diff options
| author | Wilco Dijkstra <wdijkstr@arm.com> | 2016-09-07 14:56:59 +0000 |
|---|---|---|
| committer | Wilco Dijkstra <wilco@gcc.gnu.org> | 2016-09-07 14:56:59 +0000 |
| commit | ff0f3f1cb56ae7ee030fb9e66cdd583733a056b8 (patch) | |
| tree | ec1572aa2f7fba6f3e827f68a120f7bdd0368c78 /gcc | |
| parent | ca235a8500c9e6c5ad1983ef91bb92c6d40861a8 (diff) | |
| download | gcc-ff0f3f1cb56ae7ee030fb9e66cdd583733a056b8.zip gcc-ff0f3f1cb56ae7ee030fb9e66cdd583733a056b8.tar.gz gcc-ff0f3f1cb56ae7ee030fb9e66cdd583733a056b8.tar.bz2 | |
Improve aarch64_legitimize_address - avoid splitting the offset if it is supported.
Improve aarch64_legitimize_address - avoid splitting the offset if it is
supported. When we do split, take the mode size into account. BLKmode
falls into the unaligned case but should be treated like LDP/STP.
This improves codesize slightly due to fewer base address calculations:
int f(int *p) { return p[5000] + p[7000]; }
Now generates:
f:
add x0, x0, 16384
ldr w1, [x0, 3616]
ldr w0, [x0, 11616]
add w0, w1, w0
ret
instead of:
f:
add x1, x0, 16384
add x0, x0, 24576
ldr w1, [x1, 3616]
ldr w0, [x0, 3424]
add w0, w1, w0
ret
gcc/
* config/aarch64/aarch64.c (aarch64_legitimize_address):
Avoid use of base_offset if offset already in range.
From-SVN: r240026
Diffstat (limited to 'gcc')
| -rw-r--r-- | gcc/ChangeLog | 5 | ||||
| -rw-r--r-- | gcc/config/aarch64/aarch64.c | 14 |
2 files changed, 17 insertions, 2 deletions
diff --git a/gcc/ChangeLog b/gcc/ChangeLog index 9a62e12..c39b351 100644 --- a/gcc/ChangeLog +++ b/gcc/ChangeLog @@ -1,3 +1,8 @@ +2016-09-07 Wilco Dijkstra <wdijkstr@arm.com> + + * config/aarch64/aarch64.c (aarch64_legitimize_address): + Avoid use of base_offset if offset already in range. + 2016-09-07 Kaz Kojima <kkojima@gcc.gnu.org> * config/sh/sh-protos.h (struct sh_atomic_model, diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c index 5efad46..2be750e 100644 --- a/gcc/config/aarch64/aarch64.c +++ b/gcc/config/aarch64/aarch64.c @@ -5082,9 +5082,19 @@ aarch64_legitimize_address (rtx x, rtx /* orig_x */, machine_mode mode) /* For offsets aren't a multiple of the access size, the limit is -256...255. */ else if (offset & (GET_MODE_SIZE (mode) - 1)) - base_offset = (offset + 0x100) & ~0x1ff; + { + base_offset = (offset + 0x100) & ~0x1ff; + + /* BLKmode typically uses LDP of X-registers. */ + if (mode == BLKmode) + base_offset = (offset + 512) & ~0x3ff; + } + /* Small negative offsets are supported. */ + else if (IN_RANGE (offset, -256, 0)) + base_offset = 0; + /* Use 12-bit offset by access size. */ else - base_offset = offset & ~0xfff; + base_offset = offset & (~0xfff * GET_MODE_SIZE (mode)); if (base_offset != 0) { |