From: Marcelo Tosatti <[EMAIL PROTECTED]>
The PIIX4 ACPI controller prior to revision 0x3 contains a bug where
reading of the timer port is unreliable, so the kernel reads it three
times for consistency check.
QEMU does not suffer from that problem :)
The datasheet for PIIX4, PIIX4E, and PIIX4M is the same. I failed to
find any indication that the revision increase could affect anything
other than the PMTimer port read.
This reduces idle guest CPU consumption from 14% to 8% on 4-way KVM
guest.
Signed-off-by: Avi Kivity <[EMAIL PROTECTED]>
diff --git a/qemu/hw/acpi.c b/qemu/hw/acpi.c
index b97b37d..bfee40b 100644
--- a/qemu/hw/acpi.c
+++ b/qemu/hw/acpi.c
@@ -486,7 +486,7 @@ i2c_bus *piix4_pm_init(PCIBus *bus, int devfn, uint32_t
smb_io_base)
pci_conf[0x03] = 0x71;
pci_conf[0x06] = 0x80;
pci_conf[0x07] = 0x02;
- pci_conf[0x08] = 0x00; // revision number
+ pci_conf[0x08] = 0x03; // revision number
pci_conf[0x09] = 0x00;
pci_conf[0x0a] = 0x80; // other bridge device
pci_conf[0x0b] = 0x06; // bridge device
-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
kvm-commits mailing list
kvm-commits@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/kvm-commits
