1530 Commits

Author SHA1 Message Date
r00ster91
297b5d1074 fix [x]u65529 and above overflowing
```
$ cat overflow.zig
test {
    var a: [1]u65535 = undefined;
    _ = a;
}
$ zig-out/bin/zig test overflow.zig
thread 290266 panic: integer overflow
zig/src/type.zig:3604:55: 0xada43d in intAbiAlignment (zig)
            std.math.ceilPowerOfTwoPromote(u16, (bits + 7) / 8),
                                                      ^
zig/src/type.zig:3598:42: 0xadd4ea in intAbiSize (zig)
        const alignment = intAbiAlignment(bits, target);
                                         ^
zig/src/type.zig:3500:61: 0x92be91 in abiSizeAdvanced (zig)
                return AbiSizeAdvanced{ .scalar = intAbiSize(bits, target) };
                                                            ^
zig/src/type.zig:3385:62: 0x928933 in abiSizeAdvanced (zig)
                switch (try payload.elem_type.abiSizeAdvanced(target, strat)) {
                                                             ^
zig/src/type.zig:3268:32: 0x92c012 in abiSize (zig)
        return (abiSizeAdvanced(ty, target, .eager) catch unreachable).scalar;
                               ^
```
This is only noticed in a debug build of zig and silently does the wrong
thing and overflows in release builds.

This happened to `[x]u65529` and above because of the ` + 7` on a `u16`.
2023-05-09 11:58:34 +03:00
Jacob Young
6778da4516 x86_64: implement binary operations for f16 and f16 vectors 2023-05-08 07:36:20 -04:00
Jacob Young
f8708e2c4d x86_64: implement @floor, @ceil, and @trunc for float vectors 2023-05-08 07:36:20 -04:00
Jacob Young
ea957c4cff x86_64: implement @sqrt for f16 scalars and vectors 2023-05-08 07:36:20 -04:00
Jacob Young
05580b9453 x86_64: implement float cast from f16 to f64 2023-05-08 07:36:20 -04:00
Jacob Young
1c53f0a6b0 Dwarf: workaround crash 2023-05-08 07:36:20 -04:00
Jacob Young
cba195c117 x86_64: implement some float and float vector movement
This allows actually storing value of these supported types in
registers, and not restricting them to stack slots.
2023-05-08 07:36:20 -04:00
Jacob Young
3a5e3c52e0 x86_64: implement @mulAdd 2023-05-08 07:36:20 -04:00
Jacob Young
5d4288c5f6 x86_64: fix unordered float equality 2023-05-08 07:36:19 -04:00
Jacob Young
32ab930f1d x86_64: implement f16 conversions when supported 2023-05-08 07:36:19 -04:00
Jacob Young
1a261917ce x86_64: implement @ctz and @clz for u128 2023-05-08 07:36:19 -04:00
Dominic
5a3eca5d4c
Disallow named test decls with duplicate names 2023-05-08 10:59:06 +03:00
Jacob Young
a19faa2481 x86_64: implement movement of more types
* f16
 * f128
 * vector
2023-05-03 04:25:14 -04:00
Jacob Young
36a39267b8 x86_64: fix feature confusion 2023-05-03 04:25:14 -04:00
Jacob Young
9ccdbca635 x86_64: implement fabs 2023-05-03 04:25:14 -04:00
Jacob Young
31429a4e86 codegen: handle variable and decl_ref_mut consistently 2023-05-03 04:25:14 -04:00
Jacob Young
f894ec264b codegen: fix global nested field_ptr 2023-05-03 04:25:14 -04:00
Jacob Young
3a30b82741 x86_64: implement fieldParentPtr 2023-05-03 04:25:14 -04:00
Jacob Young
3f5592c114 x86_64: implement slice elem ptr for more MCValue tags 2023-05-03 04:25:14 -04:00
Jacob Young
f56f5af403 x86_64: fix global slices 2023-05-03 04:25:14 -04:00
mlugg
152c7b1885 Implement multi-argument @min/@max and notice bounds
Resolves: #14039
2023-05-02 16:46:27 -07:00
Jacob Young
db88b41472 x86_64: fix switch multi-prongs and mul/div flags clobber 2023-05-01 19:22:53 -04:00
Jakub Konka
5e7f3d5daa x86_64: disable advanced memset tests on Windows 2023-05-01 19:22:53 -04:00
Jacob Young
47a34d038d x86_64: implement tagName 2023-05-01 19:22:52 -04:00
Jacob Young
c388960042 x86_64: fix large not and atomicrmw 2023-05-01 19:22:52 -04:00
Jacob Young
50f96c2949 x86_64: fix stack realignment 2023-05-01 19:22:52 -04:00
Jacob Young
db76ae8260 x86_64: fix emitting f80 globals 2023-05-01 19:22:52 -04:00
Jacob Young
6de457211f behavior: update affected tests for the x86_64 backend 2023-05-01 19:22:52 -04:00
Andrew Kelley
7baf0de807
Merge pull request #15458 from koachan/sparc64-codegen
stage2: sparc64: Yet another patchset for the selfhosted backend
2023-04-29 10:55:50 -07:00
Koakuma
d339e86fb1 stage2: sparc64: Skip unimplemented tests 2023-04-28 16:45:37 -07:00
Andrew Kelley
85ffb8f18f disable 2 failing behavior tests with x86 backend 2023-04-28 13:29:39 -07:00
Andrew Kelley
73d3fb9883 C backend: fix ptr comparison of array ptrs when one is null-terminated 2023-04-28 13:24:43 -07:00
Andrew Kelley
a67dec1c9f disable not yet passing new behavior tests from this branch 2023-04-28 13:24:42 -07:00
Andrew Kelley
9295355985 LLVM backend: optimize memset with comptime-known element
When the element is comptime-known, we can check if it has a repeated
byte representation. In this case, `@memset` can be lowered with the
LLVM intrinsic rather than with a loop.
2023-04-28 13:24:42 -07:00
Andrew Kelley
00b690540e llvm backend: fix lowering of memset
The bitcast of ABI size 1 elements was problematic for some types.
2023-04-28 13:24:42 -07:00
Andrew Kelley
011bc59e8a
Merge pull request #15435 from jacobly0/x86_64-frame
x86_64: add frame indices
2023-04-27 15:45:01 -07:00
Andrew Kelley
aa3405aabc
Merge pull request #15474 from Luukdegram/wasm-atomics
wasm: implement atomic instructions
2023-04-27 08:26:35 -07:00
Jakub Konka
bc69d5a00f macho: invalidate GOT/stub relocs after segment shift in memory 2023-04-26 19:08:38 -04:00
Jacob Young
b5b0b55582 x86_64: fix tlv references 2023-04-26 19:05:17 -04:00
Jacob Young
894406b9d3 behavior: update passing cbe tests 2023-04-26 19:05:17 -04:00
Andrew Kelley
3c66850e42
Merge pull request #15278 from ziglang/memcpy-memset
change semantics of `@memcpy` and `@memset`
2023-04-26 10:01:54 -07:00
Luuk de Gram
1d97181703
wasm: enable atomics behavior tests 2023-04-26 16:28:41 +02:00
Andrew Kelley
1ba72bcf9a update test cases for new memcpy/memset semantics 2023-04-25 11:23:41 -07:00
Andrew Kelley
58fabbc903 x86_64 backend: support @memset with slices 2023-04-25 11:23:41 -07:00
Andrew Kelley
881e931ee1 x86_64 backend: implement @memset for element ABI size > 1
* make memset and memset_safe guarantee that if the length is
   comptime-known then it will be nonzero.
2023-04-25 11:23:41 -07:00
Andrew Kelley
7c56145a76 add behavior test for @memset on slices
and avoid new language feature in std.ArrayList for now, until x86_64
self-hosted backend can implement it.
2023-04-25 11:23:41 -07:00
Andrew Kelley
482a0f648c Sema: implement comptime @memcpy 2023-04-25 11:23:41 -07:00
Andrew Kelley
83a7303bbf Sema: implement comptime @memset 2023-04-25 11:23:41 -07:00
Andrew Kelley
057c950093 LLVM backend: support non-byte-sized memset
Also introduce memset_safe AIR tag and support it in C backend and LLVM
backend.
2023-04-25 11:23:41 -07:00
Andrew Kelley
a5c910adb6 change semantics of @memcpy and @memset
Now they use slices or array pointers with any element type instead of
requiring byte pointers.

This is a breaking enhancement to the language.

The safety check for overlapping pointers will be implemented in a
future commit.

closes #14040
2023-04-25 11:23:40 -07:00