13 Commits

Author SHA1 Message Date
Oli Scherer
ee3c835018 Always import all tracing macros for the entire crate instead of piecemeal by module 2022-09-01 14:54:27 +00:00
Alan Egerton
4f0a64736b
Update TypeVisitor paths 2022-07-06 06:41:53 +01:00
LegionMammal978
4937a55dfb Remove in_band_lifetimes from rustc_codegen_llvm
See #91867 for more information.
2021-12-16 14:43:32 -05:00
Guillaume Gomez
759eba0a08 Fix clippy lints 2021-10-01 23:17:19 +02:00
Manish Goregaokar
6f1e930581
Rollup merge of #88820 - hlopko:add_pie_relocation_model, r=petrochenkov
Add `pie` as another `relocation-model` value

MCP: https://github.com/rust-lang/compiler-team/issues/461
2021-10-01 09:18:16 -07:00
Marcel Hlopko
198d90786b Add pie as another relocation-model value 2021-10-01 08:06:42 +02:00
Eduard-Mihai Burtescu
c1837ef1c5 Querify fn_abi_of_{fn_ptr,instance}. 2021-09-18 04:41:33 +03:00
Eduard-Mihai Burtescu
e9b68304ef ty::layout: replicate layout_of setup for fn_abi_of_{fn_ptr,instance}. 2021-09-18 04:41:29 +03:00
Eduard-Mihai Burtescu
4ce933f13f rustc_target: move LayoutOf to ty::layout. 2021-09-02 01:17:14 +03:00
Patrick Walton
c17b1904a9 Stop emitting the dso_local LLVM attribute for external symbols under the static relocation model on macOS.
This matches Clang's behavior:

973cb2c326/clang/lib/CodeGen/CodeGenModule.cpp (L1038-L1040)

Even if `dso_local` were properly supported in this way on macOS, it seems
incorrect to add this annotation as liberally as we did. The `dso_local`
annotation is for symbols that ultimately end up in the same linkage unit, but
we were adding this annotation even for `static` values inside `extern` blocks
marked with `#[link(type="framework")]`, which should be considered dynamically
linked.  Note that Clang likewise avoids emitting `dso_local` for `dllimport`
symbols:

973cb2c326/clang/lib/CodeGen/CodeGenModule.cpp (L1005-L1007)

This issue caused breakage in the `ring` crate, which links to a symbol defined
in `Security.framework` that ultimately resolves to address `0x0`:

b94d61e044/src/rand.rs (L390)

For this symbol, the use of `dso_local` causes LLVM to emit a relocation of
type `X86_64_RELOC_SIGNED`, which is a 32-bit signed PC-relative offset. If the
binary is large enough, `0x0` might be out of range, and the link will fail.
Avoiding `dso_local` causes LLVM to use the GOT instead, emitting a relocation
of type `X86_64_RELOC_GOT_LOAD`, which will properly handle the large offset
and cause the link to succeed.

As a side note, the static relocation model is effectively deprecated for
security reasons on macOS, as it prohibits PIE. It's also completely
unsupported on Apple Silicon, so I don't think it's worth going to the effort
of properly supporting this model on that platform.
2021-08-20 17:10:41 -07:00
Boris-Chengbiao Zhou
4a76934aa7 Fix static relocation model for PowerPC64
We now also use `should_assume_dso_local()` for declarations and port two
additional cases from clang:
- Exclude PPC64 [1]
- Exclude thread-local variables [2]

[1]: 033138ea45/clang/lib/CodeGen/CodeGenModule.cpp (L1038-L1040)
[2]: 033138ea45/clang/lib/CodeGen/CodeGenModule.cpp (L1048-L1050)
2021-05-28 03:48:39 +02:00
Simonas Kazlauskas
2f000a78bf Manually set dso_local when its valid to do so
This should have no real effect in most cases, as e.g. `hidden`
visibility already implies `dso_local` (or at least LLVM IR does not
preserve the `dso_local` setting if the item is already `hidden`), but
it should fix `-Crelocation-model=static` and improve codegen in
executables.

Note that this PR does not exhaustively port the logic in [clang]. Only
the obviously correct portion and what is necessary to fix a regression
from LLVM 12 that relates to `-Crelocation_model=static`.

Fixes #83335

[clang]: 3001d080c8/clang/lib/CodeGen/CodeGenModule.cpp (L945-L1039)
2021-04-03 00:00:29 +03:00
mark
9e5f7d5631 mv compiler to compiler/ 2020-08-30 18:45:07 +03:00