2.4.3pre1: kernel BUG at page_alloc.c:73!
Mar 10 21:34:30 prototype kernel: kernel BUG at page_alloc.c:73! Mar 10 21:34:30 prototype kernel: invalid operand: 0000 Mar 10 21:34:30 prototype kernel: CPU: 0 Mar 10 21:34:30 prototype kernel: EIP: 0010:[__free_pages_ok+34/784] Mar 10 21:34:30 prototype kernel: EFLAGS: 00013082 Mar 10 21:34:30 prototype kernel: eax: 0000001f ebx: c14285d0 ecx: c3642000 edx: c7839380 Mar 10 21:34:30 prototype kernel: esi: cfa70000 edi: 0000001f ebp: 00000003 esp: c3643c68 Mar 10 21:34:30 prototype kernel: ds: 0018 es: 0018 ss: 0018 Mar 10 21:34:30 prototype kernel: Process X (pid: 6847, stackpage=c3643000) Mar 10 21:34:30 prototype kernel: Stack: c021c565 c021c6f3 00000049 ffffffff cfa70000 0000001f c3643cbc c0112c37 Mar 10 21:34:30 prototype kernel: cda57000 c88ce837 cda57000 e0000000 c012af3a c012af64 c88ce933 cda1d004 Mar 10 21:34:30 prototype kernel: c51aea40 c0044646 03050000 c3643cd0 c3643cd0 c3643cd0 c88d6dc4 cfa70000 Mar 10 21:34:30 prototype kernel: Call Trace: [<cfa70000>] [iounmap+23/32] [<cda57000>] [NVdriver:osUnmapKernelSpace+67/76] [<cda57000>] [<e0000000>] [__free_pages+26/32] Mar 10 21:34:30 prototype kernel: [free_pages+36/48] [NVdriver:osFreeContigPages+79/84] [<cda1d004>] [NVdriver:RmTeardownAGP+156/176] [<cfa70000>] [NVdriver:nv_devices+0/384] [NVdriver:nvExtEscape+2888/3100] [< cda1d004>] Mar 10 21:34:30 prototype kernel: [<cda1d004>] [<cda1d004>] [NVdriver:_nv_rmsym_01225+71/104] [<cda1d004>] [<cda21308>] [<cda212c8>] [NVdriver:_nv_rmsym_01425+446/468] [<cda1d004>] Mar 10 21:34:30 prototype kernel: [<cda212c8>] [<cda1d004>] [NVdriver:_nv_rmsym_00560+222/432] [<cda1d004>] [<cda1d004>] [NVdriver:nv_devices+0/384] [<cda1d004>] [<cda1d004>] Mar 10 21:34:30 prototype kernel: [alloc_skb+230/384] [sock_def_readable+38/80] [<cda22028>] [<cda1d004>] [<cda1d004>] [NVdriver:_nv_rmsym_00958+116/168] [<cda1d004>] [<cda1d004>] Mar 10 21:34:30 prototype kernel: [NVdriver:_nv_rmsym_00345+0/204] [<cda1d004>] [NVdriver:_nv_rmsym_01083+293/776] [<cda1d004>] [<cda1d004>] [NVdriver:nv_ioctl+449/480] [NVdriver:nv_devices+0/384] [__run_task_q ueue+76/96] Mar 10 21:34:30 prototype kernel: [<cda1d004>] [NVdriver:nv_bottom_halves+0/2560] [NVdriver:nv_bottom_halves+0/2560] [schedule+614/912] [sys_ioctl+359/384] [system_call+51/56] Mar 10 21:34:30 prototype kernel: Mar 10 21:34:30 prototype kernel: Code: 0f 0b 83 c4 0c 83 7b 08 00 74 16 6a 4b 68 f3 c6 21 c0 68 65
Linux prototype 2.4.3-pre1 #1 Sun Mar 4 14:14:54 PST 2001 i686 unknown
Gnu C 2.95.3 Gnu make 3.79.1 binutils 2.10.91.0.2 util-linux 2.11 modutils 2.4.2 e2fsprogs 1.19 reiserfsprogs 3.x.0b Linux C Library 2.2.2 Dynamic linker (ldd) 2.2.2 Procps 2.0.7 Net-tools 1.58 Kbd command Sh-utils 2.0.11 Modules Loaded NVdriver
reiserfs is the filesystem.. machine is athlon thunderbird 800mhz.
Signature · 2 lines
David Raufeisen <[email protected]> Cell: (604) 818-3596
Quoted 1 line · David Raufeisen
Mar 10 21:34:30 prototype kernel: [free_pages+36/48] [NVdriver:osFreeContigPages+79/84] [<cda1d004>] [NVdriver:RmTeardownAGP+156/176] [<cfa70000>] [NVdriver:nv_devices+0/384] [NVdriver:nvExtEscape+2888/3100] [<
Bug caused by binary only driver. Complain to nvidia, not linux-kernel.
Well, the kernel module is open source..
Quoted 5 lines · Keith Owens
Quoted 1 line · David Raufeisen
Mar 10 21:34:30 prototype kernel: [free_pages+36/48] [NVdriver:osFreeContigPages+79/84] [<cda1d004>] [NVdriver:RmTeardownAGP+156/176] [<cfa70000>] [NVdriver:nv_devices+0/384] [NVdriver:nvExtEscape+2888/3100] [<Bug caused by binary only driver. Complain to nvidia, not linux-kernel.
Signature · 2 lines
David Raufeisen <[email protected]> Cell: (604) 818-3596
Quoted 3 lines
Quoted 1 line · Keith Owens
Bug caused by binary only driver. Complain to nvidia, not linux-kernel.
Quoted 1 line · David Raufeisen
Well, the kernel module is open source..
That is just the glue code that nvidia uses to fool people. The kernel module just creates a device that the main nvidia driver abuses to bypass all the kernel checks. The real code is a 2M binary only blob.
Yeah, after looking more closely I noticed (nvrmapi.lib), i've only had two lockups ever with it .. er in 2d at least (in 6 months), today and yesterday, kinda weird :)
Quoted 3 lines · Keith Owens
That is just the glue code that nvidia uses to fool people. The kernel module just creates a device that the main nvidia driver abuses to bypass all the kernel checks. The real code is a 2M binary only blob.
Signature · 2 lines
David Raufeisen <[email protected]> Cell: (604) 818-3596
Quoted 1 line
Well, the kernel module is open source..
No the Nvidia kernel module is not. Try reading it, its obfuscated to point of being binary, it contains no permission to modify or redistribute either.
In fact if you are using patched versions of it to make it work with later kernels you may well be breaking their licensing