linaro-toolchain November 2019

linaro-toolchain@lists.linaro.org

14 participants
34 discussions

New IRC Channel for Linaro Toolchain Working Group

by Matthew Gretton-Dann

All, During Connect the suggestion was made that each working group should have its own IRC Channel for discussions and topics relating to the group in particular (as opposed to #linaro which is 'generic' Linaro conversations). Therefore I have just set up #linaro-tcwg on Freenode for the Toolchain Working Group. This channel is public and open to anyone who wants to talk with the TCWG group about anything toolchain related. Thanks, Matt -- Matthew Gretton-Dann Toolchain Working Group, Linaro

2 years, 9 months

[ACTIVITY] 2 weeks ending 1 Dec 2019

by Richard Henderson

7 working days, then Thanksgiving. [VIRT-262 # ARMv8.1-PAN Privileged Access Never] Finished, still need to post. [VIRT-273 # ARMv8.2-ATS1E1, AT S1E1R and AT S1E1W instruction variants ] Finished, still need to post. [VIRT-276 # ARMv8.2-UAO, PSTATE override of Unprivileged Load/Store ] Finished, still need to post. [VIRT-263 # ARMv8.1-VHE Virtual Host Extensions ] FIXED! Welsh sprint with AJB; found and fixed two bugs. Final bug causing guest kernel crash while booting fixed upstream by Marc Zyngier vs ptrauth. Will do some more thorough testing during rc4 and post once the development phase opens up again. [VIRT-327 # Richard's upstream QEMU work ] Review of target/hexagon skeleton. Review of arm dcpop patch set for beata. Fixed a couple of arm translator bug for clyon. Some investigation into a reported hppa-linux-user bug. While I can reproduce locally, so far I have not tracked down anything that I can prove is a translation bug. r~

4 years, 4 months

[ACTIVITY] week ending Dec. 1 2019

by Alex Bennée

QEMU Tooling ([VIRT-252]) ========================= Extend gdbstub for SVE ([VIRT-281]) - worked on [v2 rebase addressing comments] - posted {PATCH v2 00/14} gdbstub refactor and SVE support Message-Id: <20191130084602.10818-1-alex.bennee(a)linaro.org> [VIRT-281] https://projects.linaro.org/browse/VIRT-281 [working prototype] https://github.com/stsquad/qemu/tree/gdbstub/sve-registers [v2 rebase addressing comments] https://github.com/stsquad/qemu/tree/gdbstub/sve-registers-v2 QEMU ARMv8.1 VHE ([VIRT_263]) ============================= - inaugural Welsh code sprint with rth - found some new bugs, squashed some old bugs - together with recent upstream fixes SUCCESS! - can now boot a guest from a VHE enabled kernel :-) [VIRT_263] https://projects.linaro.org/browse/VIRT-263 Upstream Work ([VIRT-109]) ========================== - posted {PULL for 4.2 0/3} a few vm-test fixes Message-Id: <20191126120339.18059-1-alex.bennee(a)linaro.org> - there are still niggling netbsd failures - posted {PATCH for 4.2?} .travis.yml: drop xcode9.4 from build matrix Message-Id: <20191127132430.3681-1-alex.bennee(a)linaro.org> - investigation into [ARM HPC compiler triggered linux-user bug] - may be 64k page related as couldn't reproduce on Ubuntu - posted {PATCH v1 0/5} linux-user mmap debug cleanup Message-Id: <20191128194603.24818-1-alex.bennee(a)linaro.org> [ARM HPC compiler triggered linux-user bug] https://bugs.launchpad.net/qemu/+bug/1853826 Other Activities ================ - published [QEMU Summit and KVM Forum trip report] [QEMU Summit and KVM Forum trip report] https://collaborate.linaro.org/display/CR/20191030+QEMU+Summit+and+KVM+Foru… Absences ======== - 2nd Dec Holiday Current Review Queue ==================== * {PATCH 0/4} python/qemu: New accel module and improvements Message-Id: <20191115180829.10275-1-wainersm(a)redhat.com> Added: <2019-11-28 Thu> * {PATCH v2 0/2} Run tcg tests with tci on Travis Message-Id: <20191128153525.2646-1-thuth(a)redhat.com> Added: <2019-11-28 Thu> * {PATCH 0/2} flush CPU TB cache in breakpoint_invalidate Message-Id: <20191127220602.10827-1-jcmvbkbc(a)gmail.com> Added: <2019-11-28 Thu> * {RFC PATCH 00/10} hw/avr: Introduce the Arduino board Message-Id: <20191128015030.27543-1-f4bug(a)amsat.org> Added: <2019-11-28 Thu> -- Alex Bennée

4 years, 4 months

[weekly][dsg-models][linaro] report week ending 29 Nov

by Peter Maydell

Progress: * VIRT-65 [QEMU upstream maintainership] - code review: + finally got back to the reset-refactoring patchset and gave review on v5 of that. This is very nearly ready. + reviewed and got into 4.2 rc3 some patches from Marc Z fixing some missing emulation/bugs that newer Linux guest kernels trip over + rc3 out of the door; we will need an rc4, though - more time consumed by office-move thanks -- PMM

4 years, 4 months

[ACTIVITY] Week ending 29th November 2019

by Peter Smith

[Morello] Rebase of LLD against September CUCL update complete - Painful due to LLD changing address layout (every test expected value shifted), and a naming convention change. - No functional changes needed to patch. - Submitted static linking patches for review. Will send the dynamic ones after all static linking has been merged. Wrote up notes of Linaro Tech-leads Morello Q&A. Misc: Upstream LLD reviews

4 years, 4 months

[ACTIVITY] 25-29 Nov 2019

by Christophe Lyon

== Progress == * GCC: - -mpure-code on v6m: waiting for approval, pinged again. * BFD Linker: - GNU-629: non-contiguous memory support: RFC patch sent. * GCC upstream validation: - only a minor testcase issue to fix this week * misc: - infra fixes / troubleshooting / reviews == Next == * GCC: pure-code/v6m, handle feedback * Binutils: GNU-629: support non-contiguous memory regions in linker

4 years, 4 months

[ACTIVITY] 25 - 29 November 2019

by Diana Picus

== Progress == * Out of office on Thursday * LLVM 9.0.1 - Uploaded ARM & AArch64 binaries for rc1 - ARM: opened 2 bug reports (asan and cfi tests failing) * Triaging check-lldb failures on AArch64 [LLVM-512] - Opened a few more bug reports - Got one nasty failure that I want to look into a bit more before committing a patch XFAIL-ing everything so far * Morello - Got a VM working, built the toolchain, currently trying to build android - Setting up all sorts of gerrit accounts and other minutiae == Plan == * More of the same

4 years, 4 months

Help understanding "VFP magic dust" in cortex-strings memcpy

by Alexander Monakov

Hi! I've attempted to study the implementation of memcpy for 32-bit Arm cores in Glibc (which is also found in arm-optimized-routines and first appeared in Linaro's cortex-strings project), and I came across a peculiar snippet: #ifdef USE_VFP /* Magic dust alert! Force VFP on Cortex-A9. Experiments show that the FP pipeline is much better at streaming loads and stores. This is outside the critical loop. */ vmov.f32 s0, s0 #endif This seems to imply that this NOP-like instruction affects CPU state and makes the vldr/vstr instructions that follow use different datapaths that they might otherwise? Can anyone shed more light on this, please? I was able to trace history of this code back to revision 100 in cortex-strings repository, where it appeared as part of a large rewrite by Will Newton: https://bazaar.launchpad.net/~linaro-toolchain-dev/cortex-strings/trunk/rev… The entire memcpy.S file in Arm optimized-routines repo can be found here: https://github.com/ARM-software/optimized-routines/blob/master/string/arm/m… Thanks! Alexander

4 years, 4 months

High stack usage due ftree-ch

by Adhemerval Zanella

Hi Arnd, I took a look on the stack usage issue in the kernel snippet you provided [1], and as you have noted the most impact indeed come from -ftree-ch optimization. It is enabled in all optimization levels besides -Os (since besides possible increasing the stack usage it also might increase code side). I am still fulling grasping what free-ch optimization does, but my understanding so far is it tries to reorganize the loop for later loop optimization phases. More specifically, what it ends up doing on the specific snippet is create extra stack variables for the internal membber access in the inner loop (which in its turns increase stack usage). This is also why adding the compiler barrier inhibits the optimization, since it prevents the ftree-ch to optimize the internal loop reorganization and it is passed as is to later optimizations phases. It is also a generic pass that affects all architecture, albeit the resulting stack will depend on later passes. With GCC 9.2.1 I see the resulting stack usage using -fstack-usage along with -O2: arm 632 aarch64 448 powerpc 912 powerpc64le 560 s390 600 s390x 632 i386 1376 x86_64 784 Also, -fconserve-stack does not really help with this pass since ftree-ch does not check the flag usage. The fconserve-stack currently only seems to effect the inliner by setting both large-stack-frame and large-stack-frame-growth to some conservative values. The straightforward change I am checking is just to disable tree-ch optimization if fconserve-stack is also enabled: diff --git a/gcc/tree-ssa-loop-ch.c b/gcc/tree-ssa-loop-ch.c index b894a7e0918..b14dd66257c 100644 --- a/gcc/tree-ssa-loop-ch.c +++ b/gcc/tree-ssa-loop-ch.c @@ -291,7 +291,8 @@ public: {} /* opt_pass methods: */ - virtual bool gate (function *) { return flag_tree_ch != 0; } + virtual bool gate (function *) { return flag_tree_ch != 0 + && flag_conserve_stack == 0; } /* Initialize and finalize loop structures, copying headers inbetween. */ virtual unsigned int execute (function *); On powerpc64le with gcc master: $ /home/azanella/gcc/gcc-git-build/gcc/xgcc -B /home/azanella/gcc/gcc-git-build/gcc -O2 ../stack_usage.c -c -fstack-usage && cat stack_usage.su ../stack_usage.c:157:6:mlx5e_grp_sw_update_stats 496 static $ /home/azanella/gcc/gcc-git-build/gcc/xgcc -B /home/azanella/gcc/gcc-git-build/gcc -O2 ../stack_usage.c -c -fstack-usage -fconserve-stack && cat stack_usage.su ../stack_usage.c:157:6:mlx5e_grp_sw_update_stats 176 static The reference for minimal stack usage is with -Os: $ /home/azanella/gcc/gcc-git-build/gcc/xgcc -B /home/azanella/gcc/gcc-git-build/gcc -Os ../stack_usage.c -c -fstack-usage && cat stack_usage.su ../stack_usage.c:157:6:mlx5e_grp_sw_update_stats 32 static I will try to check if also enable the same test for -fgcse and -free-ter do make sense. [1] https://godbolt.org/z/WKa-Bd

4 years, 5 months

[Activity] Nov 18th ~ Nov 22nd 2019

by Luis Machado

# Progress # o Upstream GDB * Make remote packet length in debugging output adjustable (as opposed to fix to 512 bytes). * Investigated ARM sim build issues with the GCC default moving to -fno-common. o GDB: * GNU-644 - [GDB, AArch64] gdb.base/step-over-syscalls.exp failures - No progress yet. Waiting for Kernel feedback. * [RESOLVED] GNU-645 - gdbserver is not using SVE register descriptions properly - Pushed a fix upstream. * GNU-170 - GDB BZ #21221 - gdb hangs while stepping an empty loop - On hold for now. o Friday off # Plan # o Upstream GDB * Fox -fno-common build issues with ARM sim. o GDB * GNU-644 - [GDB, AArch64] gdb.base/step-over-syscalls.exp failures - Continue working on a fix.

4 years, 5 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

linaro-toolchain November 2019