* [LTP] [PATCH] lib/tst_test.c: Take account of tst_brk(TCONF)/tst_brk(TFAIL) in summary output @ 2020-05-18 5:43 Xiao Yang 2020-05-19 14:34 ` Cyril Hrubis 0 siblings, 1 reply; 6+ messages in thread From: Xiao Yang @ 2020-05-18 5:43 UTC (permalink / raw) To: ltp Current summary output doesn't take account of tst_brk(TCONF/TFAIL), for example: ----------------------------------------------------- [root@RHEL8U2GA_Intel64 pidfd_send_signal]# ./pidfd_send_signal01 tst_test.c:1246: INFO: Timeout per run is 0h 05m 00s ../../../../include/lapi/pidfd_send_signal.h:16: CONF: syscall(424) __NR_pidfd_send_signal not supported Summary: passed 0 failed 0 skipped 0 warnings 0 ----------------------------------------------------- Signed-off-by: Xiao Yang <yangx.jy@cn.fujitsu.com> --- lib/tst_test.c | 1 + 1 file changed, 1 insertion(+) diff --git a/lib/tst_test.c b/lib/tst_test.c index 0e58060e0..b28521a67 100644 --- a/lib/tst_test.c +++ b/lib/tst_test.c @@ -316,6 +316,7 @@ void tst_vbrk_(const char *file, const int lineno, int ttype, const char *fmt, va_list va) { print_result(file, lineno, ttype, fmt, va); + update_results(TTYPE_RESULT(ttype)); /* * The getpid implementation in some C library versions may cause cloned -- 2.21.0 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* [LTP] [PATCH] lib/tst_test.c: Take account of tst_brk(TCONF)/tst_brk(TFAIL) in summary output 2020-05-18 5:43 [LTP] [PATCH] lib/tst_test.c: Take account of tst_brk(TCONF)/tst_brk(TFAIL) in summary output Xiao Yang @ 2020-05-19 14:34 ` Cyril Hrubis 2020-05-20 2:02 ` Xiao Yang 0 siblings, 1 reply; 6+ messages in thread From: Cyril Hrubis @ 2020-05-19 14:34 UTC (permalink / raw) To: ltp Hi! > lib/tst_test.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/lib/tst_test.c b/lib/tst_test.c > index 0e58060e0..b28521a67 100644 > --- a/lib/tst_test.c > +++ b/lib/tst_test.c > @@ -316,6 +316,7 @@ void tst_vbrk_(const char *file, const int lineno, int ttype, > const char *fmt, va_list va) > { > print_result(file, lineno, ttype, fmt, va); > + update_results(TTYPE_RESULT(ttype)); > > /* > * The getpid implementation in some C library versions may cause cloned Good catch, but I guess that we should also remove the update_result() call from the run_tcases_per_fs() after this. And it also makes sense to call the function as a first thing in the tst_res_/tst_brk_ function, which simplifies the code flow. So I guess that we want something like this (not tested): diff --git a/lib/tst_test.c b/lib/tst_test.c index 0e58060e0..9d0ef672d 100644 --- a/lib/tst_test.c +++ b/lib/tst_test.c @@ -278,8 +278,6 @@ void tst_vres_(const char *file, const int lineno, int ttype, const char *fmt, va_list va) { print_result(file, lineno, ttype, fmt, va); - - update_results(TTYPE_RESULT(ttype)); } void tst_vbrk_(const char *file, const int lineno, int ttype, @@ -297,7 +295,6 @@ static void tst_cvres(const char *file, const int lineno, int ttype, } print_result(file, lineno, ttype, fmt, va); - update_results(TTYPE_RESULT(ttype)); } static void do_test_cleanup(void) @@ -337,6 +334,8 @@ void tst_res_(const char *file, const int lineno, int ttype, { va_list va; + update_results(TTYPE_RESULT(ttype)); + va_start(va, fmt); tst_vres_(file, lineno, ttype, fmt, va); va_end(va); @@ -347,6 +346,8 @@ void tst_brk_(const char *file, const int lineno, int ttype, { va_list va; + update_results(TTYPE_RESULT(ttype)); + va_start(va, fmt); tst_brk_handler(file, lineno, ttype, fmt, va); va_end(va); @@ -1316,10 +1317,8 @@ static int run_tcases_per_fs(void) mntpoint_mounted = 0; } - if (ret == TCONF) { - update_results(ret); + if (ret == TCONF) continue; - } if (ret == 0) continue; -- Cyril Hrubis chrubis@suse.cz ^ permalink raw reply related [flat|nested] 6+ messages in thread
* [LTP] [PATCH] lib/tst_test.c: Take account of tst_brk(TCONF)/tst_brk(TFAIL) in summary output 2020-05-19 14:34 ` Cyril Hrubis @ 2020-05-20 2:02 ` Xiao Yang 2020-05-22 1:40 ` [LTP] [PATCH v2] lib/tst_test.c: Take account of tst_brk(TCONF/TFAIL) " Xiao Yang 0 siblings, 1 reply; 6+ messages in thread From: Xiao Yang @ 2020-05-20 2:02 UTC (permalink / raw) To: ltp On 2020/5/19 22:34, Cyril Hrubis wrote: > Hi! >> lib/tst_test.c | 1 + >> 1 file changed, 1 insertion(+) >> >> diff --git a/lib/tst_test.c b/lib/tst_test.c >> index 0e58060e0..b28521a67 100644 >> --- a/lib/tst_test.c >> +++ b/lib/tst_test.c >> @@ -316,6 +316,7 @@ void tst_vbrk_(const char *file, const int lineno, int ttype, >> const char *fmt, va_list va) >> { >> print_result(file, lineno, ttype, fmt, va); >> + update_results(TTYPE_RESULT(ttype)); >> >> /* >> * The getpid implementation in some C library versions may cause cloned > > Good catch, but I guess that we should also remove the update_result() > call from the run_tcases_per_fs() after this. Hi Cyril, Agreed. I missed redundant update_result() in run_tcases_per_fs() after this change. > > And it also makes sense to call the function as a first thing in the > tst_res_/tst_brk_ function, which simplifies the code flow. It is actually simpler but it changes the original logic of tst_cvres() called by tst_brk(TBROK) in cleanup(). Before change, tst_cvres() changes TBROK to TWARN and then saves TWARN into summary output: ---------------------------------- e.g. Add some debug code in preadv01: [root@Fedora-30 preadv]# ./preadv01 tst_buffers.c:55: INFO: Test is using guarded buffers tst_test.c:1244: INFO: Timeout per run is 0h 05m 00s preadv01.c:80: PASS: Preadv(2) read 64 bytes successfully with content 'a' expectedly preadv01.c:80: PASS: Preadv(2) read 64 bytes successfully with content 'a' expectedly preadv01.c:80: PASS: Preadv(2) read 32 bytes successfully with content 'b' expectedly preadv01.c:99: WARN: test Summary: passed 3 failed 0 skipped 0 warnings 1 ---------------------------------- After change, tst_cvres() changes TBROK to TWARN but doesn't save TWARN into summary output: ---------------------------------- e.g. Add some debug code in preadv01: [root@Fedora-30 preadv]# ./preadv01 tst_buffers.c:55: INFO: Test is using guarded buffers tst_test.c:1245: INFO: Timeout per run is 0h 05m 00s preadv01.c:80: PASS: Preadv(2) read 64 bytes successfully with content 'a' expectedly preadv01.c:80: PASS: Preadv(2) read 64 bytes successfully with content 'a' expectedly preadv01.c:80: PASS: Preadv(2) read 32 bytes successfully with content 'b' expectedly preadv01.c:99: WARN: test Summary: passed 3 failed 0 skipped 0 warnings 0 ---------------------------------- I perfer to add a update_results() in tst_vbrk_(), do you think so? Best Regards, Xiao Yang > > So I guess that we want something like this (not tested): > > diff --git a/lib/tst_test.c b/lib/tst_test.c > index 0e58060e0..9d0ef672d 100644 > --- a/lib/tst_test.c > +++ b/lib/tst_test.c > @@ -278,8 +278,6 @@ void tst_vres_(const char *file, const int lineno, int ttype, > const char *fmt, va_list va) > { > print_result(file, lineno, ttype, fmt, va); > - > - update_results(TTYPE_RESULT(ttype)); > } > > void tst_vbrk_(const char *file, const int lineno, int ttype, > @@ -297,7 +295,6 @@ static void tst_cvres(const char *file, const int lineno, int ttype, > } > > print_result(file, lineno, ttype, fmt, va); > - update_results(TTYPE_RESULT(ttype)); > } > > static void do_test_cleanup(void) > @@ -337,6 +334,8 @@ void tst_res_(const char *file, const int lineno, int ttype, > { > va_list va; > > + update_results(TTYPE_RESULT(ttype)); > + > va_start(va, fmt); > tst_vres_(file, lineno, ttype, fmt, va); > va_end(va); > @@ -347,6 +346,8 @@ void tst_brk_(const char *file, const int lineno, int ttype, > { > va_list va; > > + update_results(TTYPE_RESULT(ttype)); > + > va_start(va, fmt); > tst_brk_handler(file, lineno, ttype, fmt, va); > va_end(va); > @@ -1316,10 +1317,8 @@ static int run_tcases_per_fs(void) > mntpoint_mounted = 0; > } > > - if (ret == TCONF) { > - update_results(ret); > + if (ret == TCONF) > continue; > - } > > if (ret == 0) > continue; > > ^ permalink raw reply [flat|nested] 6+ messages in thread
* [LTP] [PATCH v2] lib/tst_test.c: Take account of tst_brk(TCONF/TFAIL) in summary output 2020-05-20 2:02 ` Xiao Yang @ 2020-05-22 1:40 ` Xiao Yang 2020-05-26 13:43 ` Xiao Yang 0 siblings, 1 reply; 6+ messages in thread From: Xiao Yang @ 2020-05-22 1:40 UTC (permalink / raw) To: ltp Current summary output doesn't take account of tst_brk(TCONF/TFAIL), for example: ----------------------------------------------------- [root@Fedora-30 pidfd_send_signal]# ./pidfd_send_signal01 tst_test.c:1246: INFO: Timeout per run is 0h 05m 00s ../../../../include/lapi/pidfd_send_signal.h:16: CONF: syscall(424) __NR_pidfd_send_signal not supported Summary: passed 0 failed 0 skipped 0 warnings 0 ---------------------------------------------------- 1) Add update_result() in tst_vbrk_() to fix the issue. 2) Remove redundant update_result() in run_tcases_per_fs() after the fix. Signed-off-by: Xiao Yang <yangx.jy@cn.fujitsu.com> --- lib/tst_test.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/lib/tst_test.c b/lib/tst_test.c index 0e58060e0..e93c88ba5 100644 --- a/lib/tst_test.c +++ b/lib/tst_test.c @@ -316,6 +316,7 @@ void tst_vbrk_(const char *file, const int lineno, int ttype, const char *fmt, va_list va) { print_result(file, lineno, ttype, fmt, va); + update_results(TTYPE_RESULT(ttype)); /* * The getpid implementation in some C library versions may cause cloned @@ -1316,10 +1317,8 @@ static int run_tcases_per_fs(void) mntpoint_mounted = 0; } - if (ret == TCONF) { - update_results(ret); + if (ret == TCONF) continue; - } if (ret == 0) continue; -- 2.21.0 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* [LTP] [PATCH v2] lib/tst_test.c: Take account of tst_brk(TCONF/TFAIL) in summary output 2020-05-22 1:40 ` [LTP] [PATCH v2] lib/tst_test.c: Take account of tst_brk(TCONF/TFAIL) " Xiao Yang @ 2020-05-26 13:43 ` Xiao Yang 2020-05-27 7:08 ` Xiao Yang 0 siblings, 1 reply; 6+ messages in thread From: Xiao Yang @ 2020-05-26 13:43 UTC (permalink / raw) To: ltp Hi, I want to push it tomorrow if nobody has any objection. Thanks, Xiao Yang On 2020/5/22 9:40, Xiao Yang wrote: > Current summary output doesn't take account of tst_brk(TCONF/TFAIL), > for example: > ----------------------------------------------------- > [root@Fedora-30 pidfd_send_signal]# ./pidfd_send_signal01 > tst_test.c:1246: INFO: Timeout per run is 0h 05m 00s > ../../../../include/lapi/pidfd_send_signal.h:16: CONF: syscall(424) __NR_pidfd_send_signal not supported > > Summary: > passed 0 > failed 0 > skipped 0 > warnings 0 > ---------------------------------------------------- > > 1) Add update_result() in tst_vbrk_() to fix the issue. > 2) Remove redundant update_result() in run_tcases_per_fs() after the fix. > > Signed-off-by: Xiao Yang<yangx.jy@cn.fujitsu.com> > --- > lib/tst_test.c | 5 ++--- > 1 file changed, 2 insertions(+), 3 deletions(-) > > diff --git a/lib/tst_test.c b/lib/tst_test.c > index 0e58060e0..e93c88ba5 100644 > --- a/lib/tst_test.c > +++ b/lib/tst_test.c > @@ -316,6 +316,7 @@ void tst_vbrk_(const char *file, const int lineno, int ttype, > const char *fmt, va_list va) > { > print_result(file, lineno, ttype, fmt, va); > + update_results(TTYPE_RESULT(ttype)); > > /* > * The getpid implementation in some C library versions may cause cloned > @@ -1316,10 +1317,8 @@ static int run_tcases_per_fs(void) > mntpoint_mounted = 0; > } > > - if (ret == TCONF) { > - update_results(ret); > + if (ret == TCONF) > continue; > - } > > if (ret == 0) > continue; ^ permalink raw reply [flat|nested] 6+ messages in thread
* [LTP] [PATCH v2] lib/tst_test.c: Take account of tst_brk(TCONF/TFAIL) in summary output 2020-05-26 13:43 ` Xiao Yang @ 2020-05-27 7:08 ` Xiao Yang 0 siblings, 0 replies; 6+ messages in thread From: Xiao Yang @ 2020-05-27 7:08 UTC (permalink / raw) To: ltp Hi, Pushed. Best Regards, Xiao Yang On 2020/5/26 21:43, Xiao Yang wrote: > Hi, > > I want to push it tomorrow if nobody has any objection. > > Thanks, > Xiao Yang > On 2020/5/22 9:40, Xiao Yang wrote: >> Current summary output doesn't take account of tst_brk(TCONF/TFAIL), >> for example: >> ----------------------------------------------------- >> [root@Fedora-30 pidfd_send_signal]# ./pidfd_send_signal01 >> tst_test.c:1246: INFO: Timeout per run is 0h 05m 00s >> ../../../../include/lapi/pidfd_send_signal.h:16: CONF: syscall(424) __NR_pidfd_send_signal not supported >> >> Summary: >> passed 0 >> failed 0 >> skipped 0 >> warnings 0 >> ---------------------------------------------------- >> >> 1) Add update_result() in tst_vbrk_() to fix the issue. >> 2) Remove redundant update_result() in run_tcases_per_fs() after the fix. >> >> Signed-off-by: Xiao Yang<yangx.jy@cn.fujitsu.com> >> --- >> lib/tst_test.c | 5 ++--- >> 1 file changed, 2 insertions(+), 3 deletions(-) >> >> diff --git a/lib/tst_test.c b/lib/tst_test.c >> index 0e58060e0..e93c88ba5 100644 >> --- a/lib/tst_test.c >> +++ b/lib/tst_test.c >> @@ -316,6 +316,7 @@ void tst_vbrk_(const char *file, const int lineno, int ttype, >> const char *fmt, va_list va) >> { >> print_result(file, lineno, ttype, fmt, va); >> + update_results(TTYPE_RESULT(ttype)); >> >> /* >> * The getpid implementation in some C library versions may cause cloned >> @@ -1316,10 +1317,8 @@ static int run_tcases_per_fs(void) >> mntpoint_mounted = 0; >> } >> >> - if (ret == TCONF) { >> - update_results(ret); >> + if (ret == TCONF) >> continue; >> - } >> >> if (ret == 0) >> continue; > > . > ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2020-05-27 7:08 UTC | newest] Thread overview: 6+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2020-05-18 5:43 [LTP] [PATCH] lib/tst_test.c: Take account of tst_brk(TCONF)/tst_brk(TFAIL) in summary output Xiao Yang 2020-05-19 14:34 ` Cyril Hrubis 2020-05-20 2:02 ` Xiao Yang 2020-05-22 1:40 ` [LTP] [PATCH v2] lib/tst_test.c: Take account of tst_brk(TCONF/TFAIL) " Xiao Yang 2020-05-26 13:43 ` Xiao Yang 2020-05-27 7:08 ` Xiao Yang
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox