diff options
| author | Eduardo Sánchez Muñoz <eduardosm-dev@e64.io> | 2023-12-08 22:39:11 +0100 |
|---|---|---|
| committer | Eduardo Sánchez Muñoz <eduardosm-dev@e64.io> | 2023-12-08 22:39:11 +0100 |
| commit | d57125632e78474e2623697fd77b5762df5b5454 (patch) | |
| tree | 068bfacc8fa23c7b63860366ab60617980a0e98e /tests/rustdoc-js-std/osstring-to-string.js | |
| parent | 57935c35aac3f389eb406ac5d33b7953daeba7d1 (diff) | |
| download | rust-d57125632e78474e2623697fd77b5762df5b5454.tar.gz rust-d57125632e78474e2623697fd77b5762df5b5454.zip | |
Fix x86 SSE4.1 ptestnzc
`(op & mask) == 0` and `(op & mask) == mask` need each to be calculated for the whole vector. For example, given * `op = [0b100, 0b010]` * `mask = [0b100, 0b110]` The correct result would be: * `op & mask = [0b100, 0b010]` Comparisons are done on the vector as a whole: * `all_zero = (op & mask) == [0, 0] = false` * `masked_set = (op & mask) == mask = false` * `!all_zero && !masked_set = true` The previous method: `op & mask = [0b100, 0b010]` Comparisons are done element-wise: * `all_zero = (op & mask) == [0, 0] = [true, true]` * `masked_set = (op & mask) == mask = [true, false]` * `!all_zero && !masked_set = [true, false]` After folding with AND, the final result would be `false`, which is incorrect.
Diffstat (limited to 'tests/rustdoc-js-std/osstring-to-string.js')
0 files changed, 0 insertions, 0 deletions