Date: Thu, 8 Jul 2010 14:53:55 -0700 From: Jack Vogel <jfvogel@gmail.com> To: Chris Buechler <cbuechler@gmail.com> Cc: net@freebsd.org Subject: Re: page fault in igb on 8.1 Message-ID: <AANLkTilkLdIt7Z8JMK_zDyiPYxjhTKEuuDBHpRrcavQe@mail.gmail.com> In-Reply-To: <AANLkTilRMW4DTpxS_4n_37QGfdiUdJL_UTCgcq3un_mK@mail.gmail.com> References: <4C3631F4.4090504@gmail.com> <AANLkTinc2xB3CAFKZLfVd5wSss6wBQe-qwqVOih--4MN@mail.gmail.com> <AANLkTilRMW4DTpxS_4n_37QGfdiUdJL_UTCgcq3un_mK@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
There is one case in igb_dma_malloc() where its still IGB_DBA_ALIGN, but the cases where its allocating TX and RX bufs the alignment is 1, that code is the same in 8.1, or head as far as I can see. Jack On Thu, Jul 8, 2010 at 2:36 PM, Chris Buechler <cbuechler@gmail.com> wrote: > On Thu, Jul 8, 2010 at 5:03 PM, Jack Vogel <jfvogel@gmail.com> wrote: > > Am confused, does not the code in 8.1 have that fix, looked to me like it > > did? > > > > Maybe I'm confused, but it looks like RELENG_8_1 has the IGB_DBA_ALIGN > rather than 1. > > - IGB_DBA_ALIGN, 0, /* alignment, bounds */ > + 1, 0, /* alignment, bounds */ > > > http://svn.freebsd.org/viewvc/base/releng/8.1/sys/dev/e1000/if_igb.c?revision=209145&view=markup > > It was reverted back in this commit. > > http://svn.freebsd.org/viewvc/base/releng/8.1/sys/dev/e1000/if_igb.c?r1=205283&r2=206211 > > Chris >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?AANLkTilkLdIt7Z8JMK_zDyiPYxjhTKEuuDBHpRrcavQe>