From mboxrd@z Thu Jan 1 00:00:00 1970 From: Adrian McMenamin Date: Mon, 16 Mar 2009 22:58:51 +0000 Subject: Re: [PATCH] sh: Move IRQ multi definition of DMAC to defconfig Message-Id: <1237244331.6636.10.camel@localhost.localdomain> List-Id: References: <49BDC5DF.6070108@renesas.com> In-Reply-To: <49BDC5DF.6070108@renesas.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To: linux-sh@vger.kernel.org On Mon, 2009-03-16 at 22:49 +0000, Adrian McMenamin wrote: >=20 > Have just tested the current kernel in your (Paul's) git repo and > although it doesn't (seem) to be suffering from the DMA issue now, it is > still not booting - oom problems killing init: >=20 .... > =EF=BF=BDStarting the hotplug events dispatcher: udevd. > Synthesizing the initial hotplug events...udevd-event[954]: rename_netif:= error changing netif name eth0 to eth1: Device or resource busy > done. > [ 29.005585] udevd invoked oom-killer: gfp_mask=3D0x1201d2, order=3D0, = oomkilladj=3D-17 > [ 29.013003] Stack: (0x8cd51ce4 to 0x8cd52000) > [ 29.017975] 1ce0: 8c0071ae 8cd51cf4 00000000 001201d2 8c05360= 6 8cd51cfc 00000041=20 > [ 29.026904] 1d00: 8cd7c360 8c053c00 8cd51d2c 000000bc 8cd7c360 0000004= 1 00000124 8c2e4198=20 On the next three reboots the box cleanly booted twice and then failed a third time - slightly differently - will now test with DMA off