A recent change to mce_init in -mm does the following:
@@ -462,7 +463,7 @@ static void mce_init(void *dummy)
wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff);
for (i = 0; i < banks; i++) {
- wrmsrl(MSR_IA32_MC0_CTL+4*i, bank[i]);
+ wrmsrl(MSR_IA32_MC0_CTL+4*i, ~0UL);
wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0);
}
....
This change appears to ignore the fact that mce_cpu_quirks() clears a
bit in bank[4] for certain CPUs, as in:
static void __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c)
{
/* This should be disabled by the BIOS, but isn't always */
if (c->x86_vendor == X86_VENDOR_AMD) {
if(c->x86 == 15)
/* disable GART TBL walk error reporting, which trips off
incorrectly with the IOMMU & 3ware & Cerberus. */
clear_bit(10, &bank[4]);
....
Is turning off that bit still needed?
Max
--
| Jon Smirl | 463 kernel developers missing! |
| Nigel Cunningham | Re: [PATCH] Remove process freezer from suspend to RAM pathway |
| Greg KH | Re: [malware-list] [RFC 0/5] [TALPA] Intro to a linux interface for on access scan... |
| Jeff Garzik | Re: Linux 2.6.23-rc9 and a heads-up for the 2.6.24 series.. |
git: | |
| Jarek Poplawski | [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| Gerrit Renker | [PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side) |
| Linus Torvalds | Re: [GIT]: Networking |
| Evgeniy Polyakov | Re: [BUG] New Kernel Bugs |
