On Sun, 2006-04-30 at 14:59 +0200, Laurent Riffard wrote:
> [root@antares ~]# grep clocksource dmesg-2.6.17-rc*
> dmesg-2.6.17-rc1-mm1:Time: tsc clocksource has been installed.
> dmesg-2.6.17-rc1-mm1:Time: acpi_pm clocksource has been installed.
> dmesg-2.6.17-rc1-mm2:Time: tsc clocksource has been installed.
> dmesg-2.6.17-rc1-mm2:Time: acpi_pm clocksource has been installed.
> dmesg-2.6.17-rc1-mm3:Time: tsc clocksource has been installed.
> dmesg-2.6.17-rc1-mm3:Time: pit clocksource has been installed.
> dmesg-2.6.17-rc2-mm1:Time: tsc clocksource has been installed.
> dmesg-2.6.17-rc2-mm1:Time: pit clocksource has been installed.
>
> Is pit clocksource broken ? If so, how can I get back acpi_pm
> clocksource ?
Sorry, I still don't have a patch for the PIT problems you're seeing,
but I did track down why the ACPI PM disappeared.
It looks like its from the patch:
i386-x86-64-fix-acpi-disabled-lapic-handling.patch
The second chunk adds:
+ if (!cpu_has_apic)
+ return -ENODEV;
+
Right before we probe for the ACPI PM timer.
Andi, is there some way we can move that to after the ACPI PM probe?
thanks
-john
On Tuesday 02 May 2006 20:16, john stultz wrote:
> It looks like its from the patch:
> i386-x86-64-fix-acpi-disabled-lapic-handling.patch
>
>
> The second chunk adds:
>
> + if (!cpu_has_apic)
> + return -ENODEV;
> +
>
> Right before we probe for the ACPI PM timer.
>
>
> Andi, is there some way we can move that to after the ACPI PM probe?
Yes there was some merging trouble with this and some of the hunks
applied to the wrong places and I didn't remove the wrong one
in the first fixup patch. Sorry. This should fix it up.
Andrew, can you send that one to Linus please?
-Andi
Remove wrong cpu_has_apic checks that came from mismerging.
We only need to check cpu_has_apic in the IO-APIC/L-APIC parsing,
not for all of ACPI.
Signed-off-by: Andi Kleen <[email protected]>
Index: linux/arch/i386/kernel/acpi/boot.c
===================================================================
--- linux.orig/arch/i386/kernel/acpi/boot.c
+++ linux/arch/i386/kernel/acpi/boot.c
@@ -1102,9 +1102,6 @@ int __init acpi_boot_table_init(void)
dmi_check_system(acpi_dmi_table);
#endif
- if (!cpu_has_apic)
- return -ENODEV;
-
/*
* If acpi_disabled, bail out
* One exception: acpi=ht continues far enough to enumerate LAPICs
@@ -1151,9 +1148,6 @@ int __init acpi_boot_init(void)
acpi_table_parse(ACPI_BOOT, acpi_parse_sbf);
- if (!cpu_has_apic)
- return -ENODEV;
-
/*
* set sci_int and PM timer address
*/
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Le 02.05.2006 20:29, Andi Kleen a écrit :
> On Tuesday 02 May 2006 20:16, john stultz wrote:
>
>> It looks like its from the patch:
>> i386-x86-64-fix-acpi-disabled-lapic-handling.patch
>>
>>
>> The second chunk adds:
>>
>> + if (!cpu_has_apic)
>> + return -ENODEV;
>> +
>>
>> Right before we probe for the ACPI PM timer.
>>
>>
>> Andi, is there some way we can move that to after the ACPI PM probe?
>
>
> Yes there was some merging trouble with this and some of the hunks
> applied to the wrong places and I didn't remove the wrong one
> in the first fixup patch. Sorry. This should fix it up.
>
> Andrew, can you send that one to Linus please?
>
> -Andi
>
> Remove wrong cpu_has_apic checks that came from mismerging.
>
> We only need to check cpu_has_apic in the IO-APIC/L-APIC parsing,
> not for all of ACPI.
>
> Signed-off-by: Andi Kleen <[email protected]>
>
> Index: linux/arch/i386/kernel/acpi/boot.c
> ===================================================================
> --- linux.orig/arch/i386/kernel/acpi/boot.c
> +++ linux/arch/i386/kernel/acpi/boot.c
> @@ -1102,9 +1102,6 @@ int __init acpi_boot_table_init(void)
> dmi_check_system(acpi_dmi_table);
> #endif
>
> - if (!cpu_has_apic)
> - return -ENODEV;
> -
> /*
> * If acpi_disabled, bail out
> * One exception: acpi=ht continues far enough to enumerate LAPICs
> @@ -1151,9 +1148,6 @@ int __init acpi_boot_init(void)
>
> acpi_table_parse(ACPI_BOOT, acpi_parse_sbf);
>
> - if (!cpu_has_apic)
> - return -ENODEV;
> -
> /*
> * set sci_int and PM timer address
> */
Hey, nice! acpi_pm clocksource came back!
FYI, here's a diff of dmesg output:
- --- /tmp/dmesg-2.6.17-rc3-mm1 2006-05-02 19:13:18.000000000 +0200
+++ /tmp/dmesg-2.6.17-rc3-mm1-patched 2006-05-02 20:52:54.000000000 +0200
@@ -1,4 +1,4 @@
- -Linux version 2.6.17-rc3-mm1 ([email protected]) (gcc version 4.0.3 (4.0.3-0.20060215.2mdk for Mandriva Linux release 2006.1)) #13 Mon May 1 23:37:11 CEST 2006
+Linux version 2.6.17-rc3-mm1 ([email protected]) (gcc version 4.0.3 (4.0.3-0.20060215.2mdk for Mandriva Linux release 2006.1)) #14 Tue May 2 20:39:28 CEST 2006
BIOS-provided physical RAM map:
sanitize start
sanitize end
@@ -33,8 +33,14 @@ On node 0 totalpages: 131052
DMA zone: 4096 pages, LIFO batch:0
Normal zone: 126956 pages, LIFO batch:31
DMI 2.3 present.
+ACPI: RSDP (v000 ASUS ) @ 0x000f6a80
+ACPI: RSDT (v001 ASUS A7V133-C 0x30303031 MSFT 0x31313031) @ 0x1ffec000
+ACPI: FADT (v001 ASUS A7V133-C 0x30303031 MSFT 0x31313031) @ 0x1ffec080
+ACPI: BOOT (v001 ASUS A7V133-C 0x30303031 MSFT 0x31313031) @ 0x1ffec040
+ACPI: DSDT (v001 ASUS A7V133-C 0x00001000 MSFT 0x0100000b) @ 0x00000000
+ACPI: PM-Timer IO Port: 0xe408
Allocating PCI resources starting at 30000000 (gap: 20000000:dfff0000)
- -Detected 1410.372 MHz processor.
+Detected 1410.227 MHz processor.
Built 1 zonelists
Kernel command line: root=/dev/vglinux1/lvroot video=vesafb:ywrap,mtrr splash=silent resume=/dev/hdb6 [email protected]/,@192.168.0.1/00:0E:9B:91:ED:72
netconsole: local port 6665
@@ -55,7 +61,7 @@ Dentry cache hash table entries: 65536 (
Inode-cache hash table entries: 32768 (order: 5, 131072 bytes)
Memory: 516072k/524208k available (1497k kernel code, 7572k reserved, 949k data, 160k init, 0k highmem)
Checking if this processor honours the WP bit even in supervisor mode... Ok.
- -Calibrating delay using timer specific routine.. 2823.26 BogoMIPS (lpj=5646521)
+Calibrating delay using timer specific routine.. 2823.18 BogoMIPS (lpj=5646365)
Mount-cache hash table entries: 512
CPU: After generic identify, caps: 0383f9ff c1cbf9ff 00000000 00000000 00000000 00000000 00000000
CPU: After vendor identify, caps: 0383f9ff c1cbf9ff 00000000 00000000 00000000 00000000 00000000
@@ -113,6 +119,7 @@ TCP established hash table entries: 1638
TCP bind hash table entries: 8192 (order: 3, 32768 bytes)
TCP: Hash tables configured (established 16384 bind 8192)
TCP reno registered
+Simple Boot Flag at 0x3a set to 0x1
Initializing Cryptographic API
io scheduler noop registered
io scheduler anticipatory registered
@@ -166,7 +173,7 @@ TCP bic registered
NET: Registered protocol family 1
Using IPI Shortcut mode
Time: tsc clocksource has been installed.
- -Time: pit clocksource has been installed.
+Time: acpi_pm clocksource has been installed.
ACPI: (supports S0 S1 S3 S4 S5)
BIOS EDD facility v0.16 2004-Jun-25, 2 devices found
RAMDISK: Compressed image found at block 0
Thanks
- --
laurent
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFEV6yLUqUFrirTu6IRAls0AJ4geYsJQFykMbSzzCtUrlPOzy10FACcCuHN
1xqN6vxpSF9n5Je7JSMY8GY=
=rU08
-----END PGP SIGNATURE-----
On Tue, 02 May 2006 20:29:04 +0200, Andi Kleen said:
> Remove wrong cpu_has_apic checks that came from mismerging.
>
> We only need to check cpu_has_apic in the IO-APIC/L-APIC parsing,
> not for all of ACPI.
>
> Signed-off-by: Andi Kleen <[email protected]>
>
> Index: linux/arch/i386/kernel/acpi/boot.c
> ===================================================================
NTP is much happier now that it has an ACPI-PM clock source that has a
drift of 8.5ppm rather than a PIT clocksource that has a drift of 500+ppm. ;)