4
High
|
aarch64 baremetal testing
|
0
Approved
|
11
Implemented
|
Matthew Gretton-Dann
|
Michael Hope
|
Michael Hope
|
4
High
|
Add support for GProf to AArch64 backend of GCC
|
0
Approved
|
12
Informational
|
|
Matthew Gretton-Dann
|
Michael Hope
|
4
High
|
Backport initial armv8 support
|
0
Approved
|
11
Implemented
|
|
Michael Hope
|
Michael Hope
|
4
High
|
Better end of loop counter optimisation
|
0
Approved
|
12
Informational
|
Kugan Vivekanandarajah
|
Michael Hope
|
Michael Hope
|
4
High
|
Cortex-A15 Backports
|
0
Approved
|
11
Implemented
|
|
Matthew Gretton-Dann
|
Michael Hope
|
4
High
|
Cortex-A15 Theme
|
0
Approved
|
12
Informational
|
|
Matthew Gretton-Dann
|
Michael Hope
|
4
High
|
Disable peeling
|
0
Approved
|
12
Informational
|
Christophe Lyon
|
Michael Hope
|
Michael Hope
|
4
High
|
Investigate LRA in GCC for ARM
|
0
Approved
|
12
Informational
|
Brice Dobry
|
Matthew Gretton-Dann
|
Michael Hope
|
4
High
|
Initial aarch64 backport
|
0
Approved
|
11
Implemented
|
Matthew Gretton-Dann
|
Michael Hope
|
Michael Hope
|
4
High
|
Enable and tune vectorizer cost model on ARM
|
0
Approved
|
11
Implemented
|
Christophe Lyon
|
Ulrich Weigand
|
Michael Hope
|
4
High
|
Backport vectorizer cost model
|
1
Pending Approval
|
11
Implemented
|
Christophe Lyon
|
Matthew Gretton-Dann
|
Matthew Gretton-Dann
|
4
High
|
Implement LRA in GCC for ARM
|
4
Discussion
|
12
Informational
|
Brice Dobry
|
Matthew Gretton-Dann
|
Matthew Gretton-Dann
|
4
High
|
Refactor backend cost model
|
4
Discussion
|
12
Informational
|
|
Michael Hope
|
Michael Hope
|
3
Medium
|
AArch64 GCC support for Stack Protection
|
0
Approved
|
12
Informational
|
|
Matthew Gretton-Dann
|
Michael Hope
|
3
Medium
|
ARMv6 saturation instructions
|
0
Approved
|
11
Implemented
|
Ulrich Weigand
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Backport support for -mcpu=native for Cortex-A7
|
0
Approved
|
11
Implemented
|
Kugan Vivekanandarajah
|
Matthew Gretton-Dann
|
Matthew Gretton-Dann
|
3
Medium
|
Backport Cortex-A7 Pipeline Description
|
0
Approved
|
11
Implemented
|
Kugan Vivekanandarajah
|
Matthew Gretton-Dann
|
Matthew Gretton-Dann
|
3
Medium
|
Backport NEON lexer
|
0
Approved
|
11
Implemented
|
Michael Hope
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Backport the A15 NEON pipeline
|
0
Approved
|
11
Implemented
|
Zhenqiang Chen
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Backport the FMA intrinsic
|
0
Approved
|
11
Implemented
|
Brice Dobry
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Better use of 16 bit Thumb-2 instructions
|
0
Approved
|
11
Implemented
|
Andrew Stubbs
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Disable invalid autoinc forms for VFP
|
0
Approved
|
11
Implemented
|
Ramana Radhakrishnan
|
Ramana Radhakrishnan
|
Michael Hope
|
3
Medium
|
Equivalent opposite condition detection
|
0
Approved
|
12
Informational
|
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Fix partial-partial PRE
|
0
Approved
|
11
Implemented
|
Michael Hope
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Fused multiply-add support
|
0
Approved
|
11
Implemented
|
Brice Dobry
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Turn off 64-bit Bitops in Neon in GCC
|
0
Approved
|
11
Implemented
|
Christophe Lyon
|
Matthew Gretton-Dann
|
Michael Hope
|
3
Medium
|
Implement constant vec permute operation for the vext instruction
|
0
Approved
|
11
Implemented
|
Christophe Lyon
|
Ramana Radhakrishnan
|
Michael Hope
|
3
Medium
|
Improve the auto increment/decrement pass
|
0
Approved
|
12
Informational
|
Ramana Radhakrishnan
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Improve the Neon vector permute intrinsics.
|
0
Approved
|
11
Implemented
|
Ramana Radhakrishnan
|
Ramana Radhakrishnan
|
Michael Hope
|
3
Medium
|
Improve the Neon max and min intrinsics.
|
0
Approved
|
12
Informational
|
|
Ramana Radhakrishnan
|
Michael Hope
|
3
Medium
|
Investigate -funroll-loops and -fvariable-expansion-in-unroller
|
0
Approved
|
12
Informational
|
|
Ulrich Weigand
|
Michael Hope
|
3
Medium
|
Prefer movw movt over literal pools where possible.
|
0
Approved
|
12
Informational
|
Zhenqiang Chen
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Add support for the rev16 and revsh instructions.
|
0
Approved
|
11
Implemented
|
Christophe Lyon
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Tidy up ubutest
|
0
Approved
|
11
Implemented
|
Michael Hope
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Fix GCC Multi-arch Testing
|
3
Drafting
|
12
Informational
|
Matthew Gretton-Dann
|
Matthew Gretton-Dann
|
|
3
Medium
|
Investigate generating out-of-line branch tables in ARM GCC
|
4
Discussion
|
12
Informational
|
|
Matthew Gretton-Dann
|
|
3
Medium
|
Improve generation of conditional compares in gcc.
|
4
Discussion
|
12
Informational
|
|
Matthew Gretton-Dann
|
|
3
Medium
|
Faster SIMD random number generator
|
4
Discussion
|
12
Informational
|
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Improve divide and modulo
|
4
Discussion
|
12
Informational
|
|
Michael Hope
|
Michael Hope
|
3
Medium
|
Investigate (& fix) failures in GCC's guality testsuite for ARM
|
4
Discussion
|
12
Informational
|
|
Matthew Gretton-Dann
|
|
3
Medium
|
Investigate lower-subreg
|
4
Discussion
|
12
Informational
|
|
Ramana Radhakrishnan
|
Michael Hope
|
3
Medium
|
Remove Unnecessary Zero/Sign Extensions
|
4
Discussion
|
12
Informational
|
Kugan Vivekanandarajah
|
Matthew Gretton-Dann
|
|
3
Medium
|
Audit the backend and split the insn patterns in the MD file.
|
4
Discussion
|
12
Informational
|
|
Ramana Radhakrishnan
|
|
3
Medium
|
Vectorize int to float and float to int conversion
|
4
Discussion
|
11
Implemented
|
|
Ira Rosen
|
Michael Hope
|
2
Low
|
Improve CRC16
|
0
Approved
|
12
Informational
|
|
Michael Hope
|
Michael Hope
|
2
Low
|
Revisit implementations of 32-bit division routines
|
4
Discussion
|
12
Informational
|
|
Matthew Gretton-Dann
|
|
2
Low
|
Support vcond with mixed types
|
4
Discussion
|
12
Informational
|
|
Ira Rosen
|
|
1
Undefined
|
Backport AArch64 CSE cmp optimisation
|
0
Approved
|
11
Implemented
|
Kumar Venkataramanan
|
Matthew Gretton-Dann
|
Matthew Gretton-Dann
|