diff options
| author | Andy Lutomirski <luto@kernel.org> | 2016-04-26 12:23:28 -0700 |
|---|---|---|
| committer | Ingo Molnar <mingo@kernel.org> | 2016-04-29 11:56:41 +0200 |
| commit | 731e33e39a5b95ad77017811b3ced32ecf9dc666 (patch) | |
| tree | 2af8619a78d1713abee9a4bcebf0fc5b453ef536 /arch/x86/include/asm/segment.h | |
| parent | b038c842b385f1470f991078e71b7c5b084a7341 (diff) | |
x86/arch_prctl/64: Remove FSBASE/GSBASE < 4G optimization
As far as I know, the optimization doesn't work on any modern distro
because modern distros use high addresses for ASLR. Remove it.
The ptrace code was either wrong or very strange, but the behavior
with this patch should be essentially identical to the behavior
without this patch unless user code goes out of its way to mislead
ptrace.
On newer CPUs, once the FSGSBASE instructions are enabled, we won't
want to use the optimized variant anyway.
This isn't actually much of a performance regression, it has no effect
on normal dynamically linked programs, and it's a considerably
simplification. It also removes some nasty special cases from code
that is already way too full of special cases for comfort.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/dd1599b08866961dba9d2458faa6bbd7fba471d7.1461698311.git.luto@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Diffstat (limited to 'arch/x86/include/asm/segment.h')
| -rw-r--r-- | arch/x86/include/asm/segment.h | 7 |
1 files changed, 0 insertions, 7 deletions
diff --git a/arch/x86/include/asm/segment.h b/arch/x86/include/asm/segment.h index e1a4afd20223..1549caa098f0 100644 --- a/arch/x86/include/asm/segment.h +++ b/arch/x86/include/asm/segment.h @@ -208,13 +208,6 @@ #define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS*8 + 3) #define __PER_CPU_SEG (GDT_ENTRY_PER_CPU*8 + 3) -/* TLS indexes for 64-bit - hardcoded in arch_prctl(): */ -#define FS_TLS 0 -#define GS_TLS 1 - -#define GS_TLS_SEL ((GDT_ENTRY_TLS_MIN+GS_TLS)*8 + 3) -#define FS_TLS_SEL ((GDT_ENTRY_TLS_MIN+FS_TLS)*8 + 3) - #endif #ifndef CONFIG_PARAVIRT |