Messages in this thread | | | From | Bartlomiej Zolnierkiewicz <> | Subject | Re: upstream regression (IO-APIC?) | Date | Mon, 3 Nov 2008 19:57:37 +0100 |
| |
Hi,
On Monday 03 November 2008, Alok Kataria wrote: > On Sun, 2008-11-02 at 12:24 -0800, Bartlomiej Zolnierkiewicz wrote: > > On Sunday 02 November 2008, Bartlomiej Zolnierkiewicz wrote: > > > On Thursday 30 October 2008, Robert Hancock wrote: > > > > Bartlomiej Zolnierkiewicz wrote: > > > > > The current Linus tree as of commit e946217e4fdaa67681bbabfa8e6b18641921f750 > > > > > is broken for me. I get either the following panic (see log from qemu below) > > > > > or lost IRQs on ATA init... Is this a known issue? > > > > > > > > > > PS The tree that I used before and was supposedly good (sorry, I'm too tired > > > > > to verify it now) had commit 57f8f7b60db6f1ed2c6918ab9230c4623a9dbe37 at head. > > > > > > Unfortunately 57f8f7b60db6f1ed2c6918ab9230c4623a9dbe37 (v2.6.28-rc1) > > > is also bad. Bisecting it further was a real pain (i.e. I hit broken > > > build with x86 irqbalance changes, broken build with netfilter nat > > > changes and jbd journal problem). In the end it turned out that 2.6.27 > > > is bad too! However with 2.6.27 the panic occurs only once per several > > > attempts and if there is no panic kernel boots normally (no lost IRQs). > > > > > > [...] > > > > > > I finally managed to narrow it down to change making x86 use tsc_khz > > > for loops_per_jiffy -- commit 3da757daf86e498872855f0b5e101f763ba79499 > > > ("x86: use cpu_khz for loops_per_jiffy calculation"). This approach > > > seems too simplistic (as I see now Arjan & Pavel expressed concerns > > > about it back when the patch was posted initially [1][2]). Also it > > > would probably be preferred to re-use existing preset_lpj variable > > > (just like KVM does it for similar purpose [3]) instead of adding a > > > lpj_tsc one and increasing complexity. > > > > It turned out that I can boot a kernel with different config with > > HZ == 250 just fine and switching to HZ == 1000 makes it fail. > > > > > > Looking into it some more: > > > > HZ == 250 kernel (good): > > > > Calibrating delay loop (skipped), value calculated using timer frequency.. 2986.79 BogoMIPS (lpj=5973580) > > > > HZ == 1000 kernel (bad): > > > > Calibrating delay loop (skipped), using tsc calculated value.. 2990.35 BogoMIPS (lpj=1495176) > > > > HZ == 1000 kernel with hackyfix (good): > > > > Calibrating delay using timer specific routine.. 3016.68 BogoMIPS (lpj=6033376) > > > > > > Argggh... lpj is used for udelay() & friends so this bug is quite > > dangerous (since udelay() & friends are used for hardware delays)... > > > > [ The commit works for HZ == 250 because it does tsc_khz * 1000 / HZ, > > tsc_khz * 4 => lpj assumption holds true and there is no frequency > > scaling at boot. ] > > > > The quick fix would be to replace 1000 / HZ by the magic number "4" > > That's not right, the magic number 4 thing would not be correct. > On one of my systems for eg, i get this in dmesg > > Detected 2010.400 MHz processor. > ... > Calibrating delay using timer specific routine.. 4022.47 BogoMIPS > (lpj=2011235) > > This is with an earlier kernel, the HZ value is 1000. And the lpj value > that we get from the calculation of (tsc_khz * 1000)/HZ is correct in > this case. And on all the systems that i have checked this assumption > holds true.
Yes, I realized this once I tried to reproduce issue on the real h/w.
> One of the things that i suspect is that you are not using delay_tsc in > this case, i.e. tsc is not used for delay which is causing that panic > > can you please try the patch below on your system ? > > [test-patch] > > Index: linux-2.6/arch/x86/kernel/tsc.c > =================================================================== > --- linux-2.6.orig/arch/x86/kernel/tsc.c 2008-10-15 10:51:14.000000000 -0700 > +++ linux-2.6/arch/x86/kernel/tsc.c 2008-11-03 09:43:01.000000000 -0800 > @@ -847,10 +847,6 @@ > cpu_khz = calibrate_cpu(); > #endif > > - lpj = ((u64)tsc_khz * 1000); > - do_div(lpj, HZ); > - lpj_fine = lpj; > - > printk("Detected %lu.%03lu MHz processor.\n", > (unsigned long)cpu_khz / 1000, > (unsigned long)cpu_khz % 1000); > @@ -871,6 +867,10 @@ > tsc_disabled = 0; > > use_tsc_delay(); > + lpj = ((u64)tsc_khz * 1000); > + do_div(lpj, HZ); > + lpj_fine = lpj; > + > /* Check and install the TSC clocksource */ > dmi_check_system(bad_tsc_dmi_table); > check_system_tsc_reliable();
Didn't help but it looks like we should apply it anyway (so "notsc" continues to work as before).
Thanks, Bart
| |