Received-SPF: pass (google.com: domain of linux-kernel+bounces-44444-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
From: Borislav Petkov <bp@alien8.de>
To: X86 ML <x86@kernel.org>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>,
	LKML <linux-kernel@vger.kernel.org>
Subject: [PATCH 2/4] x86/alternatives: Get rid of __optimize_nops()
Date: Tue, 30 Jan 2024 11:59:39 +0100
Message-ID: <20240130105941.19707-3-bp@alien8.de>
In-Reply-To: <20240130105941.19707-1-bp@alien8.de>
References: <20240130105941.19707-1-bp@alien8.de>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable

From: "Borislav Petkov (AMD)" <bp@alien8.de>

There's no need to carve out bits of the NOP optimization functionality
and look at JMP opcodes - simply do one more NOPs optimization pass
at the end of the patching.

A lot simpler code.

Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
---
 arch/x86/kernel/alternative.c | 52 +++++++----------------------------
 1 file changed, 10 insertions(+), 42 deletions(-)

diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.=
c
index d633eb59f2b6..2dd1c7fe0949 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -216,47 +216,12 @@ static int skip_nops(u8 *buf, int offset, int len)
 	return offset;
 }
=20
-/*
- * Optimize a sequence of NOPs, possibly preceded by an unconditional ju=
mp
- * to the end of the NOP sequence into a single NOP.
- */
-static bool __init_or_module
-__optimize_nops(const u8 * const instr, u8 *buf, size_t len, struct insn=
 *insn, int *next, int *prev, int *target)
-{
-	int i =3D *next - insn->length;
-
-	switch (insn->opcode.bytes[0]) {
-	case JMP8_INSN_OPCODE:
-	case JMP32_INSN_OPCODE:
-		*prev =3D i;
-		*target =3D *next + insn->immediate.value;
-		return false;
-	}
-
-	if (insn_is_nop(insn)) {
-		int nop =3D i;
-
-		*next =3D skip_nops(buf, *next, len);
-		if (*target && *next =3D=3D *target)
-			nop =3D *prev;
-
-		add_nop(buf + nop, *next - nop);
-		DUMP_BYTES(ALT, buf, len, "%px: [%d:%d) optimized NOPs: ", instr, nop,=
 *next);
-		return true;
-	}
-
-	*target =3D 0;
-	return false;
-}
-
 /*
  * "noinline" to cause control flow change and thus invalidate I$ and
  * cause refetch after modification.
  */
 static void __init_or_module noinline optimize_nops(const u8 * const ins=
tr, u8 *buf, size_t len)
 {
-	int prev, target =3D 0;
-
 	for (int next, i =3D 0; i < len; i =3D next) {
 		struct insn insn;
=20
@@ -265,7 +230,14 @@ static void __init_or_module noinline optimize_nops(=
const u8 * const instr, u8 *
=20
 		next =3D i + insn.length;
=20
-		__optimize_nops(instr, buf, len, &insn, &next, &prev, &target);
+		if (insn_is_nop(&insn)) {
+			int nop =3D i;
+
+			next =3D skip_nops(buf, next, len);
+
+			add_nop(buf + nop, next - nop);
+			DUMP_BYTES(ALT, buf, len, "%px: [%d:%d) optimized NOPs: ", instr, nop=
, next);
+		}
 	}
 }
=20
@@ -342,8 +314,6 @@ bool need_reloc(unsigned long offset, u8 *src, size_t=
 src_len)
 static void __init_or_module noinline
 apply_relocation(const u8 * const instr, u8 *buf, size_t len, u8 *src, s=
ize_t src_len)
 {
-	int prev, target =3D 0;
-
 	for (int next, i =3D 0; i < len; i =3D next) {
 		struct insn insn;
=20
@@ -352,9 +322,6 @@ apply_relocation(const u8 * const instr, u8 *buf, siz=
e_t len, u8 *src, size_t sr
=20
 		next =3D i + insn.length;
=20
-		if (__optimize_nops(instr, buf, len, &insn, &next, &prev, &target))
-			continue;
-
 		switch (insn.opcode.bytes[0]) {
 		case 0x0f:
 			if (insn.opcode.bytes[1] < 0x80 ||
@@ -533,7 +500,8 @@ void __init_or_module noinline apply_alternatives(str=
uct alt_instr *start,
 		for (; insn_buff_sz < a->instrlen; insn_buff_sz++)
 			insn_buff[insn_buff_sz] =3D 0x90;
=20
-		apply_relocation(instr, insn_buff, a->instrlen, replacement, a->replac=
ementlen);
+		apply_relocation(instr, insn_buff, a->instrlen, replacement, insn_buff=
_sz);
+		optimize_nops(instr, insn_buff, insn_buff_sz);
=20
 		DUMP_BYTES(ALT, instr, a->instrlen, "%px:   old_insn: ", instr);
 		DUMP_BYTES(ALT, replacement, a->replacementlen, "%px:   rpl_insn: ", r=
eplacement);
--=20
2.43.0