summaryrefslogtreecommitdiff
path: root/src/sna/kgem.h
AgeCommit message (Expand)Author
2019-07-19sna: Rebalance prefer_blt to weight I915_TILING_Y higherChris Wilson
2018-09-06sna: Check for incoherent writes through the GTTChris Wilson
2017-08-22sna/gen8+: Discard any blt using a LINEAR buffer that is not 64byte alignedChris Wilson
2016-11-05sna: Mark up CPU writes for scanout flushChris Wilson
2016-05-07sna: Add a special case for fast DRI2CopyRegion and NoAccelChris Wilson
2016-04-17sna: Tweak flushing before adding a new bo into a batchChris Wilson
2016-04-06sna: Manually expand sse2 memcpy to compensate for a bad compilerChris Wilson
2016-03-16sna: Relax tiling requirements to cope with kernel errorsChris Wilson
2016-02-24sna/dri2: Add active-scanout tracking to single CRTC flipsChris Wilson
2016-02-18sna: Use dirtyfb for fallback direct renderingChris Wilson
2015-11-08sna: Simply retire-requestsChris Wilson
2015-06-01sna: Enable use of GPU tiling even when SET_TILING failsChris Wilson
2015-05-17sna: Only mark wedged checking inside operations as a WARNChris Wilson
2015-04-17sna: Enable blitting with Y-tiled surfacesChris Wilson
2015-03-31sna: Consider idleness of both rings if we need a semaphoreChris Wilson
2015-03-24sna: Detect Y-tiled scanout featureChris Wilson
2015-02-25sna: Avoid asserting that a foreign bo is idleChris Wilson
2015-02-04sna: Use local defines for compilation on older distributionsChris Wilson
2015-01-25sna/gen6+: Tweak engine placement when considering src/dst combinationsChris Wilson
2015-01-22sna: Reorganise assertions when retiring idle boChris Wilson
2015-01-17sna/gen6+: Prefer not to switch away from active RENDER pipelineChris Wilson
2014-12-16sna: Make error debug tracing available at normal debug levelsChris Wilson
2014-11-19sna: gen8 BLT broken when address has bit 4 setMika Kuoppala
2014-11-06sna: Experimental support for write-combining mmapsChris Wilson
2014-10-31sna: Treat external bo carefully when marking as idleChris Wilson
2014-10-31sna: Retire all requests when a bo is found to be not busyChris Wilson
2014-10-29sna: Reuse the same fence until retiredChris Wilson
2014-10-21sna: Avoid pwriting large batchesChris Wilson
2014-09-08sna: Avoid u16 underflow when computing reserved batch spaceChris Wilson
2014-08-14sna: Do not force creation of a linear GPU bo that we don't wantChris Wilson
2014-08-02sna: Restrict fencing to only use the low 256MiBChris Wilson
2014-07-23sna: Disable rendering with the DRM device whilst away from VTChris Wilson
2014-06-29sna: Tidy calling memcpy_from_tiledChris Wilson
2014-06-23sna: Inject a batch flush before adding a fresh boChris Wilson
2014-06-18sna: Assert that we do not try to mmap a proxyChris Wilson
2014-06-13sna: Track the pageflip bo's busynessChris Wilson
2014-06-06sna: Auto-retire upload proxiesChris Wilson
2014-06-05sna/dri2: Enable immediate buffer exchangesChris Wilson
2014-06-03sna: Allow replacements to cancel operations between both bo under a PixmapChris Wilson
2014-06-02sna: Add support for DRI3Chris Wilson
2014-06-02sna: Add DBG hints for using inplace CPU mmappingsChris Wilson
2014-05-30sna: Do not allow imported buffers to be cachedChris Wilson
2014-05-30sna: Unexport kgem_get_unique_id()Chris Wilson
2014-05-26sna: Implicit release of upload buffers considered badChris Wilson
2014-05-14sna: Add some DBG to track rogue requestsChris Wilson
2014-03-12sna: Only couple up a proxy->rq when inserting into the request->buffers listChris Wilson
2014-02-04sna: Rearrange assertion to ease use of substitute cached boChris Wilson
2014-02-01sna: Allow more inplace promotions of CPU to GPU boChris Wilson
2014-01-17sna: Cull the DBG spew to stderrChris Wilson
2013-12-08sna: Simplify kgem_bo_can_map()Chris Wilson