== This week ==
- Completed backport of 202259, 202407, 202020, 201261 and 201263
- Submitted merge requesSt for completed backports
== Next week ==
- Begin work on other assigned backports
== Future ==
Document missing features of cbuild for backports
== Progress ==
* Pragma Vectorize
- Comparing pragma OMP to mimic for vectorize
- Implementing the lexer/parser
- http://llvm.org/PR18086
* Android LLVM
- Wasting time for the last time with Odroid XU and Android
- Will focus on Nexus 4, 5 and 7
* Integrated AS
- Changed to enabled by default
- Discussing some bugs found on Android unwind library (inline asm)
- http://llvm.org/PR18231
* Release 3.4
- There were some muddled merges, folks are trying to fix
- Bero found a serious bug, seems to be result of the bad merge
- http://llvm.org/PR18201
- Tested RC2, waiting for RC3
* Background
- Discussions, lots of patch reviews
* Time
- CARD-862 8/10
- Others 2/10
== Plan ==
* Continue implementing pragma vectorize in Clang
* Wait for the next release candidate, test, benchmark
* Follow up in the IAS issues raised
* If RemoteChild is gone by then, continue the MCJIT refactoring
* Have a look at TableGen users
== Issues ==
I saw a black cat, under a ladder, with a hockey mask...
== Progress ==
* Bugfixing and testing QEMU AArch64 FP patches (7/10, VIRT-183)
* Debugging and submitting a patch for issue with Ruby and ARM pointer
encryption (2/10)
* Other miscellaneous work: glibc patch review, binutils testsuite
patch, expenses (1/10)
== Issues ==
* None
== Plan ==
* 2 day week next week then off until Jan 8th
* Wrap up qemu work to a good state to handover
--
Will Newton
Toolchain Working Group, Linaro
== Progress ==
- cross-validation: (1/10). Follow-up, minor maintenance and
discussions on a few regressions. Now able to share top-level
reports.
- backports (1/10):
* multilib_defaults for 4.7
* committed all approved backport
* helped Michael solve some conflicts
- disable peeling: (3/10)
* confirmed it is now done by default on trunk, as part
-fvect-cost-model=XXX.
* Looked at benchmarks, comparing
-fvect-cost-model=cheap (default) and
-fvect-cost-model=unlimited.
- benchmarks on trunk (1/10): observed some variations from one month to
another on a few cases (checked June-November). This confirms
the need to run benchmarks regularly on trunk, in addition to
release branches.
- cbuild2 (3/10): continued working on source release. Submitted a few
small patches for review+integration.
- misc (1/10): various conf-calls and meetings.
== Next ==
- Make 2013.12 releases
- cbuild2: continue testing, try to make 4.7 source release
- libsanitizer on AArch64: resume work
== Issues ==
* Tcpanda03 seems unusable.
== Progress ==
* LRA on AArch32:
o TCWG-345 : Analyse performance of LRA for ARM. (5/10)
- Installed Saucy on Chromebook, lot of time wasted due to defective SD card
- Frequency governor now set to performance.
- Benchmark setting ongoing
- Tcpand03 SSH access re-enabled, board setting seems ok, but respawned
benchmarks are stuck and the board is no more reachable
o TCWG-343 : Make LRA the default for the ARM backend (2/10)
- Reduce iWMMXT testcase, analysis ongoing.
- Trunk testsuite with LRA shows a couple new faillures.
* Reviewed some merge requests. (1/10)
* Loop specialization patch review started. (1/10)
* Various meetings. (1/10)
== Next ==
* Continue LRA, merge and patch reviews.
== Progress ==
* Updated, tested and submitted new patches for a couple of gdb
testsuite failure on arm. [TCWG-269] [TCWG-34] [1/10]
* Performed GDB git repository testing suite results comparison:
[TCWG-34] [2/10]
Compared results and prepared googldocsheets for the following:
arm-native Vs arm-remote
arm-native Vs x86-native
arm-remote Vs x86-remote
* Investigated difference/failures in arm-native vs arm-remote
comparison sheet. [TCWG-34] [5/10]
Re-ran timeout failures with different values of timeouts.
Figured out causes of failures.
Figured out why some tests exit without emitting any logical result.
* In response to comments upstream tried to figure out alternate fix for
dwarf2 fortran parameter test. [TCWG-267] [TCWG-34] [1/10]
* Miscellaneous activities [1/10]
== Plan ==
* Analyze new failures in arm-native Vs x86-native and arm-remote Vs
x86-remote test results.
* Fix test suite failures and submit patches if any.
* Spend some time pining/updating/re-submit previous patches.
== 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.
