brkho/mesa - mesa - Brian's Gitea

Graphique des révisions

Auteur	SHA1	Message	Date
José Fonseca	438390418d	llvmpipe: First minify the texture size, then broadcast.	il y a 15 ans
José Fonseca	f5b5fb32d3	gallivm: Move into the as much of the second level code as possible. Also, pass more stuff trhough the sample build context, instead of arguments.	il y a 15 ans
José Fonseca	6b0c79e058	gallivm: Warn when doing inefficient integer comparisons.	il y a 15 ans
Keith Whitwell	e191bf4a85	gallivm: round rather than truncate in new 4x4f->1x16ub conversion path	il y a 15 ans
José Fonseca	f91b4266c6	gallivm: Use the wrappers for SSE pack intrinsics. Fixes assertion failures on LLVM 2.6.	il y a 15 ans
Keith Whitwell	607e3c542c	gallivm: special case conversion 4x4f to 1x16ub Nice reduction in the number of operations required for final color output in many shaders.	il y a 15 ans
José Fonseca	eb605701aa	gallivm: Implement brilinear filtering.	il y a 15 ans
José Fonseca	c8179ef5e8	gallivm: Fix copy'n'paste typo in previous commit.	il y a 15 ans
José Fonseca	df7a2451b1	gallivm: Clamp mipmap level and zero mip weight simultaneously.	il y a 15 ans
José Fonseca	0d84b64a4f	gallivm: Use lp_build_ifloor_fract for lod computation. Forgot this one before.	il y a 15 ans
José Fonseca	4f2e2ca4e3	gallivm: Don't compute the second mipmap level when frac(lod) == 0	il y a 15 ans
José Fonseca	05fe33b71c	gallivm: Simplify lp_build_mipmap_level_sizes' interface.	il y a 15 ans
José Fonseca	4eb222a3e6	gallivm: Do not do mipfiltering when magnifying. If lod < 0, then invariably follows that ilevel0 == ilevel1 == 0.	il y a 15 ans
José Fonseca	321ec1a224	gallivm: Vectorize the rho computation.	il y a 15 ans
José Fonseca	33f88b3492	util: Cleanup util_pack_z_stencil and friends. - Handle PIPE_FORMAT_Z32_FLOAT packing correctly. - In the integer version z shouldn't be passed as as double. - Make it clear that the integer versions should only be used for masks. - Make integer type sizes explicit (uint32_t for now, although uint64_t will be necessary later to encode f32_s8_x24).	il y a 15 ans
José Fonseca	87dd859b34	gallivm: Compute lod as integer whenever possible. More accurate/faster results for PIPE_TEX_MIPFILTER_NEAREST. Less FP <-> SI conversion overall.	il y a 15 ans
José Fonseca	1c32583581	gallivm: Only apply min/max_lod when necessary.	il y a 15 ans
Keith Whitwell	5849a6ab64	gallivm: don't apply zero lod_bias	il y a 15 ans
José Fonseca	af05f61576	gallivm: Combined ifloor & fract helper. The only way to ensure we don't do redundant FP <-> SI conversions.	il y a 15 ans
José Fonseca	012d57737b	gallivm: Fast implementation of iround(log2(x)) Not tested yet, but should be correct.	il y a 15 ans
José Fonseca	4648846bd6	gallivm: Use a faster (and less accurate) log2 in lod computation.	il y a 15 ans
José Fonseca	df3505b193	gallivm: Take the type signedness in consideration in round/ceil/floor.	il y a 15 ans
Dave Airlie	bf21b7006c	pb: fix numDelayed accounting we weren't decreasing when removing from the list.	il y a 15 ans
Dave Airlie	49866c8f34	pb: don't keep checking buffers after first busy If we assume busy buffers are added to the list in order its unlikely we'd fine one after the first busy one that isn't busy.	il y a 15 ans
Brian Paul	66992463ac	draw: check for null sampler pointers http://bugs.freedesktop.org/show_bug.cgi?id=30516	il y a 15 ans
Brian Paul	542d6cb1b8	gallivm: added some comments	il y a 15 ans
Tom Fogal	3661f757ee	Revert "Prefer intrinsics to handrolled atomic ops." This reverts commit 5f66b340aa49c6bc8d0acb2d1a6f8e9a7ef2cb2e, quickly fixing 30514.	il y a 15 ans
Tom Fogal	5f66b340aa	Prefer intrinsics to handrolled atomic ops.	il y a 15 ans
Tom Fogal	76a60faf52	Implement x86_64 atomics for compilers w/o intrinsics. Really old gcc's (3.3, at least) don't have support for the intrinsics we need. This implements a fallback for that case.	il y a 15 ans
Brian Paul	874f3a57ce	gallivm: check for level=0 case in lp_build_minify() This lets us avoid the shift and max() operations.	il y a 15 ans
José Fonseca	4e6f5e8d43	gallivm: More comprehensive border usage logic.	il y a 15 ans
Nicolas Kaiser	b719c91c82	gallium/util: remove duplicated include Remove duplicated include. Signed-off-by: Brian Paul <brianp@vmware.com>	il y a 15 ans
Nicolas Kaiser	237fa8a81c	gallium/rtasm: remove duplicated include Remove duplicated include. Signed-off-by: Brian Paul <brianp@vmware.com>	il y a 15 ans
José Fonseca	e3ccfd4e03	gallivm: Use SSE4.1's ROUNDSS/ROUNDSD for scalar rounding.	il y a 15 ans
Brian Paul	0cb545a7f2	draw: pass sampler state down to llvm jit state Fixes a regression caused from the change to make min/max lod dynamic state. https://bugs.freedesktop.org/show_bug.cgi?id=30437	il y a 15 ans
Brian Paul	3446af0179	llvmpipe: fix swizzling of texture border color The pipe_sampler_view's swizzle terms also apply to the texture border color. Simply move the apply_sampler_swizzle() call after we fetch the border color. Fixes many piglit texwrap failures.	il y a 15 ans
Luca Barbieri	ff531c5b05	ureg: support centroid interpolation	il y a 15 ans
Joakim Sindholt	b51f6e7c23	util/u_blitter: fix leak	il y a 15 ans
Brian Paul	4e2f53bacb	gallivm: fix repeat() function for NPOT textures The trick of casting the coord to an unsigned value only works for POT textures. Add a bias instead. This fixes a few piglit texwrap failures.	il y a 15 ans
José Fonseca	a69a96d85e	gallivm: Remove dead experimental code.	il y a 15 ans
José Fonseca	2a8d1fd3ce	gallivm: Fetch the lod from the dynamic state when min_lod == max_lod.	il y a 15 ans
José Fonseca	998cf11e13	draw: Fullfil the new min_lod/max_lod/lod_bias/border_color dynamic state	il y a 15 ans
Roland Scheidegger	049a8cce76	gallivm: optimize yuv decoding this is more a proof to show vector shifts on x86 with per-element shift count are evil. Since we can avoid the shift with a single compare/select, use that instead. Replaces more than 20 instructions (and slow ones at that) with about 3, and cuts compiled shader size with mesa's yuvsqure demo by over 10% (no performance measurements done - but selection is blazing fast). Might want to revisit that for future cpus - unfortunately AVX won't have vector shifts neither, but AMD's XOP will, but even in that case using selection here is probably not slower.	il y a 15 ans
Roland Scheidegger	46d05d4ef9	gallivm: don't use URem/UDiv when calculating offsets for blocks While it's true that llvm can and will indeed replace this with bit arithmetic (since block height/width is POT), it does so (llvm 2.7) by element and hence extracts/shifts/reinserts each element individually. This costs about 16 instructions (and extract is not really fast) vs. 1...	il y a 15 ans
Roland Scheidegger	26dc60d0a3	gallivm: fix copy&paste bug looks like pot_depth should be used, not pot_height (found by accident, not verified)	il y a 15 ans
Marek Olšák	8619495790	util: fix util_pack_color for B4G4R4A4 NOTE: This is a candidate for the 7.9 branch.	il y a 15 ans
Brian Paul	d1a4dd4217	llvmpipe: make texture border_color dynamic state	il y a 15 ans
Brian Paul	61b7da074e	llvmpipe: make min/max lod and lod bias dynamic state Before, changing any of these sampler values triggered generation of new JIT code. Added a new flag for the special case of min_lod == max_lod which is hit during auto mipmap generation.	il y a 15 ans
Marek Olšák	68afbe89c7	util: make calling remove_from_list multiple times in a row safe This commit fixes an infinite loop in foreach_s if remove_from_list is used more than once on the same item with other list operations in between. NOTE: This is a candidate for the 7.9 branch because the commit "r300g: fixup long-lived BO maps being incorrectly unmapped when flushing" depends on it.	il y a 15 ans
José Fonseca	440129521c	draw: Prevent clipped vertices overflow. Some pathological triangles cause a theoritically impossible number of clipped vertices. The clipper will still assert, but at least release builds will not crash, while this problem is further investigated.	il y a 15 ans

1 2 3 4 5 ...

3061 Révisions (438390418d27838bcfcb5bbb4c486db45dbaa44d)