When the validation logic for param->type was added, the logic did not
account for the case where param could be NULL. The existing code did
take that into account as can be seen in the `if (param)` check below.
Furthermore, phpdbg_set_breakpoint_expression even calls
phpdbg_create_conditional_break with param == NULL.
Fix it by placing the validation logic inside a NULL check.
Commit 6c25413183 added the flag ZEND_JIT_EXIT_INVALIDATE which
resets the trace handlers in zend_jit_trace_exit(), but forgot to
lock the shared memory section.
This could cause another worker process who still saw the
ZEND_JIT_TRACE_JITED flag to schedule ZEND_JIT_TRACE_STOP_LINK, but
when it arrived at the ZEND_JIT_DEBUG_TRACE_STOP, the handler was
already reverted by the first worker process and thus
zend_jit_find_trace() fails.
This in turn generated a bogus jump offset in the JITed code, crashing
the PHP process.
Commit 6c25413 added the flag ZEND_JIT_EXIT_INVALIDATE which resets
the trace handlers in zend_jit_trace_exit(), but forgot to consider
that on ZEND_JIT_TRACE_STOP_LINK, this changed handler gets passed to
zend_jit_find_trace(), causing it to fail, either by returning 0
(results in bogus data) or by aborting due to ZEND_UNREACHABLE(). In
either case, this crashes the PHP process.
I'm not quite sure how to fix this multi-threading problem properly;
my suggestion is to just fail the zend_jit_trace() call. After all,
the whole ZEND_JIT_EXIT_INVALIDATE fix was about reloading modified
scripts, so there's probably no point in this pending zend_jit_trace()
call.
If zend_register_module_ex were to return NULL, then module_entry will
be set to NULL, and the if's body will load module_entry->name. Since
module_entry is NULL, loading the name would cause a NULL pointer
dereference. However, since a NULL pointer dereference is undefined
behaviour, the compiler is free to remove the check.
Fix it by using *name instead of module_entry->name.
Closes GH-10157
Signed-off-by: George Peter Banyard <girgias@php.net>
`ap_get_brigade()` may fail for different reasons, and we must not
pretend that a partially read POST payload is fine; instead we report
a content length of zero what matches all other `read_post()` callbacks
of bundled SAPIs.
Closes GH-10059.
A previous fix[1] was not sufficient to catch all potential file URIs,
because the patch did not cater to URL encoding. Properly parsing and
decoding the URI may yield a different result than the handling of
SQLite3, so we play it safe, and reject any file URIs if open_basedir
is configured.
[1] <https://bugs.php.net/bug.php?id=77967>
Closes GH-10018.
There are two issues to resolve:
1. The FCC is not refetch when trying to unregister a trampoline
2. Comparing the function pointer of trampolines is meaningless as they are reallocated, thus we need to compare the name of the function
Found while working on GH-8294
Closes GH-10033
Dialect 1 databases store and transfer `NUMERIC(15,2)` values as
doubles, which we need to cater to in `firebird_stmt_get_col()` to
avoid `ZEND_ASSUME(0)` to ever be triggered, since that may result
in undefined behavior.
Since adding a regression test would require to create a dialect 1
database, we go without it.
Closes GH-10021.
There might be a moment when the child log event is executed after
freeing a child. That could possibly happen if the child output is
triggered at the same as the terminating of the child. Then the output
event could be potentially processed after the terminating event which
would cause this kind of issue.
The issue might got more visible after introducing the log_stream on
a child because it is more likely that this cannot be dereferenced
after free. However it is very hard to reproduce this issue so there
is no test for this.
The fix basically prevents passing a child pointer and instead passes
the child PID and then looks the child up by the PID when it is being
processed. This is obviously slower but it is a safe way to do it and
the slow down should not be hopefully visible in a way that it would
overload a master process.
We need to overwrite the __toString magic method for SplFileObject, similarly to how DirectoryIterator overwrites it
Moreover, the custom cast handler is useless as we define __toString methods, so use the standard one instead.
Closes GH-9912
`.nFileIndexHigh` is a unsigned 32bit number. Casting that to `__int64`
and shifting left by 32bits triggers undefined behavior if the most
significant bit of `.nFileIndexHigh` is set. We could avoid that by
casting to `(__uint64)`, but in that case the whole clause doesn't have
an effect anymore, so we drop it altogether.
Closes GH-9958.
Directly referring to a constant of an undefined throws an exception;
there is not much point in `constant()` raising a fatal error in this
case.
Closes GH-9907.
`shm_get()` (not to be confused with `shmget()`) returns `NULL` if
reallocation fails; we need to cater to that when calling the function.
Closes GH-9872.
SaltStack uses Python subprocess and redirects stderr to stdout which is
then piped to the returned output. If php-fpm starts in daemonized mode,
it should close stderr. However a fix introduced in GH-8913 keeps stderr
around so it can be later restored. That causes the issue reported in
GH-9754. The solution is to keep stderr around only when php-fpm runs in
foreground as the issue is most likely visible only there. Basically
there is no need to restore stderr when php-fpm is daemonized.
When traversing the result array, we need to cater to `param_name`
possibly being `NULL`. Prior to PHP 7.0.0, this was implicitly done
because `param_name` was of type `char*`.
Closes GH-9739.
The fix for bug 63327[1] changed the extra size of mysqlnd allocations
from `sizeof(size_t)` to the properly aligned values; however, the
allocation in `_mysqlnd_pestrdup()` has apparently been overlooked,
which (currently) causes detectable heap corruption when running
mysqli_get_client_stats.phpt on 32bit Windows versions.
[1] <338a47bb85>
Closes GH-9724.
Changing indirect call to direct call for Jit code
benefits the branch prediction, which gets 1% performance
gain in our workload.
Similarly, we change indirect jump to direct jump.
Signed-off-by: Su, Tao <tao.su@intel.com>
Signed-off-by: Wang, Xue <xue1.wang@intel.com>
The phar wrapper needs to uncompress the file; the uncompressed file
might be compressed, so the wrapper implementation loops. This raises
potential DOS issues regarding too deep or even infinite recursion (the
latter are called compressed file quines[1]). We avoid that by
introducing a recursion limit; we choose the somewhat arbitrary limit
`3`.
This issue has been reported by real_as3617 and gPayl0ad.
[1] <https://honno.dev/gzip-quine/>
This change primarily splits SAPI deactivation to module and destroy
parts. The reason is that currently some SAPIs might bail out
on deactivation. One of those SAPI is PHP-FPM that can bail out on
request end if for example the connection is closed by the client
(web sever). The problem is that in such case the resources are not
freed and some values reset. The most visible impact can have not
resetting the PG(headers_sent) which can cause erorrs in the next
request. One such issue is described in #77780 bug which this fixes
and is also cover by a test in this commit. It seems reasonable
to separate deactivation and destroying of the resource which means
that the bail out will not impact it.
Using a lot of memory may overflow some `int` calculations; to avoid
that we make sure that the operands are promoted to `size_t`.
This issue has been analyzed by @chschneider.
Closes GH-9379.
While the reason-phrase in a HTTP response status line is usually
short, there is no actual limit specified by the RFCs. As such, we
must not assume that the line fits into the buffer (which is currently
128 bytes large).
Since there is no real need to present the complete status line, we
simply read and discard the rest of a long line.
Co-authored-by: Tim Düsterhus <timwolla@googlemail.com>
Closes GH-9319.
This fix is another solution to replace d0527427be, use zend_try and zend_catch to make sure persistent stream will be released when error occurred.
Closes GH-9332.
This reverts commit d0527427be.
This patch makes Swoole/Swow can not work anymore, because Coroutine will yield to another one during socket operation, EG(record_errors) assertion will always fail, and zend_begin_record_errors() was only used during compile time before.
Note: zend_emit_recorded_errors() and the typo fix are reserved.
This is not actually related to SSL handshake but stream socket creation
which does not clean errors if the error handler is set. This fix
prevents emitting errors until the stream is freed.
The comparator function used at ksort in SORT_REGULAR mode
need to be consistent with basic comparison rules. These rules
were changed in PHP-8.0 for numeric strings, but comparator
used at ksort kept the old behaviour. It leads to inconsistent
situations, when after ksort the first key is GREATER than some
of the next ones by according to the basic comparison operators.
Closes GH-9293.
* Remove ZEND_DVAL_TO_LVAL_CAST_OK
As far as I can see, this operation should always use the _slow method, and the results seem to be wrong when ZEND_DVAL_TO_LVAL_CAST_OK is enabled.
* update NEWS
We are removing the entry in 8.0.23, as the fix was included
in the 8.0.22 release. Given that an entry already exists, we
are just deleting extra lines.
The arguments 3 to 6 of the authorizer callback may be `NULL`[1], and
we have to properly deal with that. Instead of causing a segfault, we
deny authorization, which is still better than a crash, and apparently,
we cannot do better anyway.
[1] <https://www.sqlite.org/c3ref/set_authorizer.html>
Closes GH-9040.
If the blacklist file contains a line with a single double-quote, we
called `zend_strndup(pbuf, -1)` what causes an unnecessary bail out;
instead we just ignore that line.
If the blacklist file contains an empty line, we may have caused an OOB
read; instead we just ignore that line.
Closes GH-9036.
Passing `null` to `$encodings` is supposed to behave like passing the
result of `mb_detect_order()`. Therefore, we need to remove the non-
encodings from the `elist` in this case as well. Thus, we duplicate
the global `elist`, so we can modify it.
Closes GH-9063.
The `--CGI--` section is supposed to be just a marker, and to be empty
as such. However, a previous refactoring[1] broke that.
[1] <9140c9038a>
Closes GH-9061.
