| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
_LIBCPP_HIDE_FROM_ABI everywhere (#131156)" (#141756)
This reverts commit c861fe8a71e64f3d2108c58147e7375cd9314521.
Unfortunately, this use of hidden visibility attributes causes
user-defined specializations of standard-library types to also be marked
hidden by default, which is incorrect. See discussion thread on #131156.
...and also reverts the follow-up commits:
Revert "[libc++] Add explicit ABI annotations to functions from the block runtime declared in <__functional/function.h> (#140592)"
This reverts commit 3e4c9dc299c35155934688184319d391b298fff7.
Revert "[libc++] Make ABI annotations explicit for windows-specific code (#140507)"
This reverts commit f73287e623a6c2e4a3485832bc3e10860cd26eb5.
Revert "[libc++][NFC] Replace a few "namespace std" with the correct macro (#140510)"
This reverts commit 1d411f27c769a32cb22ce50b9dc4421e34fd40dd.
|
|
_LIBCPP_HIDE_FROM_ABI everywhere (#131156)
This patch introduces `_LIBCPP_{BEGIN,END}_EXPLICIT_ABI_ANNOTATIONS`,
which allow us to have implicit annotations for most functions, and just
where it's not "hide_from_abi everything" we add explicit annotations.
This allows us to drop the `_LIBCPP_HIDE_FROM_ABI` macro from most
functions in libc++.
|
|
929f159777bec47c80a3b302f190261d426e1c3b removed the use of
`__ANDROID_API__`
|
|
The `std::error_code`/`std::error_category` functionality is designed to
support multiple error domains. On Unix, both system calls and libc
functions return the same error codes, and thus, libc++ today treats
`generic_category()` and `system_category()` as being equivalent.
However, on Windows, libc functions return `errno.h` error codes in the
`errno` global, but system calls return the very different `winerror.h`
error codes via `GetLastError()`.
As such, there is a need to map the winerror.h error codes into generic
errno codes. In libc++, however, the system_error facility does not
implement this mapping; instead the mapping is hidden inside libc++,
used directly by the std::filesystem implementation.
That has a few problems:
1. For std::filesystem APIs, the concrete windows error number is lost,
before users can see it. The intent of the distinction between
std::error_code and std::error_condition is that the error_code return
has the original (potentially more detailed) error code.
2. User-written code which calls Windows system APIs requires this same
mapping, so it also can also return error_code objects that other
(cross-platform) code can understand.
After this commit, an `error_code` with `generic_category()` is used to
report an error from `errno`, and, on Windows only, an `error_code` with
`system_category()` is used to report an error from `GetLastError()`. On
Unix, system_category remains identity-mapped to generic_category, but
is never used by libc++ itself.
The windows error code mapping is moved into system_error, so that
conversion of an `error_code` to `error_condition` correctly translates
the `system_category()` code into a `generic_category()` code, when
appropriate.
This allows code like:
`error_code(GetLastError(), system_category()) == errc::invalid_argument`
to work as expected -- as it does with MSVC STL.
(Continued from old phabricator review [D151493](https://reviews.llvm.org/D151493))
|
|
|
|
This is a follow-up to #89178. This updates the `<__config_site>`
macros.
|
|
Currently, the library-internal feature test macros are only defined if
the feature is not available, and always have the prefix
`_LIBCPP_HAS_NO_`. This patch changes that, so that they are always
defined and have the prefix `_LIBCPP_HAS_` instead. This changes the
canonical use of these macros to `#if _LIBCPP_HAS_FEATURE`, which means
that using an undefined macro (e.g. due to a missing include) is
diagnosed now. While this is rather unlikely currently, a similar change
in `<__configuration/availability.h>` caught a few bugs. This also
improves readability, since it removes the double-negation of `#ifndef
_LIBCPP_HAS_NO_FEATURE`.
The current patch only touches the macros defined in `<__config>`. If
people are happy with this approach, I'll make a follow-up PR to also
change the macros defined in `<__config_site>`.
|
|
|
|
Also introduce `_LIBCPP_ASSERT_PEDANTIC` for assertions violating which
results in a no-op or other benign behavior, but which may nevertheless
indicate a bug in the invoking code.
|
|
This patch runs clang-format on all of libcxx/include and libcxx/src, in
accordance with the RFC discussed at [1]. Follow-up patches will format
the benchmarks, the test suite and remaining parts of the code. I'm
splitting this one into its own patch so the diff is a bit easier to
review.
This patch was generated with:
find libcxx/include libcxx/src -type f \
| grep -v 'module.modulemap.in' \
| grep -v 'CMakeLists.txt' \
| grep -v 'README.txt' \
| grep -v 'libcxx.imp' \
| grep -v '__config_site.in' \
| xargs clang-format -i
A Git merge driver is available in libcxx/utils/clang-format-merge-driver.sh
to help resolve merge and rebase issues across these formatting changes.
[1]: https://discourse.llvm.org/t/rfc-clang-formatting-all-of-libc-once-and-for-all
|
|
(#67504)
This fixes a size regression in Fuchsia when building a static libc++
multilib with exceptions disabled. Referring to `system_category` in
`__throw_system_error` brings in a relatively large amount of additional
exception classes into the link without substantially improving the
error message.
|
|
When a handle to an error_category singleton object is used during the
termination phase of a program, the destruction of the error_category
object may have occurred prior to execution of the current destructor
or function registered with atexit, because the singleton object may
have been constructed after the corresponding initialization or call
to atexit. For example, the updated tests from this patch will fail if
using a libc++ built using a compiler that updates the vtable of the
object on destruction.
This patch attempts to avoid the issue by causing the destructor to not
be called in the style of ResourceInitHelper in src/experimental/memory_resource.cpp.
This approach might not work if object lifetime is strictly enforced.
Differential Revision: https://reviews.llvm.org/D65667
Co-authored-by: Louis Dionne <ldionne.2@gmail.com>
|
|
This was mention in D150044 and D154995 that this would be useful.
This addresses the last review coment of D150044.
Reviewed By: #libc, ldionne
Differential Revision: https://reviews.llvm.org/D156019
|
|
In the event the internal function __init is called with an empty string the code will take unnecessary extra steps, in addition, the code generated might be overall greater because, to my understanding, when initializing a string with an empty `const char*` "" (like in this case), the compiler might be unable to deduce the string is indeed empty at compile time and more code is generated.
The goal of this patch is to make a new internal function that will accept just an error code skipping the empty string argument. It should skip the unnecessary steps and in the event `if (ec)` is `false`, it will return an empty string using the correct ctor, avoiding any extra code generation issues.
After the conversation about this patch matured in the libcxx channel on the LLVM Discord server, the patch was analyzed quickly with "Compiler Explorer" and other tools and it was discovered that it does indeed reduce the amount of code generated when using the latest stable clang version (16) which in turn produces faster code.
This patch targets LLVM 18 as it will break the ABI by addressing https://github.com/llvm/llvm-project/issues/63985
Benchmark tests run on other machines as well show in the best case, that the new version without the extra string as an argument performs 10 times faster.
On the buildkite CI run it shows the new code takes less CPU time as well.
In conclusion, the new code should also just appear cleaner because there are fewer checks to do when there is no message.
Reviewed By: #libc, philnik
Spies: emaste, nemanjai, philnik, libcxx-commits
Differential Revision: https://reviews.llvm.org/D155820
|
|
This provides better error messages when the program terminates due to
an exception being thrown in -fno-exceptions mode. Those seem to have
been missed in https://reviews.llvm.org/D141222.
Differential Revision: https://reviews.llvm.org/D154995
|
|
Replace most uses of `_LIBCPP_ASSERT` with
`_LIBCPP_ASSERT_UNCATEGORIZED`.
This is done as a prerequisite to introducing hardened mode to libc++.
The idea is to make enabling assertions an opt-in with (somewhat)
fine-grained controls over which categories of assertions are enabled.
The vast majority of assertions are currently uncategorized; the new
macro will allow turning on `_LIBCPP_ASSERT` (the underlying mechanism
for all kinds of assertions) without enabling all the uncategorized
assertions (in the future; this patch preserves the current behavior).
Differential Revision: https://reviews.llvm.org/D153816
|
|
Other macros that disable parts of the library are named `_LIBCPP_HAS_NO_WHATEVER`.
Reviewed By: ldionne, Mordante, #libc
Spies: libcxx-commits, smeenai
Differential Revision: https://reviews.llvm.org/D143163
|
|
When we build the library with the stable ABI, we need to include some
functions in the dylib that were made inline in later versions of the
library (to avoid breaking code that might be relying on those symbols).
However, those methods were made non-inline whenever we'd be building
the library, which means that all translation units would end up using
the old out-of-line definition of these methods, as opposed to the new
inlined version. This patch makes it so that only the translation units
that actually define the out-of-line methods use the old definition,
opening up potential optimization opportunities in other translation
units.
This should solve some of the issues encountered in D65667.
Differential Revision: https://reviews.llvm.org/D123519
|
|
This is the first step towards disentangling the debug mode and assertions
in libc++. This patch doesn't make any functional change: it simply moves
_LIBCPP_ASSERT-related stuff to its own file so as to make it clear that
libc++ assertions and the debug mode are different things. Future patches
will make it possible to enable assertions without enabling the debug
mode.
Differential Revision: https://reviews.llvm.org/D119769
|
|
Our best guess is that the two syntaxes should have exactly equivalent
effects, so, let's be consistent with what we do in libcxx/include/.
I've left `#include "include/x.h"` and `#include "../y.h"` alone
because I'm less sure that they're interchangeable, and they aren't
inconsistent with libcxx/include/ because libcxx/include/ never
does that kind of thing.
Also, use the `_LIBCPP_PUSH_MACROS/POP_MACROS` dance for `<__undef_macros>`,
even though it's technically unnecessary in a standalone .cpp file,
just so we have consistently one way to do it.
Differential Revision: https://reviews.llvm.org/D119561
|
|
We've stopped doing it in libc++ for a while now because these names
would end up rotting as we move things around and copy/paste stuff.
This cleans up all the existing files so as to stop the spreading
as people copy-paste headers around.
|
|
|
|
We always build the libraries in a Standard mode that supports noexcept,
so there's no need to use the _NOEXCEPT macro.
Differential Revision: https://reviews.llvm.org/D97700
|
|
to reflect the new license. These used slightly different spellings that
defeated my regular expressions.
We understand that people may be surprised that we're moving the header
entirely to discuss the new license. We checked this carefully with the
Foundation's lawyer and we believe this is the correct approach.
Essentially, all code in the project is now made available by the LLVM
project under our new license, so you will see that the license headers
include that license only. Some of our contributors have contributed
code under our old license, and accordingly, we have retained a copy of
our old license notice in the top-level files in each project and
repository.
llvm-svn: 351648
|
|
_LIBCPP_BUILDING_LIBRARY
Summary: As suggested by Marshall in https://reviews.llvm.org/D49914
Reviewers: mclow.lists, EricWF
Subscribers: christof, dexonsmith, cfe-commits
Differential Revision: https://reviews.llvm.org/D50008
llvm-svn: 338475
|
|
The set of #ifdefs used to handle the two incompatible variants of
strerror_r were not complete (they didn't handle newlib appropriately).
Rather than attempting to make the ifdefs more complex, make them
unnecessary by choosing which behavior to use dependent upon the
return type.
Reviewers: waltl
Differential Revision: https://reviews.llvm.org/D34294
llvm-svn: 308528
|
|
Summary: This patch corrects the build errors I encountered when building on MinGW64.
Reviewers: mati865, rnk, compnerd, smeenai, bcraig
Reviewed By: mati865, smeenai
Subscribers: martell, chapuni, cfe-commits
Differential Revision: https://reviews.llvm.org/D33082
llvm-svn: 304360
|
|
It is my opinion that libc++ should never use `<cassert>`, including in the `dylib`.
This patch remove all uses of `assert` from within libc++ and replaces most of them with `_LIBCPP_ASSERT` instead.
Additionally this patch turn `LIBCXX_ENABLE_ASSERTIONS` off by default,
because the standard library should not be aborting user programs unless explicitly asked to.
llvm-svn: 294107
|
|
This reverts commit r292883. Unfortunately <string_view> uses
_LIBCPP_ASSERT in a way which is not compatible with the C++11 dylib
build. I'll investigate more tomorrow.
llvm-svn: 292923
|
|
Summary:
It is my opinion that libc++ should never use `<cassert>`, including in the `dylib`. This patch remove all uses of `assert` from within libc++ and replaces most of them with `_LIBCPP_ASSERT` instead.
Additionally this patch turn `LIBCXX_ENABLE_ASSERTIONS` off by default, because the standard library should not be aborting user programs unless explicitly asked to.
Reviewers: mclow.lists, compnerd, smeenai
Reviewed By: mclow.lists
Subscribers: mgorny, cfe-commits
Differential Revision: https://reviews.llvm.org/D29063
llvm-svn: 292883
|
|
library.
In order to allow inlining of previously out-of-line functions without an ABI break
libc++ provides legacy definitions in the dylib that old programs can
continue to use. Unfortunatly Windows link.exe detects this hack and diagnoses the duplicate
definitions.
This patch disable the duplicate definitions on Windows by adding an ABI option
which disables all "legacy out-of-line symbols"
llvm-svn: 292190
|
|
Replace the use of _WIN32 in libc++. Replace most use with a C runtime
check _LIBCPP_MSVCRT or the new _LIBCPP_WIN32 to indicate that we are
using the Win32 API. Use a new _LIBCPP_WCHAR_IS_UCS2 to indicate that we
are on an environment that has a short wchar_t.
llvm-svn: 290910
|
|
Provide a strerror_r replacement for Windows. This is needed to build
libc++ for Windows with threading.
llvm-svn: 290851
|
|
Currently libc++ compiles a special version of error_category()
into the dylib. This definition is no longer needed, and doesn't
work on Windows due to dllimport/dllexport semantics.
For those reasons this patch introduces an option to
disable/enable this definition. By default the definition
is provided in ABI v1 except on windows. This patch
also addresses D28210.
llvm-svn: 290840
|
|
When building libc++ without threading, strerror_r is not used. Define
the code only when threading is enabled. This allows us to build
system_error for Windows, which ATM doesn't build with threading.
llvm-svn: 290791
|
|
type we define. They either construct and throw the exception, or abort() (if exceptions are disabled). Use these functions everywhere instead of assert()ing when exceptions are disabled. WARNING: This is a behavior change - but only with exceptions disabled. Reviewed as: https://reviews.llvm.org/D23855.
llvm-svn: 279744
|
|
Summary:
Android didn't gain GNU's strerror_r until Marshmallow. If we're
building libc++ against something older (we build the NDK library
against the oldest release we support, currently Gingerbread), fall
back to the POSIX version.
Reviewers: mclow.lists, EricWF
Subscribers: tberghammer, danalbert, srhines, cfe-commits
Differential Revision: http://reviews.llvm.org/D21402
llvm-svn: 272827
|
|
llvm-svn: 272641
|
|
llvm-svn: 272640
|
|
Summary:
system_error::message() uses `strerror` for the generic and system categories. This function is not thread safe.
The fix is to use `strerror_r`. It has been available since 2001 for GNU libc and since BSD 4.4 on FreeBSD/OS X.
On platforms with GNU libc the extended version is used which always returns a valid string, even if an error occurs.
In single-threaded builds `strerror` is still used.
See https://llvm.org/bugs/show_bug.cgi?id=25598
Reviewers: majnemer, mclow.lists
Subscribers: erik65536, cfe-commits, emaste
Differential Revision: http://reviews.llvm.org/D20903
llvm-svn: 272633
|
|
llvm-svn: 245354
|
|
llvm-svn: 236265
|
|
Reviewers: K-ballo, mclow.lists, EricWF
Reviewed By: EricWF
Subscribers: jfb, jroelofs, majnemer, cfe-commits
Differential Revision: http://reviews.llvm.org/D6558
llvm-svn: 225273
|
|
llvm-svn: 216943
|
|
[syserr.errcat.objects]p4 specifies that
system_category().default_error_condition(ev) map to
error_condition(posv, generic_category()) if ev could map to a POSIX
errno.
Linux reserves up to and including 4095 for errno values, use this as a
bound.
This fixes syserr.errcat.objects/system_category.pass.cpp on Linux.
llvm-svn: 209795
|
|
llvm-svn: 204077
|
|
Leave the (existing, out-of-line, non-constexpr) in the dylib for compatibility with existing programs)
llvm-svn: 188858
|
|
llvm-svn: 178267
|
|
llvm-svn: 134190
|
|
llvm-svn: 132137
|
