summary |
shortlog | log |
commit |
commitdiff |
tree
first ⋅ prev ⋅ next
TORRI Vincent [Sun, 5 Mar 2006 08:40:15 +0000 (09:40 +0100)]
fixes + valid css
TORRI Vincent [Sun, 5 Mar 2006 08:22:03 +0000 (09:22 +0100)]
fixes + valid css
TORRI Vincent [Sun, 5 Mar 2006 08:05:21 +0000 (09:05 +0100)]
lots of fixes. Thanks to Indan Zupancic
TORRI Vincent [Sun, 5 Mar 2006 06:54:20 +0000 (07:54 +0100)]
lots of fixes. Thanks to Indan Zupancic
Josh Triplett [Sun, 5 Mar 2006 02:35:53 +0000 (18:35 -0800)]
Add *.tar.{bz2,gz} to .gitignore for the benefit of "make distcheck".
Jamey Sharp [Fri, 3 Mar 2006 19:08:10 +0000 (11:08 -0800)]
API/ABI change: XCBSendRequest returns the sequence number instead of using an out-parameter. Now 0 is a special sequence number indicating failure.
Jamey Sharp [Fri, 3 Mar 2006 09:45:00 +0000 (01:45 -0800)]
Add 32-bit full_sequence fields to generic errors and events, so callers can always correlate events and errors with requests correctly.
Jamey Sharp [Fri, 3 Mar 2006 07:39:38 +0000 (23:39 -0800)]
Tweak to previous API change: Require that spare iovecs fall before vector[0]. Leave vector in well-defined state.
Jamey Sharp [Fri, 3 Mar 2006 07:31:35 +0000 (23:31 -0800)]
assert() that XCBSendRequest was handed enough space to set opcodes and short length fields.
Jamey Sharp [Fri, 3 Mar 2006 07:12:21 +0000 (23:12 -0800)]
Bugfix: The fixed-length part needs two iovecs reserved, just like all the other parts.
Jamey Sharp [Fri, 3 Mar 2006 07:05:08 +0000 (23:05 -0800)]
Use XPath position() function instead of xsl:number, to evaluate array indices while generating code for easier readability.
Jamey Sharp [Thu, 2 Mar 2006 23:35:31 +0000 (15:35 -0800)]
API/ABI change: XCBSendRequest callers must pad to 4-byte boundaries now. When not in RAW mode, a null pointer for iov_base is replaced by up to 3 padding bytes.
Jamey Sharp [Thu, 2 Mar 2006 21:49:00 +0000 (13:49 -0800)]
Separate notion of request-completed from current-request, and mark requests completed more aggressively. Detects some usage errors that are otherwise undetectable.
Jamey Sharp [Mon, 27 Feb 2006 20:12:33 +0000 (12:12 -0800)]
Buffer a couple CARD32s on the stack instead of using an extra iovec. Also a bugfix: do not hold a reference to longlen after it goes out of scope.
Jamey Sharp [Mon, 27 Feb 2006 19:03:13 +0000 (11:03 -0800)]
Add XCB_REQUEST_RAW flag for XCBSendRequest.
Jamey Sharp [Mon, 27 Feb 2006 18:53:08 +0000 (10:53 -0800)]
Bugfix: how about *not* leaking all pending_replies when no reply matches, as often happens with Xlib?
Jamey Sharp [Mon, 27 Feb 2006 10:57:36 +0000 (02:57 -0800)]
Move test for other writing threads *before* allocating a sequence number.
Jamey Sharp [Mon, 27 Feb 2006 10:14:48 +0000 (02:14 -0800)]
_xcb_list is no longer used, so remove it. Simplify _xcb_map.
Jamey Sharp [Mon, 27 Feb 2006 09:43:07 +0000 (01:43 -0800)]
Replace readers generic list with hand-written typesafe version.
Jamey Sharp [Mon, 27 Feb 2006 09:06:15 +0000 (01:06 -0800)]
Rename struct XCBReplyData to reader_list to follow my naming conventions and better reflect its purpose.
Jamey Sharp [Mon, 27 Feb 2006 08:54:59 +0000 (00:54 -0800)]
In struct XCBReplyData, change void *data to pthread_cond_t *data. That was dumb.
Jamey Sharp [Mon, 27 Feb 2006 07:43:44 +0000 (23:43 -0800)]
Delete unused xcb_list functions and refactor others.
Jamey Sharp [Mon, 27 Feb 2006 06:58:17 +0000 (22:58 -0800)]
Replace current_reply generic queue with hand-written version. No generic queues remain so delete their implementation.
Jamey Sharp [Mon, 27 Feb 2006 02:28:50 +0000 (18:28 -0800)]
Replace events generic queue with hand-written typesafe version.
Jamey Sharp [Sun, 26 Feb 2006 23:45:08 +0000 (15:45 -0800)]
API/ABI break: Add flags to XCBSendRequest, first for error-checked requests.
There's no more race condition between event and reply handling.
The *RequestBlind and *RequestChecked functions are not yet implemented.
Jamey Sharp [Sun, 26 Feb 2006 10:00:03 +0000 (02:00 -0800)]
Replace pending_replies generic queue with a hand-implemented typesafe version.
Jamey Sharp [Sun, 26 Feb 2006 09:27:01 +0000 (01:27 -0800)]
Replace my old generic map ADT with a growable array for the extension cache.
Jamey Sharp [Sun, 26 Feb 2006 07:27:47 +0000 (23:27 -0800)]
XCB has not had tracing features for a long time: remove the remnants.
Jamey Sharp [Sun, 26 Feb 2006 07:26:55 +0000 (23:26 -0800)]
Rearrange an if statement that's been bothering me.
Jamey Sharp [Fri, 24 Feb 2006 10:17:36 +0000 (02:17 -0800)]
Coalesce _xcb_writev into _xcb_out_write and simplify.
Jamey Sharp [Fri, 24 Feb 2006 09:56:18 +0000 (01:56 -0800)]
Take advantage of Requires and *.private fields for a more accurate pkg-config file.
Jamey Sharp [Fri, 24 Feb 2006 09:50:48 +0000 (01:50 -0800)]
Move _xcb_set_fd_flags to xcb_conn.c and make it static. xcb_util.c now has only public functions.
Jamey Sharp [Fri, 24 Feb 2006 09:40:45 +0000 (01:40 -0800)]
Move _xcb_read_block to xcb_in.c and make it static. Change calls in xcb_conn.c to _xcb_in_read_block instead.
Jamey Sharp [Fri, 24 Feb 2006 09:17:03 +0000 (01:17 -0800)]
Remove XCB_CEIL and use a simpler definition for XCB_PAD.
Jamey Sharp [Fri, 24 Feb 2006 09:04:48 +0000 (01:04 -0800)]
Quit using "-include config.h": use #ifdef HAVE_CONFIG_H etc. like everyone else.
Jamey Sharp [Fri, 24 Feb 2006 08:59:08 +0000 (00:59 -0800)]
Coalesce readn into _xcb_in_read and simplify.
Jamey Sharp [Fri, 24 Feb 2006 08:48:18 +0000 (00:48 -0800)]
Move _xcb_readn to xcb_in.c and make it static. Minor change to _xcb_read_block to not depend on _xcb_readn.
Jamey Sharp [Fri, 24 Feb 2006 08:25:34 +0000 (00:25 -0800)]
Bugfix: protect the output queue from being written while another thread is flushing it.
Jamey Sharp [Fri, 24 Feb 2006 08:02:43 +0000 (00:02 -0800)]
Simplify: Always use writev. (In _xcb_out_flush, convert the output queue to a single iovec if needed.)
Jamey Sharp [Fri, 24 Feb 2006 06:41:59 +0000 (22:41 -0800)]
Factor padding out of _xcb_out_write_block and into its callers, XCBSendRequest and write_setup.
This requires dynamically allocating memory in XCBSendRequest, but this
malloc/free pair turns out to cause a 30% speed hit for the 'x11perf -noop'
test -- so for the moment I use alloca where available and fall back to malloc
on other platforms. Later I think I'll change the contract of XCBSendRequest
so the caller is responsible for memory allocation, because the caller ought
to always be able to stack-allocate here.
Jamey Sharp [Fri, 24 Feb 2006 01:50:53 +0000 (17:50 -0800)]
Add XCBGetQueuedRequestRead for Xlib that does no syscalls, just returns whatever XCB already knows about.
Jamey Sharp [Thu, 23 Feb 2006 23:29:40 +0000 (15:29 -0800)]
Minor performance fix: Change the contract on XCBGetRequestSent so that it does not waste time re-locking. This is for Xlib, and Xlib has already locked.
Jamey Sharp [Thu, 23 Feb 2006 22:57:46 +0000 (14:57 -0800)]
Minor performance fix: Only rearrange buffers for BIG-REQUESTs.
Jamey Sharp [Thu, 23 Feb 2006 22:32:11 +0000 (14:32 -0800)]
Move request_written update back where it was for now: doing it early can cause XCBWaitForReply to wrongly believe that the request has been flushed. Eventually, we should fix bug #6021.
Jamey Sharp [Thu, 23 Feb 2006 22:30:08 +0000 (14:30 -0800)]
Remove _xcb_assert_valid_sequence. One test is trivially true, and the other may be temporarily violated without anything bad happening.
Jamey Sharp [Thu, 23 Feb 2006 20:48:27 +0000 (12:48 -0800)]
Move _xcb_write and _xcb_writev to xcb_out.c and make them static, since only _xcb_out_write calls them.
Jamey Sharp [Thu, 23 Feb 2006 20:34:08 +0000 (12:34 -0800)]
Ignore test byproducts.
Jamey Sharp [Thu, 23 Feb 2006 20:15:09 +0000 (12:15 -0800)]
More return value changes, and make _xcb_in_read_packet static since it is not called from outside xcb_in.c.
Jamey Sharp [Thu, 23 Feb 2006 19:50:12 +0000 (11:50 -0800)]
More return value changes.
Jamey Sharp [Thu, 23 Feb 2006 18:17:40 +0000 (10:17 -0800)]
Make the return value of _xcb_conn_wait boolean, instead of syscall-like.
Jamey Sharp [Thu, 23 Feb 2006 07:38:16 +0000 (23:38 -0800)]
Bugfix: move request_written update *before* _xcb_conn_wait in _xcb_out_flush. Otherwise a reply may be read before we record that we have sent the request, and then XCB gets... confused.
Jamey Sharp [Thu, 23 Feb 2006 07:11:36 +0000 (23:11 -0800)]
Remove USE_THREAD_ASSERT compile option. _xcb_conn_wait can no longer be re-entered, since we have no callback hooks now.
Jamey Sharp [Thu, 23 Feb 2006 07:09:29 +0000 (23:09 -0800)]
Remove vestiges of adjacent request combining implementation. It is very dead.
Jamey Sharp [Mon, 20 Feb 2006 08:41:39 +0000 (00:41 -0800)]
Control debugging, optimization, and warning flags all independently.
Jamey Sharp [Sun, 19 Feb 2006 07:57:18 +0000 (23:57 -0800)]
Merge branch 'master' of git+ssh://git.freedesktop.org/git/xcb
Jamey Sharp [Sun, 19 Feb 2006 07:57:14 +0000 (23:57 -0800)]
Update .gitignores for .o files and autofoo stuff.
Eric Anholt [Sun, 19 Feb 2006 07:05:32 +0000 (23:05 -0800)]
Add missing entries to .gitignore files.
Eric Anholt [Sun, 19 Feb 2006 06:57:42 +0000 (22:57 -0800)]
Merge branch 'master' of git+ssh://anholt@git.freedesktop.org/git/xcb
Eric Anholt [Sun, 19 Feb 2006 06:56:03 +0000 (22:56 -0800)]
Move .cvsignore to .gitignore.
Jamey Sharp [Sun, 19 Feb 2006 04:03:35 +0000 (20:03 -0800)]
Quit defining _XOPEN_SOURCE. We never needed it...
Jamey Sharp [Sun, 19 Feb 2006 03:23:33 +0000 (19:23 -0800)]
Remove the old ChangeLog file. GIT makes that obsolete.
Jamey Sharp [Sun, 19 Feb 2006 02:12:40 +0000 (18:12 -0800)]
Workaround X server bug, fd.o #3210: if a GLXGetFBConfigs request is sent, fix the length field in the reply.
Josh Triplett [Sun, 19 Feb 2006 00:49:41 +0000 (16:49 -0800)]
Remove xcl and CVSROOT.