From patchwork Thu Jun 18 13:07:30 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: arun.siluvery@linux.intel.com X-Patchwork-Id: 6637001 Return-Path: X-Original-To: patchwork-intel-gfx@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork1.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork1.web.kernel.org (Postfix) with ESMTP id 6F78E9F52D for ; Thu, 18 Jun 2015 13:08:14 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id DB53F2066A for ; Thu, 18 Jun 2015 13:08:07 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) by mail.kernel.org (Postfix) with ESMTP id 81F0920654 for ; Thu, 18 Jun 2015 13:08:06 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 0977C7A14D; Thu, 18 Jun 2015 06:08:06 -0700 (PDT) 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 ESMTP id 12D0D6ECB2 for ; Thu, 18 Jun 2015 06:08:02 -0700 (PDT) Received: from orsmga001.jf.intel.com ([10.7.209.18]) by orsmga102.jf.intel.com with ESMTP; 18 Jun 2015 06:07:41 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.13,638,1427785200"; d="scan'208";a="713266544" Received: from asiluver-linux.isw.intel.com ([10.102.226.85]) by orsmga001.jf.intel.com with ESMTP; 18 Jun 2015 06:07:39 -0700 From: Arun Siluvery To: intel-gfx@lists.freedesktop.org Date: Thu, 18 Jun 2015 14:07:30 +0100 Message-Id: <1434632855-7080-2-git-send-email-arun.siluvery@linux.intel.com> X-Mailer: git-send-email 2.3.0 In-Reply-To: <1434632855-7080-1-git-send-email-arun.siluvery@linux.intel.com> References: <1434632855-7080-1-git-send-email-arun.siluvery@linux.intel.com> Subject: [Intel-gfx] [PATCH v5 1/6] drm/i915/gen8: Add infrastructure to initialize WA batch buffers 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: , MIME-Version: 1.0 Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" X-Spam-Status: No, score=-4.5 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 Some of the WA are to be applied during context save but before restore and some at the end of context save/restore but before executing the instructions in the ring, WA batch buffers are created for this purpose and these WA cannot be applied using normal means. Each context has two registers to load the offsets of these batch buffers. If they are non-zero, HW understands that it need to execute these batches. v1: In this version two separate ring_buffer objects were used to load WA instructions for indirect and per context batch buffers and they were part of every context. v2: Chris suggested to include additional page in context and use it to load these WA instead of creating separate objects. This will simplify lot of things as we need not explicity pin/unpin them. Thomas Daniel further pointed that GuC is planning to use a similar setup to share data between GuC and driver and WA batch buffers can probably share that page. However after discussions with Dave who is implementing GuC changes, he suggested to use an independent page for the reasons - GuC area might grow and these WA are initialized only once and are not changed afterwards so we can share them share across all contexts. The page is updated with WA during render ring init. This has an advantage of not adding more special cases to default_context. We don't know upfront the number of WA we will applying using these batch buffers. For this reason the size was fixed earlier but it is not a good idea. To fix this, the functions that load instructions are modified to report the no of commands inserted and the size is now calculated after the batch is updated. A macro is introduced to add commands to these batch buffers which also checks for overflow and returns error. We have a full page dedicated for these WA so that should be sufficient for good number of WA, anything more means we have major issues. The list for Gen8 is small, same for Gen9 also, maybe few more gets added going forward but not close to filling entire page. Chris suggested a two-pass approach but we agreed to go with single page setup as it is a one-off routine and simpler code wins. Moved around functions to simplify it further, add comments. One additional option is offset field which is helpful if we would like to have multiple batches at different offsets within the page and select them based on some criteria. This is not a requirement at this point but could help in future (Dave). (Thanks to Chris, Dave and Thomas for their reviews and inputs) Signed-off-by: Rafael Barbalho Signed-off-by: Arun Siluvery --- drivers/gpu/drm/i915/intel_lrc.c | 204 +++++++++++++++++++++++++++++++- drivers/gpu/drm/i915/intel_ringbuffer.h | 18 +++ 2 files changed, 218 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 0413b8f..ad0b189 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -211,6 +211,7 @@ enum { FAULT_AND_CONTINUE /* Unsupported */ }; #define GEN8_CTX_ID_SHIFT 32 +#define CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT 0x17 static int intel_lr_context_pin(struct intel_engine_cs *ring, struct intel_context *ctx); @@ -1077,6 +1078,173 @@ static int intel_logical_ring_workarounds_emit(struct intel_engine_cs *ring, return 0; } +#define wa_ctx_emit(batch, cmd) { \ + if (WARN_ON(index >= (PAGE_SIZE / sizeof(uint32_t)))) { \ + return -ENOSPC; \ + } \ + batch[index++] = (cmd); \ + } + +/** + * gen8_init_indirectctx_bb() - initialize indirect ctx batch with WA + * + * @ring: only applicable for RCS + * @wa_ctx_batch: page in which WA are loaded + * @offset: This is for future use in case if we would like to have multiple + * batches at different offsets and select them based on a criteria. + * @num_dwords: The number of WA applied are known at the beginning, it returns + * the no of DWORDS written. This batch does not contain MI_BATCH_BUFFER_END + * so it adds padding to make it cacheline aligned. MI_BATCH_BUFFER_END will be + * added to perctx batch and both of them together makes a complete batch buffer. + * + * Return: non-zero if we exceed the PAGE_SIZE limit. + */ + +static int gen8_init_indirectctx_bb(struct intel_engine_cs *ring, + uint32_t **wa_ctx_batch, + uint32_t offset, + uint32_t *num_dwords) +{ + uint32_t index; + uint32_t *batch = *wa_ctx_batch; + + index = offset; + + /* FIXME: fill one cacheline with NOOPs. + * Replace these instructions with WA + */ + while (index < (offset + 16)) + wa_ctx_emit(batch, MI_NOOP); + + /* + * MI_BATCH_BUFFER_END is not required in Indirect ctx BB because + * execution depends on the length specified in terms of cache lines + * in the register CTX_RCS_INDIRECT_CTX + */ + + *num_dwords = index - offset; + + return 0; +} + +/* + * gen8_init_perctx_bb() - initialize per ctx batch with WA + * + * This function doesn't add any padding at the end as it contains + * MI_BATCH_BUFFER_END and padding after it is redundant. + */ +static int gen8_init_perctx_bb(struct intel_engine_cs *ring, + uint32_t **wa_ctx_batch, + uint32_t offset, + uint32_t *num_dwords) +{ + uint32_t index; + uint32_t *batch = *wa_ctx_batch; + + index = offset; + + /* FIXME: fill one cacheline with NOOPs. + * Replace these instructions with WA + */ + while (index < (offset + 16)) + wa_ctx_emit(batch, MI_NOOP); + + batch[index - 1] = MI_BATCH_BUFFER_END; + + *num_dwords = index - offset; + + return 0; +} + +static int lrc_setup_wa_ctx_obj(struct intel_engine_cs *ring, u32 size) +{ + int ret; + + WARN_ON(ring->id != RCS); + + ring->wa_ctx.obj = i915_gem_alloc_object(ring->dev, PAGE_ALIGN(size)); + if (!ring->wa_ctx.obj) { + DRM_DEBUG_DRIVER("alloc LRC WA ctx backing obj failed.\n"); + return -ENOMEM; + } + + ret = i915_gem_obj_ggtt_pin(ring->wa_ctx.obj, PAGE_SIZE, 0); + if (ret) { + DRM_DEBUG_DRIVER("pin LRC WA ctx backing obj failed: %d\n", + ret); + drm_gem_object_unreference(&ring->wa_ctx.obj->base); + return ret; + } + + return 0; +} + +static void lrc_destroy_wa_ctx_obj(struct intel_engine_cs *ring) +{ + WARN_ON(ring->id != RCS); + + i915_gem_object_ggtt_unpin(ring->wa_ctx.obj); + drm_gem_object_unreference(&ring->wa_ctx.obj->base); + ring->wa_ctx.obj = NULL; +} + +static int intel_init_workaround_bb(struct intel_engine_cs *ring) +{ + int ret = 0; + uint32_t *batch; + uint32_t num_dwords; + struct page *page; + struct i915_ctx_workarounds *wa_ctx = &ring->wa_ctx; + + WARN_ON(ring->id != RCS); + + if (ring->scratch.obj == NULL) { + DRM_ERROR("scratch page not allocated for %s\n", ring->name); + return -EINVAL; + } + + ret = lrc_setup_wa_ctx_obj(ring, PAGE_SIZE); + if (ret) { + DRM_DEBUG_DRIVER("Failed to setup context WA page: %d\n", ret); + return ret; + } + + page = i915_gem_object_get_page(wa_ctx->obj, 0); + batch = kmap_atomic(page); + + if (INTEL_INFO(ring->dev)->gen == 8) { + wa_ctx->indctx_batch_offset = 0; + + ret = gen8_init_indirectctx_bb(ring, + &batch, + wa_ctx->indctx_batch_offset, + &num_dwords); + if (ret) + goto out; + + wa_ctx->indctx_batch_size = round_up(num_dwords, CACHELINE_DWORDS); + wa_ctx->perctx_batch_offset = wa_ctx->indctx_batch_size; + + ret = gen8_init_perctx_bb(ring, + &batch, + wa_ctx->perctx_batch_offset, + &num_dwords); + if (ret) + goto out; + } else { + WARN(INTEL_INFO(ring->dev)->gen >= 8, + "WA batch buffer is not initialized for Gen%d\n", + INTEL_INFO(ring->dev)->gen); + } + +out: + kunmap_atomic(batch); + if (ret) + lrc_destroy_wa_ctx_obj(ring); + + return ret; +} + static int gen8_init_common_ring(struct intel_engine_cs *ring) { struct drm_device *dev = ring->dev; @@ -1411,6 +1579,9 @@ void intel_logical_ring_cleanup(struct intel_engine_cs *ring) kunmap(sg_page(ring->status_page.obj->pages->sgl)); ring->status_page.obj = NULL; } + + if (ring->wa_ctx.obj) + lrc_destroy_wa_ctx_obj(ring); } static int logical_ring_init(struct drm_device *dev, struct intel_engine_cs *ring) @@ -1474,7 +1645,21 @@ static int logical_render_ring_init(struct drm_device *dev) if (ret) return ret; - return intel_init_pipe_control(ring); + if (INTEL_INFO(ring->dev)->gen >= 8) { + ret = intel_init_workaround_bb(ring); + if (ret) { + DRM_ERROR("WA batch buffers are not initialized: %d\n", + ret); + } + } + + ret = intel_init_pipe_control(ring); + if (ret) { + if (ring->wa_ctx.obj) + lrc_destroy_wa_ctx_obj(ring); + } + + return ret; } static int logical_bsd_ring_init(struct drm_device *dev) @@ -1754,15 +1939,26 @@ populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_o reg_state[CTX_SECOND_BB_STATE] = ring->mmio_base + 0x118; reg_state[CTX_SECOND_BB_STATE+1] = 0; if (ring->id == RCS) { - /* TODO: according to BSpec, the register state context - * for CHV does not have these. OTOH, these registers do - * exist in CHV. I'm waiting for a clarification */ reg_state[CTX_BB_PER_CTX_PTR] = ring->mmio_base + 0x1c0; reg_state[CTX_BB_PER_CTX_PTR+1] = 0; reg_state[CTX_RCS_INDIRECT_CTX] = ring->mmio_base + 0x1c4; reg_state[CTX_RCS_INDIRECT_CTX+1] = 0; reg_state[CTX_RCS_INDIRECT_CTX_OFFSET] = ring->mmio_base + 0x1c8; reg_state[CTX_RCS_INDIRECT_CTX_OFFSET+1] = 0; + if (ring->wa_ctx.obj) { + reg_state[CTX_RCS_INDIRECT_CTX+1] = + (i915_gem_obj_ggtt_offset(ring->wa_ctx.obj) + + ring->wa_ctx.indctx_batch_offset * sizeof(uint32_t)) | + (ring->wa_ctx.indctx_batch_size / CACHELINE_DWORDS); + + reg_state[CTX_RCS_INDIRECT_CTX_OFFSET+1] = + CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT << 6; + + reg_state[CTX_BB_PER_CTX_PTR+1] = + (i915_gem_obj_ggtt_offset(ring->wa_ctx.obj) + + ring->wa_ctx.perctx_batch_offset * sizeof(uint32_t)) | + 0x01; + } } reg_state[CTX_LRI_HEADER_1] = MI_LOAD_REGISTER_IMM(9); reg_state[CTX_LRI_HEADER_1] |= MI_LRI_FORCE_POSTED; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 39f6dfc..1f38af3 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -12,6 +12,7 @@ * workarounds! */ #define CACHELINE_BYTES 64 +#define CACHELINE_DWORDS (CACHELINE_BYTES / sizeof(uint32_t)) /* * Gen2 BSpec "1. Programming Environment" / 1.4.4.6 "Ring Buffer Use" @@ -119,6 +120,22 @@ struct intel_ringbuffer { struct intel_context; +/* + * we use a single page to load ctx workarounds so all of these + * values are referred in terms of dwords + * + * offset field - helpful in case if we want to have multiple batches + * at different offsets based on some conditions. It is not a requirement + * at the moment but provides an option for future use. + * indctx_batch_size - HW expects this value in terms of cachelines + */ +struct i915_ctx_workarounds { + u32 indctx_batch_offset; + u32 indctx_batch_size; + u32 perctx_batch_offset; + struct drm_i915_gem_object *obj; +}; + struct intel_engine_cs { const char *name; enum intel_ring_id { @@ -142,6 +159,7 @@ struct intel_engine_cs { struct i915_gem_batch_pool batch_pool; struct intel_hw_status_page status_page; + struct i915_ctx_workarounds wa_ctx; unsigned irq_refcount; /* protected by dev_priv->irq_lock */ u32 irq_enable_mask; /* bitmask to enable ring interrupt */