Date: Sat, 21 Jan 2017 22:36:15 +0100 From: Mateusz Guzik <mjguzik@gmail.com> To: Konstantin Belousov <kostikbel@gmail.com> Cc: Mateusz Guzik <mjg@FreeBSD.org>, src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: Re: svn commit: r312600 - head/sys/kern Message-ID: <20170121213614.GB20930@dft-labs.eu> In-Reply-To: <20170121195114.GA2349@kib.kiev.ua> References: <201701211838.v0LIcHIv072626@repo.freebsd.org> <20170121195114.GA2349@kib.kiev.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Jan 21, 2017 at 09:51:14PM +0200, Konstantin Belousov wrote:
> On Sat, Jan 21, 2017 at 06:38:17PM +0000, Mateusz Guzik wrote:
> > + if (flags & LK_RETRY) {
> Stylish test is
> if ((flags & LK_RETRY) != 0) {
> > + if ((error != 0))
> Too many ().
>
> > + goto retry;
> > + if ((vp->v_iflag & VI_DOOMED)) {
> Too many braces again, or missed != 0.
This was a side effect of putting these into __predict_false and then
removing it, albeit not fully.
> > VOP_UNLOCK(vp, 0);
> > error = ENOENT;
> > - break;
> Also, this does the functional change, it seems to completely break LK_RERY
> logic. If LK_RETRY is specified, VI_DOOMED must not result in ENOENT.
>
Yea, this was a weird-ass brainfart on my part. Fixed in r312606
The gist of the actual change I wanted to make is postponing the read
from the vnode.
> > }
> > - } while (flags & LK_RETRY && error != 0);
> > + }
> > return (error);
> > }
> >
--
Mateusz Guzik <mjguzik gmail.com>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20170121213614.GB20930>