== This Week ==
* TCWG-835 (2/10)
- Testing on SPEC CPU2006
- Found ICE while testing on spec unrelated to my patch, created
reduced test-case.
* TCWG-777 (2/10)
- Working on RTL pass - enhanced it so it only replicates insn when required.
- Prototype patch works for the test-case but ICE's during toolchain build.
- Trunk generates expected output for -O2, git-bisect shows from r226516
* TCWG-871 (1/10)
- Built for x86_64 non LTO
- Trying to cross compile for ARM
* Sick Leave (2/10)
* Public Holiday (2/10)
* Misc (1/10)
- Meetings
== Next Week ==
- Continue with TCWG-777, TCWG-835, TCWG-871
# Progress #
* TCWG-806, aarch64 remote debugging multi-arch
support. Patches are pushed in. Done. [2/10]
* TCWG-857, HW breakpoint/watchpoint in multi-arch. [3/10]
Both GDB and linux
kernel needs some fixes. Fix one bug in GDB. Read ARMv8 manual about
byte address select for thumb code (2-byte aligned instruction) and
the implementation in kernel.
* TCWG-757 some patches review, [1/10]
* Misc [4/10]
** ARM new starter training on Wed. afternoon and Thu.
** Read early-debug https://gcc.gnu.org/wiki/early-debug
# Plan #
* TCWG-857.
* Update the state of some old linaro tickets on GDB.
--
Yao
Hi,
I compiled the Linaro cross tool chain for arm v7 SOC and using the buildroot to create the images. When I use the downloaded precompiled tool chain from linaro releases, I am not facing any issues. But with compiled tool chain, multiple packages are failing with similar to below mentioned error. I tried building cross tool chain on both redhat and Ubuntu machines. When I see the difference of gcc -v, the cross compile is with "--host=x86_64-build_unknown-linux-gnu" and the pre compiled is with "--host=i686-build_pc-linux-gnu". I tried with "crosstool-ng-linaro-1.13.1-4.9-2014.08.tar.bz2" and "crosstool-ng-linaro-1.13.1-4.9-2014.09.tar.bz2".
configure:25287: checking for MD5 in -lcrypto
configure:25312: /projects/broadcom-linux/dhananjay/nas/iproc/buildroot/output/host/usr/bin/arm-linux-gnueabihf-gcc -std=gnu99 -o conftest -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -pipe -Os -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 conftest.c -lcrypto >&5
/projects/broadcom-linux/dhananjay/toolchain/tool9/toolchain/lib/ct-ng-linaro-1.13.1-4.9-2014.08/install/lib/gcc/arm-linux-gnueabihf/4.9.2/../../../../arm-linux-gnueabihf/bin/ld: warning: libdl.so.2, needed by /projects/broadcom-linux/dhananjay/nas/iproc/buildroot/output/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/lib/libcrypto.so, not found (try using -rpath or -rpath-link)
-bash-4.1$ uname -a
Linux lab-rtp-gw-05 2.6.32-358.23.2.el6.x86_64 #1 SMP Sat Sep 14 05:32:37 EDT 2013 x86_64 x86_64 x86_64 GNU/Linux
Can somebody help me here.
Thanks
Dhananjay
Hi Linaro Toolchain Group,
I am trying to compiling cross aarch64-linux-gnu toolchain using local
directory (tar file) for gcc.
I am using file:/// as mentioned in the https://wiki.linaro.org/ABE.
../abe/abe.sh --target aarch64-linux-gnu --build all --release
20150819 --tarbin
gcc=file:///home/vpathak/arm/toolchain/build/snapshots/gcc-2015.11-5.tar.bz2
But I get following error:
ERROR (#146): get_URL (not supported for .tar.* files.)
ERROR (#533): get_source
(file:///home/vpathak/arm/toolchain/build/snapshots/gcc-2015.11-5.tar.bz2
not a valid sources.conf identifier.)
TRACE(#190): checkout ()
ERROR (#193): checkout (No URL given!)
ERROR (#161): checkout_all (Failed checkout out of gcc.)
Am I missing something ? How can we use a local directory (e.g. gcc
source code) for building toolchain using abe ?
Please help.
Thanks.
--
with regards,
Virendra Kumar Pathak
== Progress ==
LLDB development
-- Testing and bug fixing lldb on hikey AArch64 [TCWG-886] [7/10]
-- Submitted following fixes and got them reviewed
-- http://reviews.llvm.org/D11899
-- http://reviews.llvm.org/D11902
-- http://reviews.llvm.org/D11987
Miscellaneous [3/10]
-- Meetings, emails, discussions etc.
-- 14th August Public Holiday
== Plan ==
LLDB development
-- Further progress on fixing test failues on hikey AArch64.
-- Testsuite status compilation for armel and armhf
== Progress ==
* Short week due to GNU Cauldron Travel
* Widening pass (TCWG-547) - 3/10
- Looked at the false positive warning for uninitialized variable
- Have a reduced test-case and looking at ways to fix this
- Split the patch for easy review
- Fixed test-cases and going through all the test-case failures
* Misc - 2/10
- gcc-patches, gcc-bugs list
- Looked at gcc vectorization
== Plan ==
- Widening pass
== Progress ==
o Linaro GCC validation (4/10)
* Validation issues investigation
* Austin's APM board seems more reliable
* New validation instability on x86_64 discovered:
- Analysis on-going but it is related to fstack-protector
* Waiting for VPN access
o Upstream GCC (4/10)
* Reworked OOM issue on armeb-linux-gnu target fix
* Patch committed on trunk
* Backport on gcc-5 validated and about to be committed
o Misc (2/10)
* Various meetings
* Team member support
== Plan ==
o Continue on validation
o Redo FSF branch merge after big-endian fix committed.
Benchmark infrastructure - TCWG-360 [6/10]
* Fixed remaining known critical issues in multinode
** Testing stymied by LAVA oversubscription
** Attempted to implement some workarounds for oversubscription
* Some discussion/investigation on generating filesystems for benchmarking
* Started brain-dumping state of infrastructure into cards.linaro.org
Bringing up Juno for noise control experiments [1/10]
* Some struggles with firmware/bootloader configuration
Misc [3/10]
Featuring some ARM management duties
= Plan =
* Get benchmarking tests through, if LAVA permits
* Fix any critical issues discovered
* Finish brain-dump into cards.l.o
* Bring up Juno for noise control experiments
* Some more ARM management
== This week ==
* TCWG-140 - Transform end of loop conditions to min_expr (4/10)
- Wrote initial test case as requested by maintainers
- Need to determine how to not run test case on targets with no
MIN_EXPR/MAX_EXPR
* TCWG-833 - Exploit Wide Add operations when appropriate (4/10)
- Fixed internal error by using unspec for vector sign/zero-extend
- Fixed multiple issues with test cases
- Investigating why gcc.dg/vect/slp-reduc-3.c with lto is regressing
for aarch32 and aarc64
- Investigating two additional regression failures on aarch64
* TCWG-834 - Use non-unit stride loads by preference when applicable (1/10)
- Spoke with Richard Biener at GNU cauldron.
- He indicated this was a problem with a interaction problem with
vectorizer not attempting multiple strategies
- He indicated that he planned to fix this for GCC 6 and not to
attempt a fix
- My plan is to write a test case that fails until the issue is
addressed in GCC 6
* Misc (1/10)
- Conference calls
== Next week ==
- Resolve testsuite regressions for TCWG-833
- Create bugzilla reports for TCWG-833 and TCWG-834
- Write testcase for TCWG-834 and submit upstream
- Investigate TCWG-832 as time permits
== This Week ==
* TCWG-777 (4/10)
- Working on RTL pass, prototype patch works for the test-case.
- updating the trunk to r226907 , it appears combine can fold
arm_andsi3_insn/arm_cmpsi_insn
into zeroextractsi_compare0_scratch at -O2 while not at -O1.
.* TCWG-835 (2/10)
- Testing patch with Spec 2006
- Found ICE unrelated to my patch
* Misc (4/10)
- Travel to home from GNU Tools Cauldron 2015, Prague.
== Next Week ==
- Continue working on TCWG-777
- Continue testing TCWG-835 patch with SPEC 2006
== Progress ==
* Buildbots (CARD-1823 6/10)
- Setting up new APMs
- Running around for a decent UEFI, Kernel, Distro
- Testing different builds, configurations
- Managed to get one APM doing the test-suite *only*! sigh...
- http://buildmaster.tcwglab.linaro.org/builders/clang-cmake-aarch64-lnt
* Background (4/10)
- Code review, meetings, discussions, etc.
- Doing more reviews than usual due to buildbot work being erradic
- Laptop is playing stupid again
== Plan ==
* Moar bots
* Moar code review
* Maybe some vodka, to wash it all up
== Progress ==
LLDB development
-- Testing and bug fixing lldb on hikey AArch64 [TCWG-886] [9/10]
-- Debugging of watchpoint test failures and problems with multiple
watchpoints.
-- Clean-up of AArch64 watchpoint code and fixing watchpoint cache bugs.
Miscellaneous [1/10]
-- Meetings, emails, discussions etc.
== Plan ==
LLDB development
-- Submit and get review on bug fixes for lldb on hikey AArch64 code changes.
-- Investigate further lldb test failues on hikey AArch64.
Miscellaneous
-- Friday 14th August, Independence day public holiday in Pakistan.
== Progress ==
o Linaro GCC validation (4/10)
* Look at the various validation issues we have
* Dug into the logs, restarted jobs, asked to restart
builders/testers. etc ...
o Upstream GCC (5/10)
* Investigate OOM issue on armeb-linux-gnu target
* Found the issue and raised an upstream bugzilla:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67127
* Submitted a fix for it
o Misc (1/10)
* Various meetings
== Plan ==
o Continue validation babysitting
== Progress ==
LLDB development
-- Setup lldb test debug environment for testing failing tests on
chromebook. [1/10]
-- Debug various lldb hang scnarios on chromebook for armhf support
[TCWG-855] [4/10]
-- Figure out alternate fix for http://reviews.llvm.org/D11129.
[TCWG-855] [3/10]
Miscellaneous [2/10]
-- Fix chromebook malfunction due to battery issues
-- Meetings, emails, discussions etc.
== Plan ==
LLDB development
-- Debug lldb test cases on armhf to figure out problems in
individual test-cases
-- Create an updated lldb development project plan on JIRA.
-- Update collaborate with steps on LLDB development process.
== Progress ==
* Widening pass (TCWG-547) - 8/10
- Handled review comments.
- Improved CONVERT_EXPR handling.
- Re based and retested - found some failures.
- This was due to VRP reusing the range info computed in VRP1 after
type promotion is applied. Invalidating the range info when type is
promoted.
- ARM with type promotion now improves (-O2) about ~2.8% with one tiny
benchmark (where it used to regress).
- Still some test cases failures (not execution failures but scanning
certain patterns in the dump/asm).
* Misc - 2/10
- Connect slides.
- gcc-patches, gcc-bgs list
- Meetings
== Plan ==
- Widening pass
== This week ==
* TCWG-146 - Detect smin/umin idiom (1/10)
- Incorporated final feedback and submitted code in SVN
* TCWG-140 - Transform end of loop conditions to min_expr (1/10)
- Misc. discussion upstream about concerning where optimization
should be performed
- Upstream maintainers asked me to create gimple ir test case
* TCWG-833 - Exploit Wide Add operations when appropriate (1/10)
- Unable to validate yet
* TCWG-834 - Use non-unit stride loads by preference when applicable (5/10)
- Further Aarch32 investigation to determine where decision to forgo
vld3 decision is being made
* Linaro connect preparation (1/10)
* Misc (1/10)
- Conference calls
== Next week ==
- Validate patches for TCWG-833 and submit upstream
- Further TCWG-834 investigation
- GNU Cauldron conference and travel
== This Week ==
* TCWG-835 (4/10)
- Validated and submitted patch upstream for review.
- Made changes to patches according to upstream reviews.
- Microbenchmarks: http://pastebin.com/tDnHZuG5
* TCWG-777 (5/10)
- Investigating different ICE's caused by my gimple remove-temps pass
- Looked at expansion of GIMPLE_COND
- Trying to write rtl version of remove-temps pass
* Misc (1/10)
- Conference calls
== Next Week ==
- Continue with TCWG-777
- GNU Tools Cauldron 2015
== Progress ==
* Performance (CARD-1832 1/10)
- Checking differences of PostRAListSched on OOO ARM cores
- Not many changes, ignoring for now
* Maintenance (CARD-1833 4/10)
- Building libc++/abi/unwind in LLVM/Clang tree
- Getting -Wa,-mfpu patches in, last important Clang driver ARM bug
- Some patches to get libunwind and libc++ to compile in-tree on ARM
- Fixed native sub-features detection (http://llvm.org/PR12794)
* Background (5/10)
- Code review, meetings, discussions, etc.
- Long discussions about TargetTuple/TargetParser/Triple
- Lots of patch reviews this week (I mean, *A LOT*)
- Moving some machines around, checking for Chromebook batteries
- Setting up cross-builder using multiarch / QEMU
- Some future planning
== Plan ==
* Look for some more performance issues in 3.7
* Try to hook up the cross-builder
* Investigate libc++ check-all failures
