summaryrefslogtreecommitdiff
path: root/lib/mesa/src
AgeCommit message (Collapse)Author
2021-02-27util: enable futex usage on BSDs after 7dc2f4788288Jonathan Gray
From Jan Beich 46c368907fcf333a19881d28c46e997845d00faf in mainline Mesa We had a related local patch in 19.2 which wasn't needed when 20.1 was imported as the above commit was backported to the upstream 20.1 branch. When Mesa 20.0 was imported after issues with 20.1 on Haswell the changes to use futexes for simple_mtx.h and u_queue.h were lost. Noticed by otto@ and kettenis@ when looking for memory leaks.
2021-01-18gallivm: Add missing header for powf.Jonathan Gray
From Vinson Lee ff8daa013621019f1606dc0c188b16f1ce34fea7 in mainline Mesa fixes clang 11 build mortimer@ had almost the same diff
2015-11-22import Mesa 11.0.6Jonathan Gray
2020-09-22glsl: Use 'using' to be explicit about visitor overloadsJonathan Gray
From Kristian Hoegsberg e3dfa8f4d694e7d64a6401752af1f973b0852aab in mainline Mesa reduces clang warnings
2020-09-22spirv/opencl: Cast opcode up front to avoid warningsJonathan Gray
From Kristian Hoegsberg b40281d8306367e68dde6b723d2114d9cb5fca5a in mainline Mesa
2020-01-22Import Mesa 19.2.8Jonathan Gray
2016-05-29Import Mesa 11.2.2Jonathan Gray
2020-09-22Merge Mesa 20.0.8Jonathan Gray
With Mesa 20.1 even after the kernel change to do wbinvd on all cpus sthen@ reported that hard hangs still occurred on his Haswell system with inteldrm. Mark Kane also reported seeing hangs on Ivy Bridge on bugs@. Some systems/workloads seem to be more prone to triggering this than others as I have not seen any hangs on Ivy Bridge and the only hangs I saw on Haswell when running piglit went away with the wbinvd change. It seems something is wrong with drm memory attributes or coherency in the kernel and newer Mesa versions expect behaviour we don't have.
2020-09-22Import Mesa 20.0.8Jonathan Gray
2020-09-14go back to brw_eu.h rev 1.9 matching upstream 20.1 branchJonathan Gray
haswell hangs still occurred just less frequently with the 1.10 revert of 'intel/eu: Use non-coherent mode (BTI=253) for stateless A64 messages' The recent kernel change to do wbinvd across all cpus instead of just one is a better approach and avoids hangs seen with piglit on haswell.
2020-08-26Import Mesa 20.1.6Jonathan Gray
2020-09-11partially revert a Mesa commit to avoid hard hangs on HaswellJonathan Gray
intel/eu: Use non-coherent mode (BTI=253) for stateless A64 messages d23bbc8c28b6a5cd7f4d3d03c74d8319da5d47d5 on 20.1 branch 4985e380dd776ac65c4ae5627138211f9d9f03ce on mainline thanks to gnezdo@ sthen@ and espie@ for reports and testing
2020-09-03correct merge errorsJonathan Gray
2020-09-03Merge Mesa 20.1.7Jonathan Gray
2020-09-03Import Mesa 20.1.7Jonathan Gray
2020-09-03Import Mesa 20.1.7Jonathan Gray
2020-09-01get available memory when assert() is disabled via NDEBUGJonathan Gray
problem spotted by Jason Ekstrand reviewing proposed patches upstream
2020-08-30disable -Wmissing-prototypes via pragma for 64 bit atomics fallbackJonathan Gray
avoids error: no previous prototype for function '__sync_sub_and_fetch_8_c'
2020-08-30correct type for oldlenp argument to sysctl(2) for machdep.altivecJonathan Gray
2020-08-27drop local change for 32 bit atomics via pthread locksJonathan Gray
Mesa is no longer built on hppa or sh
2020-08-27drop local workarounds for gcc 4.2Jonathan Gray
2020-08-27radv: Explicitly cast TIMESTAMP_NOT_READY value to uin32_t where needed.Jonathan Gray
From Oschowa c310677a7563b1e2d97f8216be1d60cb21204eae in mainline mesa reduces clang warnings
2020-08-27aco: Use correct reference type in for-range-loop.Jonathan Gray
From Oschowa 663e8cb4e67f8b85186631c6a3719ed83da32151 in mainline mesa reduces clang warnings
2020-08-27aco: Don't std::move temporary object.Jonathan Gray
From Oschowa 7b1bc460fd6ae9bf5efeca62227bb05e0c50ee15 in mainline mesa reduces clang warnings
2020-08-27aco: Don't declare 'Block' as class, but define as struct.Jonathan Gray
From Oschowa 536339b0dda33241d21a0e045681419ca46fc812 in mainline mesa reduces clang warnings
2020-08-27radv: Don't take absolute value of unsigned type.Jonathan Gray
From Oschowa c2a778ef0f1720f9fb28afd40a791488648218d0 in mainline mesa reduces clang warnings
2020-08-26Merge Mesa 20.1.6Jonathan Gray
2020-08-26Import Mesa 20.1.6Jonathan Gray
2019-05-23Import Mesa 19.0.5Jonathan Gray
2019-01-29Import Mesa 18.3.2Jonathan Gray
2018-10-23Import Mesa 17.3.9Jonathan Gray
2018-10-23Import Mesa 17.3.9Jonathan Gray
2018-10-23Import Mesa 17.3.9Jonathan Gray
2016-12-11Import Mesa 13.0.2Jonathan Gray
2016-12-11Import Mesa 13.0.2Jonathan Gray
2016-05-29Import Mesa 11.2.2Jonathan Gray
2015-11-22import Mesa 11.0.6Jonathan Gray
2015-11-22import Mesa 11.0.6Jonathan Gray
2015-11-22import Mesa 11.0.6Jonathan Gray
2020-08-01Fix Mesa build with clang 10 on mips64.Visa Hankala
On mips64, the compiler does not allow use of non-zero argument with __builtin_frame_address(). However, the returned frame address is only used when PIPE_ARCH_X86 is defined. The compile error can be avoided by making #ifdef PIPE_ARCH_X86 cover the getting of frame address too. The argument checking of __builtin_frame_address() has been present as a debug assert in clang 8. In clang 10, there is a proper runtime check for the argument. This is why the build has not failed before. OK jsg@
2020-06-18llvmpipe: use LLVM version string instead of re-computing itJonathan Gray
From Eric Engestrom 7f5ef97a07d4054efb96f0d644344644023af82c in Mesa git fixes llvmpipe renderer string showing "LLVM 16.0" with LLVM 10
2020-06-16ac: fix build with recent LLVMJonathan Gray
from Samuel Pitoiset ee9811a0bb86d3d75fafeece368f6182048807d0 in mainline Mesa
2020-04-03Remove #ifdef that doesn't make sense anymore; extend clang workaround toMark Kettenis
conver __sync_val_compare_and_swap. ok jsg@
2020-04-03Make this build on clang architectures that don't have 64-bit atomicMark Kettenis
instructions. Clang doesn't allow redeclaration (and therefore redefinition) of the __sync_* builtins. Use #pragma redefine_extname to work around that restriction. Clang also turns __sync_add_and_fetch into __sync_fetch_and_add (and __sync_sub_and_fetch into __sync_fetch_and_sub) in certain cases, so provide these functions as well. ok jsg@
2020-02-20use futexes in Mesa on archs with atomicsJonathan Gray
tested by matthieu@ naddy@ and myself
2020-02-05Add sanity checks to futex_wait()Matthieu Herrb
- handle the NULL timeout pointer case - avoid negative cases. Suggested by cheloha@ ok cheloha@, jsg@
2020-02-04Fix absolute to relative timeout computation. ok jsg@, mpi@Matthieu Herrb
2020-01-22fix some things which got broken in mergeJonathan Gray
2020-01-22add files generated by python scripts which stopped being shipped withJonathan Gray
distfiles when upstream removed autotools
2020-01-22add back autotools files removed upstreamJonathan Gray
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-09 07:57:13 +0000