Minimum version of binutils required to compile the kernel is 2.25.
This version correctly handles the "lock" prefix, so it is possible
to remove the semicolon, which was used to support ancient versions
of GNU as.
Due to the semicolon, the compiler considers "lock; insn" as two
separate instructions. Removing the semicolon makes asm length
calculations more accurate, consequently making scheduling and
inlining decisions of the compiler more accurate.
Removing the semicolon also enables assembler checks involving lock
prefix. Trying to assemble e.g. "lock andl %eax, %ebx" results in:
Error: expecting lockable instruction after `lock'
Signed-off-by: Uros Bizjak <ubiz...@gmail.com>
Cc: Richard Weinberger <rich...@nod.at>
Cc: Anton Ivanov <anton.iva...@cambridgegreys.com>
Cc: Johannes Berg <johan...@sipsolutions.net>
Cc: Thomas Gleixner <t...@linutronix.de>
Cc: Ingo Molnar <mi...@kernel.org>
Cc: Borislav Petkov <b...@alien8.de>
Cc: Dave Hansen <dave.han...@linux.intel.com>
Cc: "H. Peter Anvin" <h...@zytor.com>
---
arch/x86/um/asm/barrier.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/arch/x86/um/asm/barrier.h b/arch/x86/um/asm/barrier.h
index 4da336965698..b51aefd6ec2b 100644
--- a/arch/x86/um/asm/barrier.h
+++ b/arch/x86/um/asm/barrier.h
@@ -12,9 +12,9 @@
*/
#ifdef CONFIG_X86_32
-#define mb() alternative("lock; addl $0,0(%%esp)", "mfence",
X86_FEATURE_XMM2)
-#define rmb() alternative("lock; addl $0,0(%%esp)", "lfence",
X86_FEATURE_XMM2)
-#define wmb() alternative("lock; addl $0,0(%%esp)", "sfence", X86_FEATURE_XMM)
+#define mb() alternative("lock addl $0,0(%%esp)", "mfence", X86_FEATURE_XMM2)
+#define rmb() alternative("lock addl $0,0(%%esp)", "lfence", X86_FEATURE_XMM2)
+#define wmb() alternative("lock addl $0,0(%%esp)", "sfence", X86_FEATURE_XMM)
#else /* CONFIG_X86_32 */
--
2.48.1
