Skip to contentSkip to conversation

pty04.c/pidfd_send_signal03.c: Drop redundant tst_reap_children()

PATCH luoxiaoyu (C) via ltp 2 participants 2 messages Sep 13, 2022

lore
Share & tools
lore
#0 lore raw

Maybe I have a question to ask for.
As my understand, tst_reap_children() has been called in API even test hit tst_brk, right?

Regards,
Luo
-----origin mail-----
Re: [LTP] [PATCH] pty04.c/pidfd_send_signal03.c: Drop redundant tst_reap_children()

Quoted 7 lines · Petr Vorel
Quoted 5 lines

Hi Luo,

Quoted 2 lines

Drop redundant calling of tst_reap_children() in cleanup() since it has been called in the API (run_tests()).

Quoted 8 lines
Quoted 1 line

LGTM, I suppose that was a mistake.

Hmm, I slightly think that reflact a cautious thought from the author, in case of test hit tst_brk unexpectly with invoke cleanup it can also do reap children.

So, it's probably no harmful to keep tst_reap_children there.

Ah, thx for info. I set status rejected in patchwork.

Kind regards,
Petr

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
#1 lore raw
Quoted 4 lines · luoxiaoyu (C)

Maybe I have a question to ask for.
As my understand, tst_reap_children() has been called in API even test hit tst_brk, right?

To precisely, when hitting a tst_brk, the main process will invoke do_test_cleanup()
and finally call tst_test->cleanup function from the specific testcase, to recover the
original environment as much as they can.

In pidfd_send_signal03, the tst_reap_children will be invoked in that break situation.
But we shouldn't say "it has been called in API", they're totally different concepts.

see:
https://github.com/linux-test-project/ltp/blob/master/lib/tst_test.c#L339

Quoted 28 lines

Regards,
Luo
-----origin mail-----
Re: [LTP] [PATCH] pty04.c/pidfd_send_signal03.c: Drop redundant tst_reap_children()

Quoted 7 lines · Petr Vorel
Quoted 5 lines

Hi Luo,

Quoted 2 lines

Drop redundant calling of tst_reap_children() in cleanup() since it has been called in the API (run_tests()).

Quoted 8 lines
Quoted 1 line

LGTM, I suppose that was a mistake.

Hmm, I slightly think that reflact a cautious thought from the author, in case of test hit tst_brk unexpectly with invoke cleanup it can also do reap children.

So, it's probably no harmful to keep tst_reap_children there.

Ah, thx for info. I set status rejected in patchwork.

Kind regards,
Petr

Signature · 2 lines
Regards,
Li Wang
Signature · 2 lines
-- 
Mailing list info: https://lists.linux.it/listinfo/ltp