From: Eric Biggers <ebiggers@google.com>
Date: Sat, 6 Apr 2024 00:26:09 +0000 (-0400)
Subject: crypto: x86/sha256-avx2 - add missing vzeroupper
X-Git-Tag: v6.8.12~374
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=00f7cc2fa2c0732642f238f9d7b6f4e56f6f7031;p=thirdparty%2Fkernel%2Fstable.git

crypto: x86/sha256-avx2 - add missing vzeroupper

[ Upstream commit 57ce8a4e162599cf9adafef1f29763160a8e5564 ]

Since sha256_transform_rorx() uses ymm registers, execute vzeroupper
before returning from it.  This is necessary to avoid reducing the
performance of SSE code.

Fixes: d34a460092d8 ("crypto: sha256 - Optimized sha256 x86_64 routine using AVX2's RORX instructions")
Signed-off-by: Eric Biggers <ebiggers@google.com>
Acked-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---

diff --git a/arch/x86/crypto/sha256-avx2-asm.S b/arch/x86/crypto/sha256-avx2-asm.S
index 9918212faf914..0ffb072be9561 100644
--- a/arch/x86/crypto/sha256-avx2-asm.S
+++ b/arch/x86/crypto/sha256-avx2-asm.S
@@ -716,6 +716,7 @@ SYM_TYPED_FUNC_START(sha256_transform_rorx)
 	popq	%r13
 	popq	%r12
 	popq	%rbx
+	vzeroupper
 	RET
 SYM_FUNC_END(sha256_transform_rorx)