Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
On Tue, 23 Oct 2007, Rafael J. Wysocki wrote: > On Monday, 22 October 2007 16:11, Mark Lord wrote: > > Rafael, > > > > What happens to the jiffies variable on resume from RAM, and from DISK? > > Do we restore it to the value it had at suspend, > > or just leave it be with whatever? > > > > The answer has to be "restore the value it had at suspend time", > > but I figured I'd check here anyway. > > > > ?? > > Well, frankly, I've lost track of that recently, but it seems that we just use > the pre-suspend jiffies (at least in the current -git). > > Thomas knows better, I guess. :-) We use the pre-suspend value if nothing else fiddled in the variable. tglx - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
On Tue, 23 Oct 2007, Rafael J. Wysocki wrote: On Monday, 22 October 2007 16:11, Mark Lord wrote: Rafael, What happens to the jiffies variable on resume from RAM, and from DISK? Do we restore it to the value it had at suspend, or just leave it be with whatever? The answer has to be restore the value it had at suspend time, but I figured I'd check here anyway. ?? Well, frankly, I've lost track of that recently, but it seems that we just use the pre-suspend jiffies (at least in the current -git). Thomas knows better, I guess. :-) We use the pre-suspend value if nothing else fiddled in the variable. tglx - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
On Monday, 22 October 2007 16:11, Mark Lord wrote: > Rafael, > > What happens to the jiffies variable on resume from RAM, and from DISK? > Do we restore it to the value it had at suspend, > or just leave it be with whatever? > > The answer has to be "restore the value it had at suspend time", > but I figured I'd check here anyway. > > ?? Well, frankly, I've lost track of that recently, but it seems that we just use the pre-suspend jiffies (at least in the current -git). Thomas knows better, I guess. :-) Greetings, Rafael - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
Rafael, What happens to the jiffies variable on resume from RAM, and from DISK? Do we restore it to the value it had at suspend, or just leave it be with whatever? The answer has to be "restore the value it had at suspend time", but I figured I'd check here anyway. ?? - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
Rafael, What happens to the jiffies variable on resume from RAM, and from DISK? Do we restore it to the value it had at suspend, or just leave it be with whatever? The answer has to be restore the value it had at suspend time, but I figured I'd check here anyway. ?? - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
On Monday, 22 October 2007 16:11, Mark Lord wrote: Rafael, What happens to the jiffies variable on resume from RAM, and from DISK? Do we restore it to the value it had at suspend, or just leave it be with whatever? The answer has to be restore the value it had at suspend time, but I figured I'd check here anyway. ?? Well, frankly, I've lost track of that recently, but it seems that we just use the pre-suspend jiffies (at least in the current -git). Thomas knows better, I guess. :-) Greetings, Rafael - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
Pavel Machek wrote: Hi! Since upgrading to 2.6.23.1 from 2.6.23-rc9, resume-from-RAM has been misbehaving here. It takes much (+5-7 seconds) longer to resume *sometimes*, but not all/most of the time. I suspend those long delays may have something to do with USB, as it takes longer for my hub + mouse to come back to life during the sequence. But I have since then re-applied the powertop patches, and the long delays vanished. Back with -rc9 I normally also had those powertop fixes, and so this problem could be much older and was never noticed until I ran without them. I'm no longer actively investigating, as the delays are gone (powertop patches), and the crashes are much less frequent since patching. Cheers - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
Hi! > Since upgrading to 2.6.23.1 from 2.6.23-rc9, > resume-from-RAM has been misbehaving here. > > It takes much (+5-7 seconds) longer to resume > *sometimes*, but not all/most of the time. > And sometimes I get get flashing keyboard LEDs and have > to hold the power button > in for a full hard reset. > > With 2.6.23-rc8/rc9, no such troubles. > > Difficult to reproduce, other than perhaps once a day. > Anybody want to fess up with a likely candidate? Are there any .config differences between rc8 and .1? Can you try disabling nohz? -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
Hi! Since upgrading to 2.6.23.1 from 2.6.23-rc9, resume-from-RAM has been misbehaving here. It takes much (+5-7 seconds) longer to resume *sometimes*, but not all/most of the time. And sometimes I get get flashing keyboard LEDs and have to hold the power button in for a full hard reset. With 2.6.23-rc8/rc9, no such troubles. Difficult to reproduce, other than perhaps once a day. Anybody want to fess up with a likely candidate? Are there any .config differences between rc8 and .1? Can you try disabling nohz? -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
Pavel Machek wrote: Hi! Since upgrading to 2.6.23.1 from 2.6.23-rc9, resume-from-RAM has been misbehaving here. It takes much (+5-7 seconds) longer to resume *sometimes*, but not all/most of the time. I suspend those long delays may have something to do with USB, as it takes longer for my hub + mouse to come back to life during the sequence. But I have since then re-applied the powertop patches, and the long delays vanished. Back with -rc9 I normally also had those powertop fixes, and so this problem could be much older and was never noticed until I ran without them. I'm no longer actively investigating, as the delays are gone (powertop patches), and the crashes are much less frequent since patching. Cheers - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
On Wednesday, 17 October 2007 00:10, Mark Lord wrote: > Mark Lord wrote: > > Rafael J. Wysocki wrote: > >> On Sunday, 14 October 2007 22:13, Mark Lord wrote: > >>> Since upgrading to 2.6.23.1 from 2.6.23-rc9, resume-from-RAM has been > >>> misbehaving here. > >>> > >>> It takes much (+5-7 seconds) longer to resume *sometimes*, but not > >>> all/most of the time. > >>> And sometimes I get get flashing keyboard LEDs and have to hold the > >>> power button > >>> in for a full hard reset. > >>> > >>> With 2.6.23-rc8/rc9, no such troubles. > >>> > >>> Difficult to reproduce, other than perhaps once a day. > >>> Anybody want to fess up with a likely candidate? > >> > >> Not really, but if you rule out all of the POWERPC and MIPS patches, > >> there's > >> not much left ... > > > > Yeah, I didn't see much there either. > > > > I'll keep an eye on things over the next few days, > > and post again if it persists. > > > > I was using the powertop patches with -rc9, but not with 2.6.23.1. > > Maybe they helped (???). > > It still happens. Well, I have no idea, sorry. I also have not been able to reproduce it here ... - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
Mark Lord wrote: Rafael J. Wysocki wrote: On Sunday, 14 October 2007 22:13, Mark Lord wrote: Since upgrading to 2.6.23.1 from 2.6.23-rc9, resume-from-RAM has been misbehaving here. It takes much (+5-7 seconds) longer to resume *sometimes*, but not all/most of the time. And sometimes I get get flashing keyboard LEDs and have to hold the power button in for a full hard reset. With 2.6.23-rc8/rc9, no such troubles. Difficult to reproduce, other than perhaps once a day. Anybody want to fess up with a likely candidate? Not really, but if you rule out all of the POWERPC and MIPS patches, there's not much left ... Yeah, I didn't see much there either. I'll keep an eye on things over the next few days, and post again if it persists. I was using the powertop patches with -rc9, but not with 2.6.23.1. Maybe they helped (???). It still happens. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
Mark Lord wrote: Rafael J. Wysocki wrote: On Sunday, 14 October 2007 22:13, Mark Lord wrote: Since upgrading to 2.6.23.1 from 2.6.23-rc9, resume-from-RAM has been misbehaving here. It takes much (+5-7 seconds) longer to resume *sometimes*, but not all/most of the time. And sometimes I get get flashing keyboard LEDs and have to hold the power button in for a full hard reset. With 2.6.23-rc8/rc9, no such troubles. Difficult to reproduce, other than perhaps once a day. Anybody want to fess up with a likely candidate? Not really, but if you rule out all of the POWERPC and MIPS patches, there's not much left ... Yeah, I didn't see much there either. I'll keep an eye on things over the next few days, and post again if it persists. I was using the powertop patches with -rc9, but not with 2.6.23.1. Maybe they helped (???). It still happens. - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
On Wednesday, 17 October 2007 00:10, Mark Lord wrote: Mark Lord wrote: Rafael J. Wysocki wrote: On Sunday, 14 October 2007 22:13, Mark Lord wrote: Since upgrading to 2.6.23.1 from 2.6.23-rc9, resume-from-RAM has been misbehaving here. It takes much (+5-7 seconds) longer to resume *sometimes*, but not all/most of the time. And sometimes I get get flashing keyboard LEDs and have to hold the power button in for a full hard reset. With 2.6.23-rc8/rc9, no such troubles. Difficult to reproduce, other than perhaps once a day. Anybody want to fess up with a likely candidate? Not really, but if you rule out all of the POWERPC and MIPS patches, there's not much left ... Yeah, I didn't see much there either. I'll keep an eye on things over the next few days, and post again if it persists. I was using the powertop patches with -rc9, but not with 2.6.23.1. Maybe they helped (???). It still happens. Well, I have no idea, sorry. I also have not been able to reproduce it here ... - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
Rafael J. Wysocki wrote: On Sunday, 14 October 2007 22:13, Mark Lord wrote: Since upgrading to 2.6.23.1 from 2.6.23-rc9, resume-from-RAM has been misbehaving here. It takes much (+5-7 seconds) longer to resume *sometimes*, but not all/most of the time. And sometimes I get get flashing keyboard LEDs and have to hold the power button in for a full hard reset. With 2.6.23-rc8/rc9, no such troubles. Difficult to reproduce, other than perhaps once a day. Anybody want to fess up with a likely candidate? Not really, but if you rule out all of the POWERPC and MIPS patches, there's not much left ... Yeah, I didn't see much there either. I'll keep an eye on things over the next few days, and post again if it persists. I was using the powertop patches with -rc9, but not with 2.6.23.1. Maybe they helped (???). -ml - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
On Sunday, 14 October 2007 22:13, Mark Lord wrote: > Since upgrading to 2.6.23.1 from 2.6.23-rc9, resume-from-RAM has been > misbehaving here. > > It takes much (+5-7 seconds) longer to resume *sometimes*, but not all/most > of the time. > And sometimes I get get flashing keyboard LEDs and have to hold the power > button > in for a full hard reset. > > With 2.6.23-rc8/rc9, no such troubles. > > Difficult to reproduce, other than perhaps once a day. > Anybody want to fess up with a likely candidate? Not really, but if you rule out all of the POWERPC and MIPS patches, there's not much left ... Greetings, Rafael - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
On Sunday, 14 October 2007 22:13, Mark Lord wrote: Since upgrading to 2.6.23.1 from 2.6.23-rc9, resume-from-RAM has been misbehaving here. It takes much (+5-7 seconds) longer to resume *sometimes*, but not all/most of the time. And sometimes I get get flashing keyboard LEDs and have to hold the power button in for a full hard reset. With 2.6.23-rc8/rc9, no such troubles. Difficult to reproduce, other than perhaps once a day. Anybody want to fess up with a likely candidate? Not really, but if you rule out all of the POWERPC and MIPS patches, there's not much left ... Greetings, Rafael - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Regression: 2.6.23-rc9 okay, 2.6.23.1 resume problems
Rafael J. Wysocki wrote: On Sunday, 14 October 2007 22:13, Mark Lord wrote: Since upgrading to 2.6.23.1 from 2.6.23-rc9, resume-from-RAM has been misbehaving here. It takes much (+5-7 seconds) longer to resume *sometimes*, but not all/most of the time. And sometimes I get get flashing keyboard LEDs and have to hold the power button in for a full hard reset. With 2.6.23-rc8/rc9, no such troubles. Difficult to reproduce, other than perhaps once a day. Anybody want to fess up with a likely candidate? Not really, but if you rule out all of the POWERPC and MIPS patches, there's not much left ... Yeah, I didn't see much there either. I'll keep an eye on things over the next few days, and post again if it persists. I was using the powertop patches with -rc9, but not with 2.6.23.1. Maybe they helped (???). -ml - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
