Hi Jan,

After commit 2ca9fbd739b8a72b16dd790d0fff7b75f5488fb8 AMD IOMMU: allocate IRTE entries instead of using a static mapping, booting dom0 stalls several times.
Sometimes this results in RCU stall warnings from the dom0 kernel, hitting the "any" key, on normal or serial console, makes the boot continue for a while but it stalls several times.
(It also stalls on shutdown BTW)

I have bisected it to this commit running kernel 3.10.2 and Debian wheezy.

xl-dmesg and dmesg from a boot with stall are attached, snippet from dmesg with RCU stall appended below.

Note that i do have to run with iommu='no-amd-iommu-perdev-intremap' due to the bios bug that probably will never get fixed.
When i run without the 'no-amd-iommu-perdev-intremap' the iommu is disabled and stalls do not occur.

--
Sander


[   13.066802] udevd[1548]: starting version 175
[   13.250159] usb 4-5: New USB device found, idVendor=0a12, idProduct=0001
[   13.255702] usb 4-5: New USB device strings: Mfr=0, Product=2, SerialNumber=0
[   13.261247] usb 4-5: Product: EDRClassone
[   13.700775] bio: create slab <bio-2> at 2
[   14.332086] EXT4-fs (dm-0): mounted filesystem with ordered data mode. Opts: (null)
[   17.210269] udevd[1959]: starting version 175
[  130.161724] INFO: rcu_preempt detected stalls on CPUs/tasks:
[  130.172305]  (detected by 3, t=33837 jiffies, g=107, c=106, q=126)
[  130.180226] INFO: Stall ended before state dump start
[  131.052732] ata1.00: configured for UDMA/133
[  131.061174] ata1: EH complete
[  132.143528] ata1.00: configured for UDMA/133
[  132.149962] ata1: EH complete
[  133.214683] EXT4-fs (dm-0): re-mounted. Opts: (null)
[  133.747387] EXT4-fs (dm-0): re-mounted. Opts: barrier=1,errors=remount-ro
[  141.751979] Adding 2097148k swap on /dev/mapper/serveerstertje-swap.  Priority:-1 extents:1 across:2097148k 
[  154.078205] EXT4-fs (sda1): mounted filesystem with ordered data mode. Opts: barrier=1,errors=remount-ro
[  308.571758] INFO: rcu_preempt self-detected stall on CPU
[  308.572077] INFO: rcu_preempt self-detected stall on CPU
[  308.572105]  1: (1 ticks this GP) idle=b03/1/0 softirq=2415/2415 last_accelerate: 4e40/0059, nonlazy_posted: 0, ..
[  308.575441] INFO: rcu_preempt self-detected stall on CPU
[  308.575446] 
[  308.575451]   (t=45593 jiffies g=1527 c=1526 q=67)
[  308.575461]  2: (1 GPs behind) idle=d1b/1/0 softirq=2276/2277 last_accelerate: 4e3e/0059, nonlazy_posted: 0, L.
[  308.575473] sending NMI to all CPUs:
[  308.575477] INFO: rcu_preempt self-detected stall on CPU
[  308.575483] xen: vector 0x2 is not implemented
[  308.575490]   (t=45594 jiffies g=1527 c=1526 q=67)
[  308.575494] 
[  308.575507]  3: (1 ticks this GP) idle=9fb/1/0 softirq=1945/1945 last_accelerate: 4e3d/005a, nonlazy_posted: 0, ..
[  308.575515]   (t=45594 jiffies g=1527 c=1526 q=67)
[  308.672071]  5: (1 ticks this GP) idle=c8b/1/0 softirq=1858/1858 last_accelerate: 0014/0074, nonlazy_posted: 0, L.
[  308.679459]   (t=45625 jiffies g=1527 c=1526 q=38)
[  308.795873] r8169 0000:0a:00.0 eth1: link down
[  308.803468] r8169 0000:0a:00.0 eth1: link down
[  310.542284] r8169 0000:0a:00.0 eth1: link up
[  318.576834] Hangcheck: hangcheck value past margin!
[  318.864711] r8169 0000:0b:00.0 eth0: link down
[  318.864739] r8169 0000:0b:00.0 eth0: link down
[  320.568031] r8169 0000:0b:00.0 eth0: link up