On Sunday, 18 June 2023 at 02:55:32 UTC, Cecil Ward wrote:
On Friday, 16 June 2023 at 13:12:06 UTC, Iain Buclaw wrote:
Redundant jump? Yes, arguably. Leads to wrong runtime? Doesn't
look that way.
Completely agree with Iain, it’s not incorrect code, I wasn’t
intending to suggest that. I’d just say suboptimal, and not the
very best code generation possible.
By the way, I suspect it's related to this problem report.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96435
Taken from upstream DMD comment (I suspect is still not fixed in
either DMD, possibly LDC too but haven't checked).
https://issues.dlang.org/show_bug.cgi?id=20148#c3
Essentially, every read of a boolean gets turned into
`(*(ubyte*)&b) & 1`.
This *could* be limited to `@safe` code, or just accessing field
members of a union I suppose.
