lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Mon, 31 Jul 2006 15:59:45 +0200
From:	Takashi Iwai <tiwai@...e.de>
To:	"Rafael J. Wysocki" <rjw@...k.pl>
Cc:	Jiri Slaby <jirislaby@...il.com>, Andrew Morton <akpm@...l.org>,
	linux-pm@...l.org, alsa-devel@...a-project.org,
	linux-kernel@...r.kernel.org, Pavel Machek <pavel@....cz>,
	mingo@...e.hu
Subject: Re: [Alsa-devel] swsusp regression (s2dsk) [Was: 2.6.18-rc2-mm1]

At Sun, 30 Jul 2006 13:34:38 +0200,
Rafael J. Wysocki wrote:
> 
> On Sunday 30 July 2006 12:54, Jiri Slaby wrote:
> > Uncc: linux-mm
> > Cc: alsa-devel
> > Cc: mingo
> > 
> > Rafael J. Wysocki wrote:
> > > On Sunday 30 July 2006 10:08, Jiri Slaby wrote:
> > >> Rafael J. Wysocki napsal(a):
> > >>> On Sunday 30 July 2006 02:06, Pavel Machek wrote:
> > >>>> Hi!
> > >>>>
> > >>>>>>>>> I have problems with swsusp again. While suspending, the very last thing kernel
> > >>>>>>>>> writes is 'restoring higmem' and then hangs, hardly. No sysrq response at all.
> > >>>>>>>>> Here is a snapshot of the screen:
> > >>>>>>>>> http://www.fi.muni.cz/~xslaby/sklad/swsusp_higmem.gif
> > >>>>>>>>>
> > >>>>>>>>> It's SMP system (HT), higmem enabled (1 gig of ram).
> > >>>>>>>> Most probably it hangs in device_power_up(), so the problem seems to be
> > >>>>>>>> with one of the devices that are resumed with IRQs off.
> > >>>>>>>>
> > >>>>>>>> Does vanila .18-rc2 work?
> > >>>>>>> Yup, it does.
> > >>>>>> Can you try up kernel, no highmem? (mem=512M)?
> > >>>>> It writes then:
> > >>>>> p16v: status 0xffffffff, mask 0x00001000, pvoice f7c04a20, use 0
> > >>>>> in endless loop when resuming -- after reading from swap.
> > >>>> Okay, so we have two different problems here.
> > 
> > [snip]
> > 
> > >>>> and one is probably driver problem with p16v (whatever it is).
> > >>>>
> > >>>> /data/l/linux/sound/pci/emu10k1/irq.c:
> > >>>> snd_printk(KERN_ERR "p16v: status: 0x%08x, mask=0x%08x, pvoice=%p,
> > >>>> use=%d\n", status2, mask, pvoice, pvoice->use);
> > >>>>
> > >>>> ...aha, so you may want to unload emu10k1 for testing.
> > >> Sure, this helped.
> > > 
> > > So, we have two different regressions here.
> > > 
> > > Please try to revert git-alsa.patch and see if the emu10k1-related problem
> > > goes away.
> > 
> > Wow, it didn't helped, I find out there is a difference between in-kernel and 
> > modules version of the driver. When compiled as modules (loaded/unloaded) 
> > suspending (and resuming) is working ok (enabled higmem and preempt back -- 
> > still no smp), when compiled in-kernel (see the config diff below), it doesn't 
> > resume.
> > @@ -1263,8 +1263,8 @@
> >   CONFIG_SND=y
> >   CONFIG_SND_TIMER=y
> >   CONFIG_SND_PCM=y
> > -CONFIG_SND_HWDEP=m
> > -CONFIG_SND_RAWMIDI=m
> > +CONFIG_SND_HWDEP=y
> > +CONFIG_SND_RAWMIDI=y
> >   CONFIG_SND_SEQUENCER=y
> >   # CONFIG_SND_SEQ_DUMMY is not set
> >   CONFIG_SND_OSSEMUL=y
> > @@ -1282,8 +1282,8 @@
> >   #
> >   # Generic devices
> >   #
> > -CONFIG_SND_AC97_CODEC=m
> > -CONFIG_SND_AC97_BUS=m
> > +CONFIG_SND_AC97_CODEC=y
> > +CONFIG_SND_AC97_BUS=y
> >   # CONFIG_SND_DUMMY is not set
> >   # CONFIG_SND_VIRMIDI is not set
> >   # CONFIG_SND_MTPAV is not set
> > @@ -1347,7 +1347,7 @@
> >   # CONFIG_SND_INDIGO is not set
> >   # CONFIG_SND_INDIGOIO is not set
> >   # CONFIG_SND_INDIGODJ is not set
> > -CONFIG_SND_EMU10K1=m
> > +CONFIG_SND_EMU10K1=y
> >   # CONFIG_SND_EMU10K1X is not set
> >   # CONFIG_SND_ENS1370 is not set
> >   # CONFIG_SND_ENS1371 is not set
> 
> If the driver is compiled in, its .suspend() routine gets called before the
> suspend image is restored and puts the card in a state that confuses
> the .resume() called after the image has been restored.
> 
> I think snd_emu10k1_suspend() should reset the device if state == PMSG_PRETHAW .

Hm, do we need such inconsitent behavior?  I mean, for most drivers,
it doesn't matter whether it's built-in or a module: simply shutdown
at suspend, and initialize at resume.


Takashi
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists