From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Fri, 3 Feb 2017 07:53:26 -0800 From: Matthew Wilcox To: Jan Kara Cc: Al Viro , linux-fsdevel@vger.kernel.org, linux-mm@kvack.org, Andrew Morton , lustre-devel@lists.lustre.org, cluster-devel@redhat.com Subject: Re: [PATCH] mm: Avoid returning VM_FAULT_RETRY from ->page_mkwrite handlers Message-ID: <20170203155326.GE2267@bombadil.infradead.org> References: <20170203150729.15863-1-jack@suse.cz> <20170203151356.GB2267@bombadil.infradead.org> <20170203154640.GA25586@quack2.suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170203154640.GA25586@quack2.suse.cz> Sender: owner-linux-mm@kvack.org List-ID: On Fri, Feb 03, 2017 at 04:46:40PM +0100, Jan Kara wrote: > On Fri 03-02-17 07:13:59, Matthew Wilcox wrote: > > On Fri, Feb 03, 2017 at 04:07:29PM +0100, Jan Kara wrote: > > > Some ->page_mkwrite handlers may return VM_FAULT_RETRY as its return > > > code (GFS2 or Lustre can definitely do this). However VM_FAULT_RETRY > > > from ->page_mkwrite is completely unhandled by the mm code and results > > > in locking and writeably mapping the page which definitely is not what > > > the caller wanted. Fix Lustre and block_page_mkwrite_ret() used by other > > > filesystems (notably GFS2) to return VM_FAULT_NOPAGE instead which > > > results in bailing out from the fault code, the CPU then retries the > > > access, and we fault again effectively doing what the handler wanted. > > > > Reading this commit message makes me wonder if this is the best fix. > > It would seem logical that if I want the fault to be retried that I should > > return VM_FAULT_RETRY, not VM_FAULT_NOPAGE. Why don't we have the MM > > treat VM_FAULT_RETRY the same way that it treats VM_FAULT_NOPAGE and give > > driver / filesystem writers one fewer way to shoot themselves in the foot? > > VM_FAULT_RETRY is special, it may be used only if FAULT_FLAG_ALLOW_RETRY > was set in page fault flags and it means - we have dropped mmap_sem, we > loaded page needed to satisfy the fault and now we need to try again (have > a look at __lock_page_or_retry()). I have my reservations about this > interface but it works... Oh, I understand what it's *supposed* to be used for ;-) It's just a bit of an attractive nuisance. Maybe renaming it to something like VM_FAULT_PAGE_RETRY would stop people from thinking that it meant "retry the fault". And we could #define VM_FAULT_RETRY VM_FAULT_NOPAGE so that people who want to retry the fault in a normal way could use a return value that sounds like it does what they want instead of a return value that is supposed to be used to indicate that we put a PFN into the page table? -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org