[tip: x86/entry] x86/entry/32: Clean up syscall fast exit tests

[tip-bot2 for Brian Gerst]

The following commit has been merged into the x86/entry branch of tip:

Commit-ID:     1a09a27153f91cd7676b2d4ca574577572a8c999
Gitweb:        https://git.kernel.org/tip/1a09a27153f91cd7676b2d4ca574577572a8c999
Author:        Brian Gerst <brgerst@xxxxxxxxx>
AuthorDate:    Wed, 11 Oct 2023 18:43:51 -04:00
Committer:     Ingo Molnar <mingo@xxxxxxxxxx>
CommitterDate: Fri, 13 Oct 2023 13:05:28 +02:00

x86/entry/32: Clean up syscall fast exit tests

Merge compat and native code and clarify comments.

No change in functionality expected.

Signed-off-by: Brian Gerst <brgerst@xxxxxxxxx>
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Andy Lutomirski <luto@xxxxxxxxxx>
Cc: Borislav Petkov <bp@xxxxxxxxx>
Cc: Brian Gerst <brgerst@xxxxxxxxx>
Cc: Denys Vlasenko <dvlasenk@xxxxxxxxxx>
Cc: H. Peter Anvin <hpa@xxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Josh Poimboeuf <jpoimboe@xxxxxxxxxx>
Cc: Uros Bizjak <ubizjak@xxxxxxxxx>
Link: https://lore.kernel.org/r/20231011224351.130935-4-brgerst@xxxxxxxxx
---
 arch/x86/entry/common.c | 48 ++++++++++++++++++----------------------
 1 file changed, 22 insertions(+), 26 deletions(-)

diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
index 4c7154d..d813160 100644
--- a/arch/x86/entry/common.c
+++ b/arch/x86/entry/common.c
@@ -255,34 +255,30 @@ __visible noinstr bool do_fast_syscall_32(struct pt_regs *regs)
 	if (!__do_fast_syscall_32(regs))
 		return false;
 
-#ifdef CONFIG_X86_64
 	/*
-	 * Opportunistic SYSRETL: if possible, try to return using SYSRETL.
-	 * SYSRETL is available on all 64-bit CPUs, so we don't need to
-	 * bother with SYSEXIT.
-	 *
-	 * Unlike 64-bit opportunistic SYSRET, we can't check that CX == IP,
-	 * because the ECX fixup above will ensure that this is essentially
-	 * never the case.
+	 * Check that the register state is valid for using SYSRETL/SYSEXIT
+	 * to exit to userspace.  Otherwise use the slower but fully capable
+	 * IRET exit path.
 	 */
-	return regs->cs == __USER32_CS && regs->ss == __USER_DS &&
-		regs->ip == landing_pad &&
-		(regs->flags & (X86_EFLAGS_RF | X86_EFLAGS_TF)) == 0;
-#else
-	/*
-	 * Opportunistic SYSEXIT: if possible, try to return using SYSEXIT.
-	 *
-	 * Unlike 64-bit opportunistic SYSRET, we can't check that CX == IP,
-	 * because the ECX fixup above will ensure that this is essentially
-	 * never the case.
-	 *
-	 * We don't allow syscalls at all from VM86 mode, but we still
-	 * need to check VM, because we might be returning from sys_vm86.
-	 */
-	return regs->cs == __USER_CS && regs->ss == __USER_DS &&
-		regs->ip == landing_pad &&
-		(regs->flags & (X86_EFLAGS_RF | X86_EFLAGS_TF | X86_EFLAGS_VM)) == 0;
-#endif
+
+	/* XEN PV guests always use the IRET path */
+	if (cpu_feature_enabled(X86_FEATURE_XENPV))
+		return false;
+
+	/* EIP must point to the VDSO landing pad */
+	if (unlikely(regs->ip != landing_pad))
+		return false;
+
+	/* CS and SS must match the values set in MSR_STAR */
+	if (unlikely(regs->cs != __USER32_CS || regs->ss != __USER_DS))
+		return false;
+
+	/* If the TF, RF, or VM flags are set, use IRET */
+	if (unlikely(regs->flags & (X86_EFLAGS_RF | X86_EFLAGS_TF | X86_EFLAGS_VM)))
+		return false;
+
+	/* Use SYSRETL/SYSEXIT to exit to userspace */
+	return true;
 }
 
 /* Returns true to return using SYSEXIT/SYSRETL, or false to use IRET */