April 2023 - linux-morello-ltp - op-lists.linaro.org

by Tudor Cretu

This small series updates the io_uring tests and adds them to the morello_transitional_extended list. This series depends on the following Morello Linux series: "Support io_uring for Purecap apps" v2: - Remove changes to unused io_uring_sqe struct from the lapi header, and add a comment on why it's left unmodified - Remove the io_uring01 from relevant skip file - Update the io_uring dependency tag to "> morello-release-1.6.0" as it missed the boat Review branch: https://git.morello-project.org/tudcre01/morello-linux-ltp/-/commits/morell… Tudor Cretu (2): syscalls/io_uring: Align struct io_uring_sqe with new uABI runtest: Add io_uring tests to the extended PCuABI syscall list include/lapi/io_uring.h | 4 ++++ runtest/morello_transitional_extended | 4 ++++ runtest/syscalls_morello_purecap_skip | 4 ---- testcases/kernel/syscalls/io_uring/io_uring01.c | 4 ++-- testcases/kernel/syscalls/io_uring/io_uring02.c | 6 +++--- 5 files changed, 13 insertions(+), 9 deletions(-) -- 2.34.1

1 year, 3 months

3
4
0 0

[PATCH v2] syscalls/io_cancel01: Setup context before running the test

by Tudor Cretu

The kernel might check for a valid context before checking the access of the pointers. In such a case, it will return EINVAL instead of EFAULT. Add an io_setup step to the test to ensure the io_cancel syscall fails because of the intended cause. Signed-off-by: Tudor Cretu <tudor.cretu(a)arm.com> --- This patch doesn't have any dependency, but it's a dependency to the following Morello Linux series: "Support aio shared memory usage in the Purecap apps" Review branch: https://git.morello-project.org/tudcre01/morello-linux-ltp/-/commits/morell… --- .../kernel/syscalls/io_cancel/io_cancel01.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) diff --git a/testcases/kernel/syscalls/io_cancel/io_cancel01.c b/testcases/kernel/syscalls/io_cancel/io_cancel01.c index f7e8bd061ed7..f31ed07fac93 100644 --- a/testcases/kernel/syscalls/io_cancel/io_cancel01.c +++ b/testcases/kernel/syscalls/io_cancel/io_cancel01.c @@ -18,11 +18,22 @@ #include "tst_test.h" #include "lapi/syscalls.h" -static void run(void) +static aio_context_t ctx; + +static void setup(void) { - aio_context_t ctx; + TST_EXP_PASS_SILENT(tst_syscall(__NR_io_setup, 1, &ctx)); +} - memset(&ctx, 0, sizeof(ctx)); +static void cleanup(void) +{ + if (ctx) + TST_EXP_PASS_SILENT(tst_syscall(__NR_io_destroy, ctx)); +} + + +static void run(void) +{ TST_EXP_FAIL(tst_syscall(__NR_io_cancel, ctx, NULL, NULL), EFAULT); } @@ -31,5 +42,7 @@ static struct tst_test test = { "CONFIG_AIO=y", NULL }, + .setup = setup, + .cleanup = cleanup, .test_all = run, }; -- 2.34.1

1 year, 5 months

4
4
0 0

[PATCH 0/3] Revert setpgid02 hack and cherry-pick upstream

by Teo Couprie Diaz

Commit 41348659371e ("setpgid02: Temporarily disable EPERM test case") disabled a test case that failed in CI due to its configuration. This temporary measure is now suprseded by the fixes upstream. Revert this commit and cherry-pick the relevant upstream commits. These commits will be in the upcoming (in a few weeks) LTP release, so it might not be needed to cherry-pick them now. Teo Couprie Diaz (3): Revert "setpgid02: Temporarily disable EPERM test case" setpgid02: Use pid_max as PGID for EPERM setpgid03: Add test for PGID in different session testcases/kernel/syscalls/setpgid/setpgid02.c | 19 +++++++------------ testcases/kernel/syscalls/setpgid/setpgid03.c | 4 ++++ 2 files changed, 11 insertions(+), 12 deletions(-) -- 2.34.1

1 year, 6 months

3
7
0 0

[PATCH] syscalls/io_cancel01: Setup context before running the test

by Tudor Cretu

The kernel might check for a valid context before checking the access of the pointers. In such a case, it will return EINVAL instead of EFAULT. Add an io_setup step to the test to ensure the io_cancel syscall fails because of the intended cause. Signed-off-by: Tudor Cretu <tudor.cretu(a)arm.com> --- This patch doesn't have any dependency, but it's a dependency to the following Morello Linux series: "Support aio shared memory usage in the Purecap apps" Review branch: https://git.morello-project.org/tudcre01/morello-linux-ltp/-/commits/morell… --- .../kernel/syscalls/io_cancel/io_cancel01.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) diff --git a/testcases/kernel/syscalls/io_cancel/io_cancel01.c b/testcases/kernel/syscalls/io_cancel/io_cancel01.c index f7e8bd061..31670fbb5 100644 --- a/testcases/kernel/syscalls/io_cancel/io_cancel01.c +++ b/testcases/kernel/syscalls/io_cancel/io_cancel01.c @@ -18,11 +18,21 @@ #include "tst_test.h" #include "lapi/syscalls.h" -static void run(void) +static aio_context_t ctx; + +static void setup(void) { - aio_context_t ctx; + TST_EXP_PASS_SILENT(tst_syscall(__NR_io_setup, 1, &ctx)); +} - memset(&ctx, 0, sizeof(ctx)); +static void cleanup(void) +{ + TST_EXP_PASS_SILENT(tst_syscall(__NR_io_destroy, ctx)); +} + + +static void run(void) +{ TST_EXP_FAIL(tst_syscall(__NR_io_cancel, ctx, NULL, NULL), EFAULT); } @@ -31,5 +41,7 @@ static struct tst_test test = { "CONFIG_AIO=y", NULL }, + .setup = setup, + .cleanup = cleanup, .test_all = run, }; -- 2.34.1

1 year, 7 months

2
2
0 0

[PATCH v2] setpgid02: Temporarily disable EPERM test case

by Teo Couprie Diaz

setpgid02 was changed upstream in commit f2797fa44893 ("setpgid02: Convert to new LTP API") and now assumes that LTP runs in a different session from init (PID 1) for the EPERM test case. This breaks in our CI as the busybox rootfs runs sh as PID 1. Skip the EPERM test case for the morello release 1.6.0 while waiting for a fix to be decided with upstream. This mostly doesn't impact test coverage as EPERM is also tested for in setpgid03. Signed-off-by: Teo Couprie Diaz <teo.coupriediaz(a)arm.com> --- testcases/kernel/syscalls/setpgid/setpgid02.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/testcases/kernel/syscalls/setpgid/setpgid02.c b/testcases/kernel/syscalls/setpgid/setpgid02.c index 4b63afee8..4dcb7613d 100644 --- a/testcases/kernel/syscalls/setpgid/setpgid02.c +++ b/testcases/kernel/syscalls/setpgid/setpgid02.c @@ -31,7 +31,12 @@ static struct tcase { } tcases[] = { {&pid, &negative_pid, EINVAL}, {&ppid, &pgid, ESRCH}, - {&pid, &init_pgid, EPERM} + /* + * FIXME: This test case makes an assumption that is invalid in our CI. + * Skip it for now as the error case is covered in setpgid03 anyway. + * Should be restored when the fix is decided with upstream. + */ +/* {&pid, &init_pgid, EPERM} */ }; static void setup(void) -- 2.34.1

1 year, 7 months

1
1
0 0

[PATCH] syscalls/setpgid02: Disable EPERM test case

by Teo Couprie Diaz

setpgid02 was changed upstream in commit f2797fa44893 ("setpgid02: Convert to new LTP API") and now assumes that LTP runs in a different session from init (PID 1) for the EPERM test case. This breaks in our CI as the busybox rootfs runs sh as PID 1. Skip the EPERM test case for the morello release 1.6.0 while waiting for a fix to be decided with upstream. This mostly doesn't impact test coverage as EPERM is also tested for in setpgid03. Signed-off-by: Teo Couprie Diaz <teo.coupriediaz(a)arm.com> --- testcases/kernel/syscalls/setpgid/setpgid02.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/testcases/kernel/syscalls/setpgid/setpgid02.c b/testcases/kernel/syscalls/setpgid/setpgid02.c index 4b63afee8..fe68361aa 100644 --- a/testcases/kernel/syscalls/setpgid/setpgid02.c +++ b/testcases/kernel/syscalls/setpgid/setpgid02.c @@ -31,7 +31,12 @@ static struct tcase { } tcases[] = { {&pid, &negative_pid, EINVAL}, {&ppid, &pgid, ESRCH}, - {&pid, &init_pgid, EPERM} + /* + * This test case makes an assumption that is invalid in our CI. + * Skip it for now as the error case is covered in setpgid03 anyway. + * Should be restored when the fix is decided with upstream. + */ +/* {&pid, &init_pgid, EPERM} */ }; static void setup(void) -- 2.34.1

1 year, 7 months

3
3
0 0

[PATCH] morello_transitional: Skip setpgid02 for Morello 1.6

by Teo Couprie Diaz

setpgid02 was changed upstream in commit f2797fa44893 ("setpgid02: Convert to new LTP API") and now assumes that LTP runs in a different session from init (PID 1). This breaks in our CI as the busybox rootfs runs sh as PID 1. Don't run this test in CI for the morello release 1.6.0 while waiting for a fix to be decided with upstream. Signed-off-by: Teo Couprie Diaz <teo.coupriediaz(a)arm.com> --- runtest/morello_transitional | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/runtest/morello_transitional b/runtest/morello_transitional index d851948..8632615 100644 --- a/runtest/morello_transitional +++ b/runtest/morello_transitional @@ -328,7 +328,10 @@ setgid02 setgid02 setgid03 setgid03 setpgid01 setpgid01 -setpgid02 setpgid02 +# This test has been updated upstream and makes incorrect assumptions breaking +# CI. Skip it for Morello release 1.6.0, but should be run again once a fix is +# decided with upstream. +#setpgid02 setpgid02 setpgid03 setpgid03 setregid01 setregid01 -- 2.34.1

1 year, 7 months

4
4
0 0

[PATCH] syscalls_morello_skip: Skip setpgid02 for morello 1.6

by Teo Couprie Diaz

setpgid02 was changed upstream in commit f2797fa44893 ("setpgid02: Convert to new LTP API") and now assumes that LTP runs in a different session from init (PID 1). This breaks in our CI as the busybox rootfs runs sh as PID 1. Skip this test for the morello release 1.6.0 while waiting for a fix to be decided with upstream. Signed-off-by: Teo Couprie Diaz <teo.coupriediaz(a)arm.com> --- runtest/syscalls_morello_skip | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/runtest/syscalls_morello_skip b/runtest/syscalls_morello_skip index 1eb10f6..80a4f47 100644 --- a/runtest/syscalls_morello_skip +++ b/runtest/syscalls_morello_skip @@ -2,6 +2,11 @@ # because of either the build setup for Morello or the system the tests run on. # KERN - *: depends on Morello Linux kernel release >= morello-release-1.6.0 +# setpgid02 assumes incorrectly that LTP will always run in a different session +# from init (PID 1). This breaks in the kernel CI. +# This is to be removed when a fix has been implemented. +setpgid02 + # leapsec01 is skipped as systemd-timesyncd interfers with it and it doesn't # test any worthwhile functionality. leapsec01 -- 2.34.1

1 year, 7 months

1
1
0 0

[PATCH] runtest: Update lsit of testcases for Morello

by Beata Michalska

Drop getitimer03/setitimer03 from the list as those are no longer available as per upstream commits: d4e993089819 "getitimer03.c: convert to new LTP API" and da2f0a6513bb "setitimer03: convert to new API" Signed-off-by: Beata Michalska <beata.michalska(a)arm.com> --- runtest/morello_transitional | 2 -- 1 file changed, 2 deletions(-) diff --git a/runtest/morello_transitional b/runtest/morello_transitional index 458e6287c..d851948be 100644 --- a/runtest/morello_transitional +++ b/runtest/morello_transitional @@ -131,7 +131,6 @@ gethostid01 gethostid01 getitimer01 getitimer01 getitimer02 getitimer02 -getitimer03 getitimer03 getpgid01 getpgid01 getpgid02 getpgid02 @@ -366,7 +365,6 @@ set_tid_address01 set_tid_address01 setitimer01 setitimer01 setitimer02 setitimer02 -setitimer03 setitimer03 sigaltstack01 sigaltstack01 sigaltstack02 sigaltstack02 -- 2.25.1

1 year, 7 months

2
2
0 0

[PATCH 0/2] Add io_uring support

by Tudor Cretu

This small series updates the io_uring tests and adds them to the morello_transitional_extended list. This series depends on the following Morello Linux series: "Support io_uring for Purecap apps" Review branch: https://git.morello-project.org/tudcre01/morello-linux-ltp/-/commits/morell… Tudor Cretu (2): syscalls/io_uring: Update uAPI structs runtest: Add io_uring tests to extended PCuABI syscall list include/lapi/io_uring.h | 10 +++++----- runtest/morello_transitional_extended | 4 ++++ testcases/kernel/syscalls/io_uring/io_uring01.c | 4 ++-- testcases/kernel/syscalls/io_uring/io_uring02.c | 6 +++--- 4 files changed, 14 insertions(+), 10 deletions(-) -- 2.34.1

1 year, 7 months

4
14
0 0

2024

2023

2022

linux-morello-ltp April 2023