From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
To: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Cc: Xen-devel <xen-devel@lists.xensource.com>,
kurt.hackel@oracle.com,
Glauber de Oliveira Costa <gcosta@redhat.com>,
Avi Kivity <avi@redhat.com>,
the arch/x86 maintainers <x86@kernel.org>,
Chris Mason <chris.mason@oracle.com>,
Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>,
Ingo Molnar <mingo@elte.hu>
Subject: [PATCH 02/12] x86/vgetcpu: ignore tcache in common code.
Date: Wed, 14 Oct 2009 12:28:26 -0700 [thread overview]
Message-ID: <1255548516-15260-3-git-send-email-jeremy.fitzhardinge@citrix.com> (raw)
In-Reply-To: <1255548516-15260-1-git-send-email-jeremy.fitzhardinge@citrix.com>
The vdso implementation of vgetcpu has been ignoring the tcache
parameter since change 4307d1e5ada595c87f9a4d16db16ba5edb70dcb1,
but that overlooked the vsyscall implementation of vgetcpu.
This change ignores tcache in the common code, but passes the
pointer in from vdso so that it can be still used if there's a
new tcache implementation which addresses the problems mentioned
in 4307d1e5ada595c87f.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Ingo Molnar <mingo@elte.hu>
---
arch/x86/include/asm/vsyscall.h | 19 ++++---------------
arch/x86/vdso/vgetcpu.c | 3 +--
2 files changed, 5 insertions(+), 17 deletions(-)
diff --git a/arch/x86/include/asm/vsyscall.h b/arch/x86/include/asm/vsyscall.h
index 2fcd505..bb90047 100644
--- a/arch/x86/include/asm/vsyscall.h
+++ b/arch/x86/include/asm/vsyscall.h
@@ -48,27 +48,16 @@ int __vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache,
unsigned int p;
unsigned long j = 0;
- /* Fast cache - only recompute value once per jiffies and avoid
- relatively costly rdtscp/cpuid otherwise.
- This works because the scheduler usually keeps the process
- on the same CPU and this syscall doesn't guarantee its
- results anyways.
- We do this here because otherwise user space would do it on
- its own in a likely inferior way (no access to jiffies).
- If you don't like it pass NULL. */
- if (tcache && tcache->blob[0] == (j = jiffies)) {
- p = tcache->blob[1];
- } else if (vgetcpu_mode == VGETCPU_RDTSCP) {
+ /* tcache is ignored - it is too unreliable */
+
+ if (vgetcpu_mode == VGETCPU_RDTSCP) {
/* Load per CPU data from RDTSCP */
native_read_tscp(&p);
} else {
/* Load per CPU data from GDT */
asm("lsl %1,%0" : "=r" (p) : "r" (__PER_CPU_SEG));
}
- if (tcache) {
- tcache->blob[0] = j;
- tcache->blob[1] = p;
- }
+
if (cpu)
*cpu = p & 0xfff;
if (node)
diff --git a/arch/x86/vdso/vgetcpu.c b/arch/x86/vdso/vgetcpu.c
index 1f19f74..2c4f9ab 100644
--- a/arch/x86/vdso/vgetcpu.c
+++ b/arch/x86/vdso/vgetcpu.c
@@ -16,8 +16,7 @@
notrace long
__vdso_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
{
- /* Ignore tcache to preserve old behaviour */
- return __vgetcpu(cpu, node, NULL, *vdso_jiffies, *vdso_vgetcpu_mode);
+ return __vgetcpu(cpu, node, tcache, *vdso_jiffies, *vdso_vgetcpu_mode);
}
long getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
--
1.6.2.5
next prev parent reply other threads:[~2009-10-14 19:29 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-14 19:28 [GIT PULL RFC] pvclock cleanups and pvclock vsyscall support Jeremy Fitzhardinge
2009-10-14 19:28 ` [PATCH 01/12] x86/vsyscall: use common implementation for vgetcpu Jeremy Fitzhardinge
2009-10-14 19:28 ` Jeremy Fitzhardinge [this message]
2009-10-14 19:28 ` [PATCH 03/12] x86/vsyscall: use common implementation for vgettimeofday Jeremy Fitzhardinge
2009-10-14 19:28 ` [PATCH 04/12] kvm/pvclock: add monotonicity check Jeremy Fitzhardinge
2009-10-14 19:28 ` [PATCH 05/12] xen/pvclock: " Jeremy Fitzhardinge
2009-10-15 3:26 ` [Xen-devel] " Dan Magenheimer
2009-10-15 5:24 ` Jeremy Fitzhardinge
2009-10-15 13:27 ` Dan Magenheimer
2009-10-15 19:19 ` Jeremy Fitzhardinge
2009-10-16 1:32 ` john stultz
2009-10-16 3:10 ` Jeremy Fitzhardinge
2009-10-16 17:58 ` john stultz
2009-10-18 8:18 ` Jeremy Fitzhardinge
2009-10-14 19:28 ` [PATCH 06/12] x86/pvclock: make sure rdtsc doesn't speculate out of region Jeremy Fitzhardinge
2009-10-14 19:28 ` [PATCH 07/12] pvclock: there's no need to copy time_info into shadow Jeremy Fitzhardinge
2009-10-14 19:28 ` [PATCH 08/12] x86: create specific X86_VSYSCALL config variable Jeremy Fitzhardinge
2009-10-14 19:28 ` [PATCH 09/12] sched: add notifier for cross-cpu migrations Jeremy Fitzhardinge
2009-10-14 19:28 ` [PATCH 10/12] x86/pvclock: add vsyscall implementation Jeremy Fitzhardinge
2009-10-14 19:28 ` [PATCH 11/12] x86/fixmap: add a predicate for usermode fixmaps Jeremy Fitzhardinge
2009-10-14 19:28 ` [PATCH 12/12] xen/time: add pvclock_clocksource_vread support Jeremy Fitzhardinge
2009-10-15 6:51 ` [GIT PULL RFC] pvclock cleanups and pvclock vsyscall support Avi Kivity
2009-10-15 19:46 ` Jeremy Fitzhardinge
2009-10-18 6:43 ` Avi Kivity
2009-10-18 8:18 ` Jeremy Fitzhardinge
2009-10-18 8:23 ` Avi Kivity
2009-10-18 8:44 ` Jeremy Fitzhardinge
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=1255548516-15260-3-git-send-email-jeremy.fitzhardinge@citrix.com \
--to=jeremy.fitzhardinge@citrix.com \
--cc=avi@redhat.com \
--cc=chris.mason@oracle.com \
--cc=gcosta@redhat.com \
--cc=kurt.hackel@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=x86@kernel.org \
--cc=xen-devel@lists.xensource.com \
/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
all inboxes | Powered by JetHome®