From mboxrd@z Thu Jan 1 00:00:00 1970 From: Cyril Hrubis Date: Thu, 13 Jul 2023 17:12:15 +0200 Subject: [Cluster-devel] [LTP] [linus:master] [iomap] 219580eea1: ltp.writev07.fail In-Reply-To: <202307132107.2ce4ea2f-oliver.sang@intel.com> References: <202307132107.2ce4ea2f-oliver.sang@intel.com> Message-ID: List-Id: To: cluster-devel.redhat.com MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hi! The test description: Verify writev() behaviour with partially valid iovec list. Kernel <4.8 used to shorten write up to first bad invalid iovec. Starting with 4.8, a writev with short data (under page size) is likely to get shorten to 0 bytes and return EFAULT. This test doesn't make assumptions how much will write get shortened. It only tests that file content/offset after syscall corresponds to return value of writev(). See: [RFC] writev() semantics with invalid iovec in the middle https://marc.info/?l=linux-kernel&m=147388891614289&w=2 -- Cyril Hrubis chrubis@suse.cz