This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug rtl-optimization/77499] [7 Regression] Regression after code-hoisting, due to combine pass failing to evaluate known value range
- From: "kugan at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Wed, 14 Sep 2016 00:29:26 +0000
- Subject: [Bug rtl-optimization/77499] [7 Regression] Regression after code-hoisting, due to combine pass failing to evaluate known value range
- Auto-submitted: auto-generated
- References: <bug-77499-4@http.gcc.gnu.org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77499
--- Comment #13 from kugan at gcc dot gnu.org ---
(In reply to avieira from comment #12)
> I heard Kugan was working on getting rid of superfluous zero_extends. Adding
> him to the watch list.
>
> @Kugan: Could your work help this case? And when do you plan to have it
> submitted?
Thanks for the testcase.
With my type promotion pass, I am getting:
cmp r1, r2
ble .L10
- push {r4, r5, r6, r7}
- ldr r7, .L14
+ push {r4, r5, r6, lr}
+ ldr r5, .L14
movw r6, #45345
.L4:
- smull r5, r4, r7, r1
+ smull lr, r4, r5, r1
lsrs r0, r0, #1
sub r4, r4, r1, asr #31
- eor r5, r0, r6
add r4, r4, r4, lsl #1
cmp r1, r4
sub r1, r1, r3
it ne
- uxthne r0, r5
+ eorne r0, r0, r6
cmp r2, r1
blt .L4
- pop {r4, r5, r6, r7}
- bx lr
+ uxth r0, r0
+ pop {r4, r5, r6, pc}
.L10:
+ uxth r0, r0
bx lr
.L15:
.align 2
.L14:
.word 1431655766
.size foo, .-foo
Even though, extension is removed from loop, it has an extra uxth. I have some
cases like this to look at before I post the patch again. I am afraid, I will
not be able to do it during this stage1.