kernel 2.5.10 problems

This is the first time I've posted to this forum. Usually when I have a problem with a kernel other people are having the same problem and it gets fixed relatively quickly. But I've been experiencing some problems that haven't been going away. If more information is needed then what I have given here please let me know. I am using a SOYO K7VTA-B with a VIA 82C686B chipset and a Duron 750mhz processor.

1. Cannot mount IDE-SCSI cd drive. This problem has been mentioned previously by others. Errors occur when trying to mount it.

2. When ACPI support is installed the kernel gives a "Keyboard not found" error when booting and I have to push the reset switch to reboot. This problem has also been mentioned before but I don't think anyone has related it to the ACPI support.

3. Starting sometime after kernel 2.5.1 (I couldn't compile any kernels from then up until 2.5.5) the Evolution email program locks up whenever Calender, Tasks, or Contacts is selected. I have to go to another terminal and kill it.

4. Starting with kernel 2.5.6 (kernels 2.5.5 through 2.5.6-pre3 work) the KDE program krecord closes right after it starts.

5. With the new ALSA routines built-in is it possible to get the midi output port on a Soundblaster PCI 512 to work? I haven't been able to get it to function yet.

Thank you in advance for your help

David Macbanay

Interesting. last few lines of strace output may show up what's going on. Can you do a before/after strace on a working & non-working kernel?

| Dave Jones.        http://www.codemonkey.org.uk
| SuSE Labs

Oh, and does this still apply to 2.5.10 ?
Again, last few lines from an strace to find out what its doing when it locks up may be useful.

| Dave Jones.        http://www.codemonkey.org.uk
| SuSE Labs

Here's what shows in kernel 2.5.10. I'm using a Soundblaster PCI 512 and I have ALSA support for it installed.

gettimeofday({1019717263, 731947}, NULL) = 0
select(8, [3 4 6 7], NULL, NULL, {0, 0}) = 2 (in [6 7], left {0, 0})

read(7, 0x809fa74, 8192)                = -1 EBADFD (File descriptor in bad state)
dup(2)                                  = 8
fcntl64(8, F_GETFL)                     = 0x2 (flags O_RDWR)

fstat64(8, {st_mode=S_IFCHR|0620, st_rdev=makedev(3, 1), ...}) = 0
ioctl(8, SNDCTL_TMR_TIMEBASE, {B38400 opost isig icanon echo ...}) = 0 old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40014000
_llseek(8, 0, 0xbffff5e4, SEEK_CUR) = -1 ESPIPE (Illegal seek)
write(8, "read sound: File descriptor in b"..., 41read sound: File descriptor in bad state ) = 41

close(8)                                = 0
munmap(0x40014000, 4096)                = 0
_exit(1)                                = ? 

Here's what shows in kernel 2.5.6-pre3 until I pressed control-C.

gettimeofday({1019718466, 876979}, NULL) = 0
select(8, [3 4 6 7], NULL, NULL, {10, 631024}) = 1 (in [7], left {10, 600000}) read(7, "\377\377\1\0\377\377\2\0\2\0\376\377\376\377\0\0\375\377"..., 8192) = 8192 gettimeofday({1019718466, 922638}, NULL) = 0
ioctl(3, 0x541b, [0]) = 0
gettimeofday({1019718466, 923415}, NULL) = 0
select(8, [3 4 6 7], NULL, NULL, {10, 584588}) = 1 (in [7], left {10, 540000}) read(7, "\0\0\377\377\0\0\6\0\2\0\377\377\1\0\3\0\376\377\375\377"..., 8192) = 8192 gettimeofday({1019718466, 969320}, NULL) = 0
ioctl(3, 0x541b, [0]) = 0
gettimeofday({1019718466, 970109}, NULL) = 0
select(8, [3 4 6 7], NULL, NULL, {10, 537894} <unfinished ...>

David Macbanay

Here's the last lines from the strace output until had to kill it from another terminal. And this occurs in any kernel from 2.5.5 through 2.5.10. It may have happened before that also but I wasn't able to compile any kernels after 2.5.1 up until 2.5.5.

David Macbanay

select(19, [5 8 10 11 12 13 14 15 16 17 18], NULL, [5 8 10 11 12 13 14 15 16 17 18], NULL) = 1 (in [15])
read(15, "GIOP\1\0\1\0008\0\0\0", 12) = 12
read(15, "\0\0\0\0\4\366\377\277\1\0\0\0\30\0\0\0\0\0\0\0009\351"..., 56) = 56
writev(15, [{"GIOP\1\0\1", 7}, {"\1\f\0\0\0", 5}, {"\0\0\0\0", 4}, {"\4\366\377\277\0\0\0\0", 8}], 4) = 24 select(19, [5 8 10 11 12 13 14 15 16 17 18], NULL, [5 8 10 11 12 13 14 15 16 17 18], NULL) = 1 (in [17])
read(17, "GIOP\1\0\1\0004\0\0\0", 12) = 12
read(17, "\0\0\0\0\364\366\377\277\1\0\0\0\30\0\0\0\0\0\0\0009\351"..., 52) = 52
writev(17, [{"GIOP\1\0\1", 7}, {"\1\f\0\0\0", 5}, {"\0\0\0\0", 4}, {"\364\366\377\277\0\0\0\0", 8}], 4) = 24 select(19, [5 8 10 11 12 13 14 15 16 17 18], NULL, [5 8 10 11 12 13 14 15 16 17 18], NULL) = 1 (in [17])
read(17, "GIOP\1\0\1\0008\0\0\0", 12) = 12
read(17, "\0\0\0\0\364\366\377\277\1\0\0\0\30\0\0\0\0\0\0\0009\351"..., 56) = 56
writev(17, [{"GIOP\1\0\1", 7}, {"\1\f\0\0\0", 5}, {"\0\0\0\0", 4}, {"\364\366\377\277\0\0\0\0", 8}], 4) = 24 select(19, [5 8 10 11 12 13 14 15 16 17 18], NULL, [5 8 10 11 12 13 14 15 16 17 18], NULL) = ? ERESTARTNOHAND (To be restarted) +++ killed by SIGKILL +++

That one likely is a known bug in the ALSA OSS emulation, Takashi recently fixed it.

Gerd

===== sound/core/oss/pcm_oss.c 1.4 vs edited =====

@@ -557,7 +557,7 @@
 			if (ret < 0)
 				break;
 		}
-		ret = snd_pcm_lib_read(substream, ptr, frames);
+//		ret = snd_pcm_lib_read(substream, ptr, frames);
 		if (in_kernel) {
 			mm_segment_t fs;
 			fs = snd_enter_user();

I would like to confirm this bug on 2.5.8, 2.4.x is ok, I have to try 2.5.10 yet ...

Tommy

I have been aware of this bug since about 2.5.5. It is still present in 2.5.10. I have been attempting to gather more useful debugging information. So far, I haven't come up with anything that points clearly to the problem.

Regarding using strace to identify the bug, it should be noted that Evolution uses multiple processes that all communicate with the shell process (evolution). The process that handles access to much of Evolution's storage for the most of the Evolution components is called wombat.

My plan is to run "strace wombat", "strace evolution-mail" and "strace evolution-addressbook" in separate terminal windows. The evolution process can then be started normally.

My observations so far lead me to suspect a problem with wombat. If any of you other Evolution uses would care to help me with this testing, please do!

Thanks,
Miles

Hi!

Try acpi list.

Try strace to see what it does before it dies.

Philips Velo 1: 1"x4"x8", 300gram, 60, 12MB, 40bogomips, linux, mutt,
details at http://atrey.karlin.mff.cuni.cz/~pavel/velo/index.html.

What state are the tasks in when it hangs, S, R, D, N, T, what? I have an intermittent problem on 2.4 where an entire process group goes into T state even though nothing is tracing it. Killing the offending process then sending SIGCONT to the rest of the process group restarts the group. The offending process is usually the last one on the tree.