summaryrefslogtreecommitdiff
path: root/src/sna/kgem.h
AgeCommit message (Expand)Author
2012-02-06sna: Search all active buckets for a temporary allocationChris Wilson
2012-02-04sna: Apply redirection for the render copy into large pixmapsChris Wilson
2012-01-31sna: Split the tiling limits between upload and copyingChris Wilson
2012-01-30sna: Track large objects and limit prefer-gpu hint to small objectsChris Wilson
2012-01-29sna: Limit the tile size for uploading into large pixmapsChris Wilson
2012-01-29sna: Determine whether to use a partial proxy based on the pitchChris Wilson
2012-01-29sna: Allow ridiculously large bo, up to half the total GATTChris Wilson
2012-01-26sna: Allow gen4+ to use larger GPU boChris Wilson
2012-01-26sna: Prevent mapping through the GTT for large boChris Wilson
2012-01-26sna: Apply any previous transformation when downsamplingChris Wilson
2012-01-26sna: Experiment with creating large objects as CPU boChris Wilson
2012-01-24sna: Track a proxy's dirty status upon itself rather than the targetChris Wilson
2012-01-24sna: Split the active bo cache into per-tiling cachesChris Wilson
2012-01-21sna/gen3: Add DBG markup for tracking target stateChris Wilson
2012-01-20sna: Move the flush to the backendsChris Wilson
2012-01-18sna: Prevent switching rings with render disabledChris Wilson
2012-01-16sna: On LLC systems quietly replace all linear mmappings using the CPUChris Wilson
2012-01-15sna: Use indirect uploads rather than teardown existing CPU mapsChris Wilson
2012-01-15sna: Do not assume that the mappable aperture size is a power of twoChris Wilson
2012-01-14sna: Be more lenient wrt switching rings if the kernel supports semaphoresChris Wilson
2012-01-14sna: Ensure that the batch mode is always declared before emitting dwordsChris Wilson
2012-01-14sna: Experiment with GTT mmapped upload buffersChris Wilson
2012-01-12sna: Store damage-all in the low bit of the damage pointerChris Wilson
2012-01-11sna: Use a minimum alignment of 64Chris Wilson
2012-01-11sna: Explicitly retire the bo following a serialisation pointChris Wilson
2012-01-08sna: Tune cache size for cpu bo cacheChris Wilson
2012-01-07sna: Pad upload buffers to ensure there are an even number of rowsChris Wilson
2012-01-05sna: Only force a batch continuation if the scanout is written toChris Wilson
2012-01-04sna: Limit batch to a single page on 865gChris Wilson
2011-12-25sna: Move the is-mappable check into the callers of region_inplace()Chris Wilson
2011-12-24sna: Use shadow if the GPU is busy or not immediately mappableChris Wilson
2011-12-19sna: Tune the inplace cross-over point to be half-cache sizeChris Wilson
2011-12-19sna/gen[23]: We need to check the batch before doing an inline flushChris Wilson
2011-12-18sna: Distinguish between GTT and CPU maps when searching for VMAChris Wilson
2011-12-18sna: Purge all evicted boChris Wilson
2011-12-17sna: Simplify write domain trackingChris Wilson
2011-12-17sna: Upload to large pixmaps inplaceChris Wilson
2011-12-17sna: Search through the inactive VMA cache for potential upload boChris Wilson
2011-12-17sna: Map the upload buffer using an LLC boChris Wilson
2011-12-13sna: Experiment with creating the CPU pixmap using an LLC BOChris Wilson
2011-12-11sna: Implement a VMA cacheChris Wilson
2011-12-10sna: Only use the 64-byte pitch alignment for scanoutChris Wilson
2011-12-10sna/gen[23]: Fixup render targets with pitches below hw minimumChris Wilson
2011-12-09sna: Use a single definition for the inactive cache timeoutChris Wilson
2011-12-01sna: Move the preservation of the io handle into the common destroy pathChris Wilson
2011-11-16sna: And keep unity happyChris Wilson
2011-11-09sna: Track purgeability separately from reusabilityChris Wilson
2011-11-06sna: Beware unsigned promotion of int16_t to uint32_tChris Wilson
2011-11-05sna: For a 32k max window size, we need to handle up to 128k stridesChris Wilson
2011-11-04sna: Ensure operations on a ShmPixmap are synchronous with clientsChris Wilson
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-27 05:44:29 +0000