== This week ==
- Completed merge of crypto backports from trunk to Linaro 4.8 branch:
- 64-bit backports: 205092, 205383, 205384, 206114, 206115, 206117, 206118,
206149
- 32-bit backports: 206518 AND 206519
- Built and tested 32 and 64-bit compilers with crypto changes
== Next week ==
- Submit backports for review using git review
- New backports
== Future ==
== Progress ==
Deferred US Holiday (MLK day) from Monday to Tuesday
* Debugged what was thought to be a cbuild2 problem but ended up being
due to default enablement of multilib in gcc. (2/10)
* Started working on adding mulit-lib support to cbuild2 (minimal
configuration for installing headers). (8/10) [Jira card forthcoming]
- Moved kernel headers installation to arch based directory in
sysroot and then symlinked appropriate host include directories to
common arch directories.
- Figured out how to build multilib arm libraries for
arm-none-linux-gnueabihf and arm-none-linux-gnueabi.
== Plan ==
* Add cbuild2 scripting to build multilib libraries based on gcc
-print-multi-lib
* Figure out how to get gcc to select the correct multilib headers
directory so that we don't need a merged include directory.
== Issues ==
* None
The Linaro Toolchain Working Group is pleased to announce the
release of both Linaro GCC 4.7 and Linaro GCC 4.8.
As announced at Linaro Connect USA 2013 Linaro GCC is moving to a
pattern of quarterly stable releases, with engineering releases in the
intervening months. This is the first stable release, and contains no
known regressions compared to the 2013.12 release.
The next release of GCC 4.7 will be the 2014.04 stable release. There
will be no engineering releases of GCC 4.7 in 2013.02 or 2013.03.
The next stable release of GCC 4.8 will be the 2014.04 release. Next
month's release - 2013.02 - will be an engineering build.
Linaro GCC 4.7 2014.01 is the twenty second release in the 4.7
series. Based off the latest GCC 4.7.4+svn206380 release, this is the
ninth release after entering maintenance.
Interesting changes include:
* Updates to GCC 4.7.4+svn206380
Linaro GCC 4.8 2014.01 is the tenth release in the 4.8 series. Based
off the latest GCC 4.8.3+svn206350 release, it includes performance
improvements and bug fixes.
Interesting changes include:
* Updates to GCC 4.8.3+svn206350
* Enhanced multilib support
The source tarballs are available from:
http://releases.linaro.org/14.01/components/toolchain/gcc-linaro/4.7http://releases.linaro.org/14.01/components/toolchain/gcc-linaro/4.8
Downloads are available from the Linaro Releases website:
http://www.linaro.org/downloads/
More information on the features and issues are available from the
release pages:
https://launchpad.net/gcc-linaro/4.7/4.7-2014.01https://launchpad.net/gcc-linaro/4.8/4.8-2014.01
Mailing list: http://lists.linaro.org/mailman/listinfo/linaro-toolchain
Bugs: https://bugs.launchpad.net/gcc-linaro/
Questions? https://ask.linaro.org/
Interested in commercial support? Inquire at support(a)linaro.org
== Progress ==
* Got got Beagle Board blacks up, installed all necessary
dependencies to build toolchains. (TCWG-379 - 1/10)
* Installed QEMU and Foundation model on all TCWG x86_64 build
slaves. (TCWG-380 - 1/10)
* Minor bug fixes for Cbuildv2 produced source tarballs to be
hopefully now 100% match with what we usually release. (2/10)
* More major bug fixes and some refactoring for Cbuildv2 produced
binary tarballs to match with what we usually release. (3/10)
* Meetings. (1/10)
== Plan ==
* Build static gdbserver for target via Cbuildv2.
* More work on binary tarball building.
* Add Win32 installer for Canadian cross built binary
releases to Cbuildv2. (TCWG-381)
* Once Jenkins has a node_selector for the new Beagle Board Blacks,
get those working for native builds. (TCWG-379)
* Get QEMU and Foundation Model running on TCWG x86_64 build slaves
so Jenkins can use them for native builds. (TCWG-380)
== Issues ==
* I got invited to Mt Everest last week... :-)
* 4 days week (still some times spent at university).
== Issues ==
* none
== Progress ==
* LRA on AArch32:
o TCWG-343 : Make LRA the default for the ARM backend (3/10)
- Most of the issues fixed or about to be fixed by Vladimir :
* iWMMXT issue and Thumb1 udivmoddi4 : fixed.
* Thumb1 code size issue : partial patch proposed.
- Looking backend places (ARM and AArch64) where reload_in_progress
is used to verify if lra_in_progress is needed as well.
Validation ongoing.
o TCWG-345 : Analyse performance of LRA for ARM. (0/10)
- No progress this week.
* Loop specialisation patch review: (1/10)
o Sent feedback to Samsung.
* 2014.01 release: (1/10)
o Branch merge review
* [Bug 1258013] Re: Cannot build arm64 kernel with CONFIG_DEBUG_INFO=y (1/10)
o Issue due to a wrong .align directive in an assembly source file which is
already fixed upstream.
* GMP library sAArch64 port review: (1/10)
o Started to review Anil work.
* Misc. (1/10)
o Various meetings.
== Next ==
* Still some time spend at university
* Continue on LRA and lib GMP.
- 2014.01 releases: (2/10)
* 4.7 ready, 4.8 needed a short daily to include our multilib patch
to support our binary releases.
* retried cbuildv2 and sent feedback. Should be mostly OK by now.
- cross-validations: (2/10)
* followup-up, bugzilla
* thinking about short term Neon intrinsics testsuite use without dejagnu
- libsanitizer (2/10)
* resumed work for aarch64
- misc (2/10): conf-call and meetings
- internal support (2/10)
* issue:
boards used for benchmarking have been off for several weeks. Pending
jobs disappeared from the list, but I got no log.
== Progress ==
* Implement fork over HW breaks support for arm native targets.
[TCWG-177] [6/10]
* ARM Process Record:
- All patches accepted and upstream.
- Spent time making changes to VFP, SIMD etc patch. [2/10]
- Configuration and Git ssh setup for gdb commit rights . [1/10]
* Submitted a new fix and worked with community for the fix. [1/10]
== Plan ==
* Continue to work on fork over HW breaks support for arm native
targets. [TCWG-177]
* Patch updates and submission to all in progress upstream submissions.
* Travel to Islamabad for Macau visa application.
== Progress ==
* Libc probes support for Aarch64.
Testing the patch for "libc" probes support to "__longjump".
Native compiling in openemebedded.
Obtained a Linaro open embedded image with system tap enabled.
Getting Assembler Error during "malloc" module linking in "glibc"
configured with --enable-systemtap.
In libc_mallopt invalid "asm", invalid expression as operand for
LIBC_PROBE call.
Cross compiling.
Tried copying header files from system tap source to SYSROOT for cross
compiling glibc. The cross compiled GCC does not recognize the header
files.
* Investigate "PGO" for Aarch64.
Started bootstap testing GCC with PGO for GCC.
Stage 1 completed. stage profile in progress.
== Plan ==
- Run EMBCC benchmark with -PGO enabled.
- Work on "glibc" probe.
== Issues ==
Waiting for feedback on "libssp" machine descriptions changes.
Misc
------
14-Jan-2014 Local holiday.
== Issues ==
* None.
== Progress ==
* Test Linaro cbuildv2.
- Write a reference script to support windows installer in cbuildv2.
- Do some basic validation work against the binaries built from cbuildv2.
* Reproduce and work out a patch for "ICE when building
linux-linaro-tracking" (lp:1268893, PR59837) .
* Test to merge Linaro local armv4t multilib patch to Linaro 4.8.
* Investigate CCMP chance by taking a Bool_ var as var != 0.
== Plans ==
* cbuildv2 validation
* Tuning ccmp performance
== Planed leaves ==
* Jan. 31 - Feb. 9.
== This week ==
- Completed merge of crypto backports 206128, 206130, 206131, 206132 and
206149 from trunk to Linaro 4.8 branch
- Out sick Friday
== Next week ==
- Build and test crypto backports
- Begin merging aarch64 crypto backport to Linaro 4.8 branch
== Future ==
== Progress ==
* Cbuildv2 fix for *_release() functions which were changing out of
the build directory but then not changing back to the builddir, which
caused problems with paths relative to the builddir. (1/10)
* TCWG-340 Enable Glibc remote testsuite integration - (7/10)
- Working with upstream community on scoping and designing remote
testsuite integration.
- Setting up arm chroot environment on chromebook with documentation
on how to make this reproduceable.
* TCWG-352 AArch64 clone is missing some CFI markup - (2/10)
- Setup aarch64 environment for testing and built static clone tests
to test the patch.
== Plan ==
* Start implementing (code) for glibc remote testsuite integration.
== Issues ==
None
== Progress ==
* Buildbots
- Odroid U2 died after a few days on USB stick, too. (I guess we're stuck
with Chromebooks)
- Upgrading chromebooks (to Saucy) due to a new GCC/libstdc++ 4.7+
restriction
- Shuffling chromebooks, we now have three buildbots:
- llvm-chrome-01: quick-build+check
- llvm-chrome-02: self-host+check
- llvm-chrome-03: test-suite
* MCJIT Remote
- Re-factored the communication protocol of MCJIT for remote processes
- Still seeing one random failure on ARM, investigating
* Compiler-rt
- Investigating how to get it compiling all libraries and sanitizers on ARM
* Background
- Patch reviews, especially IAS and EH
* Time
- CARD-862 8/10
- Others 2/10
== Plan ==
* Continue MCJIT and Compiler-RT investigations
* Continue reviewing IAS and EH patches
* Try to run the EH tests on the test-suite with the current changes
== Progress ==
* Add support for AArch32 ARMv8 VRINT to QEMU (4/10, TCWG-50)
* Add support for AArch32 ARMv8 VCVT to QEMU (4/10, TCWG-49)
* Add support for AArch32 ARMv8 16<->64bit VCVT to QEMU (1/10, TCWG-51)
* Fixed gas ARM VCVT encoding bug (1/10)
== Issues ==
* None
== Plan ==
* Complete outstanding QEMU work
* ARM setjmp/longjmp SystemTap probes
* malloc
--
Will Newton
Toolchain Working Group, Linaro
Hi,
I've been playing with linux-linaro-tracking and the 2013.12 (and
2013.10, 2013.11) toolchain release and when turning on the NEON in
kernel option I get:
CC lib/raid6/neon4.o
lib/raid6/neon4.c: In function 'raid6_neon4_gen_syndrome_real':
lib/raid6/neon4.c:113:1: internal compiler error: in
dwarf2out_frame_debug_adjust_cfa, at dwarf2cfi.c:1078
}
^
Use the attached .txt as .config and do 'make zImage' to replicate the
problem. You will probably need to attached patch as well since some of
the linaro patches break namespace support :(
regards,
--
Koen Kooi
Builds and Baselines | Release Manager
Linaro.org | Open source software for ARM SoCs
* Two days week (due to compilation courses given at university).
== Issues ==
* none
== Progress ==
* LRA on AArch32:
o TCWG-343 : Make LRA the default for the ARM backend (1/10)
- iWMMXT issue : back on the issue.
- Thumb1 issues : looked at bugzilla's reports.
o TCWG-345 : Analyse performance of LRA for ARM. (0/10)
- No progress this week.
* Loop specialization patch review: (1/10)
o Back on the task.
* [Bug 1258013] Re: Cannot build arm64 kernel with CONFIG_DEBUG_INFO=y (1/10)
o Tried to reproduce the issue.
* Misc. (1/10)
o Catching up with email, etc
o Meeting
== Next ==
* Still some time spend at university
* Continue the ongoing tasks.
* Branch merge reviews.
== Progress ==
* ARM Process Record:
- Follow up on patches, almost all patches accepted. [0.5/10]
- Got GDB committ rights and set up git for pushing patches. [0.5/10]
* Spent time to read through fortran functions prologue to find a new fix
for TCWG-267. [1/10]
* Initial investigation of handling breakpoint over fork [TCWG-177] [1/10]
* Getting documentation ready for Macau visa application [2/10]
== Plan ==
* Update and send patches where required and committ all approved patches.
* Work on breakpoints issues [TCWG-177] and reverse stepping through solib
issue [TCWG-315]
* Apply for Macau visa, given all documents get ready.
== Progress ==
* Libc probes support for Aarch64.
Posted patch for "libc" probes support to "__longjump". Worked on
review comments from Maintainers. Upstream discussions in progress.
Working on testing "glibc" by using --enable-systemtap.
* Experimented with "Cbuildv2" native building.
== Plan ==
- Continue Cbuildv2.
- Investigate "PGO" for Aarch64.
- Work on "glibc" probe.
== Issues ==
Waiting for feedback on "libssp" machine descriptions changes.
misc
===
14-Jan-2014 leave.
== Progress ==
* Draft wiki page (under ACL--let me know if you want to see) for
instructions for setting up chroots on chromebooks for glibc testing:
https://collaborate.linaro.org/display/~ryan.arnold@linaro.org/Configuring+…
* Investigated expanding remote testing possibilities in glibc.
* RFC email to libc-alpha discussing proposal for glibc host testsuite
in chroots:
https://sourceware.org/ml/libc-alpha/2014-01/msg00222.html
== Issues ==
* Shortened week (3-days) because of unplanned school closures on
Monday and Tuesday due to freezing temperatures, during which I had to
watch my kids.
== Plan ==
* Start code scaffolding for glibc host testsuite in chroot.
* Start cbuild2 scaffolding for building remote chroots for glibc testing.
* Add glibc multi-lib builds and IFUNC system libraries.
* Work on cbuild2 support for local tar files.
== Progress ==
* Two day week
* Catching up on email etc. after break (0.5/10)
* Investigate setjmp/longjmp SystemTap probe support in ARM glibc
(1/10, TCWG-378)
* Investigate toolchain bug reports for ARM and AArch64 (0.5/10)
== Issues ==
* None
== Plan ==
* Resume QEMU ARMv8 AArch32 work
* Hopefully push glibc ARM pointer encryption fix for 2.19
--
Will Newton
Toolchain Working Group, Linaro
- 2013.12 releases (1/10)
* Sent announcements for 4.7 and 4.8 releases.
- 2014.01 releases (2/10)
* Created 4.7 and 4.8 branch merge requests (had handle some conflicts in 4.8)
* Looked at backporting the crypto intrinsics support into 4.8, but
there are dependencies with several other patches we haven't
backported yet.
- cross validation (4/10)
* Restarted them since I had to stop them during the holidays. Still
catching up with the backlog, should be OK during the week-end.
* Analyzed and reported several regressions or new tests failing in
some configurations.
- Neon intrinsics tests (1/10)
* Looked at Rob's neon-intrinsics branch and sent him some feedback.
- misc (2/10): conf-calls and meetings.
== Next ==
- 2014.01 releases
- cross-validation: look at results and report regressions if any
- resume libsanitizer work for AArch64
== Progress ==
* Release 3.4 went out
* Buildbots
- Using powered USB hub for the external harddrive, but all boards fail
one time or another...
- Maybe it's the hard-drive type (mobile, power-saving), reverting to USB
sticks for now
- Leaving an odroid U2 running on USB stick for a week, used to work...
- C++11 features will need GCC 4.7+, looking for an easy way to upgrade
the Chromebooks
* MCJIT
- Finished work on remote MCJIT to re-enable the failing tests
- http://llvm-reviews.chandlerc.com/D2527
* Pragma Vectorize
- New discussions hint that my parser patch is moot, for now
- Agreement is to use C++11 attributes for now, and add pragmas later
- I can't continue implementing them, though...
* Background
- Huge number of patch reviews during holidays & this week
- Good progress on new features for the IAS
- More progress on EHABI, including IAS support
* Time
- CARD-862 8/10
- Others 2/10
== Plan ==
* Continue reviewing IAS features, maybe implement some
* Work out how to create a libcxx bot
* re-introduce llvm-chrome-01 as second self-host
* Migrate all buildbots to GCC 4.7+
* Check out EHABI tests and assess progress
