From: tip-bot for Brian Gerst <brgerst@gmail.com>
To: linux-tip-commits@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, hpa@zytor.com, mingo@redhat.com,
brgerst@gmail.com, penberg@kernel.org, suresh.b.siddha@intel.com,
tglx@linutronix.de, hpa@linux.intel.com
Subject: [tip:x86/fpu] x86-32, fpu: Rewrite fpu_save_init()
Date: Fri, 10 Sep 2010 01:49:41 GMT [thread overview]
Message-ID: <tip-58a992b9cbaf449aeebd3575c3695a9eb5d95b5e@git.kernel.org> (raw)
In-Reply-To: <1283563039-3466-12-git-send-email-brgerst@gmail.com>
Commit-ID: 58a992b9cbaf449aeebd3575c3695a9eb5d95b5e
Gitweb: http://git.kernel.org/tip/58a992b9cbaf449aeebd3575c3695a9eb5d95b5e
Author: Brian Gerst <brgerst@gmail.com>
AuthorDate: Fri, 3 Sep 2010 21:17:18 -0400
Committer: H. Peter Anvin <hpa@linux.intel.com>
CommitDate: Thu, 9 Sep 2010 14:17:31 -0700
x86-32, fpu: Rewrite fpu_save_init()
Rewrite fpu_save_init() to prepare for merging with 64-bit.
Signed-off-by: Brian Gerst <brgerst@gmail.com>
Acked-by: Pekka Enberg <penberg@kernel.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1283563039-3466-12-git-send-email-brgerst@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
arch/x86/include/asm/i387.h | 47 ++++++++++++++++++++----------------------
1 files changed, 22 insertions(+), 25 deletions(-)
diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h
index 907967e..b45abef 100644
--- a/arch/x86/include/asm/i387.h
+++ b/arch/x86/include/asm/i387.h
@@ -73,6 +73,11 @@ static __always_inline __pure bool use_xsave(void)
return static_cpu_has(X86_FEATURE_XSAVE);
}
+static __always_inline __pure bool use_fxsr(void)
+{
+ return static_cpu_has(X86_FEATURE_FXSR);
+}
+
extern void __sanitize_i387_state(struct task_struct *);
static inline void sanitize_i387_state(struct task_struct *tsk)
@@ -211,6 +216,12 @@ static inline int fxrstor_checking(struct i387_fxsave_struct *fx)
return 0;
}
+static inline void fpu_fxsave(struct fpu *fpu)
+{
+ asm volatile("fxsave %[fx]"
+ : [fx] "=m" (fpu->state->fxsave));
+}
+
/* We need a safe address that is cheap to find and that is already
in L1 during context switch. The best choices are unfortunately
different for UP and SMP */
@@ -226,36 +237,24 @@ static inline int fxrstor_checking(struct i387_fxsave_struct *fx)
static inline void fpu_save_init(struct fpu *fpu)
{
if (use_xsave()) {
- struct xsave_struct *xstate = &fpu->state->xsave;
- struct i387_fxsave_struct *fx = &fpu->state->fxsave;
-
fpu_xsave(fpu);
/*
* xsave header may indicate the init state of the FP.
*/
- if (!(xstate->xsave_hdr.xstate_bv & XSTATE_FP))
- goto end;
-
- if (unlikely(fx->swd & X87_FSW_ES))
- asm volatile("fnclex");
-
- /*
- * we can do a simple return here or be paranoid :)
- */
- goto clear_state;
+ if (!(fpu->state->xsave.xsave_hdr.xstate_bv & XSTATE_FP))
+ return;
+ } else if (use_fxsr()) {
+ fpu_fxsave(fpu);
+ } else {
+ asm volatile("fsave %[fx]; fwait"
+ : [fx] "=m" (fpu->state->fsave));
+ return;
}
- /* Use more nops than strictly needed in case the compiler
- varies code */
- alternative_input(
- "fnsave %[fx] ;fwait;" GENERIC_NOP8 GENERIC_NOP4,
- "fxsave %[fx]\n"
- "bt $7,%[fsw] ; jnc 1f ; fnclex\n1:",
- X86_FEATURE_FXSR,
- [fx] "m" (fpu->state->fxsave),
- [fsw] "m" (fpu->state->fxsave.swd) : "memory");
-clear_state:
+ if (unlikely(fpu->state->fxsave.swd & X87_FSW_ES))
+ asm volatile("fnclex");
+
/* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception
is pending. Clear the x87 state here by setting it to fixed
values. safe_address is a random variable that should be in L1 */
@@ -265,8 +264,6 @@ clear_state:
"fildl %[addr]", /* set F?P to defined value */
X86_FEATURE_FXSAVE_LEAK,
[addr] "m" (safe_address));
-end:
- ;
}
#endif /* CONFIG_X86_64 */
next prev parent reply other threads:[~2010-09-10 1:54 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-09-04 1:17 [PATCH 0/12] x86 FPU cleanups (v2) Brian Gerst
2010-09-04 1:17 ` [PATCH 01/12] x86: Use correct type for %cr4 Brian Gerst
2010-09-04 6:45 ` Pekka Enberg
2010-09-10 1:46 ` [tip:x86/fpu] " tip-bot for Brian Gerst
2010-09-04 1:17 ` [PATCH 02/12] x86: Merge fpu_init() Brian Gerst
2010-09-04 6:45 ` Pekka Enberg
2010-09-10 1:46 ` [tip:x86/fpu] x86, fpu: " tip-bot for Brian Gerst
2010-09-04 1:17 ` [PATCH 03/12] x86: Merge tolerant_fwait() Brian Gerst
2010-09-04 6:46 ` Pekka Enberg
2010-09-10 1:46 ` [tip:x86/fpu] x86, fpu: " tip-bot for Brian Gerst
2010-09-04 1:17 ` [PATCH 04/12] x86: Merge __save_init_fpu() Brian Gerst
2010-09-04 6:47 ` Pekka Enberg
2010-09-10 1:47 ` [tip:x86/fpu] x86, fpu: " tip-bot for Brian Gerst
2010-09-04 1:17 ` [PATCH 05/12] x86-64: Disable preemption when using TS_USEDFPU Brian Gerst
2010-09-04 6:47 ` Pekka Enberg
2010-09-10 1:47 ` [tip:x86/fpu] x86-64, fpu: " tip-bot for Brian Gerst
2010-09-04 1:17 ` [PATCH 06/12] x86-64: Fix %cs value in convert_from_fxsr() Brian Gerst
2010-09-04 6:51 ` Pekka Enberg
2010-09-10 1:47 ` [tip:x86/fpu] x86-64, fpu: " tip-bot for Brian Gerst
2010-09-04 1:17 ` [PATCH 07/12] x86-64: Simplify constraints for fxsave/fxtstor Brian Gerst
2010-09-04 6:48 ` Pekka Enberg
2010-09-09 21:22 ` H. Peter Anvin
2010-09-09 23:43 ` Brian Gerst
2010-09-10 1:48 ` [tip:x86/fpu] x86-64, fpu: " tip-bot for Brian Gerst
2010-09-04 1:17 ` [PATCH 08/12] x86-32: Remove math_emulate stub Brian Gerst
2010-09-04 6:48 ` Pekka Enberg
2010-09-10 1:48 ` [tip:x86/fpu] x86-32, fpu: " tip-bot for Brian Gerst
2010-09-04 1:17 ` [PATCH 09/12] x86: Remove unnecessary ifdefs from i387 code Brian Gerst
2010-09-04 6:49 ` Pekka Enberg
2010-09-10 1:48 ` [tip:x86/fpu] x86, fpu: " tip-bot for Brian Gerst
2010-09-04 1:17 ` [PATCH 10/12] x86: Remove PSHUFB_XMM5_* macros Brian Gerst
2010-09-04 6:49 ` Pekka Enberg
2010-09-10 1:49 ` [tip:x86/fpu] x86, fpu: " tip-bot for Brian Gerst
2010-09-04 1:17 ` [PATCH 11/12] x86-32: Rewrite fpu_save_init() Brian Gerst
2010-09-04 6:51 ` Pekka Enberg
2010-09-10 1:49 ` tip-bot for Brian Gerst [this message]
2010-09-04 1:17 ` [PATCH 12/12] x86: Merge fpu_save_init() Brian Gerst
2010-09-04 6:51 ` Pekka Enberg
2010-09-10 1:50 ` [tip:x86/fpu] x86, fpu: " tip-bot for Brian Gerst
2010-09-04 6:45 ` [PATCH 0/12] x86 FPU cleanups (v2) Pekka Enberg
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=tip-58a992b9cbaf449aeebd3575c3695a9eb5d95b5e@git.kernel.org \
--to=brgerst@gmail.com \
--cc=hpa@linux.intel.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tip-commits@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=penberg@kernel.org \
--cc=suresh.b.siddha@intel.com \
--cc=tglx@linutronix.de \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.