== Progress ==
- libssp gcc
Initiated mails for upstream discussion on supporting global and TLS
based libssp ABI for Aarch64. Now waiting for feedback from ARM
maintainers.
- Built few benchmarks with Linaro tool chain binaries. Faced some
openmp failures, compiler disabled for libgomp. Looking to rebuild
the linaro cross tool chain.
- Rebased glibc trunk and ran cross glibc tests on V8 foundation
model. With TLS SSP patch glibc's libresolve.so fails to build
because the stage 1 GCC compiler emits global stack_chk_guard. The
glibc dos not export global stack_chk_gaurd, when TLS based stack
guard is set. Now looking at supporting and test both TLS and global
stack guard ABI, but before that waiting for confirmation from ARM
maintainers.
- Reinstalled linux 13.04 ubuntu on my laptop
- Tested the patch
http://gcc.gnu.org/ml/gcc-patches/2013-12/msg00170.html for
aarch64-none-elf target. No regressions found.
== Plan ==
- Continue upstream discussions and post RFC patches for glibc and gcc libssp
- Explore on cbuildv2
- Cross Build few openmp benchmarks
== Issues ==
* None.
== Progress ==
* 1.5 days off.
* Misc update for Linaro crosstool-ng
- Re-enable parallel build for do_cc_core.
- Add linaro-armeb-none-eabi config.
- Update all CT_TOOLCHAIN_PKGVERSION to "Linaro GCC 2013.11"
* Share toolchain binaries configures to cbuild2 group.
* Backporting r200103 and r205509 to linaro 4.8.
* Collect Spec2k INT results with -fno-move-loop-invariants and
-fira-loop-pressure on X86-64 and ARM Chrome book. Performance of
different cases waves from -2% - 2%. But the overall performance
changes < 0.5%.
== Plan ==
* Continue on CCMP.
== Progress ==
* Setup new git mirror of new binutils-gdb.git repository, and
added a cron job to update it every 10 minutes. (#369, .5/10)
* Started modifying Cbuildv2 to use new binutils-gdb
repository, which changes how we parse URLs. (.5/10)
* Modified Cbuildv2 to produce fully statically linked binaries so
32bit ones run on any distro, including 64 bit. (#368, 3.5/10)
* Worked with LAVA team to set up a private Jenkins for TCWG use
only to protect test and benchmark data. (#379, 2/10)
* Fixed Cbuildv2 bug where the dryrun function wasn't properly
trapping configure and build errors. (.5/10)
* Meetings & Card maintainance. (2/10)
* Quick winter camping trip at the hot springs for my bday, hit -38C !
== Plan ==
* More tweaks to source and binary tarball creation.
* Finish modifying Cbuildv2 to use new binutils-gdb repository.
* Fix all the minor bugs caused by properly trapping errors from
dryrun function.
* Make the new Jenkins build in a 32 bit chroot on the TCWG build
machines
* Figure out why Jenkins doesn't import the Junit files.
* Copy .sum files to remote host.
* If new Arndale or APM boards arrive this week like promised,
start evaluating them.
* Get the S&M team to start using Gerrit for merge requests for
Cbuildv2.
== Progress ==
* TCWG-367 drop --dostep, add --checkout (1/10)
Completed with following git commits:
commit 2d4d7588c1b05801ee5e1a924c8641ff7c322c07
Author: Ryan S. Arnold <ryan.arnold(a)linaro.org>
Date: Fri Dec 6 11:46:40 2013 -0600
cbuild2.sh: remove --dostep. It's no longer supported.
Also added testcase to make sure it returns an error.
Author: Ryan S. Arnold <ryan.arnold(a)linaro.org>
Date: Fri Dec 6 10:57:34 2013 -0600
Add cbuild2 --checkout <package>|all support with tests
* misc (1/10)
commit 9fb1c4404f8065ef844154e614ff4e7086fce176
Author: Ryan S. Arnold <ryan.arnold(a)linaro.org>
Date: Thu Dec 5 12:21:20 2013 -0600
lib/package.sh: git status should be run in the current directory.
commit 7bb4a9ce8ef440655fe45e6819b57da8d61e8728
Author: Ryan S. Arnold <ryan.arnold(a)linaro.org>
Date: Wed Dec 4 11:47:56 2013 -0600
lib/package.sh: git st should be git status.
* TCWG-356 Unify GIT URL parsing (5/10)
* TCWG-359 Change branch character (2/10)
Closed with following git commits:
commit c9368930e6fb1ad46d08c2dbcc34d8167344ad28
Author: Ryan S. Arnold <ryan.arnold(a)linaro.org>
Date: Tue Dec 3 16:29:00 2013 -0600
Finish git parser refactoring, use stamp functions, complete
dryrun wrapping.
commit eabbc74e48bc8c3a4addbc53ec6a492521193a0c
Author: Ryan S. Arnold <ryan.arnold(a)linaro.org>
Date: Wed Nov 27 16:31:20 2013 -0600
Working additions for git parser refactor and stamping.
== Issues ==
*misc (1/10)
- Discovered a bug in glibc when using upstream (4.9) gcc that isn't
present in 4.8. This might have something to do with how binutils
interacts with the compiler instead of simply a compiler issue:
elf/dl-addr.c:152: multiple definition of `_dl_addr_inside_object'
elf/dl-open.c:746: first defined here
== Plan ==
* Create card for new glibc multiple definition problem,
* Start working on glibc related cards
== This week ==
- Completed backport of 197997
- Completed backport of 203774
- Worked on GCC 4.8 FSF Release Branch Merge and initiated merge request
and build
== Next week ==
- Work on GCC 4.7 Release branch Merge
- Begin work on other assigned backports
== Future ==
Document missing features of cbuild for backports
== Progress ==
* MCJIT (2/10)
- Disabling Remote MCJIT for now, just to add the self-hosting bot
- Working on a more robust message passing interface
- http://llvm.org/bugs/show_bug.cgi?id=18057
- Progress stalled, waiting for changes to child process handler
* Buildbots (2/10)
- Adding Chromebook self-hosting buildbot
- http://lab.llvm.org:8011/buildslaves/linaro-chrome-02
- Adding Odroid XU check-all buildbot
- http://lab.llvm.org:8011/buildslaves/linaro-odroidxu-01
* Vectorizer (2/10)
- Implemented #pragma vectorize enable in LLVM
- http://llvm.org/bugs/show_bug.cgi?id=18086
- Adding support in Clang to add metadata to IR for all three pragmas
* Android (1/10)
- Trying LLVM kernel on XU again
- Managed to get adb/astboot to work
- Custom kernels won't boot, Tinti will create a full image (CM+LLVM)
* Release 3.4 RC2 (1/10)
- Testing, no regressions
- Some benchmarking regressions, investigating
- http://llvm.org/pre-releases/3.4/rc2/
* Background (2/10)
- Reviewing patches, discussions, meetings, etc
== Plan ==
* Continue implementing pragma vectorize on Clang
* Further release 3.4 testing and benchmarking
* Continue investigating benchmark results for 3.4
* Re-start MCJIT MPI once changes to the child process handlers are in
Hi,
According the debian bug report [1], it is not possible to use std::future
on armv5 targetting toolchains. This is because libstdc++ will only enable
std::future if ATOMIC_INT_LOCK_FREE > 1. There is no LDREX for armv5 and
older, so this definition is set to ATOMIC_INT_LOCK_FREE when compiling for
ARMv4t or ARMv5.
My impression is that you should be able to use the kernel helpers for
atomic operations in lockfree (?) manner, so the ATOMIC_INT_LOCK_FREE
definition is probably incorrect on older arm archs then?
Riku
[1] http://bugs.debian.org/727621
== Progress ==
* Disable peeling:
- spawned benchmarking jobs using recent trunk and another older
one to check for performance regressions.
- locally generated the code for the same benchmark to analyze the
differences
* cbuildv2:
- retried to create 4.8-2013.11 source release. Identified a couple
of minor issues, reported to Rob.
- for 4.7-2013.11, need support for subdir in git URLs, fix on-ongoing by Ryan
* Backports:
- gave some support to Michael
- a few have been reviewed/accepted.
* libsanitizer: little progress this week; advised to take a look at
recent LLVM sources.
* trunk validation: filed some bugzilla reports; 2 of them fixed by Kyrill.
== Next ==
* Disable peeling:
- check benchmarks results
- check code generation
* cbuildv2: re-do source releases 4.8 and 4.7 2013.11
* Backports:
- commit already approved ones, once cbuildv2 releases are OK
- see which ones can be easily added
- work on MULTILIB_DEFAULTS backport for 4.7
== Issues ==
* None.
== Progress ==
* LRA on AArch32:
o TCWG-342 : UPSTREAM: LRA issue in gfortran with store minmaxsi
- Committed as rev205526 .
- Card resolved.
o TCWG-344 : UPSTREAM: Thumb bootstrap failure with LRA
- Committed THUMB_SECONDARY_INPUT_RELOAD_CLASS patch as rev205527.
- Committed PR58785 patch as rev205581.
- Card resolved.
o TCWG-345 : Analyse performance of LRA for ARM
- Gathered results and analysis still ongoing.
o Working on a new failure with iWMMXT (LRA loops during constraints solving).
* Reviewed some merge requests.
== Next ==
* Continue on LRA
* Loop specialization patch review.
