Most recent kernel where this bug did not occur: Distribution: Hardware Environment: Software Environment: Problem Description: Steps to reproduce:
Reply-To: akpm@linux-foundation.org On Sat, 8 Sep 2007 13:08:37 -0700 (PDT) bugme-daemon@bugzilla.kernel.org wrote: > http://bugzilla.kernel.org/show_bug.cgi?id=8996 > > Summary: atl1 driver cause kernel oops IF ram > 4Gyte and a lot > of network load > Product: Networking > Version: 2.5 > KernelVersion: 2.6.22 > Platform: All > OS/Version: Linux > Tree: Mainline > Status: NEW > Severity: normal > Priority: P1 > Component: Other > AssignedTo: acme@ghostprotocols.net > ReportedBy: micheleschi@gmail.com > > > Most recent kernel where this bug did not occur: > Distribution: > Hardware Environment: > Software Environment: > Problem Description: > > Steps to reproduce: > > (switching to email. Please respond via emailed reply-to-all, not via the bugzilla UI, thanks). We'd really need to see the kernel log output from that oops, please.
On Sat, 8 Sep 2007 17:36:21 -0700 Andrew Morton <akpm@linux-foundation.org> wrote: > On Sat, 8 Sep 2007 13:08:37 -0700 (PDT) > bugme-daemon@bugzilla.kernel.org wrote: > > > http://bugzilla.kernel.org/show_bug.cgi?id=8996 > > > > Summary: atl1 driver cause kernel oops IF ram > 4Gyte > > and a lot of network load The submitter emailed me privately today asking about the atl1 DMA bug. The instant bugzilla is the apparent result of our exchange. I replied to the private inquiry thusly: [quote] A workaround is in place in the -mm kernel tree. http://marc.info/?l=linux-mm-commits&m=118764750020317&w=2 I expect it will be pushed into the 2.6.24 kernel series. Fedora intends to pick up the patch for F7 in the near future. https://bugzilla.redhat.com/show_bug.cgi?id=249511 Even without the patch, you can boot with mem=3900 on the kernel command line and you won't hit the problem. [/quote] In related news, Jeff appears poised to push the patch (with loathing) into netdev.
Patch submitted to -stable today (20071124). http://lkml.org/lkml/2007/11/24/81
2.6.22.15 released today, including the atl1 patch to fix the reported bug. http://lkml.org/lkml/2007/12/14/320
(In reply to Jay Cliburn from comment #4) > 2.6.22.15 released today, including the atl1 patch to fix the reported bug. > > http://lkml.org/lkml/2007/12/14/320 Experienced kernel crashes on a MB with this controller when copying very large files. using this optino in the grub file with 3900M completely resolved the issue. 100% reproducable with very large files (> 30GB). Without the option transferring just one large file from a remote PC crashed the kernel, Whole system was hung up. Only recoverable by power cycling. Ubuntu server LTS 22.04. System has 8GB, Have not tested it yet with an option of 7900 ref: https://answers.launchpad.net/ubuntu/+source/age/+question/701878