lkml.org 
[lkml]   [2015]   [Apr]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [PATCH 8/8] x86: Use rd/wr fs/gs base in arch_prctl
On Fri, Apr 10, 2015 at 8:50 AM, Andi Kleen <andi@firstfloor.org> wrote:
> From: Andi Kleen <ak@linux.intel.com>
>
> Convert arch_prctl to use the new instructions to
> change fs/gs if available, instead of using MSRs.
>
> This is merely a small performance optimization,
> no new functionality.
>
> With the new instructions the syscall is really obsolete,
> as everything can be set directly in ring 3. But the syscall
> is widely used by existing software, so we still support it.

It's also necessary on my poor obsolete Sandy Bridge machines :)

>
> The syscall still enforces that the addresses are not
> in kernel space, even though that is not needed more.
> This is mainly so that the programs written for new CPUs
> do not suddenly fail on old CPUs.
>
> With the new instructions available it prefers to use
> them in the context switch, instead of using the old
> "use GDT segment rewrite" trick.
>
> v2: Make kprobes safe
> Signed-off-by: Andi Kleen <ak@linux.intel.com>
> ---
> arch/x86/kernel/process_64.c | 48 ++++++++++++++++++++++++++++++++++++--------
> 1 file changed, 40 insertions(+), 8 deletions(-)
>
> diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
> index 3019c51..1fe4d79 100644
> --- a/arch/x86/kernel/process_64.c
> +++ b/arch/x86/kernel/process_64.c
> @@ -534,20 +534,38 @@ unsigned long get_wchan(struct task_struct *p)
> return 0;
> }
>
> +static noinline __kprobes void reload_user_gs(unsigned long addr)
> +{
> + local_irq_disable();
> + swapgs();
> + loadsegment(gs, 0);
> + wrgsbase(addr);
> + swapgs();
> + local_irq_enable();
> +}

These names are terrifying. How about
write_user_gsbase_and_set_gs_to_zero? (Yes, it's a mouthful.) Or
write_user_gs_and_gsbase(unsigned short gs, unsigned long gsbase),
which we'll need if we make __switch_to fully correct.

Also, at some point we should fix the fsindex vs fs crap in
thread_struct. They should be fsindex and fsbase.

--Andy


\
 
 \ /
  Last update: 2015-04-10 22:41    [W:0.152 / U:1.924 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site