fix regression in i386 inline syscall asm producing invalid code

commit 22e5bbd0deadcbd767864bd714e890b70e1fe1df inlined the i386
syscall mechanism, but wrongly assumed memory operands to the 5- and
6-argument syscall asm would be esp-based. however, nothing in the
constraints prevented them from being ebx- or ebp-based, and in those
cases, ebx and ebp could be clobbered before use of the memory operand
was complete. in the 6-argument case, this prevented restoration of
the original register values before the end of the asm block, breaking
the asm contract since ebx and ebp are not marked as clobbered. (they
can't be, because lots of compilers don't accept these registers in
constraints or clobbers if PIC or frame pointer is enabled).

doing this right is complicated by the fact that, after a single push,
no operands which might be memory operands are usable. if they are
esp-based, the value of esp has changed, rendering them invalid.

introduce some new dances to load the registers. for the 5-arg case,
push the operand that may be a memory operand first, and after that,
it doesn't matter if the operand is invalid, since we'll just use the
newly pushed value. for the 6-arg case, we need to put both operands
in memory to begin with, like the old non-inline code prior to commit
22e5bbd0deadcbd767864bd714e890b70e1fe1df accepted, so that there's
only one potentially memory-based operand to the asm. this can then be
saved with a single push, and after that the values can be read off
into the registers they're needed in.

there's some size overhead, but still a lot less execution overhead
than the old out-of-line code. doing it better depends on a modern
compiler that lets you use ebx and ebp in asm constraints without
restriction. the failure modes on compilers where this doesn't work
are inconsistent and dangerous (on at least some gcc versions 4.x and
earlier, wrong codegen!), so this is a delicate matter. it can be
addressed later if needed.

diff --git a/arch/i386/syscall_arch.h b/arch/i386/syscall_arch.h
index 8fe354242905d22f5579df51e124b4a4224d8eba..29e141f2ea383ecacd357b731e16a8063c4bd7de 100644 (file)
--- a/arch/i386/syscall_arch.h
+++ b/arch/i386/syscall_arch.h
@@ -49,17 +49,17 @@ static inline long __syscall4(long n, long a1, long a2, long a3, long a4)
 
 static inline long __syscall5(long n, long a1, long a2, long a3, long a4, long a5)
 {
-       unsigned long __ret, __tmp;
-       __asm__ __volatile__ ("mov %%ebx,%1 ; mov %3,%%ebx ; " SYSCALL_INSNS " ; mov %1,%%ebx"
-               : "=a"(__ret), "=m"(__tmp) : "a"(n), "g"(a1), "c"(a2), "d"(a3), "S"(a4), "D"(a5) : "memory");
+       unsigned long __ret;
+       __asm__ __volatile__ ("pushl %2 ; push %%ebx ; mov 4(%%esp),%%ebx ; " SYSCALL_INSNS " ; pop %%ebx ; add $4,%%esp"
+               : "=a"(__ret) : "a"(n), "g"(a1), "c"(a2), "d"(a3), "S"(a4), "D"(a5) : "memory");
        return __ret;
 }
 
 static inline long __syscall6(long n, long a1, long a2, long a3, long a4, long a5, long a6)
 {
-       unsigned long __ret, __tmp1, __tmp2;
-       __asm__ __volatile__ ("mov %%ebx,%1 ; mov %%ebp,%2 ; mov %4,%%ebx ; mov %9,%%ebp ; " SYSCALL_INSNS " ; mov %2,%%ebp ; mov %1,%%ebx"
-               : "=a"(__ret), "=m"(__tmp1), "=m"(__tmp2) : "a"(n), "g"(a1), "c"(a2), "d"(a3), "S"(a4), "D"(a5), "g"(a6) : "memory");
+       unsigned long __ret, a1a6[2] = { a1, a6 };
+       __asm__ __volatile__ ("pushl %1 ; push %%ebx ; push %%ebp ; mov 8(%%esp),%%ebx ; mov 4(%%ebx),%%ebp ; mov (%%ebx),%%ebx ; " SYSCALL_INSNS " ; pop %%ebp ; pop %%ebx ; add $4,%%esp"
+               : "=a"(__ret) : "g"(&a1a6), "a"(n), "c"(a2), "d"(a3), "S"(a4), "D"(a5) : "memory");
        return __ret;
 }