I downloaded the aarch64 binaries to a ubuntu machine:
wink@ssi-primary:~$ uname -a
Linux ssi-primary 3.5.0-21-generic #32-Ubuntu SMP Tue Dec 11 18:51:59 UTC
2012 x86_64 x86_64 x86_64 GNU/Linux
And when I try to run gcc-4.7.3:
wink@ssi-primary:~$ ls -al
~/aarch64-toolchain/gcc-linaro-aarch64-linux-gnu-4.7+bzr115029-20121015+bzr2506_linux/bin/aarch64-linux-gnu-gcc-4.7.3
-rwxr-xr-x 1 wink wink 553068 Oct 18 14:21
/home/wink/aarch64-toolchain/gcc-linaro-aarch64-linux-gnu-4.7+bzr115029-20121015+bzr2506_linux/bin/aarch64-linux-gnu-gcc-4.7.3
I get a file not found:
wink@ssi-primary:~$ strace
/home/wink/aarch64-toolchain/gcc-linaro-aarch64-linux-gnu-4.7+bzr115029-20121015+bzr2506_linux/bin/aarch64-linux-gnu-gcc-4.7.3
-v
execve("/home/wink/aarch64-toolchain/gcc-linaro-aarch64-linux-gnu-4.7+bzr115029-20121015+bzr2506_linux/bin/aarch64-linux-gnu-gcc-4.7.3",
["/home/wink/aarch64-toolchain/gcc"..., "-v"], [/* 19 vars */]) = -1
ENOENT (No such file or directory)
dup(2) = 3
fcntl(3, F_GETFL) = 0x8002 (flags
O_RDWR|O_LARGEFILE)
fstat(3, {st_mode=S_IFCHR|0620, st_rdev=makedev(136, 0), ...}) = 0
mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0)
= 0x7fa688be5000
lseek(3, 0, SEEK_CUR) = -1 ESPIPE (Illegal seek)
write(3, "strace: exec: No such file or di"..., 40strace: exec: No such
file or directory
) = 40
close(3) = 0
munmap(0x7fa688be5000, 4096) = 0
exit_group(1) = ?
I must have done something real stupid, any help appreciated.
-- Wink
== Progress ==
* Very short working week.
* Discussions about next steps in Lava integration.
* Followed up on Arndale/GCC build issues.
== Next week ==
* Review pending Backports
* Investigate bootstrap ICE PR56184 further
== Future ==
* Run HOT/COLD partitioning benchmarks
* Analyse ARM results
* On x86_64 to see what the actual benefit we could get
* fix-gcc-multiarch-testing
* Come up with strawman proposal for updating testsuite to handle
testing with varying command-line options.
--
Matthew Gretton-Dann
Linaro Toolchain Working Group
matthew.gretton-dann(a)linaro.org
== Progress ==
* 64-bits ops in Neon: upstream accepted the patch for gcc-4.9-stage1
* smin-umin: several benchmarks runs necessary to confirm that the
generic patch I suspected actually caused the regression. However, it
also brings improvements in other benches (office-type).
* looked at libasan: very little configuration seems necessary, but
running a sample program under qemu leads to asan runtime complains.
* internal tasks
== Next ==
* smin-umin: run benchmarks on trunk to confirm the regression is also present.
* vectorizer cost model: handle Richard's feedback.
* get more codecs sample codes.
* libasan: understand error messages
Christophe.
== Progress ==
* February merge 4.6 and 4.7
- backport a patch from upstream to resolve conflicts
- this backport introduced some merge issues
- fix pushed and merge request ongoing
* Boehm GC AArch64 support:
- libatomic_ops maintainer rework the patch to make the usage of
gcc atomic builtins available on the other targets
- test exhibit some failures
* libunwind aarch64 support
- bug status moved to critical
- configuration and machine description done
- implementation ongoing
* Aarch64 porting meeting:
- Cancelled this week
== Next ==
* February merge 4.6 and 4.7
- log new testsuite failures in launchpad
* Boehm GC AArch64 support:
- fix libatomic_ops
- validate GCC's Boehm gc integration
* libunwind aarch64 support
- conitnue.
== Progress ==
* Implementing GC sections support in bintuils
Looking at aarch64 relocations to consider while sweeping.
* Cbuild experiments to test gcc svn aarch64 4.7 branch.
Connection problems to the cbuild machines are solved now.
Thanks to Matt. Spawned the build.
Misc
------
* Attend tool chain weekly meet, and stand up call
* Had a look at one x86_64 related patch that gone into GCC trunk.
== Next week ==
* Continue Implementing GC sections support in binutils.
* Start analysis on jump threading in GCC.
== Progress ==
-Attended 1:1 and other team meetings
-Read linaro wiki and on boarding instructions
-Backported Cortex A7 pipeline description to Linaro 4.7 gcc from fsf
trunk and got it reviewed
== Plan for next week ==
- Look into CoreMark regresses in Thumb-2 mode when using the LR regnum.
- Start with research for Removing unnecessary zero/sign extends
Progress:
* Linaro's KVM/QEMU mini-summit, aimed at summarising current status,
identifying what needs to be done and what Linaro should be doing:
https://wiki.linaro.org/LEG/Engineering/Virtualization/201302MiniSummit
* implemented support for variable-length qdev array properties in QEMU
and sent out patch for this plus vexpress fixes which use it
* investigated how to run the Linux Test Project syscall tests under
QEMU to improve the quality of our linux-user mode testing
(wrote up a howto for the QEMU wiki: http://wiki.qemu.org/Testing/LTP
many thanks to fabo for whipping up a Jenkins job to run these nightly)
* couple of trivial linux-user patches as example low-hanging fruit
revealed by the LTP test results
-- PMM
== Progress ==
* Buildbots
- Setting up Chromebook with a buildbot
- http://llvm.org/viewvc/llvm-project?view=revision&revision=174421
- http://lab.llvm.org:8011/builders/clang-native-arm-cortex-a15
- 1h25min builds (3x faster than Pandas)
* LAVA
- Trying LLVM builds on Arndale
- GCC segfaults:
- "The bug is not reproducible, so it is likely a hardware or OS problem."
* Distributed Panda
- Setting up the boards, getting distcc on them
- Speedup is squat (1.9%) because of too many parallel linking
- Need to add LD to a group with less resources on Ninja
* Vectorization
- Reviewing some patches, discussing implementation details
- Vectorization vs. O3 speedup: 2.5x ~ 3.0x (on both ARM and x86_64)
- Working on global structure alias analysis
* LLVM
- Campaigning for integrates-as to be turned on by defaultfor ARM
- ARM seems interested in fixing the eventual bugs
- Adding 'armv7l' triple to default to 'cortex-a8'
- http://llvm.org/viewvc/llvm-project?view=revision&revision=174466
- Campaining for buildbots to use CMAke/Ninja
* EuroLLVM
- Followup meeting, securing funding
== Plan ==
* Change buildbots to not "make clean", try cmake
* Try to reduce parallel linking flood, try one more panda
* Continue with global structure alias analysis
* Try to setup a test-suite buildbot
