From: Greg KH <gregkh@suse.de>
To: linux-kernel@vger.kernel.org, stable@kernel.org
Cc: stable-review@kernel.org, torvalds@linux-foundation.org,
akpm@linux-foundation.org, alan@lxorguk.ukuu.org.uk,
Andreas Herrmann <andreas.herrmann3@amd.com>,
Ingo Molnar <mingo@elte.hu>
Subject: [patch 24/35] x86: Set cpu_llc_id on AMD CPUs
Date: Tue, 30 Jun 2009 17:14:20 -0700 [thread overview]
Message-ID: <20090701001552.431397625@mini.kroah.org> (raw)
In-Reply-To: <20090701002825.GA6518@kroah.com>
[-- Attachment #1: x86-set-cpu_llc_id-on-amd-cpus.patch --]
[-- Type: text/plain, Size: 2059 bytes --]
2.6.29-stable review patch. If anyone has any objections, please let us know.
------------------
From: Andreas Herrmann <andreas.herrmann3@amd.com>
commit 99bd0c0fc4b04da54cb311953ef9489931c19c63 upstream.
This counts when building sched domains in case NUMA information
is not available.
( See cpu_coregroup_mask() which uses llc_shared_map which in turn is
created based on cpu_llc_id. )
Currently Linux builds domains as follows:
(example from a dual socket quad-core system)
CPU0 attaching sched-domain:
domain 0: span 0-7 level CPU
groups: 0 1 2 3 4 5 6 7
...
CPU7 attaching sched-domain:
domain 0: span 0-7 level CPU
groups: 7 0 1 2 3 4 5 6
Ever since that is borked for multi-core AMD CPU systems.
This patch fixes that and now we get a proper:
CPU0 attaching sched-domain:
domain 0: span 0-3 level MC
groups: 0 1 2 3
domain 1: span 0-7 level CPU
groups: 0-3 4-7
...
CPU7 attaching sched-domain:
domain 0: span 4-7 level MC
groups: 7 4 5 6
domain 1: span 0-7 level CPU
groups: 4-7 0-3
This allows scheduler to assign tasks to cores on different sockets
(i.e. that don't share last level cache) for performance reasons.
Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
LKML-Reference: <20090619085909.GJ5218@alberich.amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
arch/x86/kernel/cpu/amd.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -207,13 +207,15 @@ static void __cpuinit amd_detect_cmp(str
{
#ifdef CONFIG_X86_HT
unsigned bits;
+ int cpu = smp_processor_id();
bits = c->x86_coreid_bits;
-
/* Low order bits define the core id (index of core in socket) */
c->cpu_core_id = c->initial_apicid & ((1 << bits)-1);
/* Convert the initial APIC ID into the socket ID */
c->phys_proc_id = c->initial_apicid >> bits;
+ /* use socket ID also for last level cache */
+ per_cpu(cpu_llc_id, cpu) = c->phys_proc_id;
#endif
}
next prev parent reply other threads:[~2009-07-01 0:44 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20090701001356.007288418@mini.kroah.org>
2009-07-01 0:28 ` [patch 00/35] 2.6.29-stable review Greg KH
2009-07-01 0:13 ` [patch 01/35] parport: netmos 9845 & 9855 1P4S fixes Greg KH
2009-07-01 0:13 ` [patch 02/35] 8250: Fix oops from setserial Greg KH
2009-07-01 0:13 ` [patch 03/35] char: mxser, fix ISA board lookup Greg KH
2009-07-01 0:14 ` [patch 04/35] jbd: fix race in buffer processing in commit code Greg KH
2009-07-01 0:14 ` [patch 05/35] r8169: fix crash when large packets are received Greg KH
2009-07-01 0:14 ` [patch 06/35] fs: remove incorrect I_NEW warnings Greg KH
2009-07-01 0:14 ` [patch 07/35] firmware_map: fix hang with x86/32bit Greg KH
2009-07-01 0:14 ` [patch 08/35] PCI: disable ASPM on VIA root-port-under-bridge configurations Greg KH
2009-07-01 0:14 ` [patch 09/35] atkbd: add forced release quirks for four more keyboard models Greg KH
2009-07-01 0:14 ` [patch 10/35] atmel_lcdfb: correct fifo size for some products Greg KH
2009-07-01 0:14 ` [patch 11/35] bonding: fix multiple module load problem Greg KH
2009-07-01 0:14 ` [patch 12/35] char: moxa, prevent opening unavailable ports Greg KH
2009-07-01 0:14 ` [patch 13/35] ISDN: Fix DMA alloc for hfcpci Greg KH
2009-07-01 0:14 ` [patch 14/35] USB: usbtmc: fix switch statment Greg KH
2009-07-01 0:14 ` [patch 15/35] x86: Add quirk for reboot stalls on a Dell Optiplex 360 Greg KH
2009-07-01 0:14 ` [patch 16/35] ALSA: ca0106 - Add missing registrations of vmaster controls Greg KH
2009-07-01 0:14 ` [patch 17/35] floppy: provide a PNP device table in the module Greg KH
2009-07-01 0:14 ` [patch 18/35] IB/mlx4: Add strong ordering to local inval and fast reg work requests Greg KH
2009-07-01 0:14 ` [patch 19/35] x86: handle initrd that extends into unusable memory Greg KH
2009-07-01 0:14 ` [patch 20/35] lockdep: Select frame pointers on x86 Greg KH
2009-07-01 0:14 ` [patch 21/35] mac80211: fix minstrel single-rate memory corruption Greg KH
2009-07-01 0:14 ` [patch 22/35] md/raid5: add missing call to schedule() after prepare_to_wait() Greg KH
2009-07-01 0:14 ` [patch 23/35] vt_ioctl: fix lock imbalance Greg KH
2009-07-01 0:14 ` Greg KH [this message]
2009-07-01 0:14 ` [patch 25/35] parport_pc: after superio probing restore original register values Greg KH
2009-07-01 0:14 ` [patch 26/35] parport_pc: set properly the dma_mask for parport_pc device Greg KH
2009-07-01 0:14 ` [patch 27/35] PCI PM: Fix handling of devices without PM support by pci_target_state() Greg KH
2009-07-01 0:14 ` [patch 28/35] PCI PM: Follow PCI_PM_CTRL_NO_SOFT_RESET during transitions from D3 Greg KH
2009-07-01 0:14 ` [patch 29/35] pcmcia/cm4000: fix lock imbalance Greg KH
2009-07-01 0:14 ` [patch 30/35] qla2xxx: Correct (again) overflow during dump-processing on large-memory ISP23xx parts Greg KH
2009-07-01 0:14 ` [patch 31/35] sound: seq_midi_event: fix decoding of (N)RPN events Greg KH
2009-07-01 0:14 ` [patch 32/35] mm: fix handling of pagesets for downed cpus Greg KH
2009-07-01 0:14 ` [patch 33/35] dm mpath: validate table argument count Greg KH
2009-07-01 0:14 ` [patch 34/35] dm mpath: validate hw_handler " Greg KH
2009-07-01 0:14 ` [patch 35/35] dm: sysfs skip output when device is being destroyed Greg KH
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20090701001552.431397625@mini.kroah.org \
--to=gregkh@suse.de \
--cc=akpm@linux-foundation.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=andreas.herrmann3@amd.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=stable-review@kernel.org \
--cc=stable@kernel.org \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox