From 8d19215be8254f4f75e9c5a0d28345947b0382db Mon Sep 17 00:00:00 2001 From: Zou Nan hai Date: Tue, 2 Nov 2010 16:31:01 +0800 Subject: [PATCH] drm/i915: SNB BLT workaround On some stepping of SNB cpu, the first command to be parsed in BLT command streamer should be MI_BATCHBUFFER_START otherwise the GPU may hang. Signed-off-by: Zou Nan hai [ickle: rebased for -next] Signed-off-by: Chris Wilson --- drivers/gpu/drm/i915/intel_ringbuffer.c | 123 +++++++++++++++++++++++- drivers/gpu/drm/i915/intel_ringbuffer.h | 3 + 2 files changed, 123 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 390aa21edbe4..a0702b6fb631 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -658,6 +658,9 @@ void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring) drm_gem_object_unreference(ring->gem_object); ring->gem_object = NULL; + if (ring->cleanup) + ring->cleanup(ring); + cleanup_status_page(ring); } @@ -877,19 +880,133 @@ blt_ring_put_user_irq(struct intel_ring_buffer *ring) /* do nothing */ } + +/* Workaround for some stepping of SNB, + * each time when BLT engine ring tail moved, + * the first command in the ring to be parsed + * should be MI_BATCH_BUFFER_START + */ +#define NEED_BLT_WORKAROUND(dev) \ + (IS_GEN6(dev) && (dev->pdev->revision < 8)) + +static inline struct drm_i915_gem_object * +to_blt_workaround(struct intel_ring_buffer *ring) +{ + return ring->private; +} + +static int blt_ring_init(struct intel_ring_buffer *ring) +{ + if (NEED_BLT_WORKAROUND(ring->dev)) { + struct drm_i915_gem_object *obj; + u32 __iomem *ptr; + int ret; + + obj = to_intel_bo(i915_gem_alloc_object(ring->dev, 4096)); + if (obj == NULL) + return -ENOMEM; + + ret = i915_gem_object_pin(&obj->base, 4096, true, false); + if (ret) { + drm_gem_object_unreference(&obj->base); + return ret; + } + + ptr = kmap(obj->pages[0]); + iowrite32(MI_BATCH_BUFFER_END, ptr); + iowrite32(MI_NOOP, ptr+1); + kunmap(obj->pages[0]); + + ret = i915_gem_object_set_to_gtt_domain(&obj->base, false); + if (ret) { + i915_gem_object_unpin(&obj->base); + drm_gem_object_unreference(&obj->base); + return ret; + } + + ring->private = obj; + } + + return init_ring_common(ring); +} + +static int blt_ring_begin(struct intel_ring_buffer *ring, + int num_dwords) +{ + if (ring->private) { + int ret = intel_ring_begin(ring, num_dwords+2); + if (ret) + return ret; + + intel_ring_emit(ring, MI_BATCH_BUFFER_START); + intel_ring_emit(ring, to_blt_workaround(ring)->gtt_offset); + + return 0; + } else + return intel_ring_begin(ring, 4); +} + +static void blt_ring_flush(struct intel_ring_buffer *ring, + u32 invalidate_domains, + u32 flush_domains) +{ + if (blt_ring_begin(ring, 4) == 0) { + intel_ring_emit(ring, MI_FLUSH_DW); + intel_ring_emit(ring, 0); + intel_ring_emit(ring, 0); + intel_ring_emit(ring, 0); + intel_ring_advance(ring); + } +} + +static int +blt_ring_add_request(struct intel_ring_buffer *ring, + u32 *result) +{ + u32 seqno; + int ret; + + ret = blt_ring_begin(ring, 4); + if (ret) + return ret; + + seqno = i915_gem_get_seqno(ring->dev); + + intel_ring_emit(ring, MI_STORE_DWORD_INDEX); + intel_ring_emit(ring, I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT); + intel_ring_emit(ring, seqno); + intel_ring_emit(ring, MI_USER_INTERRUPT); + intel_ring_advance(ring); + + DRM_DEBUG_DRIVER("%s %d\n", ring->name, seqno); + *result = seqno; + return 0; +} + +static void blt_ring_cleanup(struct intel_ring_buffer *ring) +{ + if (!ring->private) + return; + + i915_gem_object_unpin(ring->private); + drm_gem_object_unreference(ring->private); + ring->private = NULL; +} + static const struct intel_ring_buffer gen6_blt_ring = { .name = "blt ring", .id = RING_BLT, .mmio_base = BLT_RING_BASE, .size = 32 * PAGE_SIZE, - .init = init_ring_common, + .init = blt_ring_init, .write_tail = ring_write_tail, - .flush = gen6_ring_flush, - .add_request = ring_add_request, + .flush = blt_ring_flush, + .add_request = blt_ring_add_request, .get_seqno = ring_status_page_get_seqno, .user_irq_get = blt_ring_get_user_irq, .user_irq_put = blt_ring_put_user_irq, .dispatch_execbuffer = gen6_ring_dispatch_execbuffer, + .cleanup = blt_ring_cleanup, }; int intel_init_render_ring_buffer(struct drm_device *dev) diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index acd23374fe89..68043f1a186e 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -55,6 +55,7 @@ struct intel_ring_buffer { struct drm_i915_gem_execbuffer2 *exec, struct drm_clip_rect *cliprects, uint64_t exec_offset); + void (*cleanup)(struct intel_ring_buffer *ring); /** * List of objects currently involved in rendering from the @@ -90,6 +91,8 @@ struct intel_ring_buffer { wait_queue_head_t irq_queue; drm_local_map_t map; + + void *private; }; static inline u32 -- 2.34.1