If a list is preceded by an align-pad, then
accessor for the end-iterator returned a wrong
value.
Reason: the length of the align-iterator was added
to a pointer of list-member type. Therefore, the length
was multiplied by the size of the list-member type,
due to C pointer arithmetic rules.
This has looked like the following, e.g., in
xcb_randr_get_crtc_transform_pending_params_end:
i.data = ((xcb_render_fixed_t *) prev.data) + ((-prev.index) & (4 - 1)) + (R->pending_nparams);
This bug was introduced with the following commit:
http://cgit.freedesktop.org/xcb/libxcb/commit/?id=4033d39d4da21842bb1396a419dfc299591c3b1f
The fix handles this by casting to char* before adding the align,
and then casting the result to the member type.
Signed-off-by: Christian Linhart <chris@demorecorder.com>
instead of using the lower bits of the pointer address.
This fixes a bug reported by Peter Hutterer in off-list communication
back in June 2015.
This requires the alignment-checker patches in xcb/proto.
Signed-off-by: Christian Linhart <chris@demorecorder.com>
Handle align-pads when generating an end-function
in the same way as handling them when generating
an accessor or iterator function.
Signed-off-by: Christian Linhart <chris@demorecorder.com>
and make it disabled by default with an EXPERIMENTAL warning
reason: this feature is unfinished and we want to have flexibility for
ABI/API changes, while still being able to make a release soon
Signed-off-by: Christian Linhart <chris@demorecorder.com>
Consider the following:
- Two threads are calling xcb_wait_for_special_event() and xcb_wait_for_reply()
concurrently.
- The thread doing xcb_wait_for_reply() wins the race and poll()s the socket for
readability.
- The other thread will be put to sleep on the special_event_cond of the special
event (this is done in _xcb_conn_wait() via the argument
xcb_wait_for_special_event() gives it).
- The first thread gets its reply, but does not yet receive any special event.
In this case, the first thread will return to its caller. On its way out, it
will call _xcb_in_wake_up_next_reader(), but that function cannot wake up
anything since so far it did not handle xcb_wait_for_special_event().
Thus, the first thread stays blocked on the condition variable and no thread
tries to read from the socket.
A test case demonstrating this problem is available at the bug report.
Fix this similar to how we handle this with xcb_wait_for_reply():
The function wait_for_reply() adds an entry into a linked list of threads that
wait for a reply. Via this list, _xcb_in_wake_up_next_reader() can wake up this
thread so that it can call _xcb_conn_wait() again and then poll()s the socket.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=84252
Signed-off-by: Uli Schlachter <psychon@znc.in>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
All functions calling _xcb_conn_wait() must make sure that waiting
readers are woken up when we read a reply or event that they are waiting
for. xcb_wait_for_special_event() did not do so. This adds the missing
call to_xcb_in_wake_up_next_reader().
Fixes deadlock when waiting for a special event and concurrently
processing the display connection queue in another thread.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=84252
Tested-by: Thomas Daede <bztdlinux@gmail.com>
Tested-by: Clément Guérin <geecko.dev@free.fr>
Reviewed-by: Uli Schlachter <psychon@znc.in>
Signed-off-by: Michel Dänzer <michel@daenzer.net>
Signed-off-by: Uli Schlachter <psychon@znc.in>
Before this patch, the following code caused an endless loop in send_fds(),
because the queue of FDs to send was eventually full, but _xcb_out_flush_to()
didn't make any progress, since there was no request to send:
while (1) { xcb_send_fd(conn, dup(some_fd)); }
Fix this by sending a sync when flushing didn't make any progress. That way we
actually have something to send and can attach the pending FDs.
Because send_fds() can now send requests, the code in
xcb_send_request_with_fds64() has to be changed. It has to call send_fds()
before it establishes a good sequence number for the request it wants to send.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Doing xcb_send_fd(), xcb_send_request() is racy. If two threads do this at the
same time, they could mix up their file descriptors. This commit makes it
possibly to fix this race by providing a single function which does everything
that is needed.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Two threads trying to send fds at the same time could interfere. To guarantee a
correct ordering, we have to use correct locking. The code in send_fds() missed
one case: If there was another thread already writing requests, we slept on the
"done with writing" condition variable (c->out.cond). This would allow other
threads to re-acquire the iolock before us and could cause fds to be sent out of
order.
To fix this, at the beginning of send_fds() we now make sure that no other
thread is already writing requests. This is what prepare_socket_request() does.
Additionally, it gets the socket back in case xcb_take_socket() was called,
which is a good thing, too, since fds are only sent with corresponding requests.
The API docs for xcb_send_fd() says "After this function returns, the file
descriptor given is owned by xcb and will be closed eventually".
Let the implementation live up to its documentation. We now also close fds if fd
passing is unavailable (!HAVE_SENDMSG) and when the connection is in an error
state.
(This also does sneak in some preparatory functions for follow-up commits and
thus does things in a more complicated way than really necessary.)
Signed-off-by: Uli Schlachter <psychon@znc.in>
(This does not change doxygen's output or warnings).
Signed-off-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Christian Linhart <chris@demorecorder.com>
While XCB uses 64-bit sequence number internally, it only exposes
"unsigned int" so that, on 32-bit architecture, Xlib based applications
may see their sequence number wrap which causes the connection to the X
server to be lost.
Expose 64-bit sequence number from XCB API so that Xlib and others can
use it even on 32-bit environment.
This implies the following API addition:
xcb_send_request64()
xcb_discard_reply64()
xcb_wait_for_reply64()
xcb_poll_for_reply64()
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=71338
Reviewed-by: Uli Schlachter <psychon@znc.in>
Signed-off-by: Christian Linhart <chris@demorecorder.com>
Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
The name can be understood from the type of S already.
For examples, look for 'S->' in xkb.c or xinput.c.
Signed-off-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Christian Linhart <chris@demorecorder.com>
Reviewed-by: Rémi Cardona <remi@gentoo.org>
It is implied already inside the function by the `void` argument.
Signed-off-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Christian Linhart <chris@demorecorder.com>
(Also remove unnecessary parens around the condition).
Signed-off-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Christian Linhart <chris@demorecorder.com>
This works for all python>=2.6, which is what configure requires.
Signed-off-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Christian Linhart <chris@demorecorder.com>
c_client.py:2: 'from xml.etree.cElementTree import *' used; unable to detect undefined names
c_client.py:3: 'basename' imported but unused
c_client.py:9: 'time' imported but unused
c_client.py:1437: local variable 'list_obj' is assigned to but never used
c_client.py:1745: local variable 'varfield' is assigned to but never used
c_client.py:2050: local variable 'length' is assigned to but never used
c_client.py:2416: local variable 'R_obj' is assigned to but never used
c_client.py:2441: local variable 'S_obj' is assigned to but never used
Signed-off-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Christian Linhart <chris@demorecorder.com>
The results are not used, and the function doesn't have side effects.
Signed-off-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Christian Linhart <chris@demorecorder.com>
Added accessor functions for requests the same way they were added for
structs,events and replies.
Lists for replies have accessor functions now.
Signed-off-by: Jaya Tiwari <tiwari.jaya18@gmail.com>
Reviewed-by: Christian Linhart <chris@demorecorder.com>
Comment from the Reviewer Christian Linhart:
I have tested your patch after fixing the issues with the patch-format.
It looks good:
* only adds new functions, and does not modify existing functions.
Therefore it is API and ABI compatible.
* adds accessors for varsized-stuff in requests.
This is needed for server-side XCB and may be useful for implementing X11-protocol proxies.
Parametrized structs contain paramref expressions which
refer to the value of a field defined in the context
where the struct is used.
Implementing the parametrized structs turned out
to be somewhat easier than previously thought
because the generator already had some support for type-parametrization
because this is needed when case or bitcase refers to fields outside
of the switch.
So I decided to go with the flow and to implement the solution
which best fits the current implementation.
I did the following:
* I provided a way to specify fieldref with an explicitely given type:
This resulted in <paramref type="CARD8>fieldname</paramref>
A paramref is just a fieldref with an explicit type.
The type is necessary because there is no local field of that
name where the type can be derived from.
* then I tested it and made several changes in the generator
such that it really works.
Basically the generated code is as follows:
* The parameter appears on the parameter list of the
sizeof-function of the parametrized struct.
When that function gets called, an appropriate argument is supplied.
* The parameter also appears as an additional member of the iterator-struct
for the iterator of lists of that parametrized struct.
This way, the next-function can get the value of that parameter from the iterator.
When the iterator is created, this iterator-member is set accordingly.
* When the paramref appears in the length-expression of a list, then
the parameter appears on the parameterlist of the "length" and "end" functions.
When these functions get called, an appropriate argument is supplied.
Some comments:
* I did not implement inline structs.
This would probably have been more complicated, and at least some additional effort.
But that can be implemented later if needed.
(Inline structs could probably use some code from switch-case/bitcase which is already kind of
an inlined struct but one has to be careful not to break the functionality
of switch-case/bitcase. Support for inline structs inside lists must probably
be implemented from scratch...)
* The paramref expression refers to a field of the same name in the struct/request/...
where it is used.
So it is not possible to pass the value of arbitrary fields or even expressions
to the parametrized struct.
This would have been possible with the previously discussed <typearg>.
That can be added later, if needed.
( Wont be too complicated )
* So this is pretty much like the proposal from Ran Benita.
changes for V2 of this patch, according to suggestions from Ran Benita:
* replace map with list comprehension
because map returns an iterator instead of a list from Python 3 on,
so it cannot be added to a list anymore.
* removed "self" parameter of function additional_params_to_str
and accessed the variable additional_params from the outer
function directly.
changes for V2 of this patch:
* adapt to revision 2 of patchset ListInputDevices
* style fixes for similar things that Ran Benita has found in my previous patches
Message-ID: <54574397.4060000@DemoRecorder.com>
Patch-Thread-Subject: [Xcb] parametrized structs implemented
Patch-Set: ParametrizedStruct
Patch-Number: libxcb 1/1
Patch-Version: V3
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
Support for listelement-ref needs the following three changes
(in the order as they appear in the patch):
* making the current list-element accessible with the variable
xcb_listelement which is a pointer to the list-element
* supporting lists of simple-type for sumof with a nested expression
* using the variable for resolving a listelement-ref expression
Changes for V2 of this patch:
- adapt to removal of patch "libxcb 2/6" from patchset "ListInputDevices".
Changes for V3 of this patch:
- adapt to V2 of patch "libxcb 5/6" from patchset "ListInputDevices"
Changes for V4 of this patch:
- adapt to revision 2 of the patchset "ListInputDevices"
Message-ID: <545743A0.50907@DemoRecorder.com>
Patch-Thread-Subject: [Xcb] support popcount of a list and associated xml changes
Patch-Set: PopcountList
Patch-Number: libxcb 4/4
Patch-Version: V4
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
The function _c_accessor_get_length had a special case handling
for intermixed var and fixed size fields.
However:
* The implementation of that special case was buggy:
It tried to call a python-dict as a function which causes
Python to abort the program with a stacktrace and error message.
So this code was never used.
* The case it tried to handle is handeled elsewhere in the
meantime: in _c_helper_absolute_name by previous patches
made by me.
Message-ID: <1409845851-38950-3-git-send-email-chris@demorecorder.com>
Patch-Thread-Subject: [Xcb] support popcount of a list and associated xml changes
Patch-Set: PopcountList
Patch-Number: libxcb 3/4
Patch-Version: V1
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
Accessors are generally needed for var-sized fields
and fields after var-sized fields.
Generic events can have ver-sized fields.
Therefore they need accessors.
Message-ID: <1409845851-38950-2-git-send-email-chris@demorecorder.com>
Patch-Thread-Subject: [Xcb] support popcount of a list and associated xml changes
Patch-Set: PopcountList
Patch-Number: libxcb 2/4
Patch-Version: V1
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
_c_type_setup is not called for eventcopies anymore:
Reasons:
* the type-setup of an eventcopy would overwrite members of the original
event object such as c_type, ...
* it is needed for the next patch, i.e., generating accessors:
type_setup would create sizeof-etc funtions which called
undefined accessor functions.
Sizeof-functions are generated for compatibility:
Reason:
* Type-setup of eventcopies has previously generated
sizeof-functions for eventcopies.
So, we still need to generate these functions.
These new sizeof-functions simply call the sizeof-function
of the defining event of the eventcopy.
Message-ID: <1409845851-38950-1-git-send-email-chris@demorecorder.com>
Patch-Thread-Subject: [Xcb] support popcount of a list and associated xml changes
Patch-Set: PopcountList
Patch-Number: libxcb 1/4
Patch-Version: V1
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
The loop-variable "sep" is never empty in function
"_c_helper_fieldaccess_expr", after a fix elsewhere.
Therefore I removed the handling of the case of "sep" being empty.
Thanks to Ran Benita for the hint that this can be removed.
Signed-off-by: Christian Linhart <chris@demorecorder.com>
Reviewed-by: Ran Benita <ran234@gmail.com>
Message-ID: <545627C2.3050608@DemoRecorder.com>
Patch-Thread-Subject: [Xcb] [PATCHSET] ListInputDevices revision 2
Patch-Set: ListInputDevices
Patch-Number: libxcb 9/9
Patch-Version: V1
The function _c_helper_absolute_name was named in
a misleading way.
It computes a C-expression for accessing a field of an xcb-type.
Therefore the name _c_helper_fieldaccess_expr is more appropriate.
Note: Patch 6 of this series has been removed during the review process.
Signed-off-by: Christian Linhart <chris@demorecorder.com>
Reviewed-by: Ran Benita <ran234@gmail.com>
Message-ID: <545627AE.2040200@DemoRecorder.com>
Patch-Thread-Subject: [Xcb] [PATCHSET] ListInputDevices revision 2
Patch-Set: ListInputDevices
Patch-Number: libxcb 7/9
Patch-Version: V1
Support sumof with a nested expression.
The nested expression is computed for every list-element
and the result of the computation is added to the sum.
This way, sumof can be applied to a list of structs,
and, e.g., compute the sum of a specific field of that struct.
example:
<struct name="SumofTest_Element">
<field type="CARD16" name="foo" />
<field type="CARD16" name="bar" />
</struct>
<struct name="SumofTest_FieldAccess">
<field type="CARD32" name="len" />
<list type="SumofTest_Element" name="mylist1">
<fieldref>len</fieldref>
</list>
<list type="CARD16" name="mylist2">
<sumof ref="mylist1">
<fieldref>bar</fieldref>
</sumof>
</list>
</struct>
generated tmpvar:
int xcb_pre_tmp_1; /* sumof length */
int xcb_pre_tmp_2; /* sumof loop counter */
int64_t xcb_pre_tmp_3; /* sumof sum */
const xcb_input_sumof_test_element_t* xcb_pre_tmp_4; /* sumof list ptr */
generated code:
/* mylist2 */
/* sumof start */
xcb_pre_tmp_1 = _aux->len;
xcb_pre_tmp_3 = 0;
xcb_pre_tmp_4 = xcb_input_sumof_test_field_access_mylist_1(_aux);
for ( xcb_pre_tmp_2 = 0; xcb_pre_tmp_2 < xcb_pre_tmp_1; xcb_pre_tmp_2++) {
xcb_pre_tmp_3 += xcb_pre_tmp_4->bar;
xcb_pre_tmp_4++;
}
/* sumof end. Result is in xcb_pre_tmp_3 */
xcb_block_len += xcb_pre_tmp_3 * sizeof(uint16_t);
changes for V2 of this patch:
* explicitely set the member access operator in the prefix-tuple
passed to function _c_helper_field_mapping.
This enables us to simplify function "_c_helper_absolute_name"
(which will be renamed "_c_helper_fieldaccess_expr" soon)
V3: Changed style and formatting according to suggestions from Ran Benita
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
Reviewed-by: Ran Benita <ran234@gmail.com>
Message-ID: <54562798.8040500@DemoRecorder.com>
Patch-Thread-Subject: [Xcb] [PATCHSET] ListInputDevices revision 2
Patch-Set: ListInputDevices
Patch-Number: libxcb 5/9
Patch-Version: V3
A sumof-expression now generates explicit code ( for-loop etc )
instead of calling xcb_sumof.
This way, it supports any type which can be added.
Previously, only uint_8 was supported.
Here's an example and the generated code:
xml:
<struct name="SumofTest">
<field type="CARD32" name="len" />
<list type="CARD16" name="mylist1">
<fieldref>len</fieldref>
</list>
<list type="CARD8" name="mylist2">
<sumof ref="mylist1"/>
</list>
</struct>
declaration of tempvars at the start of enclosing function:
int xcb_pre_tmp_1; /* sumof length */
int xcb_pre_tmp_2; /* sumof loop counter */
int64_t xcb_pre_tmp_3; /* sumof sum */
const uint16_t* xcb_pre_tmp_4; /* sumof list ptr */
code:
/* mylist2 */
/* sumof start */
xcb_pre_tmp_1 = _aux->len;
xcb_pre_tmp_3 = 0;
xcb_pre_tmp_4 = xcb_input_sumof_test_mylist_1(_aux);
for ( xcb_pre_tmp_2 = 0; xcb_pre_tmp_2 < xcb_pre_tmp_1; xcb_pre_tmp_2++) {
xcb_pre_tmp_3 += *xcb_pre_tmp_4;
xcb_pre_tmp_4++;
}
/* sumof end. Result is in xcb_pre_tmp_3 */
xcb_block_len += xcb_pre_tmp_3 * sizeof(uint8_t);
This patch is also a preparation for sumof which can access
fields of lists of struct, etc.
V2: Changed style and formatting according to suggestions from Ran Benita
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
Reviewed-by: Ran Benita <ran234@gmail.com>
Message-ID: <54562774.8030306@DemoRecorder.com>
Patch-Thread-Subject: [Xcb] [PATCHSET] ListInputDevices revision 2
Patch-Set: ListInputDevices
Patch-Number: libxcb 4/9
Patch-Version: V2
This patch provides a mechanism for generating
preparatory code for expressions.
This is e.g. necessary when an expression needs computations
which cannot be done in a C-Expression, like for-loops.
This will be used for sumof expressions but may be useful
elsewhere.
Note: Patch 2 of this series has been removed during the review process.
V2: adapt to changes in previous patches
V3: some style and formatting changes according to suggestions from Ran Benita.
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
Reviewed-by: Ran Benita <ran234@gmail.com>
Message-ID: <54562769.3090405@DemoRecorder.com>
Patch-Thread-Subject: [Xcb] [PATCHSET] ListInputDevices revision 2
Patch-Set: ListInputDevices
Patch-Number: libxcb 3/9
Patch-Version: V3
Fix _c_helper_absolute_name for fields which cannot be accessed
as a struct/union member but which can be accessed by an
accessor function.
The fix generates calls to the accessor function in these cases.
Example:
<struct name="AbsnameTest">
<field type="CARD32" name="len" />
<list type="CARD8" name="mylist1">
<fieldref>len</fieldref>
</list>
<list type="CARD8" name="mylist2">
<sumof ref="mylist1"/>
</list>
</struct>
The sumof-expression ( <sumof ref="mylist1"/> ) refers to mylist1
which is only acessible by an accessor function.
Previously, sumof was only used inside bitcases,
where such lists are accessible by members of the
deserialized parent struct.
(there is a difference between deserialization of switches
and structs.)
V2 of this patch:
* replaced "!= None" with "is not None" because that's more pythonic.
(according to suggestion from Ran Benita)
V3 of this patch: simplification:
* fixed the recursion in _c_type_setup
so that _c_helper_absolute_name does not need check
a gazillion things as a workaround anymore.
* simplified _c_helper_absolute_name
- remove unneeded check for empty string before
append of last_sep to prefix_str
- removed those if-conditions which are not
needed anymore after fixing the recursion
in _c_type_setup.
- extract functionality for checking whether a field
needs an accessor ( and which type of accessor )
in functions.
(also extracted from _c_accessors)
- rearrange the condition branches and actions for
more readability.
V3 generates exactly the same *.c and *.h files as V2.
V4 of this patch:
* improve formatting as per suggestions of Ran
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
Reviewed-by: Ran Benita <ran234@gmail.com>
Message-ID: <54562758.5090107@DemoRecorder.com>
Patch-Thread-Subject: [Xcb] [PATCHSET] ListInputDevices revision 2
Patch-Set: ListInputDevices
Patch-Number: libxcb 1/9
Patch-Version: V4
Nested structs which are generated for named case and bitcase
do not get a typename anymore, i.e., they are anonymous structs.
Reasons for this change:
* Prior typenames have caused nameclashes
* Prior typenames introduced names in the global namespace which
did not start with the xcb prefix.
This change is safe with respect to API compatibility because:
I have searched for instances of named bitcases and there's only one place
where they are used, and that's in xkb.xml: reply GetKbdByName.
( no need to search for <case> because it was introduced after the last release )
The reply GetKbdByName is broken in its current form in the xkb.xml anyways,
so it is most probably not used anywhere.
So, my conclusion is that we can safely omit named types for nested structs.
No need for an attribute.
Message-ID: <1409731849-51897-1-git-send-email-chris@demorecorder.com>
Patch-Thread-Subject: Re: [Xcb] names of nested structs of named bitcase/case are prone to nameclashes. Solution?
Patch-Set: NestedStructTypenames
Patch-Number: libxcb 1/1
Patch-Version: V1
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
Reviewed-By: Ran Benita <ran234@gmail.com>
Fix the alignment computation inside switches which start at
an unaligned pos.
This affects both explicit and implicit align pads.
The alignment offset is derived from the lowest 3 bits of
the pointer to the protocol-data at the start of the switch.
This is sufficient for correcting all alignments up to 8-byte alignment.
As far as I know there is no bigger alignment than 8-byte for the
X-protocol.
Example:
struct InputState, where the switch starts after two 1-byte fields,
which is a 2 byte offset for 4-byte and 8-byte alignment.
The previous problem can be demonstrated when adding a
<pad align="4"/> at the end of case "key".
(Or when finding a testcase which reports the case "valuator" not
at the last position of the QueryDeviceState-reply.
I didn't find such a testcase, so I have used the pad align
as described above.)
V2: patch modified in order to fix bugs which I found when working on the
next issue:
* xcb_padding_offset has to be set 0 when xcb_block_len is set 0
* xcb_padding_offset cannot be "const" therefore
* for unpack and unserialize, the padding_offset must computed
from _buffer instead of from the aux_var.
V3: patch revised according to suggestion by Ran Benita:
* only create and use xcb_padding_offset for switch
Message-ID: <1410298000-24734-1-git-send-email-chris@demorecorder.com>
Patch-Thread-Subject: [Xcb] xinput:QueryDeviceState: full-support: generator and xml-changes
Patch-Set: QueryDeviceState
Patch-Number: libxcb 4/4
Patch-Version: V3
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
Reviewed-By: Ran Benita <ran234@gmail.com>
This essentially requires to have a correct sizeof-function
for the struct.
This in turn requires a sizeof-function for the switch, too.
Making a sizeof-function for the switch is triggered by
replacing "elif" by "if" in the first change of this patch.
This way, c_need_sizeof is also set to True for switches if appropriate.
The _c_serialize_helper_switch_field function has to support
the context "sizeof":
This is done in the second change of this patch
The third change of this patch fixes an alignment error:
It does not make sense to base the padding on the struct-type
which is generated for switch because this struct does not
represent the protocol. Rather it is the output of deserialization.
( The implicit padding for var-sized fields has other issues, IMHO,
but I am not touching these now...)
The effect on the generated code for the current xml-files
is as follows:
* several additional sizeof-functions are generated
* the fix of the alignment error only changes one place
in the XKB-extension for the GetKbdByName-reply.
This is no problem because that reply in its current form
is broken/unfinished anyways.
Note:
This patch also fixes a problem in the generator when
a fixed-size list is the last field of a case or bitcase.
Message-ID: <1408653356-21191-2-git-send-email-chris@demorecorder.com>
Patch-Thread-Subject: [Xcb] xinput:QueryDeviceState: full-support: generator and xml-changes
Patch-Set: QueryDeviceState
Patch-Number: libxcb 2/3
Patch-Version: V1
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
Reviewed-By: Ran Benita <ran234@gmail.com>
V2: patch revised according to suggestions from Ran Benita:
* removed blanks before an after parentheses of function-calls or tuples
* replaced if by elif in "if field.type.is_list". ( this fixes old code )
Message-ID: <540B4D17.1080908@DemoRecorder.com>
Patch-Thread-Subject: [Xcb] xinput:QueryDeviceState: full-support: generator and xml-changes
Patch-Set: QueryDeviceState
Patch-Number: libxcb 1/3
Patch-Version: V2
Signed-off-by: Christian Linhart <chris@DemoRecorder.com>
Reviewed-By: Ran Benita <ran234@gmail.com>
The documentation doesn't mention it and it's unlikely that a lot of code out
there handles this case correctly. So, instead of returning NULL, let
xcb_get_setup() return a pointer to a static, invalid, all-zero setup
information structure.
Signed-off-by: Uli Schlachter <psychon@znc.in>
There is no technical reason why xcb_get_setup() and xcb_get_file_descriptor()
shouldn't work on non-static error connections. They cannot be used for many
useful things, but at least they work.
This works around bugs in lots of programs out there which assume that
xcb_get_setup() does not return NULL and which just happily dereference the
results. Since xcb_connect() never returns NULL, it's a bit weird that
xcb_get_setup() can do so. xcb_get_file_descriptor() is just modified since this
can be done here equally easily and because the fd isn't closed until the final
xcb_disconnect() on the error connection.
Non-static error connections are connections which entered an error state after
xcb_connect() succeeded. If something goes wrong in establishing a connection,
xcb_connect() will return a static error connection which doesn't have the
fields used here.
Signed-off-by: Uli Schlachter <psychon@znc.in>
The implementation is rather simple:
When a <case> is used instead of a <bitcase>
then operator "==" is used instead of "&" in the if-condition.
So it creates a series of "if" statements
(instead of a switch-case statement in C )
In practice this does not matter because a good
optimizing compiler will create the same code
as for a switch-case.
With this simple implementation we get additional
flexibility in the following forms:
* a case value may appear in multiple case branches.
for example:
case C1 will be selected by values 1, 4, or 5
case C2 will be selected by values 3, 4, or 7
* mixing of bitcase and case is possible
(this will usually make no sense but there may
be protocol specs where this is needed)
details of the impl:
* replaced "is_bitcase" with "is_case_or_bitcase" in all places
so that cases are treated like bitcases.
* In function "_c_serialize_helper_switch": write operator "=="
instead of operator "&" if it is a case.
These structs are typedef'ed in xproto.h, so in xcb.h these types
(without 'struct') are actually undefined.
GCC reports this as error when building precompiled header.
Signed-off-by: Alexander Mezin <mezin.alexander@gmail.com>
Reviewed-by: Peter Harris <pharris@opentext.com>
They are bloated, don't add anything over the signature, in some cases
duplicate the doxygen comments, and are not integrated with the <doc>
tags in any way. Remove them and cut the generated LOC by half.
Signed-off-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Daniel Martin <consume.noise@gmail.com>
Some parts of the man pages (SEE ALSO and ERRORS) are generated by
iterating a Python dict. But the iteration order in a dict is random,
so each build the output is ordered differently. Avoid that by iterating
in sorted order.
Signed-off-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Daniel Martin <consume.noise@gmail.com>
The script adds many fields to the objects coming from xcbgen. To
distinguish them, a c_ prefix is used, but for some it was missing.
Signed-off-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Daniel Martin <consume.noise@gmail.com>
AIX <sys/poll.h> does redefine 'events' to 'reqevents' eventually.
To not have this cause compilation errors, need to include the local
header files after any system header file.
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Documentation was previously unclear that these always return a non-NULL
pointer, and that callers need to check it for error values, instead of
checking for a NULL return value.
Triggered by having to dig through code to answer a user's question on
the #xcb irc channel, since neither of us found it covered in the docs.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Uli Schlachter <psychon@znc.in>
Without this patch we end up with invalid C code if we've a
<pad align="n" /> between two variadic lists. Check for such a condition
and take the alignment pad into account.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=79808
Signed-off-by: Daniel Martin <consume.noise@gmail.com>
Signed-off-by: Peter Harris <pharris@opentext.com>
This changes away from hard-coding the /tmp/launch-* path to now
supporting a generic <path to unix socket>[.<screen>] format for
$DISPLAY.
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
send_request may only write to out.queue if no other thread is busy
writing to the network (as that thread may be writing from out.queue).
send_request may only allocate request sequence numbers if XCB owns
the socket.
Therefore, send_request must make sure that both conditions are true
when it holds iolock, which can only be done by looping until both
conditions are true without having dropped the lock waiting for the
second condition.
We choose to get the socket back from Xlib first as get_socket_back
has a complicated test and checking for other threads writing is a
simple in-lined check.
This also changes the sequence number checks (64k requests with no
reply, 4M request wrapping) to ensure that both conditions are true
before queueing the request.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Uli Schlachter <psychon@znc.in>
This avoids having the nested header files also included at the top
level, which is more efficient.
Signed-off-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Uli Schlachter <psychon@znc.in>
Code can be simplified if the deallocation functions can always be called in
cleanup code. So if you have some code that does several things that can go
wrong, one of which is xcb_connect(), after this change, the xcb_connection_t*
variable can be initialized to NULL and xcb_disconnect() can always be called on
the connection object.
References: http://lists.freedesktop.org/archives/xcb/2013-September/008659.html
Signed-off-by: Uli Schlachter <psychon@znc.in>
Reviewed-by: Julien Cristau <jcristau@debian.org>
There are two kind of error connections in XCB. First, if something goes wrong
while the connection is being set up, _xcb_conn_ret_error() is used to return a
static connection in an error state. If something goes wrong later,
_xcb_conn_shutdown() is used to set c->has_error.
This is important, because the static object that _xcb_conn_ret_error() returns
must not be freed, while the dynamically allocated objects that go through
_xcb_conn_shutdown() must obviously be properly deallocated.
This used to work correctly, but in 769acff0da, xcb_disconnect() was made to
ignore all connections in an error state completely. Fix this by only ignoring
the few static error connections that we have.
This was tested with the following hack:
xcb_connection_t *c = xcb_connect(NULL, NULL);
close(xcb_get_file_descriptor(c));
xcb_discard_reply(c, xcb_get_input_focus(c).sequence);
xcb_flush(c);
xcb_disconnect(c);
Valgrind confirms that xcb has a memory leak before this patch that this patch
indeed fixes.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Reviewed-by: Julien Cristau <jcristau@debian.org>
If xcb_connect() fails, it doesn't return NULL. Instead, it always
returns an xcb_connection_t*, and the user should check for errors with
the xcb_connection_has_error() function. What this function does is
check if conn->has_error contains a non-zero error code, and returns it.
If an error did occur, xcb doesn't actually return a full
xcb_connection_t though, it just returns (xcb_connection_t *)
error_code. Since the 'has_error' field is the first, it is still
possible to check conn->has_error.
That last trick was not immediately obvious to me, so add some guiding
comments. This also ensures no one obliviously rearranges the struct.
Signed-off-by: Ran Benita <ran234@gmail.com>
Signed-off-by: Uli Schlachter <psychon@znc.in>
The section number is no longer hard-coded
The left footer is now "X Version 11".
The center footer is the package name with the version, "libxcb 1.9"
The three values above are provided through xorg-macros. They are passed-in
to the python c_client code.
Example of footer (last line, above dotted line)
[...]
AUTHOR
Generated from xproto.xml. Contact xcb@lists.freedesktop.org for cor‐
rections and improvements.
X Version 11 libxcb 1.9 xcb_send_event(3)
------------------------------------------------------------------------------
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
The automake MAN primary requires a hard coded extension to build
man pages. Let's avoid that as the extension number may vary by platform.
Take advantage of the fact that the man directory only contains man pages.
Wildcards are not supported by Automake but it happens to work
sufficiently well here.
Normally xorg build man pages by converting a source .man file to a
target file with the extension number. That would be too many files
in this case.
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
The section number is no longer hard-coded, supplied by xorg-macros.
The left footer is now "X Version 11".
The center footer is the package name with the version, "libxcb 1.9"
The man directory is a sibbling to the doc directory. One can build
or clean the man pages without disturbing the library code.
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
The BASE_CFLAGS variable contains only warnings, just like the XCB
version of CWARNFLAGS. This will result in no changes in the binaries
produced. Xorg was missing -fd for SUNCC so it has been added to util-macros
v 1.18.
Do not get confused with the xorg deprecated CWARNFLAGS variable which
contains an option that is not a warning, -fno-strict-aliasing. This
option, should it be needed, can be added using the XORG_TESTSET_CFLAG
macro.
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
With the advent of the Present extension, some events (such as
PresentCompleteNotify) now use native 64-bit types on the wire.
For XGE events, we insert an extra "uint32_t full_sequence" field
immediately after the first 32 bytes of data. Normally, this causes
the subsequent fields to be shifted over by 4 bytes, and the structure
to grow in size by 4 bytes. Everything works fine.
However, if event contains 64-bit extended fields, this may result in
the compiler adding an extra 4 bytes of padding so that those fields
remain aligned on 64-bit boundaries. This causes the structure to grow
by 8 bytes, not 4. Unfortunately, XCB doesn't realize this, and
always believes that the length only increased by 4. read_packet()
then fails to malloc enough memory to hold the event, and the event
processing code uses the wrong offsets.
To fix this, mark any event structures containing 64-bit extended
fields with __attribute__((__packed__)).
v2: Use any(...) instead of True in (...), as suggested by
Daniel Martin.
v3 (Alan Coopersmith): Fix build with Solaris Studio 12.3 by moving the
attribute to after the structure definition.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Keith Packard <keithp@keithp.com> [v1]
Reviewed-by: Josh Triplett <josh@joshtriplett.org> [v1]
Reviewed-by: Daniel Martin <consume.noise@gmail.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
As RFC 2292 points out, some platforms (e.g. Darwin 9.8.0) provide
CMSG_FIRSTHDR(msg) which just returns msg.msg_control without first
checking if msg.msg_controllen is non-zero. We need a workaround for
such platforms not to let _xcb_in_read() segfault.
https://bugs.freedesktop.org/show_bug.cgi?id=72253
Signed-off-by: Julien Cristau <jcristau@debian.org>
Sync: Due to commit e6a246e50e62cbcba3 "sync: Change value list param of
CreateAlarm and ChangeAlarm into switch", various symbols disappeared,
for example xcb_sync_{change,create}_alarm_sizeof.
xinput: This extension was updated from version 1.4 to 2.3. This means
that lots of new things are generated. However, this change is
backwards-compatible and thus age gets set to 1.
xkb: In commit 37d0f55392d6 "xkb: Work around alignment problems in
GetNames and GetMap replies", some padding fields were introduced into
structures for which an _unpack() function is generated. This changed
the size of the struct and caused offsets into this struct to change.
https://bugs.freedesktop.org/show_bug.cgi?id=71507
Signed-off-by: Uli Schlachter <psychon@znc.in>
Signed-off-by: Julien Cristau <jcristau@debian.org>
This reverts commit f4d5b84800.
The version of this struct that the code generator produces breaks the API,
because it gives the fields different (albeit better) names. Thus, we need to
restore the old version of this struct.
Additionally to the revert, this struct is documented as being deprecated. The
replacement was added to xcb-proto.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=71502
Signed-off-by: Uli Schlachter <psychon@znc.in>
Signed-off-by: Julien Cristau <jcristau@debian.org>
Currently, it is not possible to correctly iterate over the replies of
some requests. For example, the list of XIDeviceInfo returned by
the XIQueryDevice request from xinput2 is read as garbage starting from
the second entry.
The culprits are the _sizeof() used by the iterators. In the above case:
int
xcb_input_xi_device_info_sizeof (const void *_buffer /**< */)
{
char *xcb_tmp = (char *)_buffer;
[...]
unsigned int xcb_block_len = 0;
[...]
xcb_block_len += sizeof(xcb_input_xi_device_info_t);
xcb_tmp += xcb_block_len;
/* name */
xcb_block_len += (((_aux->name_len + 3) / 4) * 4) * sizeof(char);
xcb_tmp += xcb_block_len;
[...]
}
The problem here is that `xcb_block_len` is not zero'd right above the
`/* name */` comment, causing `xcb_tmp` to be incremented by
`sizeof(xcb_input_xi_device_info_t)` twice. The returned size is too
large.
https://bugs.freedesktop.org/show_bug.cgi?id=68387
Tested-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Ran Benita <ran234@gmail.com>
Reviewed-by: Daniel Martin <consume.noise@gmail.com>
Signed-off-by: Ran Benita <ran234@gmail.com>
Signed-off-by: Julien Cristau <jcristau@debian.org>
This reverts commit 9ae84ad187.
After this patch was merged, there were complaints about it not being a good
idea. Revert this for now until we can agree on this.
References: http://lists.freedesktop.org/archives/xcb/2013-June/008340.html
Signed-off-by: Uli Schlachter <psychon@znc.in>
Conflicts:
src/xcbint.h
A char array on the stack is not guaranteed to have more than byte alignment.
This means that casting it to a 'struct cmsghdr' and accessing its members
may result in unaligned access. This will generate SIGBUS on struct
alignment architectures like OpenBSD/sparc64. The canonical solution is to
use a union to force proper alignment.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
Signed-off-by: Uli Schlachter <psychon@znc.in>
Use these instead of computing the values directly so that it might
work on BSD or other non-Linux systems
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
This allows apps to peel off certain XGE events into separate queues
for custom handling. Designed to support the Present extension
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-By: Uli Schlachter <psychon@znc.in>
Requests signal which replies will have fds, and the replies report
how many fds they expect in byte 1.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-By: Uli Schlachter <psychon@znc.in>
This uses sendmsg to transmit file descriptors from the application to
the X server
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-By: Uli Schlachter <psychon@znc.in>
When new XML files get installed, make sure the C files are regenerated
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-By: Uli Schlachter <psychon@znc.in>
It seems like POLLIN is specified as equivalent to POLLRDNORM | POLLRDBAND. Some
systems (e.g. QNX and HP-UX) take this literaly and have POLLIN defined as the
above bit combination. Other systems (e.g. Linux) have POLLIN as just a single
bit.
This means that if no out-of-band data is available (which should never be the
case), the result of poll() will not fulfil (fd.revents & POLLIN) == POLLIN on
QNX, because the POLLRDBAND bit is not set.
In other words, even though poll() signaled that the fd is readable, xcb would
not read from the file descriptor.
Fix this by checking if any bits from POLLIN are set in the result of poll(),
instead of all of them.
(This change was independently done by seanb@qnx.com as well)
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=38001
Acked-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Uli Schlachter <psychon@znc.in>
In commit 8eba8690ad, the API documentation for xcb_poll_for_event() was
fixed to remove an argument that was previously removed in commit 34168ab549.
However, that commit only removed the first line of the documentation, leaving
behind a spurious half-sentence. That commit happened seven years ago and now
finally someone noticed...
Thanks to Benjamin Herr for reporting this on IRC.
v2: Thanks again to Benjamin Herr for noticing that my commit message blamed the
wrong commit.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Do not create pointers in unions for fields of variadic length.
Signed-off-by: Daniel Martin <consume.noise@gmail.com>
Reviewed-by: Ran Benita <ran234@gmail.com>
to get rid of:
warning: 'xcb_align_to' may be used uninitialized in this function
Signed-off-by: Daniel Martin <consume.noise@gmail.com>
Reviewed-by: Peter Harris <pharris@opentext.com>
Fixes Solaris Studio compiler warning:
"xcb_list.c", line 50: warning: old style function definition
and gcc warning:
xcb_list.c: In function '_xcb_map_new':
xcb_list.c:50:11: warning: old-style function definition [-Wold-style-definition]
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
The generic event structure xcb_ge_event_t has the full_sequence field
at the 32byte boundary. That's why we've to inject this field into GE
events while generating the structure for them. Otherwise we would read
garbage (the internal full_sequence) when accessing normal event fields
there.
Signed-off-by: Daniel Martin <consume.noise@gmail.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Peter Harris <pharris@opentext.com>
To prevent different threads from stealing the socket from each other the
caller of "xcb_take_socket" must hold a lock that is also acquired in
"return_socket". Unfortunately xcb tries to prevent calling return_socket
from multiple threads and this can lead to a deadlock situation.
A simple example:
- X11 has taken the socket
- Thread A has locked the display.
- Thread B does xcb_no_operation() and thus ends up in libX11's return_socket(),
waiting for the display lock.
- Thread A calls e.g. xcb_no_operation(), too, ends up in return_socket() and
because socket_moving == 1, ends up waiting for thread B
=> Deadlock
This patch allows calling return_socket from different threads at the same time
an so resolves the deadlock situation.
Partially fixes: https://bugs.freedesktop.org/show_bug.cgi?id=20708
v2: fixes additional pthread_cond_wait dependencies,
rework comments and patch description
v3: separate pthread_cond_wait dependencies and unrelated whitespace
change into their own patch, use unsigned for socket_seq
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Uli Schlachter <psychon@znc.in>
Ensure that when calculating the size of the incoming response from the
Xserver, we don't overflow the integer used in the calculations when we
multiply the int32_t length by 4 and add it to the default response size.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Christian Linhart
Uli Schlachter
Michel Dänzer
Ran Benita
Jaya Tiwari
Alexander Mezin
Michael Haubenwallner
Alan Coopersmith
Daniel Martin
Jeremy Huddleston Sequoia
Keith Packard
Gaetan Nadon
Peter Harris
Kenneth Graunke
PHO
Mark Kettenis
Daphne Pfister
Christian König