occasional glitches with 2.6.36+

occasional glitches with 2.6.36+

[Thomas Fjellstrom]

Over the past couple months I've noticed a couple problems that don't always 
happen.

First, the framebuffer (intel gfx) thinks its about 1/4 the size it really is, 
even after logging in.

Second, my sdhci device likes to error out:
[    1.615309] irq 17: nobody cared (try booting with the "irqpoll" option)
[    1.615365] Pid: 0, comm: kworker/0:0 Not tainted 2.6.36-nofp #25
[    1.615418] Call Trace:
[    1.615467]  <IRQ>  [<ffffffff810a6086>] __report_bad_irq+0x26/0xa0
[    1.615564]  [<ffffffff810a628c>] note_interrupt+0x18c/0x1d0
[    1.615618]  [<ffffffff810a6a3d>] handle_fasteoi_irq+0xcd/0x100
[    1.615673]  [<ffffffff810057cd>] handle_irq+0x1d/0x30
[    1.615725]  [<ffffffff81004cee>] do_IRQ+0x6e/0xf0
[    1.615779]  [<ffffffff81385dd3>] ret_from_intr+0x0/0xa
[    1.615831]  <EOI>  [<ffffffff8100b4d7>] ? mwait_idle+0xa7/0xf0
[    1.615923]  [<ffffffff8100b474>] ? mwait_idle+0x44/0xf0
[    1.615977]  [<ffffffff81388fb1>] ? atomic_notifier_call_chain+0x11/0x20
[    1.616033]  [<ffffffff810019fa>] cpu_idle+0xba/0x180
[    1.616088]  [<ffffffff810743b9>] ? clockevents_register_device+0xd9/0xf0
[    1.616144]  [<ffffffff8137d6f5>] start_secondary+0x1e5/0x1ec
[    1.616197] handlers:
[    1.616245] [<ffffffffa02a8100>] (sdhci_irq+0x0/0x5f0 [sdhci])
[    1.616306] Disabling IRQ #17

Neither of these always happen. I've tested with memtest86+ a couple times, 
nothing so far.

But they both seem to happen at the same time.

-- 
Thomas Fjellstrom
thomas@fjellstrom.ca