Messages in this thread Patch in this message | | | Subject | NFS bug in 2.2.9-ac? found (was Re: Kernel Panic: 2.2.9-ac1) | From | Trond Myklebust <> | Date | 07 Jun 1999 17:37:30 +0200 |
| |
Theo Van Dinter <felicity@kluge.net> writes:
> On Friday afternoon, I upgraded the news server from 2.2.4-ac? to 2.2.9-ac1. > By Saturday morning, the news server was no more. > > News server HW: Single PII 266 w/ 128Mb RAM. 2x3Com 3x590 NICs. NCR53x875 > based SCSI adapter. > > When I came in on Monday morning, I wrote down the panic info on the console > and tried running a ksymoops on it. Here's what it found: > > Copying default arch from ksymoops, bfd_arch=8 bfd_mach=0 > > >>EIP; c011f153 <__free_page+97/d0> > Trace; c0141ec9 <nfs_updatepage+251/2fc>
Ah. I see what's happened. It looks as though somebody has added the statement 'page_cache_release(page);' to the definition of nfs_unlock_page in nfs_cluster.h in the 2.2.9-ac series. I can't find it in my tree, so I think I can claim innocence on this one...
Could people who've experienced problems with NFS under the 2.2.9-ac series, please apply the following patch.
Cheers, Trond
--- linux-2.2.9-ac3/include/linux/nfs_cluster.h-orig Mon Jun 7 17:29:29 1999 +++ linux-2.2.9-ac3/include/linux/nfs_cluster.h Mon Jun 7 17:33:17 1999 @@ -149,7 +149,6 @@ { clear_bit(PG_locked, &page->flags); wake_up(&page->wait); - page_cache_release(page); } #endif - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |