https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107757
--- Comment #4 from Michael Meissner ---
Note, this code only shows up when the target CPU is power8.
For the following code:
vector long long lsb64()
{
return vec_splats(1LL);
}
Both power9 and power10 generate:
xxspltib 34,1
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107757
--- Comment #3 from Michael Meissner ---
As Segher says, the test is not quite correct. I would write it as:
vector long long lsb64_opt()
{
vector long long a = vec_splats(~0LL);
__asm__("vsrd %0,%1,%2":"=v"(a):"v"(a),"v"(a));
return
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107757
--- Comment #2 from Jiu Fu Guo ---
For the scenario of this PR, the 2insn sequence would be faster than
non-prefixed loading.
One finding may be useful: when building 64bit constants, I notice that loading
may be faster sometimes if the constan
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107757
Segher Boessenkool changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
Last reconfirmed|
