linux-kernel - [PATCH v2 3/4] s390/bitops: Use __assume() for _

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20250916134803.874580-4-hca@linux.ibm.com>
Date: Tue, 16 Sep 2025 15:48:02 +0200
From: Heiko Carstens <hca@...ux.ibm.com>
To: Nathan Chancellor <nathan@...nel.org>, Miguel Ojeda <ojeda@...nel.org>,
        Kees Cook <kees@...nel.org>,
        Nick Desaulniers <nick.desaulniers+lkml@...il.com>,
        Vasily Gorbik <gor@...ux.ibm.com>,
        Alexander Gordeev <agordeev@...ux.ibm.com>,
        Juergen Christ <jchrist@...ux.ibm.com>
Cc: linux-kernel@...r.kernel.org, linux-s390@...r.kernel.org,
        Sven Schnelle <svens@...ux.ibm.com>,
        Christian Borntraeger <borntraeger@...ux.ibm.com>
Subject: [PATCH v2 3/4] s390/bitops: Use __assume() for __flogr() inline assembly return value

Use __assume() to tell compilers that the output operand of the __flogr()
inline assembly contains a value in the range of 0..64. This allows to
optimize the logical AND operation away.

This reduces the kernel image size by 2804 bytes (defconfig, gcc 15.2.0).

Suggested-by: Juergen Christ <jchrist@...ux.ibm.com>
Reviewed-by: Juergen Christ <jchrist@...ux.ibm.com>
Signed-off-by: Heiko Carstens <hca@...ux.ibm.com>
---
 arch/s390/include/asm/bitops.h | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/arch/s390/include/asm/bitops.h b/arch/s390/include/asm/bitops.h
index a1dd72b16f54..5ff069fe9526 100644
--- a/arch/s390/include/asm/bitops.h
+++ b/arch/s390/include/asm/bitops.h
@@ -132,9 +132,10 @@ static inline bool test_bit_inv(unsigned long nr,
  */
 static __always_inline unsigned char __flogr(unsigned long word)
 {
-	if (__builtin_constant_p(word)) {
-		unsigned long bit = 0;
+	unsigned long bit;
 
+	if (__builtin_constant_p(word)) {
+		bit = 0;
 		if (!word)
 			return 64;
 		if (!(word & 0xffffffff00000000UL)) {
@@ -169,7 +170,14 @@ static __always_inline unsigned char __flogr(unsigned long word)
 		asm volatile(
 			"       flogr   %[rp],%[rp]\n"
 			: [rp] "+d" (rp.pair) : : "cc");
-		return rp.even & 127;
+		bit = rp.even;
+		/*
+		 * The result of the flogr instruction is a value in the range
+		 * of 0..64. Let the compiler know that the AND operation can
+		 * be optimized away.
+		 */
+		__assume(bit <= 64);
+		return bit & 127;
 	}
 }
 
-- 
2.48.1