Date: Tue, 30 Jun 2015 08:54:41 +0300 From: Konstantin Belousov <kostikbel@gmail.com> To: Julian Elischer <julian@freebsd.org> Cc: Ed Schouten <ed@nuxi.nl>, src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: Re: svn commit: r284887 - in head/sys: kern sys ufs/ffs Message-ID: <20150630055441.GY2080@kib.kiev.ua> In-Reply-To: <55920AC2.3060809@freebsd.org> References: <201506270944.t5R9iEjO053344@svn.freebsd.org> <CABh_MKk-48D4FP-BgNLxvVuw4=3_YKnTX-5kSdL7LSAjCVCNOg@mail.gmail.com> <55920AC2.3060809@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Jun 30, 2015 at 11:19:30AM +0800, Julian Elischer wrote:
> On 6/29/15 8:19 PM, Ed Schouten wrote:
> > Hi Kostik,
> >
> > 2015-06-27 11:44 GMT+02:00 Konstantin Belousov <kib@freebsd.org>:
> >> @@ -2056,6 +2072,8 @@ ffs_bufwrite(struct buf *bp)
> >> if (bp->b_vflags & BV_BKGRDINPROG)
> >> panic("bufwrite: still writing");
> >> }
> >> + if ((bp->b_vflags & BV_BKGRDERR) != 0)
> >> + bp->b_vflags &= ~BV_BKGRDERR;
> >> BO_UNLOCK(bp->b_bufobj);
> >>
> >> /*
> > This if-statement could be removed, right? The bit could just be
> > cleared unconditionally.
> true but it replaces a read and write with just a read in the case of
> no change.
Th cache line is dirty already, test before the rwm operation would not
give any optimizations.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20150630055441.GY2080>