Re: [PATCH] score: Remove support for RTEMS_USE_16_BIT_OBJECT

On 16/11/18 5:08 pm, Sebastian Huber wrote: > The RTEMS_USE_16_BIT_OBJECT define is not set by an RTEMS port. Remove > support for 16-bit object identifiers. If someone really wants to use > RTEMS on a 16-bit target, then it is better to use self-contained > objects instead of playing around with

Re: Spike Fails to Build on CentOS 7 (Possible RSB Issue)

On 17/11/18 1:45 am, Joel Sherrill wrote: > Hi > > Hesham was nice enough to bump the Spike version in the RSB so others have a > chance of using it to test RISC-V executables. Unfortunately, it now doesn't > build for me. I am not sure if this is a  > > ../source-builder/sb-set-builder --log=l-s

Re: [POSIX Benchmark Test] psxtmmqrcvblock01 made (GCI 2018)

Glad to help! On Fri, Nov 16, 2018 at 10:48 PM Joel Sherrill wrote: > > Thanks for double checking the commit and reporting this. I pushed a fix > and gave you credit. :) > > --joel > > On Fri, Nov 16, 2018 at 10:56 AM Shashvat Jain > wrote: > >> Can someone change the status of the test to YES

Re: [POSIX Benchmark Test] psxtmmqrcvblock01 made (GCI 2018)

Thanks for double checking the commit and reporting this. I pushed a fix and gave you credit. :) --joel On Fri, Nov 16, 2018 at 10:56 AM Shashvat Jain wrote: > Can someone change the status of the test to YES in the plan.csv file ,as > it has been commited > > On Fri, Nov 16, 2018 at 8:22 PM Sh

Re: [POSIX Benchmark Test] psxtmmqrcvblock01 made (GCI 2018)

Can someone change the status of the test to YES in the plan.csv file ,as it has been commited On Fri, Nov 16, 2018 at 8:22 PM Shashvat Jain wrote: > ok , that makes sense > > On Fri, Nov 16, 2018 at 8:20 PM Joel Sherrill wrote: > >> I agree we want to reduce the overall number of executables b

Re: Implementation of a new Resource Sharing Protocol

Hello, now i've read through a bit of code. Isn't _Thread_queue_Operations_FIFO respectively _Thread_queue_Do_nothing_priority_actions more suitable for me than _Thread_queue_Operations_priority with priority_actions _Thread_queue_Priority_priority_actions? Best regards Malte On 15.11.18 17:05,

Re: qemu for riscv not in RSB

On Fri, Nov 16, 2018 at 4:37 AM Hesham Almatary wrote: > On Thu, 15 Nov 2018 at 06:05, Sebastian Huber > wrote: > > > > On 14/11/2018 22:07, Joel Sherrill wrote: > > > Hi > > > > > > I was trying to duplicate Sebastian's RISC-V testing for > > > rv64imafd_medany and found that > > > qemu-system-

Re: [POSIX Benchmark Test] psxtmmqrcvblock01 made (GCI 2018)

ok , that makes sense On Fri, Nov 16, 2018 at 8:20 PM Joel Sherrill wrote: > I agree we want to reduce the overall number of executables but the timing > tests which block > need a large set of tasks which do precisely the right thing. You have to > be very careful not to > accidentally measure

Re: or1k tool chain update to use GCC 9

Adding Stafford Horne since he is also likely to know what's up. On Fri, Nov 16, 2018 at 4:45 AM Hesham Almatary wrote: > Hi, > > I built or1k toolchain and RTEMS today. Hello World works fine on > qemu-system-or1k, but not on or1ksim (which hasn't been updated since > 2016). > Ticker on QEMU do

Re: [POSIX Benchmark Test] psxtmmqrcvblock01 made (GCI 2018)

I agree we want to reduce the overall number of executables but the timing tests which block need a large set of tasks which do precisely the right thing. You have to be very careful not to accidentally measure the overhead of _Thread_Handler or any odd initialization. I am happy to see the cases

[PATCH 8/9] bsps/nios2: Use interrupt stack for init stack

Update #3459. --- bsps/nios2/nios2_iss/start/start.S | 3 +++ 1 file changed, 3 insertions(+) diff --git a/bsps/nios2/nios2_iss/start/start.S b/bsps/nios2/nios2_iss/start/start.S index 1ffa4fc925..5caa441330 100644 --- a/bsps/nios2/nios2_iss/start/start.S +++ b/bsps/nios2/nios2_iss/start/start.S

Spike Fails to Build on CentOS 7 (Possible RSB Issue)

Hi Hesham was nice enough to bump the Spike version in the RSB so others have a chance of using it to test RISC-V executables. Unfortunately, it now doesn't build for me. I am not sure if this is a ../source-builder/sb-set-builder --log=l-spike.txt --prefix=/home/joel/rtems-work/tools/5 devel/spi

[PATCH 5/9] bsps/epiphany: Use interrupt stack for init stack

Update #3459. --- bsps/epiphany/epiphany_sim/start/start.S | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/bsps/epiphany/epiphany_sim/start/start.S b/bsps/epiphany/epiphany_sim/start/start.S index 7f828ae151..c79d23dc86 100644 --- a/bsps/epiphany/epiphany_sim/start/start

[PATCH 4/9] bsps/bfin: Use interrupt stack for init stack

Update #3459. --- bsps/bfin/bf537Stamp/start/start.S | 4 ++-- bsps/bfin/shared/start/start.S | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/bsps/bfin/bf537Stamp/start/start.S b/bsps/bfin/bf537Stamp/start/start.S index a835a48764..cc6c7c62fb 100644 --- a/bsps/bfin/bf

[PATCH 9/9] bsps/moxie: Use interrupt stack for init stack

Update #3459. --- bsps/moxie/moxiesim/start/linkcmds | 1 - bsps/moxie/moxiesim/start/start.S | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/bsps/moxie/moxiesim/start/linkcmds b/bsps/moxie/moxiesim/start/linkcmds index 1df8cfb3fe..843512b15d 100644 --- a/bsps/moxie/moxiesi

[PATCH 2/9] bsps/powerpc: Use interrupt stack for init stack

Move start.o to separate file. Update #3459. --- bsps/powerpc/gen5200/start/start.S | 16 - bsps/powerpc/gen83xx/start/start.S | 11 + bsps/powerpc/haleakala/start/linkcmds | 1 + .../powerpc/haleakala/start/{dlentry.S => start.S}

[PATCH 1/9] bsps/m68k: Use interrupt stack for init stack

Update #3459. --- bsps/m68k/av5282/start/start.S | 1 + bsps/m68k/csb360/start/start.S | 1 + bsps/m68k/mcf5206elite/start/start.S | 1 + bsps/m68k/mcf5235/start/start.S | 1 + bsps/m68k/uC5282/start/start.S | 1 + 5 files changed, 5 insertions(+) diff --git a/bsps/m68k/av5

[PATCH 3/9] bsps/mips: Use interrupt stack for init stack

Update #3459. --- bsps/mips/hurricane/start/start.S | 2 ++ bsps/mips/rbtx4925/start/start.S | 2 ++ bsps/mips/rbtx4938/start/start.S | 2 ++ 3 files changed, 6 insertions(+) diff --git a/bsps/mips/hurricane/start/start.S b/bsps/mips/hurricane/start/start.S index e3d97cd167..2a44a26d55 100644

[PATCH 7/9] bsps/sparc64: Use interrupt stack for init stack

Update #3459. --- bsps/sparc64/shared/start/start.S | 11 ++- 1 file changed, 2 insertions(+), 9 deletions(-) diff --git a/bsps/sparc64/shared/start/start.S b/bsps/sparc64/shared/start/start.S index 92c2177c1e..effdb4e842 100644 --- a/bsps/sparc64/shared/start/start.S +++ b/bsps/sparc64/

[PATCH 6/9] bsps/x86_64: Use interrupt stack for init stack

Update #3459. --- bsps/x86_64/amd64/start/linkcmds | 1 + bsps/x86_64/amd64/start/start.S | 41 +++ bsps/x86_64/amd64/start/start.c | 36 --- c/src/lib/libbsp/x86_64/amd64/Makefile.am | 7 -- 4 files changed, 47

Re: [POSIX Benchmark Test] psxtmmqrcvblock01 made (GCI 2018)

I asked that before starting to make the test and Dr. Joel asked me to make a separate one On Fri, Nov 16, 2018 at 12:35 PM Sebastian Huber < sebastian.hu...@embedded-brains.de> wrote: > Do we really need separate test programs for all these test cases? Can't > we just add this to psxtmmq01? >

Re: or1k tool chain update to use GCC 9

Hi, I built or1k toolchain and RTEMS today. Hello World works fine on qemu-system-or1k, but not on or1ksim (which hasn't been updated since 2016). Ticker on QEMU doesn't seem to get the timer interrupts i.e. it just prints out the first 3 tasks and hangs. Please note I haven't built or1k for a wh

Re: qemu for riscv not in RSB

On Thu, 15 Nov 2018 at 06:05, Sebastian Huber wrote: > > On 14/11/2018 22:07, Joel Sherrill wrote: > > Hi > > > > I was trying to duplicate Sebastian's RISC-V testing for > > rv64imafd_medany and found that > > qemu-system-riscv* does not seem to be in our qemu RSB configuration. > > > > Where is