From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Stancek Date: Wed, 12 Sep 2018 03:04:53 -0400 (EDT) Subject: [LTP] [PATCH v2] syscalls/execveat01: new test to verify execveat unlinked fd In-Reply-To: References: <1533111474.1176.4.camel@mtkswgap22> <1149740228.46962437.1536679477401.JavaMail.zimbra@redhat.com> Message-ID: <500978530.47285642.1536735893652.JavaMail.zimbra@redhat.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit To: ltp@lists.linux.it ----- Original Message ----- > Jan Stancek 於 2018年9月11日 週二 下午11:24寫道： > > We should fix this for release. If EINVAL is expected behaviour, > > test should give PASS. > > > > It's failing with 4.18: > > > > tst_test.c:1063: INFO: Timeout per run is 0h 05m 00s > > execveat03.c:71: FAIL: execveat() returned unexpected errno: EINVAL > > > > Hi Jan, > The fix of original regression, ie. EINVAL returns in execveat03 test case, > has been added to 4.18-stable and 4.14-stable: > https://www.spinics.net/lists/stable/msg255906.html > https://www.spinics.net/lists/stable/msg255724.html > If give EINVAL as PASS, the case will PASS with or without the regression > fix. I'm not sure which behavior of execveat03 is more appropriate. > Please advise. Hi Eddie, thanks for the links. I think I confused myself with the comment in testcase. I thought EINVAL is expected behavior with the fix. For now, I only added comment with commit id of the fix to testcase. Thanks, Jan > > Thanks, > Eddie >