On Tue, Jan 22, 2013 at 11:05 AM, Shuah Khan <shuahk...@gmail.com> wrote: > On Tue, Jan 22, 2013 at 9:36 AM, Deucher, Alexander > <alexander.deuc...@amd.com> wrote: >>> -----Original Message----- >>> From: Shuah Khan [mailto:shuahk...@gmail.com] >>> Sent: Tuesday, January 22, 2013 11:15 AM >>> To: Deucher, Alexander >>> Cc: Linus Torvalds; Linux Kernel Mailing List >>> Subject: Re: Linux 3.8-rc4 >>> >>> On Sat, Jan 19, 2013 at 9:44 AM, Deucher, Alexander >>> <alexander.deuc...@amd.com> wrote: >>> >> -----Original Message----- >>> >> From: Shuah Khan [mailto:shuahk...@gmail.com] >>> >> Sent: Friday, January 18, 2013 7:40 PM >>> >> To: Linus Torvalds; Deucher, Alexander >>> >> Cc: Linux Kernel Mailing List >>> >> Subject: Re: Linux 3.8-rc4 >>> >> >>> >> On Fri, Jan 18, 2013 at 3:37 PM, Shuah Khan <shuahk...@gmail.com> >>> wrote: >>> >> > On Fri, Jan 18, 2013 at 10:51 AM, Shuah Khan <shuahk...@gmail.com> >>> >> wrote: >>> >> >>> >> > >>> >> > ok. I bisected the DMAR faults and it points to the following commit: >>> >> > >>> >> > 909d9eb67f1e4e39f2ea88e96bde03d560cde3eb >>> >> > >>> >> > This commit as I recall, fixed the crash problem I was seeing back in >>> >> > 3.8-rc1. Reverting to see if crash problem reappears. >>> >> >>> >> Confirming that with this commit reverted crash problem re-appeared. >>> >> With this commit, DMAR faults show up. >>> > >>> > I'm not quite sure what's going on with your system. At this point it's >>> probably best to just disable the DMA ring on these cards until we sort out >>> what's going on. The attached patch disables the use of the DMA ring. Let >>> me know if it fixes the issues you are seeing. >>> > >>> >>> I applied this patch to 3.8-rc4 and didn't fix the DMAR faults. >> >> Are you getting continuous DMAR faults or just when while the module is >> being loaded? >> >> Alex > > During module initialization, I think might be actually right after > RV620 Microcode loading attempt: dmesg excerpts starting from radeoan
Here is the log from 3.7.4-rc4 without this problem. These faults definitely are showing up during Microcode loading process if I am reading the dmesg correctly. [ 20.560517] [drm] Loading RV620 Microcode [ 20.639618] [drm] PCIE GART of 512M enabled (table at 0x0000000000040000). [ 20.639670] radeon 0000:01:00.0: WB enabled [ 20.639674] radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000008000c00 and cpu addr 0xffff88006d5b3c00 [ 20.672797] [drm] ring test on 0 succeeded in 1 usecs [ 20.674597] [drm] ib test on ring 0 succeeded in 0 usecs [ 20.675083] [drm] radeon atom DIG backlight initialized [ 20.675085] [drm] Radeon Display Connectors Attaching compressed dmesg from 3.7.4-rc1 boot on this system -- Shuah
dmesg_3.7.4-rc1_radeon.gz
Description: GNU Zip compressed data
