From patchwork Wed May 25 18:43:33 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Matthew Auld X-Patchwork-Id: 12861518 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 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 smtp.lore.kernel.org (Postfix) with ESMTPS id 38632C433F5 for ; Wed, 25 May 2022 18:44:17 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 2EC0A10ED01; Wed, 25 May 2022 18:44:12 +0000 (UTC) Received: from mga14.intel.com (mga14.intel.com [192.55.52.115]) by gabe.freedesktop.org (Postfix) with ESMTPS id C678610ECE0; Wed, 25 May 2022 18:44:10 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1653504250; x=1685040250; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=ooq/8uQ+0bMqQE+OurAxOXHZ6vnOUvS3Jpu4qkOFL0A=; b=DwOshQ6s1qY1MWYWDqmjGBrWP9lhKlhqznC3pcrs91ua8vaVyadvP/p7 QND7eUTlVkvMUbZ1w+rsPvz1kpvWaVuI5NkyOyiQVh1n8DvKWltccWVap EBMacFBixRigVJtRCQBgG17jQEw4dkLRBupYOa7H0EvXohmXsZzG5FrUC mwMylcmHulRvCSlVdpZv+8fK6u7Le3rmS5l0geMOqSZWbeU3DoM5FjcpX 5FPxQzhf8qsVdm/5Yu7CMwUk6HPGjp9LsPZNXR+UmhgPR1zpXKO2sZANm ZQrrpmZce2SAeVkbH2Ai1KmAWpS5tKL2PHlKod20bnq4e4M80Jmz6ncyq A==; X-IronPort-AV: E=McAfee;i="6400,9594,10358"; a="274013469" X-IronPort-AV: E=Sophos;i="5.91,250,1647327600"; d="scan'208";a="274013469" Received: from fmsmga007.fm.intel.com ([10.253.24.52]) by fmsmga103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 25 May 2022 11:44:10 -0700 X-IronPort-AV: E=Sophos;i="5.91,250,1647327600"; d="scan'208";a="578465267" Received: from daithiby-mobl.ger.corp.intel.com (HELO mwauld-desk1.intel.com) ([10.252.5.16]) by fmsmga007-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 25 May 2022 11:44:07 -0700 From: Matthew Auld To: intel-gfx@lists.freedesktop.org Date: Wed, 25 May 2022 19:43:33 +0100 Message-Id: <20220525184337.491763-7-matthew.auld@intel.com> X-Mailer: git-send-email 2.34.3 In-Reply-To: <20220525184337.491763-1-matthew.auld@intel.com> References: <20220525184337.491763-1-matthew.auld@intel.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH 06/10] drm/i915/uapi: add NEEDS_CPU_ACCESS hint 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: , Cc: =?utf-8?q?Thomas_Hellstr=C3=B6m?= , Kenneth Graunke , dri-devel@lists.freedesktop.org, Daniel Vetter Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" If set, force the allocation to be placed in the mappable portion of I915_MEMORY_CLASS_DEVICE. One big restriction here is that system memory (i.e I915_MEMORY_CLASS_SYSTEM) must be given as a potential placement for the object, that way we can always spill the object into system memory if we can't make space. Testcase: igt@gem-create@create-ext-cpu-access-sanity-check Testcase: igt@gem-create@create-ext-cpu-access-big Signed-off-by: Matthew Auld Cc: Thomas Hellström Cc: Lionel Landwerlin Cc: Jon Bloomfield Cc: Daniel Vetter Cc: Jordan Justen Cc: Kenneth Graunke Cc: Akeem G Abodunrin Reviewed-by: Thomas Hellström --- drivers/gpu/drm/i915/gem/i915_gem_create.c | 26 ++++++--- include/uapi/drm/i915_drm.h | 61 +++++++++++++++++++--- 2 files changed, 71 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_create.c b/drivers/gpu/drm/i915/gem/i915_gem_create.c index d094cae0ddf1..33673fe7ee0a 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_create.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_create.c @@ -241,6 +241,7 @@ struct create_ext { struct drm_i915_private *i915; struct intel_memory_region *placements[INTEL_REGION_UNKNOWN]; unsigned int n_placements; + unsigned int placement_mask; unsigned long flags; }; @@ -337,6 +338,7 @@ static int set_placements(struct drm_i915_gem_create_ext_memory_regions *args, for (i = 0; i < args->num_regions; i++) ext_data->placements[i] = placements[i]; + ext_data->placement_mask = mask; return 0; out_dump: @@ -411,7 +413,7 @@ i915_gem_create_ext_ioctl(struct drm_device *dev, void *data, struct drm_i915_gem_object *obj; int ret; - if (args->flags) + if (args->flags & ~I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS) return -EINVAL; ret = i915_user_extensions(u64_to_user_ptr(args->extensions), @@ -427,13 +429,21 @@ i915_gem_create_ext_ioctl(struct drm_device *dev, void *data, ext_data.n_placements = 1; } - /* - * TODO: add a userspace hint to force CPU_ACCESS for the object, which - * can override this. - */ - if (ext_data.n_placements > 1 || - ext_data.placements[0]->type != INTEL_MEMORY_SYSTEM) - ext_data.flags |= I915_BO_ALLOC_GPU_ONLY; + if (args->flags & I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS) { + if (ext_data.n_placements == 1) + return -EINVAL; + + /* + * We always need to be able to spill to system memory, if we + * can't place in the mappable part of LMEM. + */ + if (!(ext_data.placement_mask & BIT(INTEL_REGION_SMEM))) + return -EINVAL; + } else { + if (ext_data.n_placements > 1 || + ext_data.placements[0]->type != INTEL_MEMORY_SYSTEM) + ext_data.flags |= I915_BO_ALLOC_GPU_ONLY; + } obj = __i915_gem_object_create_user_ext(i915, args->size, ext_data.placements, diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index e30f31a440b3..5b0a10e6a1b8 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -3366,11 +3366,11 @@ struct drm_i915_query_memory_regions { * struct drm_i915_gem_create_ext - Existing gem_create behaviour, with added * extension support using struct i915_user_extension. * - * Note that in the future we want to have our buffer flags here, at least for - * the stuff that is immutable. Previously we would have two ioctls, one to - * create the object with gem_create, and another to apply various parameters, - * however this creates some ambiguity for the params which are considered - * immutable. Also in general we're phasing out the various SET/GET ioctls. + * Note that new buffer flags should be added here, at least for the stuff that + * is immutable. Previously we would have two ioctls, one to create the object + * with gem_create, and another to apply various parameters, however this + * creates some ambiguity for the params which are considered immutable. Also in + * general we're phasing out the various SET/GET ioctls. */ struct drm_i915_gem_create_ext { /** @@ -3378,7 +3378,6 @@ struct drm_i915_gem_create_ext { * * The (page-aligned) allocated size for the object will be returned. * - * * DG2 64K min page size implications: * * On discrete platforms, starting from DG2, we have to contend with GTT @@ -3390,7 +3389,9 @@ struct drm_i915_gem_create_ext { * * Note that the returned size here will always reflect any required * rounding up done by the kernel, i.e 4K will now become 64K on devices - * such as DG2. + * such as DG2. The kernel will always select the largest minimum + * page-size for the set of possible placements as the value to use when + * rounding up the @size. * * Special DG2 GTT address alignment requirement: * @@ -3414,14 +3415,58 @@ struct drm_i915_gem_create_ext { * is deemed to be a good compromise. */ __u64 size; + /** * @handle: Returned handle for the object. * * Object handles are nonzero. */ __u32 handle; - /** @flags: MBZ */ + + /** + * @flags: Optional flags. + * + * Supported values: + * + * I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS - Signal to the kernel that + * the object will need to be accessed via the CPU. + * + * Only valid when placing objects in I915_MEMORY_CLASS_DEVICE, and only + * strictly required on configurations where some subset of the device + * memory is directly visible/mappable through the CPU (which we also + * call small BAR), like on some DG2+ systems. Note that this is quite + * undesirable, but due to various factors like the client CPU, BIOS etc + * it's something we can expect to see in the wild. See + * &drm_i915_memory_region_info.probed_cpu_visible_size for how to + * determine if this system applies. + * + * Note that one of the placements MUST be I915_MEMORY_CLASS_SYSTEM, to + * ensure the kernel can always spill the allocation to system memory, + * if the object can't be allocated in the mappable part of + * I915_MEMORY_CLASS_DEVICE. + * + * Also note that since the kernel only supports flat-CCS on objects + * that can *only* be placed in I915_MEMORY_CLASS_DEVICE, we therefore + * don't support I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS together with + * flat-CCS. + * + * Without this hint, the kernel will assume that non-mappable + * I915_MEMORY_CLASS_DEVICE is preferred for this object. Note that the + * kernel can still migrate the object to the mappable part, as a last + * resort, if userspace ever CPU faults this object, but this might be + * expensive, and so ideally should be avoided. + * + * On older kernels which lack the relevant small-bar uAPI support (see + * also &drm_i915_memory_region_info.probed_cpu_visible_size), + * usage of the flag will result in an error, but it should NEVER be + * possible to end up with a small BAR configuration, assuming we can + * also successfully load the i915 kernel module. In such cases the + * entire I915_MEMORY_CLASS_DEVICE region will be CPU accessible, and as + * such there are zero restrictions on where the object can be placed. + */ +#define I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS (1 << 0) __u32 flags; + /** * @extensions: The chain of extensions to apply to this object. *