ALSA segfault, buggy timer, and kernel > 2.4.20

Gregory Davis gregdavis at ieee.org
Thu Jul 3 12:53:29 PDT 2003


Hi all.  I am having some heap of trouble after building a fresh lfs
according to cvs.  So far so good, except...  I used kernel 2.4.21 and am
having a recurring error message: "i8253 count too high, resetting!!!" 
That message is generated in the file arch/i386/kernel/time.c of the kernel
tree.  Apparently, it is triggered by whacky reads from the onboard timer
chip of my ASUS A7A266 motherboard.  Also, I get (when I enable the APIC
option in the kernel config) another error stating "...8259A spurious
interrupt on IRQ7..." or something to that effect (and the 8259A is a PIC,
and IRQ7 is assigned to the PCI to IDE bridge).  All the while, any program
that tries to access a sound device (e.g. /dev/dsp) segfaults (xmms and
aplay).  

I upgraded to 2.4.22-pre2 which has different code in the time.c file,
disabled APIC in the kernel, rebooted and recompiled/installed ALSA 0.9.4
according to blf cvs.  Now I only get the i8253 message when I access sound
stuff (ala "alsactl -restore") and a segfault of whatever happened to
trigger the error message.  I don't think the message recurrs with the
newer setup, it just gets triggered once when the sound program segfaults.

FWIW, I used gcc 2.95.3 to compile the kernel, and 3.2 to compile everything
else (ALSA and xmms etc etc).  I'm fed up and am going to try downgrading
to kernel 2.4.20.  Does it matter if kernel modules and the kernel itself
are compiled using different versions of GCC?

Greg
-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe blfs-support' in the subject header of the message



More information about the blfs-support mailing list