Commits

Branch selector

 5724bf5

User selector

All users

Datepicker

All time

Commit History

Commits on Aug 14, 2016

• Revert "gold: add a cast to appease std::max NFC"

  Show description for 5724bf5
  compnerd

  committed
  5724bf5

• Revert "CodeGen: If Convert blocks that would form a diamond when tail-merged."

  Show description for 12fc232
  rovka

  committed
  12fc232

• Revert "Codegen: Don't tail-duplicate blocks with un-analyzable fallthrough."

  Show description for 281da4f
  rovka

  committed
  281da4f

• [IRCE] Change variable grouping; NFC

  Show description for d330744
  sanjoy

  committed
  d330744

• [IRCE] Create llvm::Loop instances for cloned out loops

  Show description for 39b5545
  sanjoy

  committed
  39b5545

• [IRCE] Don't iterate on loops that were cloned out

  Show description for 300cd13
  sanjoy

  committed
  300cd13

• [IRCE] Add better DEBUG diagnostic; NFC

  Show description for abf7d54
  sanjoy

  committed
  abf7d54

• Fix bitcode auto-upgrade when using bitcode lazy loading

  Show description for 2c1c0b4
  joker-eph

  committed
  2c1c0b4

Commits on Aug 13, 2016

• Fix unsupported relocation type R_HEX_6_X' for symbol .rodata

  Show description for 26d2a23
  

  Ron Lieberman

  committed
  26d2a23

• Revert "Fix bitcode auto-upgrade when using bitcode lazy loading"

  Show description for 4510860
  joker-eph

  committed
  4510860

• [IRCE] Fix test case; NFC

  Show description for 3e63425
  sanjoy

  committed
  3e63425

• [IRCE] Be resilient in the face of non-simplified loops

  Show description for c56bea6
  sanjoy

  committed
  c56bea6

• Fix bitcode auto-upgrade when using bitcode lazy loading

  Show description for c5432bd
  joker-eph

  committed
  c5432bd

• Revert "Revert "Invariant start/end intrinsics overloaded for address space""

  Show description for 804c815
  joker-eph

  committed
  804c815

• Revert "Invariant start/end intrinsics overloaded for address space"

  Show description for 2eb84a5
  joker-eph

  committed
  2eb84a5

• [IRCE] Use dyn_cast instead of explicit isa/cast; NFC

  Show description for d2e55b6
  sanjoy

  committed
  d2e55b6

• [IRCE] Use range-for; NFC

  Show description for 81232a9
  sanjoy

  committed
  81232a9

• [ADT] Add a reserve() method to DenseSet as well as an insert() for R-value

  Show description for 568382f
  joker-eph

  committed
  568382f

• Revert "[ADT] Add a reserve method to DenseSet as well as an insert() for R-value"

  Show description for 80f0510
  joker-eph

  committed
  80f0510

• Limit DenseMap::setNumEntries input to 1<<31, in accordance with the 31 bits allocated to NumEntries.

  Show description for 663c3ce
  yrnkrn

  committed
  663c3ce

• Add missing REQUIRES in sancov/print_coverage_pcs.test: it requires aarch64 as well now

  Show description for 5ac928f
  joker-eph

  committed
  5ac928f

• [ADT] Add a reserve method to DenseSet as well as an insert() for R-value

  Show description for 762481d
  joker-eph

  committed
  762481d

• [x86] add tests to show missed 64-bit immediate merging

  Show description for b0e01f0
  rotateright

  committed
  b0e01f0

• Test commit

  Show description for 3ee4558
  hiraditya

  committed
  3ee4558

• [X86] Add a check of isCommutable at the top of X86InstrInfo::findCommutedOpIndices. Most callers don't check if the instruction is commutable before calling.

  Show description for c819da3
  topperc

  committed
  c819da3

• [AVX-512] Add isCommutable to scalar FMA3 instructions.

  Show description for 780d202
  topperc

  committed
  780d202

• [AVX-512] Add commutable flags to 132 form FMA3 instructions.

  Show description for c078bbe
  topperc

  committed
  c078bbe

• [X86] Remove patterns for (vzmovl (insert_subvector undef, (scalar_to_vector))) as the (vzmovl VR256) pattern has higher priority. NFC

  Show description for 1e8f440
  topperc

  committed
  1e8f440

• [AVX-512] Remove an AddedComplexity that was prioritizing basic vzmovl patterns over more complex ones that produce better code.

  Show description for 7efbb99
  topperc

  committed
  7efbb99

• [AVX-512] Add patterns to support VZEXT_MOVL from 512-bit vectors with 64-bit and 32-bit elements.

  Show description for e95ebf5
  topperc

  committed
  e95ebf5

• [PM] Port LoopDataPrefetch to new pass manager

  Show description for f970c6e
  teresajohnson

  committed
  f970c6e

• AMDGPU: Fix not estimating MBB operand sizes correctly

  Show description for e8fc9ab
  arsenm

  committed
  e8fc9ab

• AMDGPU: Fix missing test for addressing mode with odd offsets

  Show description for b24aaff
  arsenm

  committed
  b24aaff

• AMDGPU/R600: Remove macros

  Show description for 129b27a
  arsenm

  committed
  129b27a

• Fix more dereferenced end() iterators after r278532

  Show description for 46b81ba
  zmodem

  committed
  46b81ba

Pagination

PreviousNext