From mboxrd@z Thu Jan 1 00:00:00 1970 From: Cyril Hrubis Date: Tue, 6 Nov 2018 16:54:53 +0100 Subject: [LTP] [PATCH 5/5] syscalls/posix_fadvise0[13]: Start using new library. In-Reply-To: References: <20181105235019.254846-1-sspatil@google.com> <20181105235019.254846-6-sspatil@google.com> Message-ID: <20181106155452.GD3014@rei.lan> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ltp@lists.linux.it Hi! > The only thing that should matter in the context of this test is > whether or not posix_fadvise() is supported by the kernel. > For some reason that is not clear to me (are we tinyfying the kernel syscall > by syscall now?), the support for fadvise in the kernel can be configured out > with CONFIG_ADVISE_SYSCALLS=n. So IMO, this test really needs a > runtime check, not a build time check like above. Yes, I would expect that we would get ENOSYS in a case that kernel is build without the support for the call. And we don't have to check in the setup, we can do tst_brk(TCONF, ...) in the actuall test as well. -- Cyril Hrubis chrubis@suse.cz