From: Cyril Hrubis Date: Thu, 30 May 2019 17:27:37 +0200 Subject: [LTP] [PATCH 2/2] syscalls: update tcases using dev_fs_type to use some_filesystems In-Reply-To: <20190530150500.j2wkehkf5fopt2nw@XZHOUW.usersys.redhat.com> References: <20190530102336.10898-1-xzhou@redhat.com> <20190530102336.10898-2-xzhou@redhat.com> <20190530111030.GC31468@rei.lan> <20190530150500.j2wkehkf5fopt2nw@XZHOUW.usersys.redhat.com> Message-ID: <20190530152736.GB25602@rei> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ltp@lists.linux.it Hi! > > I guess that we can enable the this test for all filesystems, there is > > in-kernel fallback for filesystems that does not support the syscall so > > it should work on all filesystems and the fallback code should be tested > > as well. > > > > > testcases/kernel/syscalls/getxattr/getxattr04.c | 7 +++++-- > > > > Well I would be slightly in favor of enabling this test for all > > filesystems as well. > > > > If we ever want to maintain list of filesystems that support extended > > attributes there should be only one somewhere in the test library so > > that we don't have to change thousand testcases later on. > > Hmm.. So we actually don't need this array. Just enable some proper > tests for all filesystems. Looks like that, I've checked the fgetxattr01.c test and we only skip fat and vfat when the test is executed, all the other filesystems do support extended attributes. Disabling fat and vfat does not seem to speed up the test at all, I guess that the mkfs + mount is fast enough so that the difference is not measureable. -- Cyril Hrubis chrubis@suse.cz