From mboxrd@z Thu Jan 1 00:00:00 1970 From: Cyril Hrubis Date: Fri, 15 Mar 2019 11:18:11 +0100 Subject: [LTP] [PATCH v2 1/3] syscalls/tgkill01: add new test In-Reply-To: References: <1552457573-1354-1-git-send-email-sumit.garg@linaro.org> <1552457573-1354-2-git-send-email-sumit.garg@linaro.org> <20190314122232.GA17823@rei.lan> <20190314135837.GA2536@rei.lan> Message-ID: <20190315101810.GB5383@rei> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ltp@lists.linux.it Hi! > Wouldn't this loop be more appropriate in > "tst_safe_checkpoint_wait()"? As at later stage we may have tests that > depends on checkpoints being interrupted by signals and could directly > use "tst_checkpoint_wait()". The tst_checkpoint_wait() has a single use in the source tree and that is testcases/lib/tst_checkpoint.c which is binary wrapper around checkpoints so that we can use them in shell scripts as well, which is pretty cool btw. And I think that we should retry on EINTR there as well. Also there does not seem to be test relying on checkpoints being interrupted by signals and I would like to avoid this pattern ideally as asynchronous events such as signals interrupting functions is something rather counter intuitive. -- Cyril Hrubis chrubis@suse.cz