Re: [PATCH v2] rtemstoolkit/path : Add listdir function

Pushed with a minor change in the wording (smaller). Thanks. On 20/5/19 9:23 pm, Vijay Kumar Banerjee wrote: > --- > rtemstoolkit/path.py | 29 - > tester/rt/coverage.py | 4 ++-- > 2 files changed, 30 insertions(+), 3 deletions(-) > > diff --git a/rtemstoolkit/path

Re: [PATCH] rtemstoolkit/path : print OSError in mkdir

Pushed. Thank you. On 20/5/19 9:24 pm, Vijay Kumar Banerjee wrote: > --- > rtemstoolkit/path.py | 12 ++-- > 1 file changed, 6 insertions(+), 6 deletions(-) > > diff --git a/rtemstoolkit/path.py b/rtemstoolkit/path.py > index 0044c7b..5cf6157 100644 > --- a/rtemstoolkit/path.py > +++ b/r

Re: SMP Testing of RISC-V

On 20/05/2019 20:23, Joel Sherrill wrote: Hi I can't seem to find the number of cores the RISC-V port has been tested on. I couldn't even find test results for riscv RTEMS in the archives. Info appreciated. I tested with up to two cores and mostly using a 64-bit target. It should work on up

Re: Odd Formatting in CPU Supplement

Hello Joel, it is in line what GCC prints: riscv-rtems5-gcc -print-multi-lib .; rv32i/ilp32;@march=rv32i@mabi=ilp32 rv32im/ilp32;@march=rv32im@mabi=ilp32 rv32imafd/ilp32d;@march=rv32imafd@mabi=ilp32d rv32iac/ilp32;@march=rv32iac@mabi=ilp32 rv32imac/ilp32;@march=rv32imac@mabi=ilp32 rv32imafc/ilp3

[PATCH] sb: Add sb-get-sources to download all references source files.

From: Chris Johns - Downloads all files in a single directory - Iterates over all support hosts to get any host dependent source no matter which host you run the command on. Closes #3532 --- source-builder/sb-get-sources| 27 ++ source-builder/sb/cmd-get-sources.py | 29 ++ source-b

SMP Testing of RISC-V

Hi I can't seem to find the number of cores the RISC-V port has been tested on. I couldn't even find test results for riscv RTEMS in the archives. Info appreciated. I see rtems-tester configurations for spike and sis. Are there any for qemu? I thought qemu worked for RISC-V even if it had to be a

Odd Formatting in CPU Supplement

Hi While paging through some docs, I notice this output for the default multilib bullet: https://docs.rtems.org/branches/master/cpu-supplement/riscv.html The ".:" for the default multilib looks odd enough where I thought it was a formatting error and looked at the source. It is definitely intent

[PATCH] rtemstoolkit/path : print OSError in mkdir

--- rtemstoolkit/path.py | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/rtemstoolkit/path.py b/rtemstoolkit/path.py index 0044c7b..5cf6157 100644 --- a/rtemstoolkit/path.py +++ b/rtemstoolkit/path.py @@ -126,17 +126,17 @@ def mkdir(path): os.maked

[PATCH v2] rtemstoolkit/path : Add listdir function

--- rtemstoolkit/path.py | 29 - tester/rt/coverage.py | 4 ++-- 2 files changed, 30 insertions(+), 3 deletions(-) diff --git a/rtemstoolkit/path.py b/rtemstoolkit/path.py index 2b569ae..0044c7b 100644 --- a/rtemstoolkit/path.py +++ b/rtemstoolkit/path.py @@ -157,6 +

[PATCH 14/18] score: Remove unused SMP_MESSAGE_TEST

All uses were replaced by per-processor jobs. --- cpukit/include/rtems/score/smpimpl.h | 32 +--- cpukit/score/src/smp.c | 2 -- 2 files changed, 1 insertion(+), 33 deletions(-) diff --git a/cpukit/include/rtems/score/smpimpl.h b/cpukit/include/rtems/sc

[PATCH 04/18] score: Add _SMP_Broadcast_action()

--- bsps/arm/shared/clock/clock-a9mpcore.c | 2 +- bsps/arm/shared/clock/clock-generic-timer.c | 2 +- bsps/shared/cache/cacheimpl.h | 4 +-- cpukit/include/rtems/score/smpimpl.h| 24 ++--- cpukit/score/src/smpmulticastaction.c | 12 ++--- testsu

[PATCH 03/18] smptests: Move SMP broadcast action test case

--- testsuites/smptests/smpcache01/init.c | 41 +- testsuites/smptests/smpmulticast01/init.c | 89 +++ 2 files changed, 79 insertions(+), 51 deletions(-) diff --git a/testsuites/smptests/smpcache01/init.c b/testsuites/smptests/smpcache01/init.c index 8

[PATCH 00/18] Rework _SMP_Multicast_action()

The goal of this patch set is to generalize the work done by _SMP_Multicast_action() so that it can be used to replace the SMP_MESSAGE_CLOCK_TICK and SMP_MESSAGE_TEST special cases. Sebastian Huber (18): score: Use dedicated lock for per-CPU jobs smpmulticast01: Use T_TEST_CASE() smptests: M

[PATCH 16/18] score: Remove superfluous include

--- cpukit/include/rtems/score/smpimpl.h | 1 - 1 file changed, 1 deletion(-) diff --git a/cpukit/include/rtems/score/smpimpl.h b/cpukit/include/rtems/score/smpimpl.h index e52a82cf25..5fbe82baad 100644 --- a/cpukit/include/rtems/score/smpimpl.h +++ b/cpukit/include/rtems/score/smpimpl.h @@ -22,

[PATCH 18/18] score: Add _SMP_Unicast_action()

--- cpukit/Makefile.am | 1 + cpukit/include/rtems/score/smpimpl.h | 15 cpukit/score/src/smpunicastaction.c| 51 +++ testsuites/smptests/smpmulticast01/init.c | 95 +++-- .../smptests/smpmulticast

[PATCH 07/18] score: Remove SMP_MESSAGE_CLOCK_TICK

Use _SMP_Multicast_action() instead. --- bsps/i386/pc386/clock/ckinit.c | 10 +- bsps/i386/pc386/start/smp-imps.c | 11 +-- cpukit/include/rtems/score/smpimpl.h | 10 -- 3 files changed, 10 insertions(+), 21 deletions(-) diff --git a/bsps/i386/pc386/clock/ckinit.

[PATCH 13/18] smpipi01: Use per-CPU jobs for IPI flood test

--- testsuites/smptests/smpipi01/init.c | 47 ++--- 1 file changed, 38 insertions(+), 9 deletions(-) diff --git a/testsuites/smptests/smpipi01/init.c b/testsuites/smptests/smpipi01/init.c index aa0f43b55a..18d96dafaf 100644 --- a/testsuites/smptests/smpipi01/init.

[PATCH 05/18] score: Add _SMP_Othercast_action()

--- bsps/arm/shared/clock/clock-a9mpcore.c | 12 +--- bsps/arm/shared/clock/clock-generic-timer.c | 10 -- cpukit/include/rtems/score/smpimpl.h| 15 +++ cpukit/score/src/smpmulticastaction.c | 12 4 files changed, 36 insertions(+), 13 del

[PATCH 11/18] smpipi01: Use per-CPU jobs for pending IPI test

--- testsuites/smptests/smpipi01/init.c | 50 ++--- 1 file changed, 35 insertions(+), 15 deletions(-) diff --git a/testsuites/smptests/smpipi01/init.c b/testsuites/smptests/smpipi01/init.c index 8f42c574a0..c923d049b0 100644 --- a/testsuites/smptests/smpipi01/init

[PATCH 09/18] score: Add _Per_CPU_Add_job()

--- cpukit/include/rtems/score/percpu.h | 15 ++- cpukit/score/src/smpmulticastaction.c | 37 ++ testsuites/smptests/smpmulticast01/init.c | 43 +++ 3 files changed, 78 insertions(+), 17 deletions(-) diff --git a/cpukit/include

[PATCH 15/18] score: Add _SMP_Synchronize()

--- cpukit/include/rtems/score/smpimpl.h | 8 cpukit/score/src/smpmulticastaction.c | 10 ++ testsuites/smptests/smpipi01/init.c | 7 --- 3 files changed, 22 insertions(+), 3 deletions(-) diff --git a/cpukit/include/rtems/score/smpimpl.h b/cpukit/include/rtems/score/smp

[PATCH 06/18] score: Simplify _SMP_Multicast_action()

Move resposibility to disable thread dispatching to the caller of _SMP_Multicast_action(). Using an interrupt disable for this purpose is questionable. --- bsps/shared/cache/cacheimpl.h | 25 +++-- cpukit/include/rtems/score/smpimpl.h | 8 +--- cpukit/sco

[PATCH 12/18] smpipi01: Ensure IPI works on all processors

--- testsuites/smptests/smpipi01/init.c | 29 + 1 file changed, 21 insertions(+), 8 deletions(-) diff --git a/testsuites/smptests/smpipi01/init.c b/testsuites/smptests/smpipi01/init.c index c923d049b0..aa0f43b55a 100644 --- a/testsuites/smptests/smpipi01/init.c +++ b/

[PATCH 02/18] smpmulticast01: Use T_TEST_CASE()

--- testsuites/smptests/smpmulticast01/init.c | 49 +-- 1 file changed, 27 insertions(+), 22 deletions(-) diff --git a/testsuites/smptests/smpmulticast01/init.c b/testsuites/smptests/smpmulticast01/init.c index f8e74e22cc..a2180bdf14 100644 --- a/testsuites/smptests/s

[PATCH 08/18] score: Move per-processor job data structures

This enables re-use for other purposes, e.g. replacement for SMP_MESSAGE_TEST. --- cpukit/include/rtems/score/percpu.h | 59 ++-- cpukit/score/src/smpmulticastaction.c | 87 --- 2 files changed, 74 insertions(+), 72 deletions(-) diff --git a/c

[PATCH 17/18] score: Add _Per_CPU_Wait_for_job()

--- cpukit/include/rtems/score/percpu.h | 14 cpukit/score/src/smpmulticastaction.c | 61 +++ 2 files changed, 48 insertions(+), 27 deletions(-) diff --git a/cpukit/include/rtems/score/percpu.h b/cpukit/include/rtems/score/percpu.h index 85e10fbbaa..c66

[PATCH 01/18] score: Use dedicated lock for per-CPU jobs

--- cpukit/score/src/smpmulticastaction.c | 24 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/cpukit/score/src/smpmulticastaction.c b/cpukit/score/src/smpmulticastaction.c index 0a73bff11f..ea430e8e7c 100644 --- a/cpukit/score/src/smpmulticastaction.c ++

[PATCH 10/18] score: Modify _Per_CPU_Perform_jobs()

Process only the jobs initially registered on the processing list. This makes it possible to add jobs for the current processor in a job handler. These jobs are processed with the next SMP_MESSAGE_PERFORM_JOBS message. The lock is only acquired and released once. --- cpukit/score/src/smpmultica