x86/entry: Do not special-case clone(2) in compat entry

With the CPU renaming registers on its own, and all the overhead of the
syscall entry/exit, it is doubtful whether the compiled output of

	mov	%r8, %rax
	mov	%rcx, %r8
	mov	%rax, %rcx
	jmpq	sys_clone

is measurably slower than the hand-crafted version of

	xchg	%r8, %rcx

So get rid of this special case.

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: luto@amacapital.net
Cc: viro@zeniv.linux.org.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>

4 files changed, 14 insertions, 13 deletions

diff --git a/arch/x86/entry/entry_64_compat.S b/arch/x86/entry/entry_64_compat.S
index e811dd9c5e99..ff61b7bb750b 100644
--- a/arch/x86/entry/entry_64_compat.S
+++ b/arch/x86/entry/entry_64_compat.S
@@ -406,15 +406,3 @@ ENTRY(entry_INT80_compat)
 	TRACE_IRQS_ON
 	jmp	swapgs_restore_regs_and_return_to_usermode
 END(entry_INT80_compat)
-
-ENTRY(stub32_clone)
-	/*
-	 * The 32-bit clone ABI is: clone(..., int tls_val, int *child_tidptr).
-	 * The 64-bit clone ABI is: clone(..., int *child_tidptr, int tls_val).
-	 *
-	 * The native 64-bit kernel's sys_clone() implements the latter,
-	 * so we need to swap arguments here before calling it:
-	 */
-	xchg	%r8, %rcx
-	jmp	sys_clone
-ENDPROC(stub32_clone)
diff --git a/arch/x86/entry/syscalls/syscall_32.tbl b/arch/x86/entry/syscalls/syscall_32.tbl
index e7fd0a76bf99..2a5e99cff859 100644
--- a/arch/x86/entry/syscalls/syscall_32.tbl
+++ b/arch/x86/entry/syscalls/syscall_32.tbl
@@ -126,7 +126,7 @@
 117	i386	ipc			sys_ipc				compat_sys_ipc
 118	i386	fsync			sys_fsync
 119	i386	sigreturn		sys_sigreturn			sys32_sigreturn
-120	i386	clone			sys_clone			stub32_clone
+120	i386	clone			sys_clone			compat_sys_x86_clone
 121	i386	setdomainname		sys_setdomainname
 122	i386	uname			sys_newuname
 123	i386	modify_ldt		sys_modify_ldt
diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c
index 3bc03446ec44..6512498bbef6 100644
--- a/arch/x86/ia32/sys_ia32.c
+++ b/arch/x86/ia32/sys_ia32.c
@@ -233,3 +233,14 @@ COMPAT_SYSCALL_DEFINE6(x86_fallocate, int, fd, int, mode,
 	return sys_fallocate(fd, mode, ((u64)offset_hi << 32) | offset_lo,
 			     ((u64)len_hi << 32) | len_lo);
 }
+
+/*
+ * The 32-bit clone ABI is CONFIG_CLONE_BACKWARDS
+ */
+COMPAT_SYSCALL_DEFINE5(x86_clone, unsigned long, clone_flags,
+		       unsigned long, newsp, int __user *, parent_tidptr,
+		       unsigned long, tls_val, int __user *, child_tidptr)
+{
+	return sys_clone(clone_flags, newsp, parent_tidptr, child_tidptr,
+			tls_val);
+}
diff --git a/arch/x86/include/asm/sys_ia32.h b/arch/x86/include/asm/sys_ia32.h
index 32831905d97a..906794aa034e 100644
--- a/arch/x86/include/asm/sys_ia32.h
+++ b/arch/x86/include/asm/sys_ia32.h
@@ -55,6 +55,8 @@ asmlinkage long compat_sys_x86_fadvise64(int, unsigned int, unsigned int,
 					 size_t, int);
 asmlinkage long compat_sys_x86_fallocate(int, int, unsigned int, unsigned int,
 					 unsigned int, unsigned int);
+asmlinkage long compat_sys_x86_clone(unsigned long, unsigned long, int __user *,
+				     unsigned long, int __user *);
 
 /* ia32/ia32_signal.c */
 asmlinkage long sys32_sigreturn(void);