From: Chris Wilson Date: Tue, 2 Nov 2010 11:17:23 +0000 (+0000) Subject: drm/i915: Drop the iomem accessors when writing to the kmapped blt batch X-Git-Url: https://git.karo-electronics.de/?a=commitdiff_plain;h=27153f72d04bcd83b3a66e219418a21d6269553b;p=linux-beck.git drm/i915: Drop the iomem accessors when writing to the kmapped blt batch I presumed that we would be writing to the batch through the GTT having bound it, so I converted it to use iomem. Even later as I spotted that we didn't even move the batch to the GTT (now an issue since we default to uncached memory on SNB) I still didn't realise that using iomem for kmapped memory was incorrect. Fix it. Signed-off-by: Chris Wilson --- diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index a0702b6fb631..85071570e1f9 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -899,7 +899,7 @@ 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; + u32 *ptr; int ret; obj = to_intel_bo(i915_gem_alloc_object(ring->dev, 4096)); @@ -913,8 +913,8 @@ static int blt_ring_init(struct intel_ring_buffer *ring) } ptr = kmap(obj->pages[0]); - iowrite32(MI_BATCH_BUFFER_END, ptr); - iowrite32(MI_NOOP, ptr+1); + *ptr++ = MI_BATCH_BUFFER_END; + *ptr++ = MI_NOOP; kunmap(obj->pages[0]); ret = i915_gem_object_set_to_gtt_domain(&obj->base, false);