From mboxrd@z Thu Jan 1 00:00:00 1970 From: Cyril Hrubis Date: Fri, 8 Dec 2017 11:43:02 +0100 Subject: [LTP] [PATCH v3 09/12] syscalls/mount03: Copy setuid_test to execute instead of 'TEST FILE' In-Reply-To: <87609hwlkr.fsf@e105922-lin.cambridge.arm.com> References: <20171204163655.8838-1-punit.agrawal@arm.com> <20171204163655.8838-10-punit.agrawal@arm.com> <20171208102658.GA13286@rei> <87609hwlkr.fsf@e105922-lin.cambridge.arm.com> Message-ID: <20171208104302.GD13286@rei> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ltp@lists.linux.it Hi! > >> case 5: > >> /* Validate MS_NOSUID flag of mount call */ > >> > >> - snprintf(file, PATH_MAX, "%ssetuid_test", path_name); > >> - SAFE_FILE_PRINTF(cleanup, file, "TEST FILE"); > >> + TST_RESOURCE_COPY(cleanup, "mount03_setuid_test", path_name); > >> + snprintf(file, PATH_MAX, "%smount03_setuid_test", path_name); > >> > >> SAFE_STAT(cleanup, file, &file_stat); > >> > >> @@ -291,9 +291,8 @@ int test_rwflag(int i, int cnt) > >> /* reset the setup_uid */ > >> if (status) > >> return 0; > >> - else > >> - return 1; > >> } > >> + return 1; > >> } > >> case 6: > >> /* Validate MS_NOATIME flag of mount call */ > >> @@ -373,6 +372,10 @@ static void setup(void) > >> strncpy(path, path_name, PATH_MAX); > >> snprintf(path_name, PATH_MAX, "%s/%s/", path, mntpoint); > >> > >> + SAFE_MOUNT(cleanup, device, mntpoint, fs_type, 0, NULL); > >> + TST_RESOURCE_COPY(cleanup, "mount03_setuid_test", path_name); > >> + SAFE_UMOUNT(cleanup, mntpoint); > > > > I'm, a bit confused now, do we copy the file here in the setup or in the > > actual test above? Surely we don't have to do both. > > In an earlier email you said - > > "... secondly you are supposed to use the TST_RESOURCE_COPY() macro > instead of the function itself and it should be done once in the test > setup as well as the code that sets the actual setuid bit." > > And now I'm confused as I updated the patch as per your request (or so I > thought). Where's the disconnect? As far as I can see, we do TST_RESOURCE_COPY() twice in this patch, once in the test setup and once in the actual test, which is what confuses me. -- Cyril Hrubis chrubis@suse.cz