Date: Sat, 30 Sep 2017 14:53:13 -0400 From: Anton Yuzhaninov <citrin+bsd@citrin.ru> To: freebsd-fs@freebsd.org Subject: Re: Panic: vm_fault: fault on nofault entry Message-ID: <1eaf07ba-1928-a162-16fe-1908d350ed46@citrin.ru> In-Reply-To: <20170930081835.GH95911@kib.kiev.ua> References: <35aef1ac-a1dd-4526-46f8-599b4fb3b0f4@citrin.ru> <20170930081835.GH95911@kib.kiev.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
On 09/30/17 04:18, Konstantin Belousov wrote: >> I have panics when I try to use nginx on my notebook. When sendfile is >> enabled it is easy to trigger panic, after disabling sendfile I've got >> no panics so far and I think that panic is triggered by using sendfile. >> >> Root filesystem - UFS on GELI whith 8k sectorsize. >> >> GELI with sectorsize large than page size is non-standard configuration, >> and I've got problem because of this in the past: >> https://lists.freebsd.org/pipermail/freebsd-fs/2016-October/023957.html >> >> but it was fixed in: >> https://svnweb.freebsd.org/changeset/base/307626 > Well, it was fixed by replacing the code path for normal page in by > the less buggy pager. The sendfile path was not touched, it still uses > the generic pager with all its bugs sitting in place. > > There is no way around it, you cannot use block size > page size for > sendfile, and there are more issues, e.g. with the simultaneous file > truncation. > Thanks for explaining. On a desktop I can easily avoid using sendfile(2). Hard part for me was to find out that panic was triggered by sendfile (and to something else).
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1eaf07ba-1928-a162-16fe-1908d350ed46>