x86/retpoline: Add SKL retthunk retpolines

Ensure that retpolines do the proper call accounting so that the return
accounting works correctly.

Specifically; retpolines are used to replace both 'jmp *%reg' and
'call *%reg', however these two cases do not have the same accounting
requirements. Therefore split things up and provide two different
retpoline arrays for SKL.

The 'jmp *%reg' case needs no accounting, the
__x86_indirect_jump_thunk_array[] covers this. The retpoline is
changed to not use the return thunk; it's a simple call;ret construct.

[ strictly speaking it should do:
	andq $(~0x1f), PER_CPU_VAR(__x86_call_depth)
  but we can argue this can be covered by the fuzz we already have
  in the accounting depth (12) vs the RSB depth (16) ]

The 'call *%reg' case does need accounting, the
__x86_indirect_call_thunk_array[] covers this. Again, this retpoline
avoids the use of the return-thunk, in this case to avoid double
accounting.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220915111147.996634749@infradead.org

arch/x86/include/asm/nospec-branch.h

@@ -301,6 +301,8 @@
 
 typedef u8 retpoline_thunk_t[RETPOLINE_THUNK_SIZE];
 extern retpoline_thunk_t __x86_indirect_thunk_array[];
+extern retpoline_thunk_t __x86_indirect_call_thunk_array[];
+extern retpoline_thunk_t __x86_indirect_jump_thunk_array[];
 
 extern void __x86_return_thunk(void);
 extern void zen_untrain_ret(void);
@@ -330,6 +332,16 @@ static inline void x86_set_skl_return_thunk(void) {}
 #include <asm/GEN-for-each-reg.h>
 #undef GEN
 
+#define GEN(reg)						\
+	extern retpoline_thunk_t __x86_indirect_call_thunk_ ## reg;
+#include <asm/GEN-for-each-reg.h>
+#undef GEN
+
+#define GEN(reg)						\
+	extern retpoline_thunk_t __x86_indirect_jump_thunk_ ## reg;
+#include <asm/GEN-for-each-reg.h>
+#undef GEN
+
 #ifdef CONFIG_X86_64
 
 /*

arch/x86/kernel/alternative.c

@@ -377,6 +377,56 @@ static int emit_indirect(int op, int reg, u8 *bytes)
 	return i;
 }
 
+static inline bool is_jcc32(struct insn *insn)
+{
+	/* Jcc.d32 second opcode byte is in the range: 0x80-0x8f */
+	return insn->opcode.bytes[0] == 0x0f && (insn->opcode.bytes[1] & 0xf0) == 0x80;
+}
+
+static int emit_call_track_retpoline(void *addr, struct insn *insn, int reg, u8 *bytes)
+{
+	u8 op = insn->opcode.bytes[0];
+	int i = 0;
+
+	/*
+	 * Clang does 'weird' Jcc __x86_indirect_thunk_r11 conditional
+	 * tail-calls. Deal with them.
+	 */
+	if (is_jcc32(insn)) {
+		bytes[i++] = op;
+		op = insn->opcode.bytes[1];
+		goto clang_jcc;
+	}
+
+	if (insn->length == 6)
+		bytes[i++] = 0x2e; /* CS-prefix */
+
+	switch (op) {
+	case CALL_INSN_OPCODE:
+		__text_gen_insn(bytes+i, op, addr+i,
+				__x86_indirect_call_thunk_array[reg],
+				CALL_INSN_SIZE);
+		i += CALL_INSN_SIZE;
+		break;
+
+	case JMP32_INSN_OPCODE:
+clang_jcc:
+		__text_gen_insn(bytes+i, op, addr+i,
+				__x86_indirect_jump_thunk_array[reg],
+				JMP32_INSN_SIZE);
+		i += JMP32_INSN_SIZE;
+		break;
+
+	default:
+		WARN("%pS %px %*ph\n", addr, addr, 6, addr);
+		return -1;
+	}
+
+	WARN_ON_ONCE(i != insn->length);
+
+	return i;
+}
+
 /*
  * Rewrite the compiler generated retpoline thunk calls.
  *
@@ -409,8 +459,12 @@ static int patch_retpoline(void *addr, struct insn *insn, u8 *bytes)
 	BUG_ON(reg == 4);
 
 	if (cpu_feature_enabled(X86_FEATURE_RETPOLINE) &&
-	    !cpu_feature_enabled(X86_FEATURE_RETPOLINE_LFENCE))
+	    !cpu_feature_enabled(X86_FEATURE_RETPOLINE_LFENCE)) {
+		if (cpu_feature_enabled(X86_FEATURE_CALL_DEPTH))
+			return emit_call_track_retpoline(addr, insn, reg, bytes);
+
 		return -1;
+	}
 
 	op = insn->opcode.bytes[0];
 
@@ -427,8 +481,7 @@ static int patch_retpoline(void *addr, struct insn *insn, u8 *bytes)
 	 *   [ NOP ]
 	 * 1:
 	 */
-	/* Jcc.d32 second opcode byte is in the range: 0x80-0x8f */
+	if (is_jcc32(insn)) {
-	if (op == 0x0f && (insn->opcode.bytes[1] & 0xf0) == 0x80) {
 		cc = insn->opcode.bytes[1] & 0xf;
 		cc ^= 1; /* invert condition */
 

arch/x86/lib/retpoline.S

@@ -14,17 +14,18 @@
 
 	.section .text.__x86.indirect_thunk
 
-.macro RETPOLINE reg
+
+.macro POLINE reg
 	ANNOTATE_INTRA_FUNCTION_CALL
 	call    .Ldo_rop_\@
-.Lspec_trap_\@:
+	int3
-	UNWIND_HINT_EMPTY
-	pause
-	lfence
-	jmp .Lspec_trap_\@
 .Ldo_rop_\@:
 	mov     %\reg, (%_ASM_SP)
 	UNWIND_HINT_FUNC
+.endm
+
+.macro RETPOLINE reg
+	POLINE \reg
 	RET
 .endm
 
@@ -54,7 +55,6 @@ SYM_INNER_LABEL(__x86_indirect_thunk_\reg, SYM_L_GLOBAL)
  */
 
 #define __EXPORT_THUNK(sym)	_ASM_NOKPROBE(sym); EXPORT_SYMBOL(sym)
-#define EXPORT_THUNK(reg)	__EXPORT_THUNK(__x86_indirect_thunk_ ## reg)
 
 	.align RETPOLINE_THUNK_SIZE
 SYM_CODE_START(__x86_indirect_thunk_array)
@@ -66,10 +66,65 @@ SYM_CODE_START(__x86_indirect_thunk_array)
 	.align RETPOLINE_THUNK_SIZE
 SYM_CODE_END(__x86_indirect_thunk_array)
 
-#define GEN(reg) EXPORT_THUNK(reg)
+#define GEN(reg) __EXPORT_THUNK(__x86_indirect_thunk_ ## reg)
 #include <asm/GEN-for-each-reg.h>
 #undef GEN
 
+#ifdef CONFIG_CALL_DEPTH_TRACKING
+.macro CALL_THUNK reg
+	.align RETPOLINE_THUNK_SIZE
+
+SYM_INNER_LABEL(__x86_indirect_call_thunk_\reg, SYM_L_GLOBAL)
+	UNWIND_HINT_EMPTY
+	ANNOTATE_NOENDBR
+
+	CALL_DEPTH_ACCOUNT
+	POLINE \reg
+	ANNOTATE_UNRET_SAFE
+	ret
+	int3
+.endm
+
+	.align RETPOLINE_THUNK_SIZE
+SYM_CODE_START(__x86_indirect_call_thunk_array)
+
+#define GEN(reg) CALL_THUNK reg
+#include <asm/GEN-for-each-reg.h>
+#undef GEN
+
+	.align RETPOLINE_THUNK_SIZE
+SYM_CODE_END(__x86_indirect_call_thunk_array)
+
+#define GEN(reg) __EXPORT_THUNK(__x86_indirect_call_thunk_ ## reg)
+#include <asm/GEN-for-each-reg.h>
+#undef GEN
+
+.macro JUMP_THUNK reg
+	.align RETPOLINE_THUNK_SIZE
+
+SYM_INNER_LABEL(__x86_indirect_jump_thunk_\reg, SYM_L_GLOBAL)
+	UNWIND_HINT_EMPTY
+	ANNOTATE_NOENDBR
+	POLINE \reg
+	ANNOTATE_UNRET_SAFE
+	ret
+	int3
+.endm
+
+	.align RETPOLINE_THUNK_SIZE
+SYM_CODE_START(__x86_indirect_jump_thunk_array)
+
+#define GEN(reg) JUMP_THUNK reg
+#include <asm/GEN-for-each-reg.h>
+#undef GEN
+
+	.align RETPOLINE_THUNK_SIZE
+SYM_CODE_END(__x86_indirect_jump_thunk_array)
+
+#define GEN(reg) __EXPORT_THUNK(__x86_indirect_jump_thunk_ ## reg)
+#include <asm/GEN-for-each-reg.h>
+#undef GEN
+#endif
 /*
  * This function name is magical and is used by -mfunction-return=thunk-extern
  * for the compiler to generate JMPs to it.

arch/x86/net/bpf_jit_comp.c

@@ -417,7 +417,10 @@ static void emit_indirect_jump(u8 **pprog, int reg, u8 *ip)
 		EMIT2(0xFF, 0xE0 + reg);
 	} else if (cpu_feature_enabled(X86_FEATURE_RETPOLINE)) {
 		OPTIMIZER_HIDE_VAR(reg);
-		emit_jump(&prog, &__x86_indirect_thunk_array[reg], ip);
+		if (cpu_feature_enabled(X86_FEATURE_CALL_DEPTH))
+			emit_jump(&prog, &__x86_indirect_jump_thunk_array[reg], ip);
+		else
+			emit_jump(&prog, &__x86_indirect_thunk_array[reg], ip);
 	} else {
 		EMIT2(0xFF, 0xE0 + reg);	/* jmp *%\reg */
 		if (IS_ENABLED(CONFIG_RETPOLINE) || IS_ENABLED(CONFIG_SLS))