From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To: Borislav Petkov <bp@alien8.de>, x86@kernel.org
Cc: "Ingo Molnar" <mingo@kernel.org>,
linux-kernel@vger.kernel.org, "Andy Lutomirski" <luto@kernel.org>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Radim Krčmář" <rkrcmar@redhat.com>,
kvm@vger.kernel.org, "Jason A. Donenfeld" <Jason@zx2c4.com>,
"Rik van Riel" <riel@surriel.com>,
"Dave Hansen" <dave.hansen@linux.intel.com>
Subject: [PATCH] x86/fpu: Disable BH while while loading FPU registers in __fpu__restore_sig()
Date: Mon, 12 Nov 2018 07:56:43 -0800 [thread overview]
Message-ID: <20181112155643.vclej44qzg3pmbow@linutronix.de> (raw)
In-Reply-To: <20181109232521.l2ll2n3coxygkxv4@linutronix.de>
The sequence
fpu->initialized = 1;
preempt_disable();
fpu__restore(fpu);
preempt_enable();
is racy in regard to a context switch. A context switch after the first
line would save the `actual' content of the FPU registers and trash away
the state that has been prepared (since fpu__drop()).
Use local_bh_disable() around the restore sequence to avoid the race. BH
needs to be disabled because BH is allowed to run (even with preemption
disabled) and might invoke kernel_fpu_begin().
This possible race has been reported by the Kernel Test Robot in FEB
2016 while there still was lazy FPU support.
Link: https://lkml.kernel.org/r/20160226074940.GA28911@pd.tnic
Cc: stable@vger.kernel.org
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
arch/x86/kernel/fpu/signal.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/fpu/signal.c b/arch/x86/kernel/fpu/signal.c
index 61a949d84dfa5..d99a8ee9e185e 100644
--- a/arch/x86/kernel/fpu/signal.c
+++ b/arch/x86/kernel/fpu/signal.c
@@ -344,10 +344,10 @@ static int __fpu__restore_sig(void __user *buf, void __user *buf_fx, int size)
sanitize_restored_xstate(tsk, &env, xfeatures, fx_only);
}
+ local_bh_disable();
fpu->initialized = 1;
- preempt_disable();
fpu__restore(fpu);
- preempt_enable();
+ local_bh_enable();
return err;
} else {
--
2.19.1
next prev parent reply other threads:[~2018-11-12 15:56 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-11-07 19:48 [PATCH v4] x86: load FPU registers on return to userland Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 01/23] x86/fpu: Use ULL for shift in xfeature_uncompacted_offset() Sebastian Andrzej Siewior
2018-11-08 11:38 ` Borislav Petkov
2018-11-09 16:56 ` Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 02/23] x86/fpu: Remove fpu->initialized usage in __fpu__restore_sig() Sebastian Andrzej Siewior
2018-11-08 14:57 ` Borislav Petkov
2018-11-09 17:35 ` Sebastian Andrzej Siewior
2018-11-09 18:52 ` Borislav Petkov
2018-11-09 23:25 ` Sebastian Andrzej Siewior
2018-11-12 15:56 ` Sebastian Andrzej Siewior [this message]
2018-11-12 17:48 ` [PATCH] x86/fpu: Disable BH while while loading FPU registers " Dave Hansen
2018-11-19 11:41 ` Sebastian Andrzej Siewior
2018-11-19 15:04 ` Dave Hansen
2018-11-19 15:06 ` Sebastian Andrzej Siewior
2018-11-19 15:08 ` Dave Hansen
2018-11-19 15:19 ` Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 03/23] x86/fpu: Remove fpu__restore() Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 04/23] x86/entry/32: Remove asm/math_emu.h include Sebastian Andrzej Siewior
2018-11-08 18:45 ` Andy Lutomirski
2018-11-07 19:48 ` [PATCH 05/23] x86/fpu: Remove preempt_disable() in fpu__clear() Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 06/23] x86/fpu: Always init the `state' " Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 07/23] x86/fpu: Remove fpu->initialized usage in copy_fpstate_to_sigframe() Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 08/23] x86/fpu: Remove fpu->initialized Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 09/23] x86/fpu: Remove user_fpu_begin() Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 10/23] x86/entry: Remove _TIF_ALLWORK_MASK Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 11/23] x86/fpu: Add (__)make_fpregs_active helpers Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 12/23] x86/fpu: Make __raw_xsave_addr() use feature number instead of mask Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 13/23] x86/fpu: Make get_xsave_field_ptr() and get_xsave_addr() " Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 14/23] x86/pkeys: Make init_pkru_value static Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 15/23] x86/fpu: Only write PKRU if it is different from current Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 16/23] x86/pkeys: Don't check if PKRU is zero before writting it Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 17/23] x86/fpu: Eager switch PKRU state Sebastian Andrzej Siewior
2018-11-08 11:12 ` Paolo Bonzini
2018-11-19 18:17 ` Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 18/23] x86/entry: Add TIF_NEED_FPU_LOAD Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 19/23] x86/fpu: Always store the registers in copy_fpstate_to_sigframe() Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 20/23] x86/fpu: Prepare copy_fpstate_to_sigframe() for TIF_NEED_FPU_LOAD Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 21/23] x86/fpu: Update xstate's PKRU value on write_pkru() Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 22/23] x86/fpu: Don't restore the FPU state directly from userland in __fpu__restore_sig() Sebastian Andrzej Siewior
2018-11-08 18:25 ` Andy Lutomirski
2018-11-09 19:09 ` Sebastian Andrzej Siewior
2018-11-07 19:48 ` [PATCH 23/23] x86/fpu: Defer FPU state load until return to userspace Sebastian Andrzej Siewior
2018-11-08 8:33 ` Sebastian Andrzej Siewior
2018-11-12 3:02 ` [PATCH v4] x86: load FPU registers on return to userland Wanpeng Li
2018-11-12 3:26 ` Jason A. Donenfeld
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=20181112155643.vclej44qzg3pmbow@linutronix.de \
--to=bigeasy@linutronix.de \
--cc=Jason@zx2c4.com \
--cc=bp@alien8.de \
--cc=dave.hansen@linux.intel.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@kernel.org \
--cc=pbonzini@redhat.com \
--cc=riel@surriel.com \
--cc=rkrcmar@redhat.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®