Commit Graph

5308 Commits

Author SHA1 Message Date
Barinzaya 33a3aab791 Added simd_extract_msbs intrinsic. 2025-02-24 08:39:32 -05:00
gingerBill d23453811d Improve semicolon checking rules when parsing 2025-02-24 13:13:36 +00:00
gingerBill ebda946d61 Move temporary array out of CheckerInfo 2025-02-22 18:12:43 +00:00
gingerBill de755f95ec Minor rearrange for parsing field lists 2025-02-22 17:58:14 +00:00
gingerBill 875dbf3140 Disallow syntax: proc(x,:T) 2025-02-22 17:56:42 +00:00
gingerBill 5bd43b94ec Improve error message for matrices with no rows or columns 2025-02-22 17:50:37 +00:00
gingerBill 748a771dad Allow indirection for swizzling on procedure parameters 2025-02-21 12:38:40 +00:00
gingerBill 55e0f945a1 Merge pull request #4860 from odin-lang/bill/typeid_hash_table
Change `typeid` definition to be based around the canonical type hash
2025-02-21 10:21:28 +00:00
Barinzaya 266e84103e Fixed an issue with SIMD vector equality.
Comparing SIMD vectors with `==` was checking that the mask of elements
that matched was not 0, meaning it succeeded if *any* element was equal,
rather than if *all* elements were equal.
2025-02-20 17:05:07 -05:00
gingerBill bf9f2e43bf Fix typeid size in LLVM 2025-02-20 18:47:41 +00:00
gingerBill e8c38e7427 Handle any correctly for 32-bit systems 2025-02-20 17:50:56 +00:00
gingerBill 614c0dd740 Fix typeid size for 32-bit platforms 2025-02-20 17:10:04 +00:00
gingerBill 5489a88983 Change typeid definition to be based around the canonical type hash
`typeid` used to be a fancy index with extra metadata stored on it. Now it is direct hash of the type.

This is safe to do in practice since any possible collisions are checked at compile time AND the chances of having a 1% collision are around 1 in 600K (see the Birthday Paradox).

Therefore accessing a `^Type_Info` is now a hash table lookup with linear probing. The table is twice the size than necessary so prevent too much probing due to an overly dense hash table.
2025-02-20 14:10:45 +00:00
gingerBill c25ac939d4 Add cast back 2025-02-20 13:27:18 +00:00
gingerBill 1d348318f2 Use PtrMap temporarily 2025-02-20 13:23:23 +00:00
gingerBill 0ab323012e Use TypeSet instead of PtrSet<Type *> 2025-02-20 11:12:59 +00:00
gingerBill 0946f6b48b Improve name canonicalization for some debuggers 2025-02-20 11:11:22 +00:00
gingerBill 90a973ccf6 Make debug symbols use the new name canonicalization rules 2025-02-20 10:27:59 +00:00
gingerBill 29456bcdea Move docs to the header 2025-02-19 11:12:32 +00:00
gingerBill 0bac34eec8 Number fields within procedures with a depth-first numbering system 2025-02-19 10:59:05 +00:00
gingerBill 827cd45f04 Add to check to fix crash 2025-02-18 17:12:58 +00:00
gingerBill 0482facdda Fix for weird builtin types 2025-02-18 15:26:59 +00:00
gingerBill e168cea670 Add offset for nested things 2025-02-18 15:17:04 +00:00
gingerBill 481745784f Remove debug print 2025-02-18 15:04:02 +00:00
gingerBill 04c1ff61f4 Update the comments 2025-02-18 15:03:49 +00:00
gingerBill 23efd1bd02 Add TypeWriter stream to allow for in-place hashing and string generation 2025-02-18 14:32:41 +00:00
gingerBill 19b59461b0 Use TypeSet for DeclInfo deps 2025-02-18 13:31:34 +00:00
gingerBill 721bcf2249 Minor code clean up 2025-02-18 13:24:08 +00:00
gingerBill d69eb57cfa Fix typos 2025-02-18 13:18:51 +00:00
gingerBill 4a29d9bb84 Simplify type info table construction 2025-02-17 16:29:42 +00:00
gingerBill b8f057951c Begin work on TypeSet 2025-02-17 13:46:17 +00:00
gingerBill 9b26bb2e6a Begin work on hash types 2025-02-17 13:10:38 +00:00
gingerBill 043f9aea61 Clean up rules for name mangling 2025-02-17 12:40:43 +00:00
gingerBill b5cf776830 Use new name canonicalization approach 2025-02-17 11:48:54 +00:00
gingerBill 99d91ccd31 Work on making name mangling deterministic 2025-02-17 11:32:49 +00:00
gingerBill 4eba3698aa Begin work on nested declarations 2025-02-17 09:47:49 +00:00
gingerBill 2a5933513c Fill in more canonical types 2025-02-15 12:32:05 +00:00
gingerBill 98201962e0 Begin work on producing a canonicalized type string for hashing types. 2025-02-14 17:29:38 +00:00
gingerBill fbee045023 Merge pull request #4797 from laytan/improve-abs-of-float
improve abs() on floats for more correct and faster results
2025-02-08 07:49:22 +00:00
gingerBill 58c4f13c82 Merge branch 'master' into bill/sdl3 2025-02-07 07:53:47 +00:00
gingerBill 667ddc1907 Merge pull request #4786 from korvahkh/fix-variants2ptrs
Preserve `#no_nil` in `intrinsics.type_convert_variants_to_pointers`
2025-02-07 07:30:14 +00:00
gingerBill 4e300b3fc8 Merge pull request #4796 from laytan/obfuscate-more
-obfuscate-source-code-locations on bounds checks and type assertions
2025-02-07 07:30:00 +00:00
Laytan Laats 5defddffd0 reorganize tests and handle endian 2025-02-06 21:44:34 +01:00
Laytan Laats de83ad2a25 apply abs fix to constant system too 2025-02-06 21:19:54 +01:00
Laytan Laats 1053ec3051 make corrected linkage with -use-separate-modules apply hidden visibility
Fixes #4798

The DLL was using the type info of the host/exe, causing crashes.
This PR tries fixing by applying hidden visibility to these corrected
symbols which makes sure that the DLL can't see the type table of the host/exe.
2025-02-06 20:45:58 +01:00
Laytan Laats b1068c7f2e improve abs() on floats for more correct and faster results 2025-02-05 21:17:48 +01:00
Laytan Laats b86d2c30b6 fix odin report macos version reporting unknown when release isn't a 3 point 2025-02-05 20:16:24 +01:00
Laytan Laats b77430bea8 -obfuscate-source-code-locations on bounds checks and type assertions 2025-02-05 19:42:20 +01:00
gingerBill f80bea5b11 Remove transmute suggestion with -vet-cast when transmuting native <-> endian-specific types 2025-02-05 10:27:17 +00:00
gingerBill ab469e657d Remove debug code 2025-02-05 10:20:18 +00:00