| Age | Commit message (Collapse) | Author | Lines | |
|---|---|---|---|---|
| 2024-04-18 | Update `checked_ops` so 32- and 64-bit gets the same checks | Scott McMurray | -62/+0 | |
| 2024-04-18 | At debuginfo=0, don't inline debuginfo when inlining | Scott McMurray | -6/+0 | |
| 2024-04-03 | Remove MIR unsafe check | Matthew Jasper | -7/+3 | |
| This also remove safety information from MIR. | ||||
| 2024-03-10 | MIR printing: print the path of uneval'd const; refer to promoteds in a ↵ | Ralf Jung | -1/+1 | |
| consistent way | ||||
| 2023-12-24 | Enable GVN by default. | Camille GILLOT | -1/+5 | |
| 2023-07-20 | Always propagate into operands. | Camille GILLOT | -19/+15 | |
| 2023-07-19 | Turn copy into moves during DSE. | Camille GILLOT | -1/+1 | |
| 2023-06-19 | Promote unchecked_add/sub/mul/shl/shr to mir::BinOp | Scott McMurray | -21/+14 | |
| 2023-06-16 | [libs] Simplify `unchecked_{shl,shr}` | Scott McMurray | -79/+20 | |
| There's no need for the `const_eval_select` dance here. And while I originally wrote the `.try_into().unwrap_unchecked()` implementation here, it's kinda a mess in MIR -- this new one is substantially simpler, as shown by the old one being above the inlining threshold but the new one being below it. | ||||
| 2023-06-15 | Remove comments from mir-opt MIR dumps | Ben Kimock | -92/+86 | |
| 2023-06-03 | Only check inlining counter after recusing. | Camille GILLOT | -25/+25 | |
| 2023-06-01 | Rebase fallout. | Camille GILLOT | -65/+49 | |
| 2023-06-01 | Restrict test to x64. | Camille GILLOT | -1/+1 | |
| 2023-06-01 | Annotate needs-unwind. | Camille GILLOT | -1/+1 | |
| 2023-06-01 | Add e2e mir test for checked arithmetic. | Camille GILLOT | -0/+160 | |
