https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110009
Bug 110009 depends on bug 111285, which changed state.
Bug 111285 Summary: [12 Regression] vector ABSU is lowered incorrectly
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111285
What|Removed |Added
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110009
--- Comment #9 from Andrew Pinski ---
So now there is an open question here.
which is more Canonical:
```
_1 = v < 0;
a_4 = (int) _1;
b_5 = -a_4;
_9 = b_5 | 1;
```
or
```
_t1 = v >> 31
_9 = _t1 | 1
```
or
```
_tb = v < 0
_9 = _tb ? -1 :
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110009
Andrew Pinski changed:
What|Removed |Added
Blocks||98907
--- Comment #8 from Andrew Pinski
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110009
--- Comment #7 from Andrew Pinski ---
Created attachment 55838
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=55838&action=edit
dg-testcase
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110009
--- Comment #6 from Andrew Pinski ---
Vector testcases too:
```
#define vector __attribute__((vector_size(4*sizeof(int
vector int
signed_f1(vector int v)
{
vector int b_5;
vector int _7;
vector int _9;
b_5 = v>>(sizeof(int)*8 - 1);
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110009
--- Comment #5 from Andrew Pinski ---
With the nop_convert added and using bitwise_equal_p :
(simplify
(mult:c @0 (nop_convert? (bit_ior:c (rshift @1 INTEGER_CST@2) integer_onep)))
(if (!TYPE_UNSIGNED (TREE_TYPE (@1))
&& wi::to_wide (@2
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110009
Andrew Pinski changed:
What|Removed |Added
Status|NEW |ASSIGNED
Assignee|unassigned
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110009
Andrew Pinski changed:
What|Removed |Added
Last reconfirmed||2023-07-12
Severity|normal
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110009
--- Comment #2 from Andrew Pinski ---
(In reply to Georg-Johann Lay from comment #1)
> (In reply to Andrew Pinski from comment #0)
> > unsigned
> > f1 (int v)
> > {
> > [...]
> > int b_5;
> >
> > b_5 = v>>(sizeof(v)*8 - 1);
>
> Does it d
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110009
--- Comment #1 from Georg-Johann Lay ---
(In reply to Andrew Pinski from comment #0)
> unsigned
> f1 (int v)
> {
> [...]
> int b_5;
>
> b_5 = v>>(sizeof(v)*8 - 1);
Does it depend on -fwrapv maybe.
10 matches
Mail list logo