cmovb %rcx, %rax
ret
-Seems like the jb branch has high likelyhood of being taken. It would have
+Seems like the jb branch has high likelihood of being taken. It would have
saved a few instructions.
//===---------------------------------------------------------------------===//
-Poor codegen:
-
-int X[2];
-int b;
-void test(void) {
- memset(X, b, 2*sizeof(X[0]));
-}
-
-llc:
- movq _b@GOTPCREL(%rip), %rax
- movzbq (%rax), %rax
- movq %rax, %rcx
- shlq $8, %rcx
- orq %rax, %rcx
- movq %rcx, %rax
- shlq $16, %rax
- orq %rcx, %rax
- movq %rax, %rcx
- shlq $32, %rcx
- movq _X@GOTPCREL(%rip), %rdx
- orq %rax, %rcx
- movq %rcx, (%rdx)
- ret
-
-gcc:
- movq _b@GOTPCREL(%rip), %rax
- movabsq $72340172838076673, %rdx
- movzbq (%rax), %rax
- imulq %rdx, %rax
- movq _X@GOTPCREL(%rip), %rdx
- movq %rax, (%rdx)
- ret
-
-//===---------------------------------------------------------------------===//
-
It's not possible to reference AH, BH, CH, and DH registers in an instruction
requiring REX prefix. However, divb and mulb both produce results in AH. If isel
emits a CopyFromReg which gets turned into a movb and that can be allocated a
if we have whole-function selectiondags.
//===---------------------------------------------------------------------===//
+
+Take the following code
+(from http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34653):
+extern unsigned long table[];
+unsigned long foo(unsigned char *p) {
+ unsigned long tag = *p;
+ return table[tag >> 4] + table[tag & 0xf];
+}
+
+Current code generated:
+ movzbl (%rdi), %eax
+ movq %rax, %rcx
+ andq $240, %rcx
+ shrq %rcx
+ andq $15, %rax
+ movq table(,%rax,8), %rax
+ addq table(%rcx), %rax
+ ret
+
+Issues:
+1. First movq should be movl; saves a byte.
+2. Both andq's should be andl; saves another two bytes. I think this was
+ implemented at one point, but subsequently regressed.
+3. shrq should be shrl; saves another byte.
+4. The first andq can be completely eliminated by using a slightly more
+ expensive addressing mode.
+
+//===---------------------------------------------------------------------===//
+
+Consider the following (contrived testcase, but contains common factors):
+
+#include <stdarg.h>
+int test(int x, ...) {
+ int sum, i;
+ va_list l;
+ va_start(l, x);
+ for (i = 0; i < x; i++)
+ sum += va_arg(l, int);
+ va_end(l);
+ return sum;
+}
+
+Testcase given in C because fixing it will likely involve changing the IR
+generated for it. The primary issue with the result is that it doesn't do any
+of the optimizations which are possible if we know the address of a va_list
+in the current function is never taken:
+1. We shouldn't spill the XMM registers because we only call va_arg with "int".
+2. It would be nice if we could scalarrepl the va_list.
+3. Probably overkill, but it'd be cool if we could peel off the first five
+iterations of the loop.
+
+Other optimizations involving functions which use va_arg on floats which don't
+have the address of a va_list taken:
+1. Conversely to the above, we shouldn't spill general registers if we only
+ call va_arg on "double".
+2. If we know nothing more than 64 bits wide is read from the XMM registers,
+ we can change the spilling code to reduce the amount of stack used by half.
+
+//===---------------------------------------------------------------------===//