This is an attempt to build TI mcsdk 3.0.3.15 with
bitbake tisdk-server-rootfs-image
It fails trying to make rpms for the linaro toolchain.
| DEBUG: Python function write_specfile finished
| DEBUG: Executing shell function BUILDSPEC
| error: line 4: invalid tag value("^[A-Za-z0-9+._]+$") Release:
Release: r2-arago3
| error: Package has no %description:
external-linaro-toolchain-2.15.armv7ahf_vfp_neon
| Building target platforms: armv7ahf_vfp_neon-oe-linux-gnueabi
| DEBUG: Python function do_package_rpm finished
| DEBUG: Python function do_package_write_rpm finished
| ERROR: Function failed: BUILDSPEC (see
/home2/ghannon/keyDSP/build/mcsdk/build/arago-tmp-external-linaro-toolch
ain/work/armv7ahf-vfp-neon-oe-linux-gnueabi/external-linaro-toolchain-20
13.03-r2-arago3/temp/log.do_package_write_rpm.8491 for further
information)
ERROR: Task 40
(/home2/ghannon/keyDSP/build/mcsdk/sources/meta-linaro/recipes-devtools/
external-linaro-toolchain/external-linaro-toolchain.bb,
do_package_write_rpm) failed with exit code '1'
If I change Release: r2-arago3 to not have a - in the spec file then I
can build the rpms manually with the same command.
Is there a place where I can change that globally?
Gary Hannon
Sr. Software Engineer
CSP Inc.
43 Manning Road Billerica, MA 01821
978-663-7598 x1509
ghannon(a)cspi.com
== Progress ==
* IAS
- Having a go at .dn/.qn aliases
- Some Dwarf 2/3 discussions, kernel build flags, etc
* EHABI
- Tables being generated even when no EH in C
- Will need some tinkering in Clang and LLVM
- Huge discussion about EHABI, Dwarf unwinding and exception tables
- Starting the refactory of EHABI to cope with the differences
* Compiler-RT
- Finding the best way to use compiler-RT with Clang (not an easy task!)
* Background
- Usual patch reviews & discussions
- Mapping AArch64 features implemented for Connect session
- Playing with the Calxedas for some benchmarks
* Time
- CARD-862 8/10
- Others 2/10
== Plan ==
* Continue with EHABI refactoring
* Implement --rtlib in Clang for compiler-rt usage
* Continue .dn/.qn implementation
== Progress ==
* 3 day week
* Further work on longjmp/setjmp Systemtap probes on ARM (1/10, TCWG-378)
* Research and analysis of malloc workloads for benchmarking (5/10, TCWG-160)
== Issues ==
* None
== Plan ==
* More work on an application level malloc benchmark framework
* Resurrect glibc benchmark graphing script
* newlib release
* gdb 7.7 release (build an rc to troubleshoot releasing from git)
--
Will Newton
Toolchain Working Group, Linaro
Hello Ulrich,
I'm on a freescale mx53 (single core, cortex a8) and trying to find a
nasty problem in our $CUSTOMER's application. Working hardware watch
point support in gdb should solve the problem, or at least brings us a
big step forward. I stumbled over your about three years old thread[1]
about the cortex a8 memory mapped debug registers problem. Do I
understand correctly, that you had a8 watch point running on at least
one board? Have you got any code and or pointers to get watch point
support running?
regards,
Marc Kleine-Budde
[1]
http://lists.linaro.org/pipermail/linaro-toolchain/2011-February/000820.html
--
Pengutronix e.K. | Marc Kleine-Budde |
Industrial Linux Solutions | Phone: +49-231-2826-924 |
Vertretung West/Dortmund | Fax: +49-5121-206917-5555 |
Amtsgericht Hildesheim, HRA 2686 | http://www.pengutronix.de |
*Hi LINARO-TEAM,*
I would like start a software development on a “Renesas RZ (ARM Cortex-A9)
MPU (R7S721010xxx” by using Eclipse IDE (Kepler - 4.3.1).
> Global objective = Use free SW development tools
I have read, that the “Linaro GCC Toolchain Release Supports Full Range of
ARM Cortex-A Processors” and it’s an open source software for the ARM
architecture, including the GCC toolchain.
> Well – I think that is what I need.
So, I have installed the Linaro Toolchain
(gcc-arm-none-eabi-4_7-2013q3-20130916-win32.exe) and test the compilation
of the “*Hello World ARM C Project*” for the Target Processor “cortex-A9”
> So, the compilation is OK.
*NOW, I would like prepare a small TEST-Project with FreeRTOS.*
FreeRTOSV8.0.0_Release_Candidate_2
> I have found the “Renesas RZ (ARM Cortex-A9) RTOS Demo”:
http://www.freertos.org/Renesas_RZ_Cortex-A9-RTOS.html
But the two projects are provided to be built with the* IAR Embedded
Workbench <http://www.iar.com/ewarm>* or the *ARM DS-5
<http://arm.com/products/tools/software-tools/ds-5/index.php> embedded
development tools*.
> I have seen that the DS-5 Development Studio is not for free but it use
also the Linaro GNU Compiler (GCC) toolchain.
So, I have tried to import the “Renesas RZ (ARM Cortex-A9) RTOS Demo”
project to Eclipse, *but the import doesn’t work – and I’m not sure that
the project uses the Linaro GNU GCC Compiler**.*
> I think that Demo project for DS-5 use also special plugins/configuration
settings from the Development Studio – so this can’t work ?!…
*HELP:*
*I’m new in the “Eclipse IDE / Compiler (GCC) toolchain” WORLD … *
So, can you help me to give me a good direction - *what I need to do *to
make running this small “Blinky-Demo for Renesas RZ (ARM Cortex-A9) with
the FreeRTOS” by using the Linaro GCC Toolchain.
> Or have you another small Demo project for the Linaro GCC Toolchain which
uses the FreeRTOS for the Renesas RZ (ARM Cortex-A9) family?
… Or you think that is not easily possible to do that for a Newcomer like
me?
I hope that you can help me to put me on the right road.
Best regards,
*Steffen SPRUNGACTIA Automotive*
=================
Software Engineer
=================
*ACTIA* Automotive
5 Rue Jorge SEMPRUN – BP 74215
31432 TOULOUSE Cedex 4 (FRANCE)
Tél.: + 33 (0)5 61 17 68 75
Fax: +33 (0)5 61 55 42 31
Skype: steffen.sprung
EMAIL: steffen.sprung(a)actia.fr
ACTIA: http://www.actia.com
[image: image001]
*P* Avant d’imprimer ce mail, demandez-vous si ceci est nécessaire.
Before printing this email, assess if it is really needed.
Hi,
The systemtap test suite compilation failed with below error.
ARCH: arm
---------------
kernel location:
kernel version: 3.13.0-1-linaro-arndale
systemtap location: /usr/local/bin/stap
systemtap version: version 2.5/0.157, non-git sources
gcc location: /usr/bin/gcc
gcc version: gcc (Ubuntu/Linaro 4.8.1-10ubuntu8) 4.8.1
**** failed systemtap kernel-devel smoke test:
In file included from /usr/local/share/systemtap/runtime/sym.c:16:0,
from /usr/local/share/systemtap/runtime/linux/runtime.h:198,
from /usr/local/share/systemtap/runtime/runtime.h:24,
from /tmp/stapvEzrD5/stap_f7468ebd0051d533d2bae853173fe5a7_892_src.c:24:
/usr/local/share/systemtap/runtime/vma.c: In function '_stp_vma_mmap_cb':
/usr/local/share/systemtap/runtime/vma.c:133:21: error: pointer targets in
initialization differ in signedness [-Werror=pointer-sign]
const char *name = (dentry != NULL) ? dentry->d_name.name : NULL;
^
cc1: all warnings being treated as errors
make[4]: ***
[/tmp/stapvEzrD5/stap_f7468ebd0051d533d2bae853173fe5a7_892_src.o] Error 1
make[3]: *** [_module_/tmp/stapvEzrD5] Error 2
WARNING: kbuild exited with status: 2
Pass 4: compilation failed. [man error::pass4]
**** aborting testing.
Please let me know if you need more information.
Best regards
Naresh kamboju
== Progress ==
* Fixed remaining issues in rework of hwbreakpoint and watchpoint
implementation for arm
native targets. Tested and submitted patch. [TCWG-177] [8/10]
* Installation and Setting up of new desktop machine. [2/10]
== Plan ==
* Run testsuites to see current status of arm gdb.
* Follow up on submitted patches.
* Public holiday on Wednesday 5th February.
* Libsanitizer for AArch64: (4/10)
- seems to be mostly working, but trouble with validation both using
the Foundation Model and qemu-aarch64.
- Some tests seems to loop forever while unwinding under qemu, but
run fine under the Foundation model
- Conversely the Foundation Model shows some random errors, and the
corresponding tests pass under qemu
- Some random timeouts with the Foundation Model, despite using ssh
multiplexing
- Added a small patch to qemu to hande missing mmap flag. Need to
send upstream.
- asked Rob to run GCC testsuite with my patch on AArch64 but it's
too unstable these days
* Cross-validation (2/10)
- script now able to validate a GCC patch and compare results with
unmodified GCC trunk
- added arm-none-eabi target
- using qemu-aarch64 for aarch64-none-linux-gnueabi but the
validations now time out. Need investigation, but I might have to
revert to no execution for this target.
- identified a problem with the recent armv7-ve patch
* Benchmarks: (1/10)
- collecting data for 4.8 tables
* Peeling: (1/10)
- backported new vectorizer cost model to check it if's OK to
include it in our next release
* Misc (2/10): conf-calls, meetings, docs (howtos for TCWG)
== Next ==
* libsanitizer on AArch64
* fix cross-validation on aarch64-none-linux-gnu
* benchmarks: complete table
* backports: chek new vectorizer cost model, monthly branch merge,
review Michael's backports of Crypto intrinsics
* 2 days week (university & child care).
== Issues ==
* none
== Progress ==
* LRA on AArch32:
o TCWG-343 : Make LRA the default for the ARM backend (0/10)
- No progress this week
o TCWG-345 : Analyse performance of LRA for ARM. (1/10)
- Configured and ran benchmarks on Cortex-a15.
* Looked at Linaro 4.8 release performance on Cortex-a15. (2/10)
* Misc. (1/10)
o Booked Hotel and flight for connect
== Next ==
* Back on LRA and lib GMP
== Progress ==
* Investigate "PGO" for Aarch64 (3/10).
Bootstap testing aarch64 with PGO for GCC.Stage 2, feedback profile in progress.
Tested a small test case for -fbranch-probabilities, works same as x86_64.
Working on checking the "gcda" files generated and profile runs for coremark.
* Cbuild2 discussed with rob, ryan on using cbuild2 for developement. (3/10)
Thanks Rob for fixing some issues and adding features that I
requested. Tested some features added. Captured it as FAQ.
* Reinstalled ubuntu OS on my laptop (2/10)
Misc (2/10)
-------------
AMD internal meetings and did some investigations
== Plan ==
- Continue run EMBCC and coremark benchmarks with -PGO enabled.
- Going on marriage vacation from 07-Feb till 06 March
== Issues===
Follow up upstream libssp GCC discussions. No response yet after 2 pings :(