset != gen8_canonical_addr(exec[i].offset) || + offset_in_page(exec[i].offset))) + return -EINVAL; + + /* From drm_mm perspective address space is continuous, + * so from this point we're always using non-canonical form + * internally. + */ + exec[i].offset &= (1ULL << 48) - 1; + if (exec[i].alignment && !is_power_of_2(exec[i].alignment)) return -EINVAL; @@ -1687,6 +1720,8 @@ i915_gem_execbuffer(struct drm_device *dev, void *data, /* Copy the new buffer offsets back to the user's exec list. */ for (i = 0; i < args->buffer_count; i++) { + exec2_list[i].offset = + gen8_canonical_addr(exec2_list[i].offset); ret = __copy_to_user(&user_exec_list[i].offset, &exec2_list[i].offset, sizeof(user_exec_list[i].offset)); @@ -1752,6 +1787,8 @@ i915_gem_execbuffer2(struct drm_device *dev, void *data, int i; for (i = 0; i < args->buffer_count; i++) { + exec2_list[i].offset = + gen8_canonical_addr(exec2_list[i].offset); ret = __copy_to_user(&user_exec_list[i].offset, &exec2_list[i].offset, sizeof(user_exec_list[i].offset));