From patchwork Tue Aug 4 08:59:53 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Lionel Landwerlin X-Patchwork-Id: 11699983 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 1B12E14B7 for ; Tue, 4 Aug 2020 09:00:08 +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 35DE620792 for ; Tue, 4 Aug 2020 09:00:08 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 35DE620792 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=intel-gfx-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 760826E451; Tue, 4 Aug 2020 09:00:07 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by gabe.freedesktop.org (Postfix) with ESMTPS id 128C96E450 for ; Tue, 4 Aug 2020 09:00:06 +0000 (UTC) IronPort-SDR: p0bX2VD1fjcW4LEVZXaPPVxPaFDHtgyV4YegCM7AFKsFLmbaoxCUVNeu5WFGaRMotJ3qJTFgG7 l7fTNnzpPceQ== X-IronPort-AV: E=McAfee;i="6000,8403,9702"; a="153442626" X-IronPort-AV: E=Sophos;i="5.75,433,1589266800"; d="scan'208";a="153442626" X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga004.fm.intel.com ([10.253.24.48]) by orsmga102.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 04 Aug 2020 02:00:05 -0700 IronPort-SDR: FG7FJg3PA0pQkjbRe5F12oVE5zwaKvwPjEo+jtFHe+2IC0o/P9NmOGcxfzBgyVzHWwUlhF8yEn V746HH3pxYNw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.75,433,1589266800"; d="scan'208";a="315261490" Received: from wrottner-mobl1.ger.corp.intel.com (HELO delly.ger.corp.intel.com) ([10.252.61.120]) by fmsmga004.fm.intel.com with ESMTP; 04 Aug 2020 02:00:03 -0700 From: Lionel Landwerlin To: intel-gfx@lists.freedesktop.org Date: Tue, 4 Aug 2020 11:59:53 +0300 Message-Id: <20200804085954.350343-2-lionel.g.landwerlin@intel.com> X-Mailer: git-send-email 2.28.0 In-Reply-To: <20200804085954.350343-1-lionel.g.landwerlin@intel.com> References: <20200804085954.350343-1-lionel.g.landwerlin@intel.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH 1/2] drm/i915: introduce a mechanism to extend execbuf2 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: Daniel Vetter , Chris Wilson Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" We're planning to use this for a couple of new feature where we need to provide additional parameters to execbuf. v2: Check for invalid flags in execbuffer2 (Lionel) v3: Rename I915_EXEC_EXT -> I915_EXEC_USE_EXTENSIONS (Chris) v4: Rebase Move array fence parsing in i915_gem_do_execbuffer() Signed-off-by: Lionel Landwerlin Reviewed-by: Chris Wilson (v1) Reviewed-by: Daniel Vetter Reviewed-by: Chris Wilson --- .../gpu/drm/i915/gem/i915_gem_execbuffer.c | 131 +++++++++++------- include/uapi/drm/i915_drm.h | 26 +++- 2 files changed, 103 insertions(+), 54 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index 07cb2dd0f795..ed8d1c2517f6 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -26,6 +26,7 @@ #include "i915_gem_ioctls.h" #include "i915_sw_fence_work.h" #include "i915_trace.h" +#include "i915_user_extensions.h" struct eb_vma { struct i915_vma *vma; @@ -281,6 +282,13 @@ struct i915_execbuffer { int lut_size; struct hlist_head *buckets; /** ht for relocation handles */ struct eb_vma_array *array; + + struct i915_eb_fence { + struct drm_syncobj *syncobj; /* Use with ptr_mask_bits() */ + } *fences; + u32 n_fences; + + u64 extension_flags; /** Available extensions parameters */ }; static inline bool eb_use_cmdparser(const struct i915_execbuffer *eb) @@ -1622,7 +1630,8 @@ static int i915_gem_check_execbuffer(struct drm_i915_gem_execbuffer2 *exec) return -EINVAL; /* Kernel clipping was a DRI1 misfeature */ - if (!(exec->flags & I915_EXEC_FENCE_ARRAY)) { + if (!(exec->flags & (I915_EXEC_FENCE_ARRAY | + I915_EXEC_USE_EXTENSIONS))) { if (exec->num_cliprects || exec->cliprects_ptr) return -EINVAL; } @@ -2189,41 +2198,41 @@ eb_pin_engine(struct i915_execbuffer *eb, } static void -__free_fence_array(struct drm_syncobj **fences, unsigned int n) +__free_fence_array(struct i915_eb_fence *fences, unsigned int n) { while (n--) - drm_syncobj_put(ptr_mask_bits(fences[n], 2)); + drm_syncobj_put(ptr_mask_bits(fences[n].syncobj, 2)); kvfree(fences); } -static struct drm_syncobj ** +static int get_fence_array(struct drm_i915_gem_execbuffer2 *args, - struct drm_file *file) + struct i915_execbuffer *eb) { const unsigned long nfences = args->num_cliprects; struct drm_i915_gem_exec_fence __user *user; - struct drm_syncobj **fences; + struct i915_eb_fence *fences; unsigned long n; int err; if (!(args->flags & I915_EXEC_FENCE_ARRAY)) - return NULL; + return 0; /* Check multiplication overflow for access_ok() and kvmalloc_array() */ BUILD_BUG_ON(sizeof(size_t) > sizeof(unsigned long)); if (nfences > min_t(unsigned long, ULONG_MAX / sizeof(*user), SIZE_MAX / sizeof(*fences))) - return ERR_PTR(-EINVAL); + return -EINVAL; user = u64_to_user_ptr(args->cliprects_ptr); if (!access_ok(user, nfences * sizeof(*user))) - return ERR_PTR(-EFAULT); + return -EFAULT; fences = kvmalloc_array(nfences, sizeof(*fences), __GFP_NOWARN | GFP_KERNEL); if (!fences) - return ERR_PTR(-ENOMEM); + return -ENOMEM; for (n = 0; n < nfences; n++) { struct drm_i915_gem_exec_fence fence; @@ -2239,7 +2248,7 @@ get_fence_array(struct drm_i915_gem_execbuffer2 *args, goto err; } - syncobj = drm_syncobj_find(file, fence.handle); + syncobj = drm_syncobj_find(eb->file, fence.handle); if (!syncobj) { DRM_DEBUG("Invalid syncobj handle provided\n"); err = -ENOENT; @@ -2249,38 +2258,31 @@ get_fence_array(struct drm_i915_gem_execbuffer2 *args, BUILD_BUG_ON(~(ARCH_KMALLOC_MINALIGN - 1) & ~__I915_EXEC_FENCE_UNKNOWN_FLAGS); - fences[n] = ptr_pack_bits(syncobj, fence.flags, 2); + fences[n].syncobj = ptr_pack_bits(syncobj, fence.flags, 2); } - return fences; + eb->fences = fences; + eb->n_fences = nfences; + + return 0; err: __free_fence_array(fences, n); - return ERR_PTR(err); -} - -static void -put_fence_array(struct drm_i915_gem_execbuffer2 *args, - struct drm_syncobj **fences) -{ - if (fences) - __free_fence_array(fences, args->num_cliprects); + return err; } static int -await_fence_array(struct i915_execbuffer *eb, - struct drm_syncobj **fences) +await_fence_array(struct i915_execbuffer *eb) { - const unsigned int nfences = eb->args->num_cliprects; unsigned int n; int err; - for (n = 0; n < nfences; n++) { + for (n = 0; n < eb->n_fences; n++) { struct drm_syncobj *syncobj; struct dma_fence *fence; unsigned int flags; - syncobj = ptr_unpack_bits(fences[n], &flags, 2); + syncobj = ptr_unpack_bits(eb->fences[n].syncobj, &flags, 2); if (!(flags & I915_EXEC_FENCE_WAIT)) continue; @@ -2298,18 +2300,16 @@ await_fence_array(struct i915_execbuffer *eb, } static void -signal_fence_array(struct i915_execbuffer *eb, - struct drm_syncobj **fences) +signal_fence_array(struct i915_execbuffer *eb) { - const unsigned int nfences = eb->args->num_cliprects; struct dma_fence * const fence = &eb->request->fence; unsigned int n; - for (n = 0; n < nfences; n++) { + for (n = 0; n < eb->n_fences; n++) { struct drm_syncobj *syncobj; unsigned int flags; - syncobj = ptr_unpack_bits(fences[n], &flags, 2); + syncobj = ptr_unpack_bits(eb->fences[n].syncobj, &flags, 2); if (!(flags & I915_EXEC_FENCE_SIGNAL)) continue; @@ -2358,12 +2358,38 @@ static void eb_request_add(struct i915_execbuffer *eb) mutex_unlock(&tl->mutex); } +static const i915_user_extension_fn execbuf_extensions[] = { +}; + +static int +parse_execbuf2_extensions(struct drm_i915_gem_execbuffer2 *args, + struct i915_execbuffer *eb) +{ + eb->extension_flags = 0; + + if (!(args->flags & I915_EXEC_USE_EXTENSIONS)) + return 0; + + /* The execbuf2 extension mechanism reuses cliprects_ptr. So we cannot + * have another flag also using it at the same time. + */ + if (eb->args->flags & I915_EXEC_FENCE_ARRAY) + return -EINVAL; + + if (args->num_cliprects != 0) + return -EINVAL; + + return i915_user_extensions(u64_to_user_ptr(args->cliprects_ptr), + execbuf_extensions, + ARRAY_SIZE(execbuf_extensions), + eb); +} + static int i915_gem_do_execbuffer(struct drm_device *dev, struct drm_file *file, struct drm_i915_gem_execbuffer2 *args, - struct drm_i915_gem_exec_object2 *exec, - struct drm_syncobj **fences) + struct drm_i915_gem_exec_object2 *exec) { struct drm_i915_private *i915 = to_i915(dev); struct i915_execbuffer eb; @@ -2393,6 +2419,9 @@ i915_gem_do_execbuffer(struct drm_device *dev, eb.batch_len = args->batch_len; eb.trampoline = NULL; + eb.fences = NULL; + eb.n_fences = 0; + eb.batch_flags = 0; if (args->flags & I915_EXEC_SECURE) { if (INTEL_GEN(i915) >= 11) @@ -2429,10 +2458,18 @@ i915_gem_do_execbuffer(struct drm_device *dev, } } - err = eb_create(&eb); + err = parse_execbuf2_extensions(args, &eb); if (err) goto err_out_fence; + err = get_fence_array(args, &eb); + if (err) + goto err_arr_fence; + + err = eb_create(&eb); + if (err) + goto err_arr_fence; + GEM_BUG_ON(!eb.lut_size); err = eb_select_context(&eb); @@ -2527,8 +2564,8 @@ i915_gem_do_execbuffer(struct drm_device *dev, goto err_request; } - if (fences) { - err = await_fence_array(&eb, fences); + if (eb.n_fences) { + err = await_fence_array(&eb); if (err) goto err_request; } @@ -2558,8 +2595,8 @@ i915_gem_do_execbuffer(struct drm_device *dev, i915_request_get(eb.request); eb_request_add(&eb); - if (fences) - signal_fence_array(&eb, fences); + if (eb.n_fences) + signal_fence_array(&eb); if (out_fence) { if (err == 0) { @@ -2587,6 +2624,8 @@ i915_gem_do_execbuffer(struct drm_device *dev, i915_gem_context_put(eb.gem_context); err_destroy: eb_destroy(&eb); +err_arr_fence: + __free_fence_array(eb.fences, eb.n_fences); err_out_fence: if (out_fence_fd != -1) put_unused_fd(out_fence_fd); @@ -2686,7 +2725,7 @@ i915_gem_execbuffer_ioctl(struct drm_device *dev, void *data, exec2_list[i].flags = 0; } - err = i915_gem_do_execbuffer(dev, file, &exec2, exec2_list, NULL); + err = i915_gem_do_execbuffer(dev, file, &exec2, exec2_list); if (exec2.flags & __EXEC_HAS_RELOC) { struct drm_i915_gem_exec_object __user *user_exec_list = u64_to_user_ptr(args->buffers_ptr); @@ -2718,7 +2757,6 @@ i915_gem_execbuffer2_ioctl(struct drm_device *dev, void *data, struct drm_i915_private *i915 = to_i915(dev); struct drm_i915_gem_execbuffer2 *args = data; struct drm_i915_gem_exec_object2 *exec2_list; - struct drm_syncobj **fences = NULL; const size_t count = args->buffer_count; int err; @@ -2746,15 +2784,7 @@ i915_gem_execbuffer2_ioctl(struct drm_device *dev, void *data, return -EFAULT; } - if (args->flags & I915_EXEC_FENCE_ARRAY) { - fences = get_fence_array(args, file); - if (IS_ERR(fences)) { - kvfree(exec2_list); - return PTR_ERR(fences); - } - } - - err = i915_gem_do_execbuffer(dev, file, args, exec2_list, fences); + err = i915_gem_do_execbuffer(dev, file, args, exec2_list); /* * Now that we have begun execution of the batchbuffer, we ignore @@ -2795,7 +2825,6 @@ end:; } args->flags &= ~__I915_EXEC_UNKNOWN_FLAGS; - put_fence_array(args, fences); kvfree(exec2_list); return err; } diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 00546062e023..0efded7b15f0 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -1046,6 +1046,10 @@ struct drm_i915_gem_exec_fence { __u32 flags; }; +enum drm_i915_gem_execbuffer_ext { + DRM_I915_GEM_EXECBUFFER_EXT_MAX /* non-ABI */ +}; + struct drm_i915_gem_execbuffer2 { /** * List of gem_exec_object2 structs @@ -1062,8 +1066,15 @@ struct drm_i915_gem_execbuffer2 { __u32 num_cliprects; /** * This is a struct drm_clip_rect *cliprects if I915_EXEC_FENCE_ARRAY - * is not set. If I915_EXEC_FENCE_ARRAY is set, then this is a - * struct drm_i915_gem_exec_fence *fences. + * & I915_EXEC_USE_EXTENSIONS are not set. + * + * If I915_EXEC_FENCE_ARRAY is set, then this is a pointer to an array + * of struct drm_i915_gem_exec_fence and num_cliprects is the length + * of the array. + * + * If I915_EXEC_USE_EXTENSIONS is set, then this is a pointer to a + * single struct drm_i915_gem_base_execbuffer_ext and num_cliprects is + * 0. */ __u64 cliprects_ptr; #define I915_EXEC_RING_MASK (0x3f) @@ -1181,7 +1192,16 @@ struct drm_i915_gem_execbuffer2 { */ #define I915_EXEC_FENCE_SUBMIT (1 << 20) -#define __I915_EXEC_UNKNOWN_FLAGS (-(I915_EXEC_FENCE_SUBMIT << 1)) +/* + * Setting I915_EXEC_USE_EXTENSIONS implies that + * drm_i915_gem_execbuffer2.cliprects_ptr is treated as a pointer to an linked + * list of i915_user_extension. Each i915_user_extension node is the base of a + * larger structure. The list of supported structures are listed in the + * drm_i915_gem_execbuffer_ext enum. + */ +#define I915_EXEC_USE_EXTENSIONS (1 << 21) + +#define __I915_EXEC_UNKNOWN_FLAGS (-(I915_EXEC_USE_EXTENSIONS<<1)) #define I915_EXEC_CONTEXT_ID_MASK (0xffffffff) #define i915_execbuffer2_set_context_id(eb2, context) \ From patchwork Tue Aug 4 08:59:54 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Lionel Landwerlin X-Patchwork-Id: 11699987 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id ABDA21392 for ; Tue, 4 Aug 2020 09:00:11 +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 C7D0220792 for ; Tue, 4 Aug 2020 09:00:11 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org C7D0220792 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=intel-gfx-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 121356E455; Tue, 4 Aug 2020 09:00:11 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by gabe.freedesktop.org (Postfix) with ESMTPS id 690846E450 for ; Tue, 4 Aug 2020 09:00:07 +0000 (UTC) IronPort-SDR: w6rnXwjBG5eJF44M1LsBRTtqhePuPmJdc67+VpG+v0SXi/SWkz+BVWKzXRhD2dR+/IxldsyhQM SEUpwqRd+H0Q== X-IronPort-AV: E=McAfee;i="6000,8403,9702"; a="153442629" X-IronPort-AV: E=Sophos;i="5.75,433,1589266800"; d="scan'208";a="153442629" X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga004.fm.intel.com ([10.253.24.48]) by orsmga102.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 04 Aug 2020 02:00:07 -0700 IronPort-SDR: TIG7GxdF/Pj7jp6bESStCnzKHlytTqrh3C7QQittpr0q38lrpxttfmPfLxw8EhJ5Bb5kaaF+6+ zD9xzlfFCGqw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.75,433,1589266800"; d="scan'208";a="315261511" Received: from wrottner-mobl1.ger.corp.intel.com (HELO delly.ger.corp.intel.com) ([10.252.61.120]) by fmsmga004.fm.intel.com with ESMTP; 04 Aug 2020 02:00:05 -0700 From: Lionel Landwerlin To: intel-gfx@lists.freedesktop.org Date: Tue, 4 Aug 2020 11:59:54 +0300 Message-Id: <20200804085954.350343-3-lionel.g.landwerlin@intel.com> X-Mailer: git-send-email 2.28.0 In-Reply-To: <20200804085954.350343-1-lionel.g.landwerlin@intel.com> References: <20200804085954.350343-1-lionel.g.landwerlin@intel.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH 2/2] drm/i915: add syncobj timeline support 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: Daniel Vetter , Chris Wilson Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" Introduces a new parameters to execbuf so that we can specify syncobj handles as well as timeline points. v2: Reuse i915_user_extension_fn v3: Check that the chained extension is only present once (Chris) v4: Check that dma_fence_chain_find_seqno returns a non NULL fence (Lionel) v5: Use BIT_ULL (Chris) v6: Fix issue with already signaled timeline points, dma_fence_chain_find_seqno() setting fence to NULL (Chris) v7: Report ENOENT with invalid syncobj handle (Lionel) v8: Check for out of order timeline point insertion (Chris) v9: After explanations on https://lists.freedesktop.org/archives/dri-devel/2019-August/229287.html drop the ordering check from v8 (Lionel) v10: Set first extension enum item to 1 (Jason) v11: Rebase v12: All multiple extension nodes of timeline syncobj (Chris) Signed-off-by: Lionel Landwerlin Co-authored-by: Chris Wilson Reviewed-by: Daniel Vetter (v11) Reviewed-by: Chris Wilson --- .../gpu/drm/i915/gem/i915_gem_execbuffer.c | 312 ++++++++++++++---- drivers/gpu/drm/i915/i915_drv.c | 3 +- drivers/gpu/drm/i915/i915_getparam.c | 1 + include/uapi/drm/i915_drm.h | 38 ++- 4 files changed, 279 insertions(+), 75 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index ed8d1c2517f6..9ce114d67288 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -223,6 +223,13 @@ struct eb_vma_array { * the batchbuffer in trusted mode, otherwise the ioctl is rejected. */ +struct eb_fence { + struct drm_syncobj *syncobj; /* Use with ptr_mask_bits() */ + struct dma_fence *dma_fence; + u64 value; + struct dma_fence_chain *chain_fence; +}; + struct i915_execbuffer { struct drm_i915_private *i915; /** i915 backpointer */ struct drm_file *file; /** per-file lookup tables and limits */ @@ -283,12 +290,8 @@ struct i915_execbuffer { struct hlist_head *buckets; /** ht for relocation handles */ struct eb_vma_array *array; - struct i915_eb_fence { - struct drm_syncobj *syncobj; /* Use with ptr_mask_bits() */ - } *fences; - u32 n_fences; - - u64 extension_flags; /** Available extensions parameters */ + struct eb_fence *fences; + unsigned long num_fences; }; static inline bool eb_use_cmdparser(const struct i915_execbuffer *eb) @@ -2198,77 +2201,222 @@ eb_pin_engine(struct i915_execbuffer *eb, } static void -__free_fence_array(struct i915_eb_fence *fences, unsigned int n) +__free_fence_array(struct eb_fence *fences, unsigned int n) { - while (n--) + while (n--) { drm_syncobj_put(ptr_mask_bits(fences[n].syncobj, 2)); + dma_fence_put(fences[n].dma_fence); + kfree(fences[n].chain_fence); + } kvfree(fences); } static int -get_fence_array(struct drm_i915_gem_execbuffer2 *args, - struct i915_execbuffer *eb) +add_timeline_fence_array(struct i915_execbuffer *eb, + const struct drm_i915_gem_execbuffer_ext_timeline_fences *timeline_fences) { - const unsigned long nfences = args->num_cliprects; - struct drm_i915_gem_exec_fence __user *user; - struct i915_eb_fence *fences; - unsigned long n; - int err; + struct drm_i915_gem_exec_fence __user *user_fences; + u64 __user *user_values; + struct eb_fence *f; + u64 nfences; + int err = 0; - if (!(args->flags & I915_EXEC_FENCE_ARRAY)) + nfences = timeline_fences->fence_count; + if (!nfences) return 0; /* Check multiplication overflow for access_ok() and kvmalloc_array() */ BUILD_BUG_ON(sizeof(size_t) > sizeof(unsigned long)); if (nfences > min_t(unsigned long, - ULONG_MAX / sizeof(*user), - SIZE_MAX / sizeof(*fences))) + ULONG_MAX / sizeof(*user_fences), + SIZE_MAX / sizeof(*f)) - eb->num_fences) return -EINVAL; - user = u64_to_user_ptr(args->cliprects_ptr); - if (!access_ok(user, nfences * sizeof(*user))) + user_fences = u64_to_user_ptr(timeline_fences->handles_ptr); + if (!access_ok(user_fences, nfences * sizeof(*user_fences))) + return -EFAULT; + + user_values = u64_to_user_ptr(timeline_fences->values_ptr); + if (!access_ok(user_values, nfences * sizeof(*user_values))) return -EFAULT; - fences = kvmalloc_array(nfences, sizeof(*fences), - __GFP_NOWARN | GFP_KERNEL); - if (!fences) + f = krealloc(eb->fences, + (eb->num_fences + nfences) * sizeof(*f), + __GFP_NOWARN | GFP_KERNEL); + if (!f) return -ENOMEM; - for (n = 0; n < nfences; n++) { - struct drm_i915_gem_exec_fence fence; + eb->fences = f; + f += eb->num_fences; + + BUILD_BUG_ON(~(ARCH_KMALLOC_MINALIGN - 1) & + ~__I915_EXEC_FENCE_UNKNOWN_FLAGS); + + while (nfences--) { + struct drm_i915_gem_exec_fence user_fence; struct drm_syncobj *syncobj; + struct dma_fence *fence = NULL; + u64 point; + + if (__copy_from_user(&user_fence, + user_fences++, + sizeof(user_fence))) + return -EFAULT; + + if (user_fence.flags & __I915_EXEC_FENCE_UNKNOWN_FLAGS) + return -EINVAL; + + if (__get_user(point, user_values++)) + return -EFAULT; + + syncobj = drm_syncobj_find(eb->file, user_fence.handle); + if (!syncobj) { + DRM_DEBUG("Invalid syncobj handle provided\n"); + return -ENOENT; + } + + fence = drm_syncobj_fence_get(syncobj); - if (__copy_from_user(&fence, user++, sizeof(fence))) { - err = -EFAULT; - goto err; + if (!fence && user_fence.flags && + !(user_fence.flags & I915_EXEC_FENCE_SIGNAL)) { + DRM_DEBUG("Syncobj handle has no fence\n"); + drm_syncobj_put(syncobj); + return -EINVAL; } - if (fence.flags & __I915_EXEC_FENCE_UNKNOWN_FLAGS) { - err = -EINVAL; - goto err; + if (fence) + err = dma_fence_chain_find_seqno(&fence, point); + + if (err && !(user_fence.flags & I915_EXEC_FENCE_SIGNAL)) { + DRM_DEBUG("Syncobj handle missing requested point %llu\n", point); + drm_syncobj_put(syncobj); + return err; + } + + /* + * A point might have been signaled already and + * garbage collected from the timeline. In this case + * just ignore the point and carry on. + */ + if (!fence && !(user_fence.flags & I915_EXEC_FENCE_SIGNAL)) { + drm_syncobj_put(syncobj); + continue; + } + + /* + * For timeline syncobjs we need to preallocate chains for + * later signaling. + */ + if (point != 0 && user_fence.flags & I915_EXEC_FENCE_SIGNAL) { + /* + * Waiting and signaling the same point (when point != + * 0) would break the timeline. + */ + if (user_fence.flags & I915_EXEC_FENCE_WAIT) { + DRM_DEBUG("Trying to wait & signal the same timeline point.\n"); + dma_fence_put(fence); + drm_syncobj_put(syncobj); + return -EINVAL; + } + + f->chain_fence = + kmalloc(sizeof(*f->chain_fence), + GFP_KERNEL); + if (!f->chain_fence) { + drm_syncobj_put(syncobj); + dma_fence_put(fence); + return -ENOMEM; + } + } else { + f->chain_fence = NULL; } - syncobj = drm_syncobj_find(eb->file, fence.handle); + f->syncobj = ptr_pack_bits(syncobj, user_fence.flags, 2); + f->dma_fence = fence; + f->value = point; + f++; + eb->num_fences++; + } + + return 0; +} + +static int add_fence_array(struct i915_execbuffer *eb) +{ + struct drm_i915_gem_execbuffer2 *args = eb->args; + struct drm_i915_gem_exec_fence __user *user; + unsigned long num_fences = args->num_cliprects; + struct eb_fence *f; + + if (!(args->flags & I915_EXEC_FENCE_ARRAY)) + return 0; + + if (!num_fences) + return 0; + + /* Check multiplication overflow for access_ok() and kvmalloc_array() */ + BUILD_BUG_ON(sizeof(size_t) > sizeof(unsigned long)); + if (num_fences > min_t(unsigned long, + ULONG_MAX / sizeof(*user), + SIZE_MAX / sizeof(*f) - eb->num_fences)) + return -EINVAL; + + user = u64_to_user_ptr(args->cliprects_ptr); + if (!access_ok(user, num_fences * sizeof(*user))) + return -EFAULT; + + f = krealloc(eb->fences, + (eb->num_fences + num_fences) * sizeof(*f), + __GFP_NOWARN | GFP_KERNEL); + if (!f) + return -ENOMEM; + + eb->fences = f; + f += eb->num_fences; + while (num_fences--) { + struct drm_i915_gem_exec_fence user_fence; + struct drm_syncobj *syncobj; + struct dma_fence *fence = NULL; + + if (__copy_from_user(&user_fence, user++, sizeof(user_fence))) + return -EFAULT; + + if (user_fence.flags & __I915_EXEC_FENCE_UNKNOWN_FLAGS) + return -EINVAL; + + syncobj = drm_syncobj_find(eb->file, user_fence.handle); if (!syncobj) { DRM_DEBUG("Invalid syncobj handle provided\n"); - err = -ENOENT; - goto err; + return -ENOENT; + } + + if (user_fence.flags & I915_EXEC_FENCE_WAIT) { + fence = drm_syncobj_fence_get(syncobj); + if (!fence) { + DRM_DEBUG("Syncobj handle has no fence\n"); + drm_syncobj_put(syncobj); + return -EINVAL; + } } BUILD_BUG_ON(~(ARCH_KMALLOC_MINALIGN - 1) & ~__I915_EXEC_FENCE_UNKNOWN_FLAGS); - fences[n].syncobj = ptr_pack_bits(syncobj, fence.flags, 2); + f->syncobj = ptr_pack_bits(syncobj, user_fence.flags, 2); + f->dma_fence = fence; + f->value = 0; + f->chain_fence = NULL; + f++; + eb->num_fences++; } - eb->fences = fences; - eb->n_fences = nfences; - return 0; +} -err: - __free_fence_array(fences, n); - return err; +static void put_fence_array(struct eb_fence *fences, int num_fences) +{ + if (fences) + __free_fence_array(fences, num_fences); } static int @@ -2277,21 +2425,17 @@ await_fence_array(struct i915_execbuffer *eb) unsigned int n; int err; - for (n = 0; n < eb->n_fences; n++) { + for (n = 0; n < eb->num_fences; n++) { struct drm_syncobj *syncobj; - struct dma_fence *fence; unsigned int flags; syncobj = ptr_unpack_bits(eb->fences[n].syncobj, &flags, 2); - if (!(flags & I915_EXEC_FENCE_WAIT)) - continue; - fence = drm_syncobj_fence_get(syncobj); - if (!fence) - return -EINVAL; + if (!eb->fences[n].dma_fence) + continue; - err = i915_request_await_dma_fence(eb->request, fence); - dma_fence_put(fence); + err = i915_request_await_dma_fence(eb->request, + eb->fences[n].dma_fence); if (err < 0) return err; } @@ -2299,13 +2443,12 @@ await_fence_array(struct i915_execbuffer *eb) return 0; } -static void -signal_fence_array(struct i915_execbuffer *eb) +static void signal_fence_array(const struct i915_execbuffer *eb) { struct dma_fence * const fence = &eb->request->fence; unsigned int n; - for (n = 0; n < eb->n_fences; n++) { + for (n = 0; n < eb->num_fences; n++) { struct drm_syncobj *syncobj; unsigned int flags; @@ -2313,10 +2456,34 @@ signal_fence_array(struct i915_execbuffer *eb) if (!(flags & I915_EXEC_FENCE_SIGNAL)) continue; - drm_syncobj_replace_fence(syncobj, fence); + if (eb->fences[n].chain_fence) { + drm_syncobj_add_point(syncobj, + eb->fences[n].chain_fence, + fence, + eb->fences[n].value); + /* + * The chain's ownership is transferred to the + * timeline. + */ + eb->fences[n].chain_fence = NULL; + } else { + drm_syncobj_replace_fence(syncobj, fence); + } } } +static int +parse_timeline_fences(struct i915_user_extension __user *ext, void *data) +{ + struct i915_execbuffer *eb = data; + struct drm_i915_gem_execbuffer_ext_timeline_fences timeline_fences; + + if (copy_from_user(&timeline_fences, ext, sizeof(timeline_fences))) + return -EFAULT; + + return add_timeline_fence_array(eb, &timeline_fences); +} + static void retire_requests(struct intel_timeline *tl, struct i915_request *end) { struct i915_request *rq, *rn; @@ -2359,14 +2526,13 @@ static void eb_request_add(struct i915_execbuffer *eb) } static const i915_user_extension_fn execbuf_extensions[] = { + [DRM_I915_GEM_EXECBUFFER_EXT_TIMELINE_FENCES] = parse_timeline_fences, }; static int parse_execbuf2_extensions(struct drm_i915_gem_execbuffer2 *args, struct i915_execbuffer *eb) { - eb->extension_flags = 0; - if (!(args->flags & I915_EXEC_USE_EXTENSIONS)) return 0; @@ -2420,7 +2586,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, eb.trampoline = NULL; eb.fences = NULL; - eb.n_fences = 0; + eb.num_fences = 0; eb.batch_flags = 0; if (args->flags & I915_EXEC_SECURE) { @@ -2439,14 +2605,24 @@ i915_gem_do_execbuffer(struct drm_device *dev, if (args->flags & I915_EXEC_IS_PINNED) eb.batch_flags |= I915_DISPATCH_PINNED; + err = parse_execbuf2_extensions(args, &eb); + if (err) + goto err_ext; + + err = add_fence_array(&eb); + if (err) + goto err_ext; + #define IN_FENCES (I915_EXEC_FENCE_IN | I915_EXEC_FENCE_SUBMIT) if (args->flags & IN_FENCES) { if ((args->flags & IN_FENCES) == IN_FENCES) return -EINVAL; in_fence = sync_file_get_fence(lower_32_bits(args->rsvd2)); - if (!in_fence) - return -EINVAL; + if (!in_fence) { + err = -EINVAL; + goto err_ext; + } } #undef IN_FENCES @@ -2458,17 +2634,9 @@ i915_gem_do_execbuffer(struct drm_device *dev, } } - err = parse_execbuf2_extensions(args, &eb); - if (err) - goto err_out_fence; - - err = get_fence_array(args, &eb); - if (err) - goto err_arr_fence; - err = eb_create(&eb); if (err) - goto err_arr_fence; + goto err_out_fence; GEM_BUG_ON(!eb.lut_size); @@ -2564,7 +2732,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, goto err_request; } - if (eb.n_fences) { + if (eb.fences) { err = await_fence_array(&eb); if (err) goto err_request; @@ -2595,7 +2763,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, i915_request_get(eb.request); eb_request_add(&eb); - if (eb.n_fences) + if (eb.fences) signal_fence_array(&eb); if (out_fence) { @@ -2624,13 +2792,13 @@ i915_gem_do_execbuffer(struct drm_device *dev, i915_gem_context_put(eb.gem_context); err_destroy: eb_destroy(&eb); -err_arr_fence: - __free_fence_array(eb.fences, eb.n_fences); err_out_fence: if (out_fence_fd != -1) put_unused_fd(out_fence_fd); err_in_fence: dma_fence_put(in_fence); +err_ext: + put_fence_array(eb.fences, eb.num_fences); return err; } diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 8b9bd929ba45..068447f565a9 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -1846,7 +1846,8 @@ static struct drm_driver driver = { */ .driver_features = DRIVER_GEM | - DRIVER_RENDER | DRIVER_MODESET | DRIVER_ATOMIC | DRIVER_SYNCOBJ, + DRIVER_RENDER | DRIVER_MODESET | DRIVER_ATOMIC | DRIVER_SYNCOBJ | + DRIVER_SYNCOBJ_TIMELINE, .release = i915_driver_release, .open = i915_driver_open, .lastclose = i915_driver_lastclose, diff --git a/drivers/gpu/drm/i915/i915_getparam.c b/drivers/gpu/drm/i915/i915_getparam.c index 421613219ae9..f96032c60a12 100644 --- a/drivers/gpu/drm/i915/i915_getparam.c +++ b/drivers/gpu/drm/i915/i915_getparam.c @@ -132,6 +132,7 @@ int i915_getparam_ioctl(struct drm_device *dev, void *data, case I915_PARAM_HAS_EXEC_BATCH_FIRST: case I915_PARAM_HAS_EXEC_FENCE_ARRAY: case I915_PARAM_HAS_EXEC_SUBMIT_FENCE: + case I915_PARAM_HAS_EXEC_TIMELINE_FENCES: /* For the time being all of these are always true; * if some supported hardware does not have one of these * features this value needs to be provided from diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 0efded7b15f0..e6cbd22308aa 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -619,6 +619,12 @@ typedef struct drm_i915_irq_wait { */ #define I915_PARAM_PERF_REVISION 54 +/* Query whether DRM_I915_GEM_EXECBUFFER2 supports supplying an array of + * timeline syncobj through drm_i915_gem_execbuffer_ext_timeline_fences. See + * I915_EXEC_USE_EXTENSIONS. + */ +#define I915_PARAM_HAS_EXEC_TIMELINE_FENCES 55 + /* Must be kept compact -- no holes and well documented */ typedef struct drm_i915_getparam { @@ -1046,8 +1052,36 @@ struct drm_i915_gem_exec_fence { __u32 flags; }; -enum drm_i915_gem_execbuffer_ext { - DRM_I915_GEM_EXECBUFFER_EXT_MAX /* non-ABI */ +/** + * See drm_i915_gem_execbuffer_ext_timeline_fences. + */ +#define DRM_I915_GEM_EXECBUFFER_EXT_TIMELINE_FENCES 0 + +/** + * This structure describes an array of drm_syncobj and associated points for + * timeline variants of drm_syncobj. It is invalid to append this structure to + * the execbuf if I915_EXEC_FENCE_ARRAY is set. + */ +struct drm_i915_gem_execbuffer_ext_timeline_fences { + struct i915_user_extension base; + + /** + * Number of element in the handles_ptr & value_ptr arrays. + */ + __u64 fence_count; + + /** + * Pointer to an array of struct drm_i915_gem_exec_fence of length + * fence_count. + */ + __u64 handles_ptr; + + /** + * Pointer to an array of u64 values of length fence_count. Values + * must be 0 for a binary drm_syncobj. A Value of 0 for a timeline + * drm_syncobj is invalid as it turns a drm_syncobj into a binary one. + */ + __u64 values_ptr; }; struct drm_i915_gem_execbuffer2 {