The dynamic loader, starting around version 1284, patches the thunk emitted for
thread local variables by the compiler, so that its format changes from
struct Thunk {
void *func;
size_t module;
size_t offset;
}
to
struct Thunk_v2 {
void *func;
uint32_t module;
uint32_t offset;
// other fields
}
which has the same size, but not the same layout.
This is mentionned in
9307719dd8/libdyld/ThreadLocalVariables.h (L90)
As a result, access to thread specific variables in JIT is broken.
Fix by using the new layout when the new dynamic loader is in use.
Closes GH-20121
The PHAR_ZIP_FAIL and PHAR_ZIP_FAIL_FREE macros are almost the same.
The reason the latter exists is because of a single error path where the
error message is on the heap and needs to be freed. Instead, use a stack
allocated variable for the error message so we can get rid of the
duplicate macro code. This stack variable is big enough as the messages
written by phar_verify_signature() are short.
Closes GH-20144.
If only we did not have the pseudo-key "@attributes", we could've just
removed the custom get_debug_info implementation and this would work out
of the box. Anyway, we just have to manually check for an override now.
Closes GH-20131.
preload_load() reads EG(class_table) and EG(function_table), but these may not
be initialized. Move these accesses out of preload_load().
Closes GH-20081
The fix fixes some other races that could result in mixed up copy and
nprocs number. It requires creating a copy in a similar way like for
request status.
This was not previously used to not impact other requests. However this
does not make that much sense as the only thing that impacts it is
holding the lock and not waiting for it. It is true that if there was a
big contention then the lock would be acquired more often but that can
be achieved by using fpm_get_status in loop so it should not make a
huge difference hopefully.
Closes GH-19974
In CGI, php_auto_globals_create_server() (i.e. auto_global_callback() here)
initializes $_ENV to reuse for $_SERVER. However, because $_SERVER is
constructed first, we have not yet initialized auto_global->armed of the $_ENV
global. Split the loop into initialization and constructor phases.
Fixes GH-19934
Closes GH-19870
* PHP-8.2:
Update NEWS with info about curl 8.16 compat fixes
Fix curl_setopt_ssl test for curl 8.16
Fix more curl 8.16 issues
Fix curl 8.16.0 compilation with zts
Fix curl build failure on macOS+curl 8.16
* PHP-8.1:
Update NEWS with info about curl 8.16 compat fixes
Fix curl_setopt_ssl test for curl 8.16
Fix more curl 8.16 issues
Fix curl 8.16.0 compilation with zts
Fix curl build failure on macOS+curl 8.16
On Linux, these two character devices are exceptions in that they can be
seeked. Check their major/minor device number.
Co-authored-by: divinity76 <hans@loltek.net>
This ensures that no useless "Failed to poll event" error messages are
logged during normal server operation, as the SOCK_EAGAIN error simply
indicates another worker is already serving the request.
Closes GH-19964
The reason this happens is because the array_unique operation happens in-place
because the input array is RC1.
At one point during comparison an exception is thrown which will capture the
arguments in the backtrace, which will increment the refcount of the RC1 array
to 2. Then a modification happens after the throw on the RC2 array causing the
assertion failure.
We shouldn't try continue work after an exception happened during the sort.
Closes GH-20059.
The #if to declare instrumented versions of strlcpy and strlcat was too
inclusive on *BSD systems where Clang already provides instrumented strong
symbols, resulting in "duplicate symbol" at link-time.
Fix GH-20002
Closes GH-20032
Change the reproducer code in `bug35916.phpt` from `stream_bucket_append` to
`stream_bucket_prepend` and you have the same bug.
Furthermore, even in the append case the check is incorrect because the bucket
can already be in the brigade at a position other than the tail.
To solve this properly, unlink the brigade first and also use that as a
condition to manage the refcount.
Closes GH-18973.
Or on Windows it is going to use either FormatMessageW or strerror_s
for compatibility with previous error messages.
It also needs to accomodate for GNU and BSD versions of strerror_r
returning different type.
Closes GH-19251
