Re: [Linaro-TCWG-CI] binutils-2_41-release-971-gf1837d9c3f2: FAIL: 1 regressions on arm

Hi Nick, Our CI system detected a regression with your binutils patch for PR 30865, reported at: https://lists.linaro.org/archives/list/linaro-toolchain@lists.linaro.org/thread/AHLZB5WEEPZEYS2FLONIEMU6W7YCI5KA/ Could you please take a look ? Thanks, Prathamesh On Mon, 6 Nov 2023 at 21:17,

[ACTIVITY] 18 - 22 Oct 2021

== This Week == * GCC - Committed a clean up patch to gimple-isel - PR93183: Committed fix - PR102376: Patch approved upstream - PR83750: Patch approved upstream but it regresses one test-case. == Next Week == - Continue with ongoing tasks ___

[ACTIVITY] 2 - 6 August 2021

== This Week == * GNU-708 (Attribute to mark param as const) - Created prototype patch - Discussions on gcc mailing list * PR66791 (replace builtins in intrinsics with vector extensions) - Fixed issue with PR98435 test-case as suggested by Christophe - Pinged patches for review. == Next Week ==

[ACTIVITY] 19 - 23 July 2021

== This Week == * PR66791 (replace builtins with vector extensions in arm_neon.h intrinsics) - vshl_n: Submitted patch for review and addressing comments by Richard. - vmul_n fp: Investigated regression on arm-none-eabi and followed up with Christophe. - vmul_n int: Waiting for feedback. - vld1:

[ACTIVITY] 14 - 25 June 2021

== This Week == * PR97906 (missed lowering to vac) - Committed patch. * PR66791 (replace builtins with vector extensions in neon intrinsics): - Committed patch for vceq - Submitted patch for vca - Submitted patch for vdup * PR98435 (missed opt in expanding vector constructor) - Patch

[ACTIVITY] 31 May - 4 June 2021

== This Week == * GCC (1) PR66791 (replace builtins with gcc vector extensions): - Waiting for feedback on vmul_n patch - Created patch for vector comparison ops. (2) PR97906 (missed lowering to vacge) - Submitted patch upstream. (3) PR98435 (missed optimization in expanding vec constructor) -

[ACTIVITY] 17 - 21 May 2021

== This Week == * PR66791 (Replace builtins with C operations in intrinsics) - vtst: Committed patch to trunk. - vshl_n: Patch results in worse code-gen, abandoned. - vmul_n: Created patch, looking into testsuite fallout * PR97906 (Missed lowering abs(x) >= abs(y) to vcage x, y) - Investigated

[ACTIVITY] 3 - 7 May 2021

== This Week == * GCC - PR97903 (missed lowering to vtst): Committed fix to trunk, - PR66791: Submitted patch to replace vtst* (a, b) by (a & b) != 0 in arm_neon.h. - Committed fix for build failure in trunk. * Validation - vect metric: Pushed all pending patches. == Next Week == - PR66791:

[ACTIVITY] 7 - 11 Dec 2020

== This Week == * PR97872 (missed optimization for comparison on vectors) - Committed fix to trunk * PR97903 (Lower a & b != 0 to vtst) - In progress * PR66791 (Replace builtins by vector extensions in arm_neon.h intrinsics) - Committed patches for vmvn - Patches approved for vcreate, vneg,

[ACTIVITY] 30 Nov - 4 Dec 2020

== This Week == * PR97872 (missed optimization on less-than vector comparison) - Upstream discussion with Richard, patch close to being approved. - Causes a regression for pr78102.c. * PR66791 (replace builtins by vector extensions in arm_neon.h intrinsics) - Submitted patches for vmvn, vcreate,

[ACTIVITY] 23 - 27 Nov 2020

== This Week == * PR66791 (Replace __builtins_* in arm_neon.h intrinsics by C vector extensions) - Submitted patch for vmvn, waiting for feedback. - Created and tested patches for vcreate and vneg. * PR97849 (segfault in ifcvt with -march=armv8.2-a+sve) - Committed fix to trunk. == Next Week ==

[ACTIVITY] 7 - 11 September 2020

== This Week == * GNU-659 (calculix) - Comparing code-gen of hoisting region doesn't show many more extra stores - Came up with two heuristics, which "fix" the issue, but aren't very general. * LLVM-611 - D79785: Created fix for expensive checks issue, and posted on llvm-dev - D87430: Posted

[ACTIVITY] 31 August - 4 September 2020

== This Week == * GNU-659 (calculix LTO regression) - Gathered perf stats for various opt levels and posted on ML * LLVM-611 (tune blx lowering heuristic) - Dave Green committed my patch which had to be reverted due to build-bot failure with expensive checks turned on. - Posted analysis of the

[ACTIVITY] 24 - 28 August 2020

== This Week == * GNU-659 (LTO calculix regression) - Started upstream discussion * LLVM-611 (Tune lowering heuristic to blx for code size) - Addressing upstream feedback for patch to fold tLDRpci, tBLx -> tBL in foldMemoryOperand hook - Added new heuristic to ISel, for follow up patch. *

[ACTIVITY] 22 - 26 June 2020

== This Week == * GCC - GNU-659 (LTO regressing calculix): Running and analysing benchmarks with different configs. * LLVM - LVM-611 (heuristic to lower calls to blx for armv6-m): Addressing upstream suggestions. - LLVM-612 (code-gen for imm8 args for Thumb1): Posted patch upstream, waiting for

[ACTIVITY] 6 - 10 April 2020

== This Week == * LLVM-611 (tune heuristic to lower blx): Working on feedback. * GNU-659: Doing benchmarking experiments * Public holiday - One day off == Next Week == - Continue with LLVM-611 and GNU-659 ___ linaro-toolchain mailing list

[ACTIVITY] 30 March - 3 April 2020

== This Week == * LLVM - LLVM-611 (Tune heuristic for blx): Posted upstream. - LLVM-612 (Redundant reg moves for 8-bit immediates): WIP patch * Public Holiday - one day off == Next Week == - Continue with LLVM tasks - Start looking at GNU-659 ___

[ACTIVITY] 20 - 24 Jan 2020

== This Week == * GCC development - PR47785: Addressed Richard's suggestions and posted patch upstream. - PR92867: Addressed Joseph's suggestions and posted patch upstream. * GCC bugzilla: - PR93397: Investigated issue. * Misc - Meetings == Next Week == - Continue ongoing tasks

[ACTIVITY] 18 - 22 Nov 2019

== This Week == * GCC - PR92554: Spent some time triaging the issue, but gave up after Richard posted better fix. - PR89007: Addressing upstream suggestions. - PR92608: Committed fix to trunk. - GNU-583: Looking thru Kugan's patch and upstream discussion. * Validation - Submitted patch to add

[ACTIVITY] 11 - 15 Nov 2019

== This Week == * GCC - PR89007: Posted patch upstream. - GNU-583: Worked with Kugan, to reproduce LTO failure and verified his patch fixes it. - Bug 5479: Investigated with Maxim, most likely a glibc issue. * Validation: - Posted abe patch to add --gcc_patch_file option. == Next Week == -

[ACTIVITY] 21 - 25 Oct 2019

== This Week == * GCC - PR91272: Addressing upstream suggestions. - PR92163: Addressing upstream suggestions. * Validation - CI monitoring * Misc - Meetings == Next Week == - GCC: Continue ongoing tasks ___ linaro-toolchain mailing list

[ACTIVITY] 14 - 18 Oct 2019

== This Week == * GCC SVE: - PR91532: Added test to PR91532 fix. - PR92085: Fix committed to trunk. - PR86753: Fix committed to trunk. - PR91272: Addressing upstream suggestions. - PR92033: Proposed patch, but Richard had a better approach. - PR89007: WIP. * Validation - Merged

[ACTIVITY] 2 - 6 September 2019

== This Week == * GCC: (a) PR86753: Addressing upstream suggestions, quite close to completion now. (b) PR78736: Fix committed to trunk. (c) PR91272: Reworked patch. (d) Bug 5464: Does not reproduce with arm-eabi-gcc-8.3.0 == Next Week == - Continue ongoing tasks

[ACTIVITY] 26 - 30 August 2019

== This Week == * GCC: (i) PR86753 - Addressing upstream suggestions, pivoted to another approach. (ii) PR91272 - Patch fixes issue but not correct approach, need to rework it. (iii) PR78736 - Submitted patch upstream. == Next Week == - Continue ongoing tasks.

[ACTIVITY] 19 - 23 August 2019

== This Week == * GCC : (i) PR83756: Addressing feedback from upstream. (ii) PR90724: Committed to trunk. (iii) PR91272: Started working on patch. (iv) PR88839: Merged fix from sve-acle-branch to trunk. * Validation (i) Patch to separate build and test steps in tcwg_gnu. (ii) SVE validation ==

[ACTIVITY] 29 July - 2 August 2019

== This Week == * PR86753 - Working on approach suggested by Richard. * PR90724 - Pinged upstream. * Validation - Merged patch to jenkins-scripts to add testsuite comparison to tcwg_gnu. - Submitted patch for separating build and test steps in tcwg_gnu. == Next Week == - Continue ongoing

[ACTIVITY] 15-19 July 2019

== This Week == * PR90724 - Pinged patch * PR91166 - Patch posted upstream. * PR86753 - Started investigating the issue. * Misc - Meetings == Next Week == - PR86753 - CI babysitting ___ linaro-toolchain mailing list linaro-toolchain@lists.linaro.org

[ACTIVITY] 8 - 12 July 2019

== This Week == * PR90723 - Committed fix to trunk in r273466. * PR90724 - Posted patch upstream, waiting for feedback. * Validation - SPEC2k6 with SVE seems to compile, but spotted couple of infra issues. - Sent abe patch to store sum files. * Misc - Meetings

[ACTIVITY] 1 - 5 July 2019

== This Week == * PR88833 - Fixed pending issues with x86 and committed fix to trunk * PR90723 - Issue seems to be infinite recursion overflowing the stack, investigating. * Misc - Meetings == Next Week == - PR90723 - Add testsuite comparison to tcwg_gnu

[ACTIVITY] 24 - 28 June 2019

== This Week == * PR88833 (6/10) - Patch approved by Richard. - Regresses pr88152.C due to a possibly latent issue with combine. * PR90722 (3/10) - Investigating the issue * Misc (1/10) - Meetings ___ linaro-toolchain mailing list

[ACITIVTY] 17 - 21 June 2019

== This Week == * GCC PR88833 (5/10) - Addressed suggestions from Richard and posted patch upstream: https://gcc.gnu.org/ml/gcc-patches/2019-06/msg01289.html * Holidays (4/10) * Misc (1/10) - Meetings == Next Week == - Address upstream suggestions for PR88833. - tcwg_gnu job: Add gcc

[ACTIVITY] 27 - 31 May 2019

== This Week == * PR88837 (7/10) - Addressed all upstream suggestions. - Found a (hackish) way to test patch with qemu (multiple issues). - Sorting thru "strange" testsuite fallout most of which seems unrelated to my patch. * PR88833 (2/10) - Looking at fwprop pass * Misc (1/10) - Meetings ==

[ACTIVITY] 20 - 24 May 2019

== This Week == * PR88833 (4/10) - Started investigating the issue, it seems one of the code-movement RTL passes like cse2 do not remove identical register copies resulting in extra register move. * PR88837 (5/10) - Patch almost approved offline by Richard, suggested me to move discussion

[ACTIVITY] 13 - 17 May 2019

== This Week == * PR88837 (9/10) - Tweaked patch to handle few more special cases with suggestions from Richard. * Misc (1/10) - Meetings == Next Week == - Continue ongoing tasks ___ linaro-toolchain mailing list linaro-toolchain@lists.linaro.org

[ACTIVITY] 6 - 10 May 2019

== This Week == * PR88837 (9/10) - Addressing review comments from Richard. Coming quite close should hopefully be committed next week. * Misc (1/10) - LLVM Buildbot babysitting - Meetings == Next Week == - Continue ongoing tasks ___ linaro-toolchain

[ACTIVITY] 29 April - 3 May 2019

== This Week == * PR88837 (8/10) - WIP patch based on Richard's feedback. * Public Holiday (2/10) == Next Week == - Continue ongoing tasks ___ linaro-toolchain mailing list linaro-toolchain@lists.linaro.org

[ACTIVITY] 22 - 26 April 2019

== This Week == * PR88837 (7/10) - Discussed and finalized algorithm for vector construction with Richard. * GNU-606 (1/10) - Experimented gdbserver memory consumption with docker. * Public Holiday (2/10) == Next Week == - Continue ongoing tasks. ___

[ACTIVITY] 15 April - 19 April 2019

== This Week == * GCC PR88837 (7/10) - Trying different approaches for shortest initialization of vector constructor * Public Holiday (2/10) * Misc (1/10) - Meetins == Next Week == - GCC PR88837 - GNU-606 ___ linaro-toolchain mailing list

[ACTIVITY] 8 April 2019 - 12 April 2019

== This Week == * PR88837 (9/10) - Addressing patch reviews from Richard. * Misc (1/10) - Committed patch to add support for bootstrap-lto-lean config to jenkins-scripts. - Meetings == Next Week == - Continue ongoing tasks ___ linaro-toolchain mailing

[ACTIVITY] 18 March - 22 March 2019

== This Week == * PR88837 (2/10) - Started investigating the bug * Validation (2/10) - tcwg_gnu: Merged into master and added profiledbootstrap. - ABE cleanup patch to remove --enable bootstrap option * TCWG-1508 (1/10) - Investigated merge failure, Uros's suggested patch fixes the issue. *

[ACTIVITY] 11 March - 15 March 2019

== This Week == * PR88839 (8/10) - Addressing suggestions by Richard. - Patch works for all cases except for VNx2DI/VNx2DF modes. I have a workaround for those two cases but investigating for a better approach. * Validation (1/10) - Submitted patches to merge tcwg_gnu branch into master for

[ACTIVITY] 4 March - 8 March 2018

== This Week == * PR88839 - Poor implementation of blend like permutes (4/10) - Addressing suggestions by Richard on patch * GCC vectorizer (2/10) - Background reading * Validation (1/10) - Added more bootstrap configs for tcwg_gnu job - Committed patch to abe to enable bootstrap implicitly

[ACTIVITY] 11 February - 15 February 2019

== This Week == * SVE ACLE (4/10) - Committed svlsl patch to branch * SVE (3/10) - Studying GCC's vectorizer - Started looking at PR88839 * Validation (2/10) - GNU-156: Working on tcwg_gnu CI job. * Misc (1/10) - Meetings - Filed PR89332 in GCC bugzilla. == Next Week == - Continue ongoing

[ACTIVITY] 4 February - 8 February 2019

== This Week == * SVE ACLE (4/10) - svlsl * Getting familiar with GCC's vectorizer (4/10) * Validation (1/10) - Upstream monitoring job * Misc (1/10) - Meetings == Next Week == - Continue ongoing tasks ___ linaro-toolchain mailing list

[ACTIVITY] 28 January 2019 - February 1 2019

== This Week == * SVE ACLE (6/10) - svlsl * Validation (2/10) - Upstream monitoring job based on round-robin - LLVM Buildbot babysitting * PR88839: Poor implementation of blend like permutes (1/10) * Misc (1/10) - Meetings == Next Week == - Continue ongoing tasks

[ACTIVITY] 21 January - 25 January 2019

== This Week == * SVE ACLE (8/10) - WIP patch for svlsl * Validation (1/10) - Upsteam monitoring job based on round-robin. - abe bug with make install * Meetings (1/10) == Next Week == - Continue ongoing tasks. ___ linaro-toolchain mailing list

[ACTIVITY] 14 January - 18 January 2019

== Progress == * SVE ACLE (6/10) - Committed patch for svlsl_wide. * GCC bugs (1/10) - PR88788: Infinite loop in malloc_candidate_p_1: Fixed in r267933 * Public holiday (2/10) * Misc (1/10) - Meetings == Next Week == - Continue ongoing tasks. ___

[ACTIVITY] 7 January - 11 January 2019

== This Week == * SVE ACLE (6/10) - Working on svlsl_wide patch. * PR887881: Infinite loop in malloc_candidate_p_1 (1/10) * Validation (2/10) - Prototype HCQC wrapper script * Misc (1/10) - Meetings == Next Week == - Continue ongoing tasks. ___

[ACTIVITY] 5 - 9 November 2018

== This Week == * GNU-405: Implement division using vrecpe/vrecps (1/10) - Committed patch * SVE ACLE (2/10) - Continue work on shift intrinsics * PR83750 (1/10) - Addressing upstream comments * Holidays (6/10) - Diwali festival == Next Week == - Continue working on SVE ACLE and PR83750

[ACTIVITY] October 29 - November 2 2018

== This Week == * PR83750: CSE erf/erfc pair (6/10) - Submitted patch upstream * SVE ACLE (2/10) - Started working on shift intrinsics * GNU-235 (1/10) - Cancelled card after I realized vrp does not support floating point ranges -:/ * Misc (1/10) - Meetings == Next Week == - PR83750: Address

[ACTIVITY] 22 - 26 October 2018

== This Week == * GNU-405: Implement division using vrecpe / vrecps (4/10) - Patch validated and posted upstream. * SVE ACLE intrinsics (4/10) - Going thru documentation. * GNU-235: Provide value-range info for erf family of functions (1/10) - Working on patch. * Misc (1/10) - Meetings ==

[ACTIVITY] 15 - 19 October 2018

== This Week == * TCWG-1234: Coremark regression (7/10) - Fixed golang regressions with the patch. - Posted patch upstream to change hoisting order and apply cost model. * Public holiday (2/10) * Misc (1/10) - Meetings == Next Week == - TCWG-319, SVE

[ACTIVITY] 20 - 24 August 2018

== This Week == * TCWG-1234: Hoisting and register pressure (4/10) - Came up with another heuristic to avoid spill - Verified it does not regress other test-cases (testsuite + private benchmarks) * TCWG-319: Use vrecpe / vrecps to implement division (1/10) - Rebased patch and validated it -

Re: [TCWG] BOLT - Binary optimization and layout tool

Please ignore, sent to the wrong list. Sorry for the noise. Regards, Prathamesh On 20 June 2018 at 00:34, Prathamesh Kulkarni wrote: > Just came across Facebook's announcement of open-sourcing BOLT, a tool > for optimising instruction cache and TLB misses. Looks interesting. > They cla

[TCWG] BOLT - Binary optimization and layout tool

Just came across Facebook's announcement of open-sourcing BOLT, a tool for optimising instruction cache and TLB misses. Looks interesting. They claim to get 2-15% perf improvement in their services with BOLT and have a port for aarch64:

[ACTIVITY] 21 - 25 May 2018

== This Week == * TCWG-1234: Code hoisting and register pressure (8/10) - Created patches for restricting hoisting out of blocks on loop exit but that was false alarm. - Investigating interaction of forwprop, code hoisting and register pressure. - Started upstream discussion. * GCC bugs (1/10) -

[ACTIVITY] 14 - 18 May 2018

== This Week == * GCC bugs (8/10) a) Committed patches for: i) TCWG-1413 / PR83648: missing -Wsuggest-attribute=malloc on a trivial malloc-like function ii) TCWG-1414 / PR85817: ICE in expand_call at gcc/calls.c:4291 iii) TCWG-1415 / PR85734: --suggest-attribute=malloc misdiagnoses static

[ACTIVITY] 19 - 23 Feb 2018

== This Week == * Coremark regression (TCWG-1234) (6/10) - Working on defining a cost model for hoisting. * Public Holiday (2/10) * Misc (2/10) - Meetings - Slides == Next Week == - Continue ongoing tasks. ___ linaro-toolchain mailing list

[ACTIVITY] 5 - 9 February 2018

== This Week == * Code hoisting (TCWG-1234) (6/10) - Working on approach to add basic block info to ANTIC sets * GCC bugs (3/10) - Extend sincos to handle sin 2x (PR83661): Patch completed. * Misc (1/10) - Meetings ___ linaro-toolchain mailing list

[ACTIVITY] 29 January - 2 February 2018

== This Week == * Code hoisting (TCWG-1234) (6/10) - Working on patch with feedback from Thomas. * GCC bugs (3/10) - Extend sincos to handle sin 2x (PR83661): WIP patch * Misc (1/10) - Meetings == Next Week == - TCWG-1234 - GCC bugs ___

[ACTIVITY] 15 - 19 January 2018

== This Week == * GCC bugs (4/10) a) Extend malloc_candidate_p to handle multiple phi's (TCWG-1331) - Created patch b) CSE erf/erfc pair (PR83750) - Created patch * TCWG-1234 (5/10) - WIP patch on second approach * Misc (1/10) - Meetings == Next Week == - GCC bugs - TCWG-1234

Re: [ACTIVITY] 8 - 12 January 2018

Oops, this weekly report was for 15-19 January 2018 not 8-12 Jan. On 23 January 2018 at 13:25, Prathamesh Kulkarni <prathamesh.kulka...@linaro.org> wrote: > == This Week == > > * GCC bugs (4/10) > a) TCWG-1331 - Created patch > b) PR83750 - Created patch > > * TCW

[ACTIVITY] 8 - 12 January 2018

== This Week == * GCC bugs (4/10) a) TCWG-1331 - Created patch b) PR83750 - Created patch * TCWG-1234 (5/10) - WIP patch on second approach * Misc (1/10) - Meetings == Next Week == - GCC bugs - TCWG-1234 ___ linaro-toolchain mailing list

[ACTIVITY] 8 - 12 January 2018

== This Week == * GCC bugs (8/10) - PR81703: Committed to trunk, restricted test-case to x86_64 for now. - PR83514: Committed to trunk. - PR82665: Upstream iteration and approved for next stage-1. - PR83648: Upstream iteration and approved for next stage-1. - TCWG-1329: Created patch. - PR83570:

[ACTIVITY] 2 - 5 January 2018

== This Week == * GCC bugs (8/10) a) PR83501 - Committed to trunk. b) PR82665 - Upstream iteration, waiting for final approval. c) PR83648 - Upstream iteration. d) PR81703 - Created a fix. e) PR83661 - Work in progress patch. * Public Holiday (2/10) == Next Week == - Continue ongoing tasks

[ACTIVITY] 11 - 15 Dec 2017

== This Week == * alloc_size propagation (6/10) - Prototype patch done * Sick leave (4/10) - Cough and Viral fevers == Next Week == - Continue ongoing tasks ___ linaro-toolchain mailing list linaro-toolchain@lists.linaro.org

[ACTIVITY] 27 November - 1 December 2017

== This Week == * TCWG-1234 (6/10) - WIP patch * Public Holiday (2/10) * Misc (2/10) - PR83142 - Meetings == Next Week == - Continue ongoing tasks ___ linaro-toolchain mailing list linaro-toolchain@lists.linaro.org

[ACTIVITY] 20 - 24 November 2017

== This Week == * TCWG-1234 (6/10) - Using dfs to compute bb dist - Reading through rest of PRE/SCCVN to better understand the infrastructure * PR82808 (2/10) - Trying to find an approach to pass parm_type from all callers. - Martin posted a patch to address the parm_type issue. * Misc (2/10) -

[ACTIVITY] 13 - 17 November 2017

== This Week == * PR82665 (3/10) - Submitted patch upstream - Blocked on POINTER_DIFF_EXPR patch by Marc Glisse. * PR82808 (1/10) - Submitted another fix but was rejected by Richard. * TCWG-1234 (4/10) - Tried tweaking with register allocator, but to no effect - Trying to think about possible

[ACTIVITY] 30 October - 3 November 2017

== This Week == * PR82808 (4/10) - Submitted patch upstream * TCWG-1234 (4/10) - Experimenting with unhoisting approach - Working through ICE's * TCWG-319 (1/10) - Rebased and locally validated patch * Misc (1/10) - Meetings == Next Week == - Continue ongoing tasks

[ACTIVITY] 23 October - 27 October 2017

== This Week == * TCWG-1005 (4/10) - Iteration on upstream feedback and validated patch - Committed to trunk in r254140. * Reverse of TCWG-1253 transform (3/10) - Adding reverse transform for div pattern in expand * TCWG-1234 (2/10) - Experimenting with "unhoisting" approach * Misc (1/10) -

[ACTIVITY] 16 - 20 October 2017

== This Week == * TCWG-1234 (4/10) - Patch inhibits if-conversion for couple of tests. * TCWG-1253 (3/10) - Fixed regressions with the patch * Took Monday off for recovering from travel (2/10) * Misc (1/10) - Meetings == Next Week == - TCWG-1234, TCWG-1253, TCWG-319

[ACTIVITY] 2nd Oct - 6th Oct 2017

== This Week == * TCWG-1234: code-hoisting regression (3/10) - Have a workaround that fixes the particular regression (but may introduce another) - Trying to cross build benchmarks - Investigating PRE and code-hoisting optimizations * TCWG-1005: malloc attr propagation (3/10) - Iteration based

[ACTIVITY] 28 August - 1st September 2017

== This Week == * Type promotion (6/10) - Benchmarking on ARM and AArch64 - Experimenting with pass order to reduce redundancies - Preparing the patch to send upstream * PR78809 (2/10) - Prototype patch done - Working through testsuite fallout * PR78736 (1/10) - Iteration on upstream feedback.

[ACTIVITY] 21 - 25 August 2017

== This Week == * Type promotion (4/10) - Looking at perf regressions on ppc64 - Enabling at -O2 with vrp removes some perf regressions - Did benchmarking for code-size - Issues with benchmarking for performance * PR78809 (2/10) - WIP patch * PR78736 (1/10) - Updated patch based on upstream

[ACTIVITY] 31st July - 4th August 2017

== This Week == * malloc propagation (2/10) - Submitted patch upstream * type promotion (4/10) - Verified ppc64 miscompilation with memcpy-bi.c is a latent issue. - Investigating another miscompilation on ppc64 with patch. - Experimenting with pass placement. * bugs (3/10) - PR78736: pinged

[ACTIVITY] 24 - 28 July 2017

== This Week == * malloc propagation (4/10) - Resolved bootstrap comparison failure - Patch passes validation with and without LTO. - Working on handling indirect calls. * type promotion (3/10) - Created patche for resolving ICE with armv3/armv5t. - Have got a hack-ish patch to address

[ACTIVITY] 10 - 14 July 2017

== This Week == * PR78736 (2/10) - Submitted patch upstream * type-promotion (3/10) - Scheduling path-splitting before type-promotion fixes the regression with path-split-1.c - Created patch to fix issues with type-promotion interfering with widening_mul and bswap optimizations *

[ACTIVITY] 3 July - 7 July 2017

== This Week == * PR78736 (4/10) - Improved patch to not warn for enums with equal value ranges - Caused fallouts in libgomp and libgfortran - Large kernel fallout! * type promotion (5/10) - Looking at interference between path-splitting and type-promotion optimizations - miscompilation of

ACTIVITY 22 - 26 May 2016

== This Week == * TCWG-1005 (4/10) - Reworking the patch with suggested improvements * PR80806 (1/10) - Rejected patch, based on discussions with Martin and Jeff - Submitted patch to remove dead call to memset in value-prof.c * type promotion (2/10) - Discussions with Kugan - Trying out

[ACTIVITY] 8 - 12 May 2017

== This Week == * PR77644 (2/10) - Committed to trunk * PR78736 (1/10) - Upstream reviews * PR80613 (1/10) - Committed to trunk * TCWG-1005 (1/10) - Looking how to handle indirect calls * Sick leave (4/10) - Eye infection * Misc (1/10) - Meetings == Next Week == - TCWG-1005, TCWG-1129, gcc

[ACTIVITY] 1 - 5 May 2017

== This Week == * TCWG-1005 (4/10) - Made some improvements to analysis, which increased candidates for malloc functions in gcc source. - Patch review from Kugan * PR78736 (2/10) - Submitted upstream for review * PR80613 (2/10) - Submitted patch but was rejected by Richard - "Fixed" the bug by

[ACTIVITY] 17 - 21 April 2017

== This Week == * PR79697 (2/10) - Updated patch on top of trunk, passes validation. * PR79715 (2/10) - Updated patch according to upstream feedback, passes native bootstrap * TCWG-1010 (4/10) - Playing with LLVM's bit-tracking DCE pass - WIP prototype * TCWG- (1/10) - WIP patch * Misc

[ACTIVITY] 3 - 7 April 2017

== This Week == * TCWG-1005 (6/10) - All ICE's resolved with firefox -;) - Few improvements still left - handling indirect calls, handle cases when return value from malloc'd function has more than single use. * Validation (1/10) - patch for adding --set buildconfig option to abe * Public

[ACTIVITY] 27-31 March 2017

== This Week == * TCWG-1005 (6/10) - Fixed firefox ICE - Unfortunately that gives rise to another ICE with ipa-icf pass :( Investigating if this caused by the patch or a latent bug in ipa-icf. - Builds well with chromium * Validation (1/10) - abe/extraconfig patches - backport review * Public

[ACTIVITY] 20 - 24th March 2017

== This Week == * TCWG-1005 (6/10) - Trying to fix ICE with firefox * Validation (2/10) - Tweaking .gn files for building chromium - patches to abe - backport reviews * Misc (2/10) - Meetings - bitwise-dce == Next Week == - Continue ongoing tasks

[ACTIVITY] 13 - 17 Feb 2017

== This Week == * gimplefe (5/10) - Committed patch to add abs_expr - Committed patch to fix PR79478 - Committed patch to add fma_expr - Committed patch to fix ICE's in c_parser_gimple_postfix_expression - Submitted patch to fix incorrect translation for an invalid test-case, but decided to drop

[ACTIVITY] 6 - 10 Feb 2017

== This Week == * gimple-fe (4/10) a) Add support for abs_expr b) Add support for fma_expr * TCWG-1010 (4/10) - Continue working on prototype, dealing with unexpected ICE's :( * Misc (2/10) - Reading up on gcc's inliner - Meetings == Next Week == - TCWG-1010, TCWG-1005, gimple-fe -

[ACTIVITY] 30 Jan - 3 Feb 2017

== This Week == * TCWG-1006 (2/10) - Created prototype patch for propagating returns_nonnull along the same lines as malloc. * TCWG-1010 (2/10) - Continuing to work on prototype for bitwise dce * ICE with __builtin_abs with -fgimple (2/10) - Posted patch upstream * Sick leave (2/10) * Misc

[ACTIVITY] 16 - 20 Jan 2017

== This Week == * TCWG-1005 (malloc attr propagation) (4/10) - Extended for pointer comparisons - Patch review from Kugan - Removed some cruft from the patch * TCWG-1010 (bitwise dce propagation) (4/10) - Started working on prototype * Misc (2/10) - Looked at Wheader-guard - Building chromium

[ACTIVITY] 9 - 13 Jan 2016

== This Week == * TCWG-1005 (malloc attr propagation) (6/10) - Worked through bootstrap failures - Patch found 35 functions that could have malloc attribute during gcc build - The current analysis to find candidate functions is too restrictive, working on improving that. - Wrote few test-cases.

[ACTIVITY] 2 Jan - 6th Jan 2017

== This Week == * TCWG-1005 (propagate malloc attribute) (7/10) - Created prototype patch for propagating malloc attribute in ipa-pure-const * TCWG-1010 (bitwise-dce) (2/10) - Looking into LLVM's implementation of demanded-bits analysis * Misc (1/10) - Meetings == Next Week == - TCWG-1005,

Re: [ACTIVITY] 5 - 9 Dec 2016

On 10 December 2016 at 01:47, Prathamesh Kulkarni <prathamesh.kulka...@linaro.org> wrote: > * TCWG-971 (6/10) > - Resolved ICE's in couple of cases. > - Got stuck on the case when split_point->entry_bb has phi whose one > of the arguments is defined > outside split

[ACTIVITY] 28 Nov to 2 Dec 2016

== This Week == * TCWG-971 (4/10) - Prototype patch can handle simple cases * GCC bugs (5/10) a) PR78529 - Cannot reproduce after several tries :( b) PR78599 - Submitted patch upstream to bail out of bits prop if param type is not integral or pointer type, but a better fix would be to

[ACTIVITY] 14 - 18 November 2016

== This Week == * PR78319 (2/10) - After some investigation, it seems issue is with uninit pass rather than r241915. - Committed patch to XFAIL the offending test on arm-none-eabi, after approval from Richard * PR78154 (2/10) - Patch iterations based on upstream review. * PR78153 (2/10) -

[ACTIVITY] 07 - 11 Nov 2016

== This Week == * PR35691 (2/10) - Fix committed to trunk for case when LOGICAL_OP_NON_SHORT_CIRCUIT is true. - Fixed PR78256, caused by the commit. * PR78335 (2/10) - Created patch * PR35503 (1/10) - Rebased and committed to trunk after validation * PR78139 (1/10) - Caused due to r241915

[ACTIVITY] 31st Oct - 4 Nov 2016

== This Week == * ipa mod/ref analysis (2/10) - Prototype patch now detects modifications to reference parameters - Working on mod/ref analysis of global variables. * PR35691 (2/10) - Patch iterations based on upstream review * PR35503 (2/10) - Patch approved by Jason - Building kernel with

[ACTIVITY] 24 - 28 Oct 2016

* TCWG-72 (2/10) - Committed to trunk finally! * ipa mod/ref analysis (6/10) - Prototype in progress based on Muchnik's text * Misc (2/10) - Committed a trivial patch to use VAR_P in arm.c. - Background reading on paper about virtual-ssa. - Meetings == Next Week == - Continue with mod/ref

[ACTIVITY] 10 - 14 October 2016

== This Week == * TCWG-72 (3/10) - Committed patch to remove optab functions for [us]divmod_optab after Bernd's approval. - My assumption that __udivmoddi4 is generically available is perhaps wrong, call to __udivmoddi4 generates undefined reference error on aarch64-linux-gnu. To be safe, I

[ACTIVITY] 3 - 7 October 2016

== This Week == * TCWG-807 (2/10) - Upstream patch iterations with Honza and Martin. - Committed to trunk as r240898. * TCWG-779 (1/10) - Changed %Z specifier to take int*, unsigned len args instead of vec for easier format checking. - Committed diagnostic.c and diagnostic-core.h changes. -

  1   2   >