Re: Linker Set Based Initialization

On 09/12/15 13:48, Sebastian Huber wrote: On 08/12/15 21:10, Sebastian Huber wrote: - Am 8. Dez 2015 um 16:03 schrieb Joel Sherrill : What BSPs/architectures have you tested? I will test the with the attached patch tomorrow on all BSPs. This patch adds a link time test to ticker.

[PATCH] m32r: Partially remove m32rsim reference

closes #2447. --- dejagnu/boards/rtems-m32r-m32rsim.exp | 40 -- gcc/RTEMS_GCC_Testing.txt | 1 - gcc/do_one | 1 - gcc/rundeja| 6 - gcc/testsuite/ada/acats/Makefile.rtems | 3 --- rtems/

[PATCH] m32r/m32rsim: Remove

updates #2447. --- c/src/lib/libbsp/m32r/acinclude.m4 | 2 - c/src/lib/libbsp/m32r/m32rsim/Makefile.am | 48 c/src/lib/libbsp/m32r/m32rsim/bsp_specs| 13 -- c/src/lib/libbsp/m32r/m32rsim/configure.ac | 20 -- c/src/lib/libbsp/m32r/m32rsim/cons

Re: [PATCH] m32r/m32rsim: Remove

This patch does not apply. The checked out repo needs to be rebaselined. I added instructions on the ticket. The rtems-testing patches include the architecture m32r and are not restricted to just the m32rsim BSP. Those patches need reducing and the remaining ones combined into a single patch. Ins

Re: [PATCH] m32rsim: Remove m32rsim references

I am working on it. Need to find the correct version of the rtems-testing one though. --joel On Wed, Dec 9, 2015 at 5:03 PM, Gedare Bloom wrote: > OK to commit by me. > > On Wed, Dec 9, 2015 at 4:16 PM, Aun-Ali Zaidi wrote: > > closes #2447. > > --- > > dejagnu/boards/rtems-m32r-m32rsim.exp |

Re: [PATCH] m32r/m32rsim: Remove

P.S. this will likely cause conflict for Sebastian's pending changeset. On Wed, Dec 9, 2015 at 6:03 PM, Gedare Bloom wrote: > OK to commit by me. > > On Wed, Dec 9, 2015 at 2:27 PM, Aun-Ali Zaidi wrote: >> updates #2447. >> --- >> c/src/lib/libbsp/m32r/acinclude.m4 | 2 - >> c

Re: [PATCH] m32r/m32rsim: Remove

OK to commit by me. On Wed, Dec 9, 2015 at 2:27 PM, Aun-Ali Zaidi wrote: > updates #2447. > --- > c/src/lib/libbsp/m32r/acinclude.m4 | 2 - > c/src/lib/libbsp/m32r/m32rsim/Makefile.am | 48 > c/src/lib/libbsp/m32r/m32rsim/bsp_specs| 13 -- > c/src/li

Re: [PATCH] m32rsim: Remove m32rsim references

OK to commit by me. On Wed, Dec 9, 2015 at 4:16 PM, Aun-Ali Zaidi wrote: > closes #2447. > --- > dejagnu/boards/rtems-m32r-m32rsim.exp | 40 > --- > rtems/bit_all_bsps| 2 +- > sim-scripts/.gitignore| 2 -- > sim-scripts/Make

[PATCH] m32rsim: Remove m32rsim references

closes #2447. --- dejagnu/boards/rtems-m32r-m32rsim.exp | 40 --- rtems/bit_all_bsps| 2 +- sim-scripts/.gitignore| 2 -- sim-scripts/Makefile | 2 +- sim-scripts/m32rsim.in| 25

[PATCH 05/15] gcc/rundeja: Remove m32r reference

updates #2447. --- gcc/rundeja | 6 -- 1 file changed, 6 deletions(-) diff --git a/gcc/rundeja b/gcc/rundeja index de037e4..d2f6ee7 100755 --- a/gcc/rundeja +++ b/gcc/rundeja @@ -47,12 +47,6 @@ case $1 in DEJABSP=m32csim BSP_SIM=m32csim ;; - m32rsim) -CPU=m32r -BSP=m3

[PATCH 14/15] rtems/bit_all_multilib: Remove m32r reference

updates #2447. --- rtems/bit_all_multilib | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rtems/bit_all_multilib b/rtems/bit_all_multilib index 300ffbf..65aea1e 100755 --- a/rtems/bit_all_multilib +++ b/rtems/bit_all_multilib @@ -4,7 +4,7 @@ # c4x is temporary out. #ALL_C

[PATCH 08/15] gcc/do_one: Remove m32r reference

updates #2447. --- gcc/do_one | 1 - 1 file changed, 1 deletion(-) diff --git a/gcc/do_one b/gcc/do_one index 4b9fc14..f1ffbee 100755 --- a/gcc/do_one +++ b/gcc/do_one @@ -362,7 +362,6 @@ if [ ${needCPU} = yes -a ${needBSP} = yes -a \ i386-pc386|i386-qemu) dejacfg=rtems-i386-qemu ;; lm3

[PATCH 11/15] simple-build-script/build_tools: Remove m32r reference

updates #2447. --- simple-build-script/build_tools | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/simple-build-script/build_tools b/simple-build-script/build_tools index 972c8e6..0afbabd 100755 --- a/simple-build-script/build_tools +++ b/simple-build-script/build_tools @@ -286

[PATCH 12/15] rtems/bit_rtems: Remove m32r reference

updates #2447. --- rtems/bit_rtems | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rtems/bit_rtems b/rtems/bit_rtems index 2e524f7..c749433 100755 --- a/rtems/bit_rtems +++ b/rtems/bit_rtems @@ -328,7 +328,7 @@ fi # Some CPUs do not have C++ support if [ ${ENABLE_RTEMS_CXX} =

[PATCH 09/15] gcc/parallelize_build: Remove m32r reference

updates #2447. --- gcc/parallelize_build | 1 - 1 file changed, 1 deletion(-) diff --git a/gcc/parallelize_build b/gcc/parallelize_build index cf55c21..4652e4e 100755 --- a/gcc/parallelize_build +++ b/gcc/parallelize_build @@ -185,7 +185,6 @@ ${BASEDIR}/rtems-testing/gcc/do_one -B1 -A ${args} h8

[PATCH 07/15] gcc/RTEMS_GCC_Testing.txt: Remove m32r references

updates #2447. --- gcc/RTEMS_GCC_Testing.txt | 1 - 1 file changed, 1 deletion(-) diff --git a/gcc/RTEMS_GCC_Testing.txt b/gcc/RTEMS_GCC_Testing.txt index 8bd886f..1c9fa98 100644 --- a/gcc/RTEMS_GCC_Testing.txt +++ b/gcc/RTEMS_GCC_Testing.txt @@ -133,7 +133,6 @@ this target | i386-rtems*

[PATCH 10/15] gcc/testsuite/ada/acats/Makefile.rtems: Remove m32r references

updates #2447. --- gcc/testsuite/ada/acats/Makefile.rtems | 3 --- 1 file changed, 3 deletions(-) diff --git a/gcc/testsuite/ada/acats/Makefile.rtems b/gcc/testsuite/ada/acats/Makefile.rtems index 58d97a1..8f2b6f6 100644 --- a/gcc/testsuite/ada/acats/Makefile.rtems +++ b/gcc/testsuite/ada/acats/

[PATCH 13/15] rtems/bit_all_bsps: Remove m32r references

updates #2447. --- rtems/bit_all_bsps | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/rtems/bit_all_bsps b/rtems/bit_all_bsps index 5a6e024..06175d8 100755 --- a/rtems/bit_all_bsps +++ b/rtems/bit_all_bsps @@ -94,7 +94,7 @@ if [ $# -ne 0 ] ; then else # leave out nios2 ti

[PATCH 01/15] dejagnu/boards/rtems-m32r-m32rsim.exp: Remove

updates #2447. --- dejagnu/boards/rtems-m32r-m32rsim.exp | 40 --- 1 file changed, 40 deletions(-) delete mode 100755 dejagnu/boards/rtems-m32r-m32rsim.exp diff --git a/dejagnu/boards/rtems-m32r-m32rsim.exp b/dejagnu/boards/rtems-m32r-m32rsim.exp deleted file mod

[PATCH 03/15] sim-scripts/.gitignore: Remove m32r references

updates #2447. --- sim-scripts/.gitignore | 2 -- 1 file changed, 2 deletions(-) diff --git a/sim-scripts/.gitignore b/sim-scripts/.gitignore index 30df64f..924fea3 100644 --- a/sim-scripts/.gitignore +++ b/sim-scripts/.gitignore @@ -42,8 +42,6 @@ lm32_evr-gdb lm3s6965 m32csim m32csim-gdb -m32

[PATCH 06/15] gcc/test_driver: Remove m32r reference

updates #2447. --- gcc/test_driver | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/gcc/test_driver b/gcc/test_driver index f9f8202..d4eafe9 100755 --- a/gcc/test_driver +++ b/gcc/test_driver @@ -116,7 +116,7 @@ do done CPUS_PART1="avr arm h8300 i386 m32c mips sh sparc64"

[PATCH 04/15] sim-scripts/m32rsim.in: Remove

updates #2447. --- sim-scripts/m32rsim.in | 25 - 1 file changed, 25 deletions(-) delete mode 100644 sim-scripts/m32rsim.in diff --git a/sim-scripts/m32rsim.in b/sim-scripts/m32rsim.in deleted file mode 100644 index fe2b7ce..000 --- a/sim-scripts/m32rsim.in +++ /dev/n

[PATCH 02/15] sim-scripts/Makefile: Remove m32r reference

updates #2447. --- sim-scripts/Makefile | 5 + 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/sim-scripts/Makefile b/sim-scripts/Makefile index da47201..e9bf8f9 100644 --- a/sim-scripts/Makefile +++ b/sim-scripts/Makefile @@ -1,6 +1,6 @@ INSTALL_DIR=../bin GDBSIM_SCRIPTS=bf537S

[PATCH] m32r/m32rsim: Remove

updates #2447. --- c/src/lib/libbsp/m32r/acinclude.m4 | 2 - c/src/lib/libbsp/m32r/m32rsim/Makefile.am | 48 c/src/lib/libbsp/m32r/m32rsim/bsp_specs| 13 -- c/src/lib/libbsp/m32r/m32rsim/configure.ac | 20 -- c/src/lib/libbsp/m32r/m32rsim/cons

Re: [PATCH 5/5] bsps: Delete superfluous bsp_pretasking_hook()

Feel free to commit this before we hit more GCI students removing bsps.. On Wed, Dec 9, 2015 at 10:14 AM, Sebastian Huber wrote: > Use the bsp_predriver_hook() instead. > > Update #2408. > --- > c/src/lib/libbsp/arm/altera-cyclone-v/Makefile.am | 1 - > c/src/lib/libbsp/arm/beagle/Makefile.am

Re: [rtems commit] bsps/sh: Disable fileio test due to lack of memory

On 09/12/15 16:01, Sebastian Huber wrote: On 09/12/15 15:58, Joel Sherrill wrote: I don't have a problem with the patch just the idea that fileio used to fit and now it doesn't. Are we going up in size? Is gcc producing worse code? Any ideas? How do other targets look from a size perspective?

[PATCH 1/5] bsps/virtex*: Use default bsp_pretasking_hook()

Update #2408. --- c/src/lib/libbsp/powerpc/virtex4/Makefile.am| 1 + c/src/lib/libbsp/powerpc/virtex4/start/start.S | 2 +- c/src/lib/libbsp/powerpc/virtex4/startup/bspstart.c | 17 - c/src/lib/libbsp/powerpc/virtex5/Makefile.am| 1 + c/src/lib/libbsp/powerp

[PATCH 4/5] bsps/powerpc: Use default bsp_pretasking_hook()

Update #2408. --- c/src/lib/libbsp/powerpc/Makefile.am | 2 +- c/src/lib/libbsp/powerpc/beatnik/Makefile.am | 4 +- .../libbsp/powerpc/motorola_powerpc/Makefile.am| 4 +- c/src/lib/libbsp/powerpc/mvme3100/Makefile.am | 3 +- c/src/lib/libbsp/powerpc/mvme3100/includ

[PATCH 3/5] bsps/bfin: Use default bsp_pretasking_hook()

Rename BSP-specific bsp_pretasking_hook() to bsp_predriver_hook(). Update #2408. --- c/src/lib/libbsp/bfin/TLL6527M/Makefile.am | 2 +- c/src/lib/libbsp/bfin/TLL6527M/startup/bspstart.c | 4 ++-- c/src/lib/libbsp/bfin/bf537Stamp/Makefile.am| 2 +- c/src/lib/libbsp/bfin/bf537Sta

[PATCH 2/5] bsp/beatnik: Delete superfluous declaration

Update #2408. --- c/src/lib/libbsp/powerpc/beatnik/startup/bspstart.c | 6 -- 1 file changed, 6 deletions(-) diff --git a/c/src/lib/libbsp/powerpc/beatnik/startup/bspstart.c b/c/src/lib/libbsp/powerpc/beatnik/startup/bspstart.c index a797451..dc0214a 100644 --- a/c/src/lib/libbsp/powerpc/bea

[PATCH 5/5] bsps: Delete superfluous bsp_pretasking_hook()

Use the bsp_predriver_hook() instead. Update #2408. --- c/src/lib/libbsp/arm/altera-cyclone-v/Makefile.am | 1 - c/src/lib/libbsp/arm/beagle/Makefile.am| 1 - c/src/lib/libbsp/arm/csb336/Makefile.am| 2 +- c/src/lib/libbsp/arm/csb337/Makefile.am| 2 +- c/s

Re: [rtems commit] bsps/sh: Disable fileio test due to lack of memory

On 09/12/15 15:58, Joel Sherrill wrote: I don't have a problem with the patch just the idea that fileio used to fit and now it doesn't. Are we going up in size? Is gcc producing worse code? Any ideas? How do other targets look from a size perspective? I didn't compare the results with 4.11,

Re: [rtems commit] bsps/sh: Disable fileio test due to lack of memory

I don't have a problem with the patch just the idea that fileio used to fit and now it doesn't. Are we going up in size? Is gcc producing worse code? Any ideas? How do other targets look from a size perspective? On Wed, Dec 9, 2015 at 7:23 AM, Sebastian Huber wrote: > Module:rtems > Branch:

Re: Some PowerPC fail to link some bootloader image

On Wed, Dec 9, 2015 at 8:47 AM, Sebastian Huber < sebastian.hu...@embedded-brains.de> wrote: > Hello, > > I have a problem with some older PowerPC BSPs: > > powerpc-rtems4.12/c/mvme2307/testsuites/samples/hello (master) > make > clean all > test -z "*.num *.nxe *.elf *.srec* *.bin *.bt *.ralf" ||

Some PowerPC fail to link some bootloader image

Hello, I have a problem with some older PowerPC BSPs: powerpc-rtems4.12/c/mvme2307/testsuites/samples/hello (master) > make clean all test -z "*.num *.nxe *.elf *.srec* *.bin *.bt *.ralf" || rm -f *.num *.nxe *.elf *.srec* *.bin *.bt *.ralf test -z "hello.exe" || rm -f hello.exe rm -f *.o pow

Re: Linker Set Based Initialization

On 08/12/15 21:10, Sebastian Huber wrote: - Am 8. Dez 2015 um 16:03 schrieb Joel Sherrill : What BSPs/architectures have you tested? I will test the with the attached patch tomorrow on all BSPs. This patch adds a link time test to ticker. This test verifies that the linker comman

Re: [PATCH 2/2] bsps: Call bsp_work_area_initialize() early

ok. Joel, you may need to tweak any slides you have on the boot sequence. ;) On Wed, Dec 9, 2015 at 3:15 AM, Sebastian Huber wrote: > Call bsp_work_area_initialize() before bsp_start(). This allows > bsp_start() to use malloc() etc. which is beneficial for systems with a > plug-and-play hardware

[PATCH 2/2] bsps: Call bsp_work_area_initialize() early

Call bsp_work_area_initialize() before bsp_start(). This allows bsp_start() to use malloc() etc. which is beneficial for systems with a plug-and-play hardware enumeration. Update #2408. --- c/src/lib/libbsp/i386/pc386/include/bsp.h | 2 - .../lib/libbsp/i386/pc386/startup/bspgetworkare

[PATCH 1/2] bsps/sparc: Use default bsp_pretasking_hook()

Move content to bsp_predriver_hook() functions of the BSPs. Update #2408. --- c/src/lib/libbsp/sparc/erc32/Makefile.am | 2 +- .../lib/libbsp/sparc/erc32/startup/bsppredriver.c | 2 ++ c/src/lib/libbsp/sparc/leon2/Makefile.am | 2 +- .../lib/libbsp/sparc/leon2/startup/bspp