== Issues ==
* None
== Progress ==
* CCMP for AARCH64.
- Design cases to cover most conditional compare combinations.
- Set up qemu environment and run regression tests. And fix all the new FAILs.
* Respawn 2014.01-01 baremental toolchains by setting
CT_TARGET_LDFLAGS=" --specs=rdimon.specs "
* Rebuild arm-linux-gnueabi toolchain and ran spec2k. But can not
reproduce the regression about Linaro 4.8 based on 2014.01 release.
== Plans ==
* Send out the CCMP patches for community review.
== Progress ==
- TCWG-394 (5/10)
- Found one more issue, regression tested and posted the patch
upstream for review
- http://gcc.gnu.org/ml/gcc-patches/2014-02/msg01282.html.
- TCWG-395 (1/10)
- Started with a google doc for wiki update
- Speck regression analysis with 4.7, 4.8 Nov and Dec releases (2/10)
- Native build of these versions and rank spec2k benchmarking for all
of them
- Not able to reproduce it on a15 as reported
== Misc ==
1 Day off sick
== Plan ==
- Get all the information required for wiki - TCWG-395
- Start working on TCWG-396
== Progress ==
* Fixed remote testing via Cbuildv2 (TCWG 324 - 3/10). Currently it
supports multiple build farms, it uses my local hardware and the
TCWG build farm, and when run via Jenkins it uses the TCWG build
farm.
* Got qemu-aarch64 built and working, added to the DejaGnu site.exp
file used by Cbuildv2, so now aarch64 execution tests actually
work. Then cloned qemu-aarch to the x86_64 tcwg build farm
machines so Jenkins can use it. (TCWG 393 - 3/10)
* Meetings and misc. (4/10)
- Worked on LCA14 presentation.
- Write documentation on how to tweak the config file for DejaGnu
used by Cbuildv2 for remote test execution.
- Got Gerritt and Jenkins integrated (thanks Paul!) so merge
requests fire up Jenkins to do builds on everything in the TCWG
build farm.
- Had ITS create a new list 'tcwg-test-results', which soon will
get buid failure notifications from Jenkins and any test
regressions.
- Added support for emailing test results to Cbuildv2.
== Plan ==
* Test the Gerrit/Jenkins integration.
* Improve new test analysis script to compare builds from more than
the same day, since now some test runs take 32hours with all the
execution tests running on a chromebook an Beagle Board.
* Travel to Connect!
== Issues ==
* Somebody needs to upgrade the crouton install on all our
chromebooks, as the current version won't build GCC native.
== Progress ==
* IAS (TCWG-377)
- Abandoning softvfp implementation, we should not use it
- Both softvfp and dwarf flags went back to kernel/android
- Macro issue has been solved, IAS is now Gold!
- Still doesn't compile the kernel, though... ;)
- Discussions about inline assembly bailing on asm garbage
* AArch64 (TCWG-387)
- Trying to get the build working and passing tests
- Some worries about config.guess updates (license)
- Commented out some old JIT tests, they will never work
- Memory allocation failures look real bugs, though
- At least Five real bugs in test-suite
* Background
- Patch reviews
- Further discussions about Dwarf vs. EH unwind tables
- Discussing vectorizer pragmas with GCC list
- Reviewing EuroLLVM14 papers
- Helping Rob upgrade TCWG Chromebooks
* Time
- CARD-862 8/10
- Others 2/10
== Plan ==
* Take a look at Compiler-RT
* Check some of the AArch64 errors
* Maybe EH/unwind issues
* Connect
Hi,
I finally got around to checking the attached patch for the
https://bugs.launchpad.net/ubuntu/+source/gcc-4.8/+bug/1270789
I noticed attached patch causes regression for pr38151.c in gcc test-suite.
A reduced test-case that triggers this is:
static unsigned long global_max_fast;
int __libc_mallopt (int param_number, int value)
{
__asm__ __volatile__ ("# %[_SDT_A2]" :: [_SDT_A2] "nor"
((global_max_fast)));
global_max_fast = 1;
}
In this regard I have couple of questions:
1. Is the in-line asm valid? Look ok to me.
2. For the pr38151.c regression, asm diff is as shown below.
< add x0, x0, :lo12:.LANCHOR0
< ldr x0, [x0]
---
> ldr x0, [x0,#:lo12:.LANCHOR0]
This causes:
pr38151.c:(.text+0x10c): relocation truncated to fit:
R_AARCH64_LDST64_ABS_LO12_NC against `.rodata'
collect2: error: ld returned 1 exit status.
If I however increase the alignment of .rodata where .LANCHOR0 is
defined, this passes. Is alignment of BITS_PER_UNIT valid for
SYMBOL_REF? If I change it as I am doing this attached patch, is there
anything else I need to do.
Thanks,
Kugan
Hi Linaro'ites
We have upgraded eglibc to 2.19 in OpenEmbedded and I started to notice
WARNING: No recipes available for:
/home/kraj/angstrom/sources/meta-linaro/meta-linaro-toolchain/recipes-core/eglibc/eglibc_2.18.bbappend
NOTE: Resolving any missing task queue dependencies
And it seems in this bbappend the SRC_URIs are overridden to point to a Linaro
release of eglibc
I wanted to know if there is a 2.19 head for linaro eglibc being created ?
in that case that bbappend needs fixing too.
Thanks
-Khem
Ganesh,
Thank you for your email - please note this question is best directed
at linaro-toolchain(a)lists.linaro.org where it is more likely to get
picked up quickly.
Our general rules for benchmarking are to use -mcpu=cortex-a## -O3 as
the compiler flags (where ## is the particular CPU you are interested
in).
Linaro's philosophy is to try and get the best results possible that
normal users will see, we don't go looking for 'magic' options that
may give better performance on one benchmark, but not in general.
Thanks,
Matt
On 20 February 2014 09:22, Gopalasubramanian, Ganesh
<Ganesh.Gopalasubramanian(a)amd.com> wrote:
> Hi,
>
> We would like to run some benchmarks in the foundation model.
> I like to know which is the best option-set (GCC compiler options) that the linaro community recommends for aarch64.
>
> Regards
> Ganesh
>
--
Matthew Gretton-Dann
Linaro Toolchain Working Group
matthew.gretton-dann(a)linaro.org
== Progress ==
- vectorizer (7/10)
- Looked into vectorizer target-hooks
- enablement of half-float for vectorising; dropped the patch after
discussing
- Started analysing code generated with unlimited model
- gcc 4.8 regression (1/10)
- Built Novemnebr and December release (native with system libc)
- Ran spec2k fp. Can see regression for ammp with -O3 (with
-march=armv7-a and -mthumb)
-Started bisecting
- Chromebook reset-up (2/10)
- SD card died and setup again
- Using hdd for gcc testing
== Plan ==
- Check ARMv5 regression for unaligned access
- Look into vectorizer cost model/benchmarking
Richard,
I found some emails about you implementing softvfp back in 2003, and
I'd like to know what is the expected behaviour when it conflicts with
the target triple, for example:
-triple arm-linux-gnueabihf + -mfpu=sofvfp+vfp
In this case, in LLVM, the triple sets "-float-abi=hard" but the fpu
would set "+soft-float-abi", which are contradictory flags.
Is that case even possible? If so, what's the expected behaviour? Soft
or hard float?
Do the extra flags always override the triple behaviour? Is it
expected that *every* compiler flag will work on a
last-seen-sets-behaviour manner?
cheers,
--renato
