From: Chuck Ebbert <76306.1226@compuserve.com>
To: Andi Kleen <ak@suse.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>,
Arjan van de Ven <arjan@infradead.org>,
Ingo Molnar <mingo@elte.hu>, Adrian Bunk <bunk@stusta.de>,
linux-kernel <linux-kernel@vger.kernel.org>,
Linus Torvalds <torvalds@osdl.org>, andrea <andrea@cpushare.com>
Subject: Re: [patch] let CONFIG_SECCOMP default to n
Date: Wed, 12 Jul 2006 17:37:08 -0400 [thread overview]
Message-ID: <200607121739_MC3-1-C4D3-28B9@compuserve.com> (raw)
In-Reply-To: <p73wtain80h.fsf@verdi.suse.de>
On 12 Jul 2006 17:43:42 +0200, Andi Kleen wrote:
Alan Cox <alan@lxorguk.ukuu.org.uk> writes:
> >
> > I really don't care about cpushare and patents for some users of the
> > code in question. On the other hand turning on performance harming code
> > for a tiny number of users is dumb. If it were a loadable module it
> > would be different.
>
> Actually there are some promising applications of seccomp outside
> cpushare.
>
> e.g. Andrea at some point proposed to run codecs which often
> have security issues in a simple cpusec jail. That's ok for
> them because they normally don't need to do any system calls.
>
> I liked the idea. While this can be done with LSM (e.g. apparmor) too
> seccomp is definitely much easier and simpler and more "obviously safe"
> than anything LSM based.
>
> If the TSC disabling code is taken out the runtime overhead
> of seccomp is also very small because it's only tested in slow
> paths.
We can just fold the TSC disable stuff into the new thread_flags test
at context-switch time:
[compile tested only; requires just-sent fix to i386 system.h]
arch/i386/kernel/process.c | 61 ++++++++++++++---------------------------
include/asm-i386/thread_info.h | 3 +-
2 files changed, 24 insertions(+), 40 deletions(-)
--- 2.6.18-rc1-nb.orig/arch/i386/kernel/process.c
+++ 2.6.18-rc1-nb/arch/i386/kernel/process.c
@@ -535,12 +535,11 @@ int dump_task_regs(struct task_struct *t
return 1;
}
-static noinline void __switch_to_xtra(struct task_struct *next_p,
- struct tss_struct *tss)
+static noinline void
+__switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
+ struct tss_struct *tss)
{
- struct thread_struct *next;
-
- next = &next_p->thread;
+ struct thread_struct *next = &next_p->thread;
if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
set_debugreg(next->debugreg[0], 0);
@@ -552,6 +551,19 @@ static noinline void __switch_to_xtra(st
set_debugreg(next->debugreg[7], 7);
}
+ /*
+ * Context switch may need to tweak the TSC disable bit in CR4.
+ * The optimizer should remove this code when !CONFIG_SECCOMP.
+ */
+ if (has_secure_computing(task_thread_info(prev_p)) ^
+ has_secure_computing(task_thread_info(next_p))) {
+ /* prev and next are different */
+ if (has_secure_computing(task_thread_info(next_p)))
+ write_cr4(read_cr4() | X86_CR4_TSD);
+ else
+ write_cr4(read_cr4() & ~X86_CR4_TSD);
+ }
+
if (!test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) {
/*
* Disable the bitmap via an invalid offset. We still cache
@@ -561,7 +573,7 @@ static noinline void __switch_to_xtra(st
return;
}
- if (likely(next == tss->io_bitmap_owner)) {
+ if (likely(tss->io_bitmap_owner == next)) {
/*
* Previous owner of the bitmap (hence the bitmap content)
* matches the next task, we dont have to do anything but
@@ -583,33 +595,6 @@ static noinline void __switch_to_xtra(st
}
/*
- * This function selects if the context switch from prev to next
- * has to tweak the TSC disable bit in the cr4.
- */
-static inline void disable_tsc(struct task_struct *prev_p,
- struct task_struct *next_p)
-{
- struct thread_info *prev, *next;
-
- /*
- * gcc should eliminate the ->thread_info dereference if
- * has_secure_computing returns 0 at compile time (SECCOMP=n).
- */
- prev = task_thread_info(prev_p);
- next = task_thread_info(next_p);
-
- if (has_secure_computing(prev) || has_secure_computing(next)) {
- /* slow path here */
- if (has_secure_computing(prev) &&
- !has_secure_computing(next)) {
- write_cr4(read_cr4() & ~X86_CR4_TSD);
- } else if (!has_secure_computing(prev) &&
- has_secure_computing(next))
- write_cr4(read_cr4() | X86_CR4_TSD);
- }
-}
-
-/*
* switch_to(x,yn) should switch tasks from x to y.
*
* We fsave/fwait so that an exception goes off at the right time
@@ -688,13 +673,11 @@ struct task_struct fastcall * __switch_t
set_iopl_mask(next->iopl);
/*
- * Now maybe handle debug registers and/or IO bitmaps
+ * Now maybe handle debug registers, IO bitmaps, TSC disable
*/
- if (unlikely((task_thread_info(next_p)->flags & _TIF_WORK_CTXSW))
- || test_tsk_thread_flag(prev_p, TIF_IO_BITMAP))
- __switch_to_xtra(next_p, tss);
-
- disable_tsc(prev_p, next_p);
+ if (unlikely(task_thread_info(prev_p)->flags & _TIF_WORK_CTXSW_PREV ||
+ task_thread_info(next_p)->flags & _TIF_WORK_CTXSW_NEXT))
+ __switch_to_xtra(prev_p, next_p, tss);
return prev_p;
}
--- 2.6.18-rc1-nb.orig/include/asm-i386/thread_info.h
+++ 2.6.18-rc1-nb/include/asm-i386/thread_info.h
@@ -164,7 +164,8 @@ static inline struct thread_info *curren
#define _TIF_ALLWORK_MASK (0x0000FFFF & ~_TIF_SECCOMP)
/* flags to check in __switch_to() */
-#define _TIF_WORK_CTXSW (_TIF_DEBUG|_TIF_IO_BITMAP)
+#define _TIF_WORK_CTXSW_NEXT (_TIF_IO_BITMAP | _TIF_SECCOMP | _TIF_DEBUG)
+#define _TIF_WORK_CTXSW_PREV (_TIF_IO_BITMAP | _TIF_SECCOMP)
/*
* Thread-synchronous status.
--
Chuck
"You can't read a newspaper if you can't read." --George W. Bush
next reply other threads:[~2006-07-12 21:43 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-07-12 21:37 Chuck Ebbert [this message]
2006-07-12 21:55 ` Linus Torvalds
2006-07-12 22:48 ` andrea
2006-07-12 21:57 ` Andi Kleen
2006-07-12 22:14 ` [patch] let CONFIG_SECCOMP default to n II Andi Kleen
[not found] <6tgj0-8ip-19@gated-at.bofh.it>
[not found] ` <6xP8s-5mc-9@gated-at.bofh.it>
[not found] ` <6xUhQ-4Wx-33@gated-at.bofh.it>
[not found] ` <6xVdX-6oH-53@gated-at.bofh.it>
[not found] ` <6xVnz-6AI-21@gated-at.bofh.it>
[not found] ` <6xZUd-4Es-13@gated-at.bofh.it>
[not found] ` <6y7yy-7ws-13@gated-at.bofh.it>
[not found] ` <6y7RK-7TX-9@gated-at.bofh.it>
2006-07-17 11:37 ` [patch] let CONFIG_SECCOMP default to n Bodo Eggert
-- strict thread matches above, loose matches on Subject: below --
2006-07-13 5:43 Albert Cahalan
2006-07-13 7:07 ` andrea
2006-06-29 19:21 [2.6 patch] " Adrian Bunk
2006-06-30 0:44 ` Lee Revell
2006-06-30 1:07 ` Andrew Morton
2006-06-30 1:40 ` Adrian Bunk
2006-06-30 4:52 ` Andrea Arcangeli
2006-06-30 9:47 ` Ingo Molnar
2006-06-30 14:58 ` andrea
2006-07-11 7:36 ` [patch] " Ingo Molnar
2006-07-11 14:17 ` andrea
2006-07-11 14:32 ` Arjan van de Ven
2006-07-11 15:31 ` andrea
2006-07-11 15:54 ` Arjan van de Ven
2006-07-11 16:13 ` andrea
2006-07-11 16:23 ` Arjan van de Ven
2006-07-11 16:57 ` Alan Cox
2006-07-11 16:25 ` Alan Cox
2006-07-11 16:02 ` Adrian Bunk
2006-07-11 16:16 ` andrea
2006-07-11 16:24 ` Alan Cox
2006-07-12 15:43 ` Andi Kleen
2006-07-12 21:07 ` Ingo Molnar
2006-07-12 22:06 ` Andi Kleen
2006-07-12 22:19 ` Ingo Molnar
2006-07-12 22:33 ` Andi Kleen
2006-07-12 22:49 ` Ingo Molnar
2006-07-13 3:16 ` Andrea Arcangeli
2006-07-13 11:23 ` Jeff Dike
2006-07-13 11:35 ` Ingo Molnar
2006-07-13 3:04 ` Andrea Arcangeli
2006-07-13 3:12 ` Linus Torvalds
2006-07-13 4:40 ` Andrea Arcangeli
2006-07-13 4:51 ` andrea
2006-07-13 5:12 ` Linus Torvalds
2006-07-13 6:22 ` andrea
2006-07-13 1:51 ` Andrew Morton
2006-07-13 2:00 ` Linus Torvalds
2006-07-13 7:44 ` James Bruce
2006-07-13 8:34 ` andrea
2006-07-13 9:18 ` Andrew Morton
2006-07-13 12:13 ` Andi Kleen
2006-07-12 21:22 ` Ingo Molnar
2006-07-12 22:11 ` Andi Kleen
2006-07-11 15:54 ` Pavel Machek
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=200607121739_MC3-1-C4D3-28B9@compuserve.com \
--to=76306.1226@compuserve.com \
--cc=ak@suse.de \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=andrea@cpushare.com \
--cc=arjan@infradead.org \
--cc=bunk@stusta.de \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=torvalds@osdl.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®