2.6.10-rc3-mm1: swsusp
Hi,
I must say I'm really impressed with the progress that swsusp made since I tried it last time (close to 2.6.10-rc1 as I recall). Now I've been using it for a couple of days on 2.6.10-rc3-mm1 and It's never refused to suspend the machine because of the lack of (contiguous) memory which happened very often before, and it seems to be much faster. Using it my notebook has reached more that 1 day of uptime recently, wich was practically impossible previously. It's a great job and vast improvement for which I thank you very much!
Still, unfortunately, today it crashed on suspend and I wasn't able to get any useful information related to the crash, because swsusp apparently does not send some of its messages to the serial console. In particular, anything from within the critical section is not printed there and that's why I think (I'm not sure though) that the crash occured in the critical section. Could you tell me please if it's possible to make all of the swsusp messages appear on the serial console and, if so, how to do this (I've already tried "dmesg -n 8" and "echo 9 > /proc/sysrq-trigger" but none of them helps)?
Greets,
RJW
Signature · 3 lines
- Would you tell me, please, which way I ought to go from here? - That depends a good deal on where you want to get to. -- Lewis Carroll "Alice's Adventures in Wonderland"
Hi!
Quoted 6 lines
I must say I'm really impressed with the progress that swsusp made since I tried it last time (close to 2.6.10-rc1 as I recall). Now I've been using it for a couple of days on 2.6.10-rc3-mm1 and It's never refused to suspend the machine because of the lack of (contiguous) memory which happened very often before, and it seems to be much faster. Using it my notebook has reached
Hmmm, unfortunately I did not changes in that area. Perhaps memory managment was fixed/improved?
Quoted 8 lines
Still, unfortunately, today it crashed on suspend and I wasn't able to get any useful information related to the crash, because swsusp apparently does not send some of its messages to the serial console. In particular, anything from within the critical section is not printed there and that's why I think (I'm not sure though) that the crash occured in the critical section. Could you tell me please if it's possible to make all of the swsusp messages appear on the serial console and, if so, how to do this (I've already tried "dmesg -n 8" and "echo 9 >
/proc/sysrq-trigger" but none of them helps)?
Using regular vga console and digital camera seems to be popular way to get dumps. I'm not sure why swsusp critical section interferes with serial, perhaps serial console support has to "know" about serial console and not suspend it during suspend() call?
Pavel
Signature · 2 lines
People were complaining that M$ turns users into beta-testers... ...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!
Quoted 2 lines · Pavel Machek
Hi!
[-- snip --]
Quoted 11 lines
Quoted 10 lines
Still, unfortunately, today it crashed on suspend and I wasn't able to get any useful information related to the crash, because swsusp apparently does not send some of its messages to the serial console. In particular, anything from within the critical section is not printed there and that's why I think (I'm not sure though) that the crash occured in the critical section. Could you tell me please if it's possible to make all of the swsusp messages appear on the serial console and, if so, how to do this (I've already tried "dmesg -n 8" and "echo 9 >
/proc/sysrq-trigger" but none of them helps)?
Quoted 2 lines
Using regular vga console and digital camera seems to be popular way to get dumps.
Well, the problem is that:
1) I don't have a DC,
2) even if I had one, my VTs go nuts after switching to X and then they only display garbage ... :-)
Quoted 3 lines
I'm not sure why swsusp critical section interferes with serial, perhaps serial console support has to "know" about serial console and not suspend it during
suspend()call?
It didn't do that before 2.6.9 or so. Something must have changed afterwards. And the serial console shows messages that appear right before writing the image to swap (like "swsusp: Version: ..." etc.).
Greets,
RJW
Signature · 3 lines
- Would you tell me, please, which way I ought to go from here? - That depends a good deal on where you want to get to. -- Lewis Carroll "Alice's Adventures in Wonderland"
Hi!
Quoted 7 lines
Quoted 3 lines
I'm not sure why swsusp critical section interferes with serial, perhaps serial console support has to "know" about serial console and not suspend it during
suspend()call?It didn't do that before 2.6.9 or so. Something must have changed afterwards. And the serial console shows messages that appear right before writing the image to swap (like "swsusp: Version: ..." etc.).
You may want to remove suspend() support from serial drivers and see what happens... Or find exact patch that broke it.
Pavel
Signature · 2 lines
People were complaining that M$ turns users into beta-testers... ...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!
Quoted 1 line · Rafael J. Wysocki
Still, unfortunately, today it crashed on suspend and I wasn't able to get any useful information related to the crash,
Have you tried an -rc3-bk13 snapshot? Some changes went in related to ALSA drivers and swsusp/ACPI suspend, perhaps it will fix the problem.
Quoted 2 lines · Ari Pollak
Quoted 1 line · Rafael J. Wysocki
Still, unfortunately, today it crashed on suspend and I wasn't able to get
any
Quoted 4 lines
Quoted 1 line
useful information related to the crash,
Have you tried an -rc3-bk13 snapshot? Some changes went in related to ALSA drivers and swsusp/ACPI suspend, perhaps it will fix the problem.
Unfortunately, I have no time for playing with -bk kernels. I'll wait for -rc4 or 2.6.10 or maybe Andrew will include this fixes into -mm before?
Greets,
RJW
Signature · 3 lines
- Would you tell me, please, which way I ought to go from here? - That depends a good deal on where you want to get to. -- Lewis Carroll "Alice's Adventures in Wonderland"