Current-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: (AGP) related panic


Aleksej Saushev <> writes:

> I'm observing reproducable panic with sources as of today,
> ca. 10:45 MSD, no crash dump is generated, system just hangs,
> handwritten backtrace follows:
> uvm_pglistalloc: page not on freelist
> uvm_pglist_add
> uvm_pglistalloc
> _bus_dmamem_alloc_range
> _bus_dmamem_alloc
> agp_generic_bind_memory
> cdev_ioctl
> spec_ioctl
> VOP_ioctl
> vn_ioctl
> sys_ioctl

Additional information: in another discussion thread,
it was pointed, that this may be connected with erroneous
"makeoptions CPUFLAGS", which allowed usage of SSE.
Yet I have never allowed SSE in CPUFLAGS.
Even more, I have rebuilt kernel to avoid that, the panic
occurred in that case too. Following options were probed:

makeoptions     CPUFLAGS="-march=prescott"
makeoptions     CPUFLAGS="-march=i486 -mtune=prescott"

Thus, hardly the problem is connected with SSE, in my opinion.
Am I wrong?

I have modified kernel source (see below), am running modified
kernel now, no panic so far. Yet the keyboard doesn't work in X
server, which renders it completely useless.

I don't observe any printout neither in dmesg, nor on console,
am I doing anything wrong?

Diff follows:

Index: dev/pci/agp.c
RCS file: /cvsroot/src/sys/dev/pci/agp.c,v
retrieving revision 1.61
diff -u -r1.61 agp.c
--- dev/pci/agp.c       22 Aug 2008 18:05:44 -0000      1.61
+++ dev/pci/agp.c       22 Sep 2008 17:27:24 -0000
@@ -549,6 +549,12 @@
        bus_addr_t pa;
        int contigpages, nseg;
+       printf("agp_generic_bind_memory: sz=%llu phys=%llu off=%lld 
+              (unsigned long long)mem->am_size,
+              (unsigned long long)mem->am_physical,
+              (long long)mem->am_offset, mem->am_is_bound);
+       printf("agp_generic_bind_memory: requested offset %ld\n", (long)offset);
        if (mem->am_is_bound) {


Home | Main Index | Thread Index | Old Index