This is an automated email from the git hooks/post-receive script.
unknown user pushed a change to branch master in repository llvm.
from bc139df [X86] Improve 64-bit shifts on 32-bit targets (PR14593) new 7882ecb [AVX512] Add X86::VR512RegClassID to X86RegisterInfo::getLarg [...] new b08f1f3 [AVX512] Stop treating VR512 specially in getLoadStoreRegOpco [...] new 64e2873 [AVX512] Move FR32X/FR64X handling in getLoadStoreRegOpcode i [...] new f015e11 [AVX512] Add VLX packed move instructions to the execution de [...] new 4022678 [AVX512] Always use EVEX encodings for 128/256-bit move instr [...] new d29f9a0 [X86] Simplify code for determing GR or FR reg classes by que [...]
The 6 revisions listed above as "new" are entirely new to this repository and will be described in separate emails. The revisions listed as "adds" were already present in the repository and have only been added to this reference.
Summary of changes: lib/Target/X86/X86ISelLowering.cpp | 59 +---- lib/Target/X86/X86InstrInfo.cpp | 89 ++++---- lib/Target/X86/X86RegisterInfo.cpp | 1 + test/CodeGen/X86/avx512-extract-subvector.ll | 6 +- test/CodeGen/X86/avx512-gather-scatter-intrin.ll | 18 +- test/CodeGen/X86/avx512-intel-ocl.ll | 8 +- test/CodeGen/X86/avx512-vbroadcasti256.ll | 4 +- test/CodeGen/X86/avx512bwvl-intrinsics.ll | 184 +++++++-------- test/CodeGen/X86/avx512dqvl-intrinsics.ll | 12 +- test/CodeGen/X86/avx512vbmivl-intrinsics.ll | 13 +- test/CodeGen/X86/avx512vl-intrinsics-upgrade.ll | 60 ++--- test/CodeGen/X86/avx512vl-intrinsics.ll | 110 ++++----- test/CodeGen/X86/avx512vl-mov.ll | 16 +- test/CodeGen/X86/masked_gather_scatter.ll | 28 +-- test/CodeGen/X86/masked_memop.ll | 8 +- test/CodeGen/X86/nontemporal-loads.ll | 36 +-- test/CodeGen/X86/vector-compare-results.ll | 22 +- test/CodeGen/X86/vector-half-conversions.ll | 278 +++++++++++------------ 18 files changed, 445 insertions(+), 507 deletions(-)