This was a "fake" type used to handle C varargs parameters, much like
generic poison. In fact, it is treated identically to generic poison in
all cases other than one (the final coercion of a call argument), which
is trivially special-cased. Thus, it makes sense to remove this special
tag and instead use `generic_poison_type` in its place. This fixes
several bugs in Sema related to missing handling of this tag.
Resolves: #19781
As of Clang 18, calling memcpy() with a misaligned pointer trips UBSAN,
even if the length is zero. This unfortunately includes any call to
`@memcpy` when source or destination are undefined and the length is
zero.
This patch makes the C backend avoid calling memcpy when the length is
zero, thereby avoiding undefined behavior.
A zig1.wasm update will be needed in the llvm18 branch to activate this
code.
wheras on NetBSD, only 2 PT_LOAD are usually produced by other compilers
(tested with host gcc and clang).
$ ldd -v main_4segs
.../main_4segs: wrong number of segments (4 != 2)
.../main_4segs: invalid ELF class 2; expected 1
* Fix the ELF binaries for freestanding target created with the self-hosted linker.
The ELF specification (generic ABI) states that ``loadable process segments must have congruent
values for p_vaddr and p_offset, modulo the page size''. Linux refuses to load binaries that
don't meet this requirement (execve() fails with EINVAL).
This is a partial revert of 105db13536b4dc2affe130cb8d2eee6c97c89bcd.
As we learned from Void Linux packaging, these options are not actually
helpful since the distribution package manager may very well want to
cross-compile the packages that it is building.
So, let's not overcomplicate things. There are already the standard
options: -Dtarget, -Dcpu, and -Ddynamic-linker.
These options are generally provided when the project generates machine
code artifacts, however, there may be a project that does no such thing,
in which case it makes sense for these options to be missing. The Zig
Build System is a general-purpose build system, after all.
Certain types (notably, `std.ComptimeStringMap`) were resulting in excessively
long type names when instantiated, which in turn resulted in excessively long
symbol names. These are problematic for two reasons:
* Symbol names are sometimes read by humans -- they ought to be readable.
* Some other applications (looking at you, xcode) trip on very long symbol names.
To work around this for now, we cap the depth of value printing at 1, as opposed
to the normal 3. This doesn't guarantee anything -- there could still be, for
instance, an incredibly long aggregate -- but it works around the issue in
practice for the time being.
The operation `undefined & 0` ought to result in the value `0`, and likewise for
zeroing only some bits. `std/packed_int_array.zig` tests were failing because
this behavior was not implemented -- this issue was previously masked by faulty
bitcast logic which turned `undefined` values into `0xAA` on pointer loads.
Ideally, we would like to be able to track the undefined bits at comptime.
This is related to #19634.
This commit reverts the handling of partially-undefined values in
bitcasting to transform these bits into an arbitrary numeric value,
like happens on `master` today.
As @andrewrk rightly points out, #19634 has unfortunate consequences
for the standard library, and likely requires more thought. To avoid
a major breaking change, it has been decided to revert this design
decision for now, and make a more informed decision further down the
line.
We've got a big one here! This commit reworks how we represent pointers
in the InternPool, and rewrites the logic for loading and storing from
them at comptime.
Firstly, the pointer representation. Previously, pointers were
represented in a highly structured manner: pointers to fields, array
elements, etc, were explicitly represented. This works well for simple
cases, but is quite difficult to handle in the cases of unusual
reinterpretations, pointer casts, offsets, etc. Therefore, pointers are
now represented in a more "flat" manner. For types without well-defined
layouts -- such as comptime-only types, automatic-layout aggregates, and
so on -- we still use this "hierarchical" structure. However, for types
with well-defined layouts, we use a byte offset associated with the
pointer. This allows the comptime pointer access logic to deal with
reinterpreted pointers far more gracefully, because the "base address"
of a pointer -- for instance a `field` -- is a single value which
pointer accesses cannot exceed since the parent has undefined layout.
This strategy is also more useful to most backends -- see the updated
logic in `codegen.zig` and `codegen/llvm.zig`. For backends which do
prefer a chain of field and elements accesses for lowering pointer
values, such as SPIR-V, there is a helpful function in `Value` which
creates a strategy to derive a pointer value using ideally only field
and element accesses. This is actually more correct than the previous
logic, since it correctly handles pointer casts which, after the dust
has settled, end up referring exactly to an aggregate field or array
element.
In terms of the pointer access code, it has been rewritten from the
ground up. The old logic had become rather a mess of special cases being
added whenever bugs were hit, and was still riddled with bugs. The new
logic was written to handle the "difficult" cases correctly, the most
notable of which is restructuring of a comptime-only array (for
instance, converting a `[3][2]comptime_int` to a `[2][3]comptime_int`.
Currently, the logic for loading and storing work somewhat differently,
but a future change will likely improve the loading logic to bring it
more in line with the store strategy. As far as I can tell, the rewrite
has fixed all bugs exposed by #19414.
As a part of this, the comptime bitcast logic has also been rewritten.
Previously, bitcasts simply worked by serializing the entire value into
an in-memory buffer, then deserializing it. This strategy has two key
weaknesses: pointers, and undefined values. Representations of these
values at comptime cannot be easily serialized/deserialized whilst
preserving data, which means many bitcasts would become runtime-known if
pointers were involved, or would turn `undefined` values into `0xAA`.
The new logic works by "flattening" the datastructure to be cast into a
sequence of bit-packed atomic values, and then "unflattening" it; using
serialization when necessary, but with special handling for `undefined`
values and for pointers which align in virtual memory. The resulting
code is definitely slower -- more on this later -- but it is correct.
The pointer access and bitcast logic required some helper functions and
types which are not generally useful elsewhere, so I opted to split them
into separate files `Sema/comptime_ptr_access.zig` and
`Sema/bitcast.zig`, with simple re-exports in `Sema.zig` for their small
public APIs.
Whilst working on this branch, I caught various unrelated bugs with
transitive Sema errors, and with the handling of `undefined` values.
These bugs have been fixed, and corresponding behavior test added.
In terms of performance, I do anticipate that this commit will regress
performance somewhat, because the new pointer access and bitcast logic
is necessarily more complex. I have not yet taken performance
measurements, but will do shortly, and post the results in this PR. If
the performance regression is severe, I will do work to to optimize the
new logic before merge.
Resolves: #19452Resolves: #19460
This was overlooked in #19458. Using the fully qualified name of each
module usually makes sense, but there is one module where it does not,
namely, the root module, since its name is `root`. The original Autodoc
tar creation logic used `comp.root_name` for the root module back when
it was the only module included in `sources.tar`, and that made sense.
Now, we get the best of both worlds, using the proper root name for the
root module while using the module name for the rest.
This allows `std.Uri.resolve_inplace` to properly preserve the fact
that `new` is already escaped but `base` may not be. I originally tried
just moving `raw_uri` around, but it made uri resolution unmanagably
complicated, so I instead added per-component information to `Uri` which
allows extra allocations to be avoided when constructing uris with
components from different sources, and in some cases, deferring the work
all the way to when the uri is printed, where an allocator may not even
be needed.
Closes#19587
Reference:
https://github.com/ziglang/zig/pull/19500#discussion_r1556476973
Arena is now used for Diagnostic (tar and git). `deinit` is not called on Diagnostic
allowing us to reference strings from Diagnostic in UnpackResult without
dupe.
That seamed reasonable to me. Instead of using gpa for Diagnostic, and
then dupe to arena. Or using arena for both and making dupe so we can deinit
Diagnostic.
Using test cases from:
https://github.com/ianprime0509/pathological-packages repository.
Depends on existence of the FAT32 file system. Folder is in FAT32 file
system because it is case insensitive and and does not support symlinks.
It is complicated test case requires internet connection, depends on
existence of FAT32 in the specific location. But it is so valuable for
development. Running `zig test Package.zig` is so much faster than
building zig binary and running `zig fetch URL`. Committing it here
although it should probably be removed.
