Enable ThinLTO with incremental compilation.
This is an updated version of #52309. This PR allows `rustc` to use (local) ThinLTO and incremental compilation at the same time. In theory this should allow for getting compile-time improvements for small changes while keeping the runtime performance of the generated code roughly the same as when compiling non-incrementally.
The difference to #52309 is that this version also caches the pre-LTO version of LLVM bitcode. This allows for another layer of caching:
1. if the module itself has changed, we have to re-codegen and re-optimize.
2. if the module itself has not changed, but a module it imported from during ThinLTO has, we don't need to re-codegen and don't need to re-run the first optimization phase. Only the second (i.e. ThinLTO-) optimization phase is re-run.
3. if neither the module itself nor any of its imports have changed then we can re-use the final, post-ThinLTO version of the module. (We might have to load its pre-ThinLTO version though so it's available for other modules to import from)
Before this addition, every delimited group like (...) [...] {...} has
only a single Span that covers the full source location from opening
delimiter to closing delimiter. This makes it impossible for a
procedural macro to trigger an error pointing to just the opening or
closing delimiter. The Rust compiler does not seem to have the same
limitation:
mod m {
type T =
}
error: expected type, found `}`
--> src/main.rs:3:1
|
3 | }
| ^
On that same input, a procedural macro would be forced to trigger the
error on the last token inside the block, on the entire block, or on the
next token after the block, none of which is really what you want for an
error like above.
This commit adds group.span_open() and group.span_close() which access
the Span associated with just the opening delimiter and just the closing
delimiter of the group. Relevant to Syn as we implement real error
messages for when parsing fails in a procedural macro.
Reduce number of syscalls in `rand`
This skips the initial zero-length `getrandom` call and
directly hands the user buffer to the operating system, saving one
`getrandom` syscall.
Fix of bug introduced by #53762 (tool_lints)
Before implementing backwards compat for tool lints, the `Tool` case when parsing cmdline lints was unreachable. This changed with #53762.
This fix is needed for rls test-pass. (@nrc)
r? @Manishearth
