From: Mel Gorman <mgorman@suse.de>
To: H Peter Anvin <hpa@zytor.com>, Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Mike Galbraith <mgalbraith@suse.de>, Linux-X86 <x86@kernel.org>,
LKML <linux-kernel@vger.kernel.org>, Mel Gorman <mgorman@suse.de>
Subject: [PATCH 1/3] x86, fpu: Do not copy fpu preload state
Date: Wed, 6 Aug 2014 13:55:46 +0100 [thread overview]
Message-ID: <1407329748-3928-2-git-send-email-mgorman@suse.de> (raw)
In-Reply-To: <1407329748-3928-1-git-send-email-mgorman@suse.de>
__switch_to declares a fpu_switch_t on the stack then calls a prepare
function which declares a second one on the stack and copies the result
back. As the functions are inline the compiler will optimise it correctly
but it looks weird. This patch makes it clear that a single instance of
fpu_switch_t is used.
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
arch/x86/include/asm/fpu-internal.h | 14 ++++++--------
arch/x86/kernel/process_32.c | 2 +-
arch/x86/kernel/process_64.c | 2 +-
3 files changed, 8 insertions(+), 10 deletions(-)
diff --git a/arch/x86/include/asm/fpu-internal.h b/arch/x86/include/asm/fpu-internal.h
index 115e368..b8771c4a 100644
--- a/arch/x86/include/asm/fpu-internal.h
+++ b/arch/x86/include/asm/fpu-internal.h
@@ -416,15 +416,14 @@ static inline int fpu_lazy_restore(struct task_struct *new, unsigned int cpu)
cpu == new->thread.fpu.last_cpu;
}
-static inline fpu_switch_t switch_fpu_prepare(struct task_struct *old, struct task_struct *new, int cpu)
+static inline void switch_fpu_prepare(struct task_struct *old, struct task_struct *new, int cpu,
+ fpu_switch_t *fpu)
{
- fpu_switch_t fpu;
-
/*
* If the task has used the math, pre-load the FPU on xsave processors
* or if the past 5 consecutive context-switches used math.
*/
- fpu.preload = tsk_used_math(new) && (use_eager_fpu() ||
+ fpu->preload = tsk_used_math(new) && (use_eager_fpu() ||
new->thread.fpu_counter > 5);
if (__thread_has_fpu(old)) {
if (!__save_init_fpu(old))
@@ -433,7 +432,7 @@ static inline fpu_switch_t switch_fpu_prepare(struct task_struct *old, struct ta
old->thread.fpu.has_fpu = 0; /* But leave fpu_owner_task! */
/* Don't change CR0.TS if we just switch! */
- if (fpu.preload) {
+ if (fpu->preload) {
new->thread.fpu_counter++;
__thread_set_has_fpu(new);
prefetch(new->thread.fpu.state);
@@ -442,16 +441,15 @@ static inline fpu_switch_t switch_fpu_prepare(struct task_struct *old, struct ta
} else {
old->thread.fpu_counter = 0;
old->thread.fpu.last_cpu = ~0;
- if (fpu.preload) {
+ if (fpu->preload) {
new->thread.fpu_counter++;
if (!use_eager_fpu() && fpu_lazy_restore(new, cpu))
- fpu.preload = 0;
+ fpu->preload = 0;
else
prefetch(new->thread.fpu.state);
__thread_fpu_begin(new);
}
}
- return fpu;
}
/*
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 7bc86bb..260ed717 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -257,7 +257,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
/* never put a printk in __switch_to... printk() calls wake_up*() indirectly */
- fpu = switch_fpu_prepare(prev_p, next_p, cpu);
+ switch_fpu_prepare(prev_p, next_p, cpu, &fpu);
/*
* Reload esp0.
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index ca5b02d..b155462 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -284,7 +284,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
unsigned fsindex, gsindex;
fpu_switch_t fpu;
- fpu = switch_fpu_prepare(prev_p, next_p, cpu);
+ switch_fpu_prepare(prev_p, next_p, cpu, &fpu);
/*
* Reload esp0, LDT and the page table pointer:
--
1.8.4.5
next prev parent reply other threads:[~2014-08-06 12:56 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-08-06 12:55 [PATCH 0/3] Reduce length of the eagerfpu path during x86 context switches Mel Gorman
2014-08-06 12:55 ` Mel Gorman [this message]
2014-08-06 12:55 ` [PATCH 2/3] x86, fpu: Split FPU save state preparation into eagerfpu and !eagerfpu parts Mel Gorman
2014-08-06 12:55 ` [PATCH 3/3] x86, fpu: Do not update fpu_counter in the eagerfpu case Mel Gorman
2014-08-27 16:03 ` [PATCH 0/3] Reduce length of the eagerfpu path during x86 context switches Mel Gorman
2014-08-27 17:03 ` H. Peter Anvin
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=1407329748-3928-2-git-send-email-mgorman@suse.de \
--to=mgorman@suse.de \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgalbraith@suse.de \
--cc=suresh.b.siddha@intel.com \
--cc=x86@kernel.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
all inboxes | Powered by JetHome®