[PATCH 5/5] x86/asm/entry/32: Simplify FLAGS_NT clearing in SYSENTER32 code.

[Denys Vlasenko]

"sysenter_fix_flags" detour does not need to be convoluted anymore,
straigten it up. However, we still use this:

        jnz     2f
        jmp     sysenter_jumps_here
2:	...

instead of this:

        jz      sysenter_jumps_here
	...

because "cold" conditional forward branch is predicted not taken
by most CPUs - exactly what we want. Latter version would get it wrong.

Reinstate "why we use SYSRETL instead of SYSEXIT" comment.

Signed-off-by: Denys Vlasenko <dvlasenk@xxxxxxxxxx>
CC: Ingo Molnar <mingo@xxxxxxxxxx>
CC: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
CC: Krzysztof A. Sobiecki <sobkas@xxxxxxxxx>
CC: Steven Rostedt <rostedt@xxxxxxxxxxx>
CC: Borislav Petkov <bp@xxxxxxxxx>
CC: "H. Peter Anvin" <hpa@xxxxxxxxx>
CC: Andy Lutomirski <luto@xxxxxxxxxxxxxx>
CC: Oleg Nesterov <oleg@xxxxxxxxxx>
CC: Frederic Weisbecker <fweisbec@xxxxxxxxx>
CC: Alexei Starovoitov <ast@xxxxxxxxxxxx>
CC: Will Drewry <wad@xxxxxxxxxxxx>
CC: Kees Cook <keescook@xxxxxxxxxxxx>
CC: x86@xxxxxxxxxx
CC: linux-kernel@xxxxxxxxxxxxxxx
---
 arch/x86/entry/entry_64_compat.S | 26 +++++++++++++++++---------
 1 file changed, 17 insertions(+), 9 deletions(-)

diff --git a/arch/x86/entry/entry_64_compat.S b/arch/x86/entry/entry_64_compat.S
index 73b56a5..bd3664f 100644
--- a/arch/x86/entry/entry_64_compat.S
+++ b/arch/x86/entry/entry_64_compat.S
@@ -98,20 +98,28 @@ ENTRY(entry_SYSENTER_compat)
 	ASM_CLAC
 
 	/*
-	 * Sysenter doesn't filter flags, so we need to clear NT
-	 * ourselves.  To save a few cycles, we can check whether
-	 * NT was set instead of doing an unconditional popfq.
+	 * Sysenter doesn't filter flags, so we need to clear NT ourselves.
 	 */
 	testl	$X86_EFLAGS_NT, EFLAGS(%rsp)
-	jnz	sysenter_fix_flags
-sysenter_flags_fixed:
+	jnz	2f
 	jmp	sysenter_jumps_here
-
-sysenter_fix_flags:
+2:
 	pushq	$(X86_EFLAGS_IF|X86_EFLAGS_FIXED)
 	popfq
-	jmp	sysenter_flags_fixed
-
+	jmp	sysenter_jumps_here
+	/*
+	 * SYSEXIT insn is not obviously safe for 64-bit kernels --
+	 * an NMI between STI and SYSEXIT has poorly specified behavior,
+	 * and NMI followed by an IRQ with usergs is fatal.
+	 * So we just pretend we're using SYSEXIT but we really use
+	 * SYSRETL instead. (Yes, SYSRETL works even on Intel CPUs.)
+	 * We do that by reusing the entire SYSCALL code path:
+	 * the jump above takes us there.
+	 *
+	 * The difference of SYSENTER 32-bit ABI versus SYSCALL
+	 * is that SYSENTER ABI does not promise to preserve EDX and EBP,
+	 * SYSCALL does.
+	 */
 ENDPROC(entry_SYSENTER_compat)
 
 /*
-- 
1.8.1.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/