https://bugzilla.novell.com/show_bug.cgi?id=374217 User novell@mirell.de added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c2 M. S. changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |novell@mirell.de --- Comment #2 from M. S. 2008-03-26 20:11:06 MST --- Does booting with the "noexec=off" option make it suspend again? If yes, this is a bug in s2ram introduced due to a bug fixed in the kernel. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User novell@mirell.de added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c4 --- Comment #4 from M. S. 2008-03-27 05:41:26 MST --- Could you switch to VT1, login as root and check the output of "s2ram --vbemode_restore --force"? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User novell@mirell.de added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c6 --- Comment #6 from M. S. 2008-03-27 11:43:29 MST --- My "working" Macbook running latest factory: - macbook 2.6.25-rc6-git5-10-pae #1 SMP - system.hardware.product = 'MacBook1,1' - suspend-0.80.20080219-14 Mind that I HAVE to use vbemode restore if using the intel driver otherwise it crashes on resume (been like that since 10.3 already) as stated on http://en.opensuse.org/s2ram. I actually wonder why the quirks in both HAL fdi's and the s2ram whitelist still have vbe_post set which clearly crashes things. Who still uses i830?? I had perfect suspend/resume on 10.3 and have it now working in 11.0/factory (again, only after setting vbemode restore) with only the resume ending up at a black screen, the appletouch driver screwing up and requiring a VT switch to restore X correctly. I am experimenting with mactel patches for appletouch but have no clue about the reason for the black screen. Without the noexec=off at boot the suspend process never gets launched due to a bugfix in the new kernel causing s2ram to segfault (see http://linux.derkeiler.com/Mailing-Lists/Kernel/2008-03/msg01674.html). You checked your /var/log/suspend.log? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User fm@opensuse.org added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c9 --- Comment #9 from Felix Möller 2008-03-27 21:34:49 MST --- Ok the window gets smaller I have been compiling the kernel for more than 10 hours now. :-( # git bisect log git-bisect start # good: [49914084e797530d9baaf51df9eda77babc98fa8] Linux 2.6.24 git-bisect good 49914084e797530d9baaf51df9eda77babc98fa8 # bad: [19af35546de68c872dcb687613e0902a602cb20e] Linux 2.6.25-rc1 git-bisect bad 19af35546de68c872dcb687613e0902a602cb20e # good: [d2e626f45cc450c00f5f98a89b8b4c4ac3c9bf5f] x86: add PAGE_KERNEL_EXEC_NOCACHE git-bisect good d2e626f45cc450c00f5f98a89b8b4c4ac3c9bf5f # bad: [d2fc0bacd5c438cb459fdf531eff00ab18422a00] Merge git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86 git-bisect bad d2fc0bacd5c438cb459fdf531eff00ab18422a00 # good: [5f7b9b457751efc9f3ad120d0ebdb19fe753e9d0] [ALSA] oxygen: add front panel capture git-bisect good 5f7b9b457751efc9f3ad120d0ebdb19fe753e9d0 # git bisect good Bisecting: 617 revisions left to test after this [b6cf160c4b788a31f6a4017a469b956ca77febf4] Merge git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6 I needed three patches so far to compile the kernel: time.h.patch:+ /* The following asm() prevents the compiler from time.h.patch:+ * optimising this loop into a modulo operation. */ time.h.patch:+ asm("" : "+r"(ns)); time.h.patch:+ core.c.patch:- pages->state.guest_tss.esp1 = cpu->esp1; core.c.patch:+ pages->state.guest_tss.sp1 = cpu->esp1; cls_flow.c.patch:+#ifdef CONFIG_NET_EMATCH cls_flow.c.patch: if (f->ematches.hdr.nmatches && cls_flow.c.patch: tcf_em_tree_dump(skb, &f->ematches, TCA_FLOW_EMATCHES) < 0) cls_flow.c.patch: goto nla_put_failure; cls_flow.c.patch:- cls_flow.c.patch:+#endif -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User fm@opensuse.org added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c11 Felix Möller changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |rjwysocki@sisk.pl --- Comment #11 from Felix Möller 2008-03-28 05:58:57 MST --- Ok I am left with 51 revisions to test. So I am getting there. But it takes quiet some time to compile a kernel ... There is a suspend merge in the commits 2008-02-02 04:29:57. With alot of commits from Rafael. CCing him. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User fm@opensuse.org added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c13 Felix Möller changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |pavel@novell.com, len.brown@intel.com --- Comment #13 from Felix Möller 2008-03-28 14:48:22 MST --- Ok I found the offending commit: commit 3c1d2b6085d75df0691cec6a4a053c0aa55fe4c9 Author: Rafael J. Wysocki Date: Tue Jan 8 00:06:16 2008 +0100 ACPI: Separate disabling of GPEs from _PTS The preparation to enter an ACPI system sleep state is now tied to the disabling of GPEs, but the GPEs should not be disabled before suspending devices. Since on ACPI 1.0x systems the _PTS global control method should be executed before suspending devices, we need to disable GPEs separately. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Len Brown This commit breaks it for me. adding acpi_new_pts_ordering to the boot prompt makes the display switch off again. CCing Pavel and Len as he acked the commit. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User fm@opensuse.org added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c15 --- Comment #15 from Felix Möller 2008-03-28 16:24:48 MST --- Without acpi_new_pts_ordering there is no way to get my MacBook to suspend (The display just does not switch of). With acpi_new_pts_ordering I can suspend. (display goes off and the the little white led blinks). So yes for me it is like "pre 2.6.25-rc". Right now I tried to get the system to resume ;): command from runlevel 1 from Xorg * s2ram -f -a 1 crash * s2ram -f -a 2 display off works * s2ram -f -a 3 crash * s2ram -f -p -m display off works * s2ram -f -p -s works crash * s2ram -f -m display off works * s2ram -f -s works crash * s2ram -f -p display off works * s2ram -f -a 1 -m crash * s2ram -f -a 1 -s crash any idea here? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User fm@opensuse.org added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c17 --- Comment #17 from Felix Möller 2008-03-28 17:14:17 MST --- (1) I tried to create an ascii art table. ;) It contains on the what happens when resuming from runlevel 1 and in the second column what happens when resuming to Xorg. (2) I am using the default openSUSE kernel again it has: <6>vesafb: framebuffer at 0x80000000, mapped to 0xf8880000, using 6144k, total 16064k <6>vesafb: mode is 1024x768x16, linelength=2048, pages=9 <6>vesafb: scrolling: redraw <6>vesafb: Truecolor: size=0:5:6:5, shift=0:11:5:0 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User fm@opensuse.org added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c19 --- Comment #19 from Felix Möller 2008-03-28 18:23:33 MST --- (In reply to comment #18 from Rafael Wysocki)

So, there are many combinations of s2ram options which allow you to successfully resume to X. Is that correct? Yes. I tried every combination once and wrote down what it did. Hopefully this is reliable...

Please try using vga=normal instead of the vesafb and see if that helps with the "blank screen after the resume" problem. Ok with vga=normal I just found out something interesting after s2ram -f -m: tty 1 was black I changed to X tty7 which was black too. Then changing to tty1 again made the console appear and after changing to tty7 X was there too.

-- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User rjwysocki@sisk.pl added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c21 --- Comment #21 from Rafael Wysocki 2008-03-28 18:55:22 MST --- (In reply to comment #20 from Felix Möller)

Ok with vga=normal just s2ram -f -a 3 works for X and tty1.

Well, I don't think we can do any better than that at the moment. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User rjwysocki@sisk.pl added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c23 --- Comment #23 from Rafael Wysocki 2008-03-30 06:05:18 MST --- Please try the patch at http://marc.info/?l=linux-acpi&m=120683999519118&w=4 . It restores the 2.6.24 suspend code ordering, so it should bring your system back to the 2.6.24 behavior without any additional kernel command line arguments. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User fm@opensuse.org added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c25 --- Comment #25 from Felix Möller 2008-04-01 08:47:10 MST --- As just reported to LKML: Rafael's patch posted there does solve my problem with current git head. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User gregkh@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c27 Greg Kroah-Hartman changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |fm@opensuse.org --- Comment #27 from Greg Kroah-Hartman 2008-05-05 11:43:23 MST --- I think this went into the final version of 2.6.25, can you try the Beta2 release to see if that solves this issue for you? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User gregkh@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c29 --- Comment #29 from Greg Kroah-Hartman 2008-05-06 11:02:40 MST --- Great! This should now be fixed then. Felix, can you try the Beta 2 to verify that this is solved? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=374217 User gregkh@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=374217#c31 Greg Kroah-Hartman changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED --- Comment #31 from Greg Kroah-Hartman 2008-05-06 12:30:30 MST --- Thanks for verifying, I'll close this bug out. Yes, please create new bugs for the other things,they are not kernel issues. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.