Andrew Kelley
3b77f23e76
Merge pull request #20194 from ziglang/musl-v1.2.5
...
upgrade musl to v1.2.5
2024-06-05 13:44:05 -04:00
Andrew Kelley
ac3faa6cb3
LLVM backend: loongarch64 support
2024-06-05 13:43:35 -04:00
Andrew Kelley
0098e650fb
update musl source files to v1.2.5
...
adds loongarch64 and riscv32
2024-06-04 16:54:14 -07:00
Veikka Tuominen
7f6ec51c8b
Merge pull request #20084 from Vexu/missing-errors
...
Add missing errors to `@ptrFromInt` and Signal calling convention validation
2024-06-03 14:20:21 +03:00
Carl Åstholm
d74180c373
Replace YES_COLOR with CLICOLOR_FORCE
...
Instead of introducing YES_COLOR, a completely new standard, into the mix
it might make more sense to instead tag along with the CLICOLOR_FORCE env var,
which dates back to at least 2000 with FreeBSD 4.1.1 and which is
supported by tools like CMake.
<https://bixense.com/clicolors/ >
2024-06-02 17:35:34 -04:00
Andrew Kelley
0c2cd83814
zig run: finish progress node before executing child
...
also lock stderr for good measure. it's generally a good idea to lock
stderr when you are spawning and waiting for a child that inherits stderr.
2024-06-02 15:27:28 -04:00
Veikka Tuominen
15791b8b1a
Sema: validate function signature for Signal calling convention
2024-06-02 21:42:13 +03:00
Veikka Tuominen
17a0458e53
Sema: add missing error for runtime @ptrFromInt to comptime-only type
...
Closes #20083
2024-06-02 21:42:13 +03:00
Veikka Tuominen
c01d8c8b20
pass -fno-builtin when testing lib/c.zig, lib/compiler_rt.zig
2024-06-02 10:10:35 +03:00
Veikka Tuominen
06d9f88f97
llvm: disable constant anons on Darwin
2024-06-01 19:32:19 +03:00
Vahur Sinijärv
4af163ac94
mark anondecls as constants in llvm ir
2024-05-30 00:32:16 +03:00
Andrew Kelley
3b77f1ed7e
rename zig-cache to .zig-cache
...
closes #20077
2024-05-29 10:20:15 -07:00
Andrew Kelley
5bdfe22092
CLI: actually fix translate-c creating root progress node twice
...
7281cc1d839da6e84bb76fadb2c1eafc22a82df7 did not solve the problem
because even when Node.index is none, it still counts as initializing
the global Progress object. Just use a normal zig optional, and all is
good.
2024-05-27 20:56:49 -07:00
Andrew Kelley
b7889f262a
zig build: respect --color argument
...
`--color off` now disables the CLI progress bar both in the parent
process and the build runner process.
2024-05-27 20:56:49 -07:00
Andrew Kelley
44389253c2
fix zig translate-c creating root progress node twice
2024-05-27 20:56:49 -07:00
Andrew Kelley
c9587d3e40
CLI: add missing call to root progress node end()
...
cleans up unwanted "LLVM Emit Object" being left on the screen
2024-05-27 20:56:49 -07:00
Andrew Kelley
849693f07c
zig build: give a root progress name
...
Now it's more clear when zig is building the build script vs building
the actual project.
2024-05-27 20:56:48 -07:00
Andrew Kelley
d3b03ed64b
frontend: fix use of undefined progress node
...
This was causing a crash when running `zig test`.
2024-05-27 20:56:48 -07:00
Andrew Kelley
e820173467
Compilation: fix sub-compilations given wrong progress node
2024-05-27 20:56:48 -07:00
Jacob Young
d77f5e7aaa
Progress: fix compile errors on windows
...
Works for `zig build-exe`, IPC still not implemented yet.
2024-05-27 20:56:48 -07:00
mlugg
d403d8cb7a
Module: fix and improve progress reporting
...
* correctly report time spent analyzing function bodies
* print fully qualified decl names
* also have a progress node for codegen
The downside of these changes is that it's a bit flickerey, but the
upside is that it's accurate; you can see what the compiler's doing!
2024-05-27 20:56:48 -07:00
Andrew Kelley
2759173452
zig build: lock stderr while executing the build runner
2024-05-27 20:56:48 -07:00
Andrew Kelley
352dc2c06a
compiler: show decl name in progress node
2024-05-27 20:56:48 -07:00
Andrew Kelley
2233d95b0f
main: avoid creating multiple std.Progress instances
2024-05-27 20:56:48 -07:00
Andrew Kelley
b27fdf92fb
Compilation: only create progress nodes for work actually being done
2024-05-27 20:56:48 -07:00
Andrew Kelley
f97c2f28fd
update the codebase for the new std.Progress API
2024-05-27 20:56:48 -07:00
Andrew Kelley
591bbafee3
Merge pull request #20049 from ziglang/std.process.Child
...
std: restructure child process namespace
2024-05-26 16:20:07 -04:00
Jacob Young
6e5e7e7b19
Compilation: fix regressed assembly diagnostics
...
Regressed by #17947
2024-05-26 12:38:31 -04:00
Andrew Kelley
f47824f24d
std: restructure child process namespace
2024-05-26 09:31:55 -07:00
Jakub Konka
42c058575e
link/macho: fix 32bit build
2024-05-24 23:06:27 +02:00
Jakub Konka
ed7073c630
link/macho: fix perf bug in DWARF parsing
2024-05-24 15:34:48 +02:00
Jakub Konka
d31eb744ce
link/macho: fix 32bit build
2024-05-23 12:04:17 +02:00
Jakub Konka
f3a503eca2
link/macho: ensure we set alignment of literals to max alignment
2024-05-23 12:04:17 +02:00
Jakub Konka
71bbc5efc9
link/macho: print error message when hitting unexpected remainder error
2024-05-23 12:04:17 +02:00
Jakub Konka
8fc0c7dce1
link/macho: apply fixes to deduping logic
...
* test non-ObjC literal deduping logic
2024-05-23 12:04:17 +02:00
Jakub Konka
434e69482e
link/macho: dedup literals in objects and internal object file
2024-05-23 12:04:17 +02:00
Andrew Kelley
9be8a9000f
Revert "implement @expect builtin ( #19658 )"
...
This reverts commit a7de02e05216db9a04e438703ddf1b6b12f3fbef.
This did not implement the accepted proposal, and I did not sign off
on the changes. I would like a chance to review this, please.
2024-05-22 09:57:43 -07:00
David Rubin
a7de02e052
implement @expect builtin ( #19658 )
...
* implement `@expect`
* add docs
* add a second arg for expected bool
* fix typo
* move `expect` to use BinOp
* update to newer langref format
2024-05-22 10:51:16 -05:00
Jacob Young
ed75f62568
Compilation: better cleanup of temporary files
...
The diag file did not get deleted on the success path and the dep file
did not get deleted on the failure path.
2024-05-22 10:42:01 -04:00
Veikka Tuominen
0fb2015fd3
llvm: fix @wasmMemory{Size,Grow} for wasm64
...
Closes #19942
2024-05-22 09:48:52 -04:00
Veikka Tuominen
5fe9f88b13
Merge pull request #20023 from Vexu/fixes
...
Minor LLVM backend fixes
2024-05-22 11:54:38 +03:00
Veikka Tuominen
eb0f871cb9
Merge pull request #19961 from wooster0/errorstuff
...
Sema: improvements to error messages related to the handling of (error) values
2024-05-21 22:37:00 +03:00
wooster0
ac55685a94
Sema: add missing declared here note
2024-05-22 02:16:56 +09:00
Veikka Tuominen
f776e70c39
llvm: fix lowering of packed structs with optional pointers
...
Closes #20022
2024-05-21 19:52:25 +03:00
Veikka Tuominen
167854c19c
llvm: lower ptr to int constants with correct address spaces
...
Closes #19915
2024-05-21 19:52:00 +03:00
Veikka Tuominen
8aae0d87b5
Target: add OpenHarmonyOS ABI
...
Closes #20009
2024-05-20 09:25:52 -04:00
Wooster
f14cf13ff8
Sema: suggest using try/catch/if on method call on error union
2024-05-14 01:13:49 +09:00
r00ster91
9ae43567a3
Sema: improve error set/union discard/ignore errors
...
Previously the error had a note suggesting to use `try`, `catch`, or
`if`, even for error sets where none of those work.
Instead, in case of an error set the way you can handle the error
depends very much on the specific case. For example you might be in a
`catch` where you are discarding or ignoring the error set capture
value, in which case one way to handle the error might be to `return`
the error.
So, in that case, we do not attach that error note.
Additionally, this makes the error tell you what kind of an error it is:
is it an error set or an error union? This distinction is very relevant
in how to handle the error.
2024-05-14 01:13:49 +09:00
r00ster91
8579904ddd
Sema: add error note for !?Type types when optional type is expected
2024-05-14 01:13:49 +09:00
r00ster91
31689d0dd9
Sema: remove periods from the few error messages that have them
...
For consistency.
2024-05-14 01:13:49 +09:00
