This vDSO code only gets used by 64-bit kernel,
not 32-bit. In 64-bit kernels, data segment is the same
for 32-bit and 64-bit userspace, and SYSRET insn does load %ss
with its selector. No need to repeat it by hand. Segment loads
are somewhat expensive: tens of cycles.
Signed-off-by: Denys Vlasenko <dvlas...@redhat.com>
CC: Linus Torvalds <torva...@linux-foundation.org>
CC: Steven Rostedt <rost...@goodmis.org>
CC: Ingo Molnar <mi...@kernel.org>
CC: Borislav Petkov <b...@alien8.de>
CC: "H. Peter Anvin" <h...@zytor.com>
CC: Andy Lutomirski <l...@amacapital.net>
CC: Oleg Nesterov <o...@redhat.com>
CC: Frederic Weisbecker <fweis...@gmail.com>
CC: Alexei Starovoitov <a...@plumgrid.com>
CC: Will Drewry <w...@chromium.org>
CC: Kees Cook <keesc...@chromium.org>
CC: x...@kernel.org
CC: linux-kernel@vger.kernel.org
---
Patch was run-tested.
arch/x86/vdso/vdso32/syscall.S | 11 +++++++++--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/arch/x86/vdso/vdso32/syscall.S b/arch/x86/vdso/vdso32/syscall.S
index 5415b56..ccdb9ef 100644
--- a/arch/x86/vdso/vdso32/syscall.S
+++ b/arch/x86/vdso/vdso32/syscall.S
@@ -19,8 +19,15 @@ __kernel_vsyscall:
.Lpush_ebp:
movl %ecx, %ebp
syscall
- movl $__USER32_DS, %ecx
- movl %ecx, %ss
+ /*
+ * Used to load __USER32_DS to %ss here,
+ * but it's not necessary: this vDSO is only used if our kernel
+ * is 64-bit one (and we are on AMD CPU).
+ * For 64-bit kernels, __USER32_DS and __USER_DS are the same.
+ * SYSRET restores %ss to the same value when returning to
+ * either 64- or 32-bit userspace, and 64-bit kernel uses the same
+ * descriptor for %ss in 64- and 32-bit userspace.
+ */
movl %ebp, %ecx
popl %ebp
.Lpop_ebp:
--
1.8.1.4
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
