* origin/master: (40 commits)
int to size_t where the underlaying API supports it
use php_socket_t instead of int
fix signed/unsigned mismatch warning
fix compilation warning
Improved specialisation $this variable accessed through IS_UNUSED operand must be IS_OBJECT, so we don't have to check for its type or perform dereference.
Add notes about get_class_entry/get_class_name to UPGRADING
Fix casts in GD
Drop redundant casting code from ext/filter
update NEWS
update NEWS
update NEWS
update NEWS
Added note to UPGRADING regarding 64-bit support in pack()/unpack()
pack(): Use SIZEOF_ZEND_LONG instead of SIZEOF_LONG
Add 64 bit formats to pack() and unpack()
Help to CPU branch predictor
Removed unused EG(orig_error_reporting)
Update get_class_name semantics
Remove Z_OBJ_CLASS_NAME_P
Improved VM stack primitives for fast paths. Slow paths are not inlined anymore.
...
TLS is already used in TSRM, the way exporting the tsrm cache through
a thread local variable is not portable. Additionally, the current
patch suffers from bugs which are hard to find, but prevent it to
be worky with apache. What is done here is mainly uses the idea
from the RFC patch, but
- __thread variable is removed
- offset math and declarations are removed
- extra macros and definitions are removed
What is done merely is
- use an inline function to access the tsrm cache. The function uses
the portable tsrm_tls_get macro which is cheap
- all the TSRM_* macros are set to placebo. Thus this opens the way
remove them later
Except that, the logic is old. TSRMLS_FETCH will have to be done once
per thread, then tsrm_get_ls_cache() can be used. Things seeming to be
worky are cli, cli server and apache. I also tried to enable bz2
shared and it has worked out of the box. The change is yet minimal
diffing to the current master bus is a worky start, IMHO. Though will
have to recheck the other previously done SAPIs - embed and cgi.
The offsets can be added to the tsrm_resource_type struct, then
it'll not be needed to declare them in the userspace. Even the
"done" member type can be changed to int16 or smaller, then adding
the offset as int16 will not change the struct size. As well on the
todo might be removing the hashed storage, thread_id != thread_id and
linked list logic in favour of the explicit TLS operations.
#NOTE: There is a very small possibility that this will further break
#extensions that access wrapper->{err_stack, err_count}. On PECL SVN, rar is the
#only one and it may leak memory after this. I say "further break" because
#extensions that do that are already broken (will segfault) under ZTS, which is
#why this patch is necessary.
#There was what I deem as tacit acceptance from 5.3/5.4 RMs on this.
effect)
#Made "from" a proper INI setting and bound it to a global variable.
#Previously, it was simply read from the hash table with the parsed ini file
#by using cfg_get_string (I wonder why this mechanism still exists...)
# To not change a PHPAPI in a point release, a new function have been added to support contexts:
# php_copy_file_ctx(), php_copy_file_ex() now simply wraps to that
# After looking at the logs, Jani did a bad merge into 5.3, so that
# the context parameter sent to copy() actually isn't used at all. This
# relatively simple patch fixes that for trunk.
#
# See FR #42965
# internals:
# This changes the php_copy_*() decls to contain an additional parameter for stream contexts
Userdefined wrappers were being registered into a global wrapper hash
which can cross threads. Termination of once instance then has the
potential to leave an active stream in another instance with no wrapper
leading to segfault.
. stream_socket_client() - similar to fsockopen(), but more powerful.
. stream_socket_server() - Creates a server socket.
. stream_socket_accept() - Accept a client connection.
. stream_socket_get_name() - Get local or remote name of socket.
Tidy up some leaks and debug printfs.
Move more streams functions into streamsfuncs.c and streamsfuncs.h.