libXv: Changes to 'master'

Alan Coopersmith alanc at kemper.freedesktop.org
Fri Jul 5 21:17:49 PDT 2013


 configure.ac   |    2 
 src/Xv.c       |  231 ++++++++++++++++++++++++++++++++-------------------------
 src/Xvlibint.h |   28 +-----
 3 files changed, 139 insertions(+), 122 deletions(-)

New commits:
commit 06d275959b5dd78c319d5200ca6d986934b18920
Author: Alan Coopersmith <alan.coopersmith at oracle.com>
Date:   Sat Jun 22 23:43:58 2013 -0700

    Refactor error handling
    
    Reduce code duplication, make error checking & cleanup more consistent
    
    Signed-off-by: Alan Coopersmith <alan.coopersmith at oracle.com>

commit 29b23d387e32d09e1b34682f01cee899a08a1176
Author: Alan Coopersmith <alan.coopersmith at oracle.com>
Date:   Sat Jun 22 21:47:34 2013 -0700

    Add missing calls to _XEatDataWords when we are skipping _XRead
    
    If we failed to allocate the buffer to _XRead into, discard the
    rest of the reply, instead of leaving it to confuse the reading
    of the next reply.
    
    Signed-off-by: Alan Coopersmith <alan.coopersmith at oracle.com>

commit 5a09a55b5b13280990465d83205e796bb8cd68ac
Author: Alan Coopersmith <alan.coopersmith at oracle.com>
Date:   Sat Jun 22 21:04:06 2013 -0700

    Replace custom copy of GetReq macro with call to Xlib 1.5's _XGetRequest
    
    xvproto.h names don't match those required by the Xlibint.h GetReq* macros,
    but at least we can rely on the _XGetRequest function for the bulk of the
    work now, instead of duplicating it.
    
    Also clears clang warnings repeated for every request function:
    
    Xv.c:137:5: warning: cast from 'char *' to 'xvQueryExtensionReq *' increases re
    quired alignment from 1 to 2 [-Wcast-align]
        XvGetReq(QueryExtension, req);
        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    ./Xvlibint.h:52:8: note: expanded from macro 'XvGetReq'
            req = (xv##name##Req *)(dpy->last_req = dpy->bufptr);\
                  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    Xv.c:137:5: warning: implicit conversion loses integer precision: 'int' to 'CAR
    D8' (aka 'unsigned char') [-Wconversion]
        XvGetReq(QueryExtension, req);
        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    ./Xvlibint.h:53:30: note: expanded from macro 'XvGetReq'
            req->reqType = info->codes->major_opcode;\
                         ~ ~~~~~~~~~~~~~^~~~~~~~~~~~
    
    Signed-off-by: Alan Coopersmith <alan.coopersmith at oracle.com>

commit ef2722e10b275d13d754d2b92e67f72b0716d070
Author: Alan Coopersmith <alan.coopersmith at oracle.com>
Date:   Sat Jun 22 20:33:13 2013 -0700

    Require ANSI C89 pre-processor, drop pre-C89 token pasting support
    
    Signed-off-by: Alan Coopersmith <alan.coopersmith at oracle.com>
    Reviewed-by: Mark Kettenis <kettenis at openbsd.org>

commit e73a2199f039e9ae772efb6be8d1db68eb346fa4
Author: Alan Coopersmith <alan.coopersmith at oracle.com>
Date:   Sat Jun 22 13:35:12 2013 -0700

    Use pad_to_int32 macro instead of repeated (x + 3) & ~3 pattern
    
    Makes code clearer, and using ~3U instead of ~3 clears some signed int
    warnings.
    
    Signed-off-by: Alan Coopersmith <alan.coopersmith at oracle.com>

commit 4ced44196d6636006c9724a373c0d2ca0c12dc1e
Author: Alan Coopersmith <alan.coopersmith at oracle.com>
Date:   Sat Jun 22 11:20:13 2013 -0700

    Fix sign conversion & comparison warnings from clang
    
    Clears up:
    Xv.c:196:21: warning: implicit conversion changes signedness: 'CARD32' (aka 'unsigned int') to 'int' [-Wsign-conversion]
      size = rep.length << 2;
           ~ ~~~~~~~~~~~^~~~
    Xv.c:212:41: warning: implicit conversion changes signedness: 'int' to 'size_t' (aka 'unsigned long') [-Wsign-conversion]
          if ((pas=(XvAdaptorInfo *)Xmalloc(size))==NULL) {
                                    ~~~~~~~~^~~~~
    Xv.c:236:43: warning: implicit conversion changes signedness: 'int' to 'unsigned long' [-Wsign-conversion]
          pa->num_adaptors = rep.num_adaptors - ii;
                           ~ ~~~~~~~~~~~~~~~~~^~~~
    Xv.c:243:40: warning: implicit conversion changes signedness: 'int' to 'size_t' (aka 'unsigned long') [-Wsign-conversion]
          if ( (name = (char *)Xmalloc(size+1)) == NULL)
                               ~~~~~~~~~~~~^~~
    Xv.c:251:37: warning: implicit conversion changes signedness: 'int' to 'size_t' (aka 'unsigned long') [-Wsign-conversion]
          (void)strncpy(name, u.string, size);
                ~~~~~~~                 ^~~~
    Xv.c:260:36: warning: implicit conversion changes signedness: 'int' to 'size_t' (aka 'unsigned long') [-Wsign-conversion]
          if ((pfs=(XvFormat *)Xmalloc(size))==NULL) {
                               ~~~~~~~~^~~~~
    Xv.c:269:20: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare]
          for (jj=0; jj<pa->num_formats; jj++) {
                     ~~^~~~~~~~~~~~~~~~
    Xv.c:259:29: warning: implicit conversion loses integer precision: 'unsigned long' to 'int' [-Wshorten-64-to-32]
          size = pa->num_formats*sizeof(XvFormat);
               ~ ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~
    Xv.c:305:16: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare]
      for (ii=0; ii<pAdaptors->num_adaptors; ii++, pa++)
                 ~~^~~~~~~~~~~~~~~~~~~~~~~~
    Xv.c:356:21: warning: implicit conversion changes signedness: 'CARD32' (aka 'unsigned int') to 'int' [-Wsign-conversion]
      size = rep.length << 2;
           ~ ~~~~~~~~~~~^~~~
    Xv.c:369:41: warning: implicit conversion changes signedness: 'int' to 'size_t' (aka 'unsigned long') [-Wsign-conversion]
      if ( (pes = (XvEncodingInfo *)Xmalloc(size)) == NULL) {
                                    ~~~~~~~~^~~~~
    Xv.c:392:45: warning: implicit conversion changes signedness: 'int' to 'unsigned long' [-Wsign-conversion]
          pe->num_encodings = rep.num_encodings - jj;
                            ~ ~~~~~~~~~~~~~~~~~~^~~~
    Xv.c:397:40: warning: implicit conversion changes signedness: 'int' to 'size_t' (aka 'unsigned long') [-Wsign-conversion]
          if ( (name = (char *)Xmalloc(size+1)) == NULL) {
                               ~~~~~~~~~~~~^~~
    Xv.c:404:31: warning: implicit conversion changes signedness: 'int' to 'size_t' (aka 'unsigned long') [-Wsign-conversion]
          strncpy(name, u.string, size);
          ~~~~~~~                 ^~~~
    Xv.c:433:16: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare]
      for (ii=0; ii<pEncodings->num_encodings; ii++, pe++) {
                 ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
    Xv.c:886:27: warning: comparison of integers of different signs: 'int' and 'CAR
    D32' (aka 'unsigned int') [-Wsign-compare]
                for (i = 0; i < rep.num_attributes; i++) {
                            ~ ^ ~~~~~~~~~~~~~~~~~~
    Xv.c:946:27: warning: comparison of integers of different signs: 'int' and 'CAR
    D32' (aka 'unsigned int') [-Wsign-compare]
                for (i = 0; i < rep.num_formats; i++) {
                            ~ ^ ~~~~~~~~~~~~~~~
    Xv.c:1100:5: warning: comparison of integers of different signs: 'int' and 'unsigned int' [-Wsign-compare]
        SetReqLen(req, len, len);
        ^~~~~~~~~~~~~~~~~~~~~~~~
    X11/Xlibint.h:530:27: note: expanded from macro 'SetReqLen'
        if ((req->length + n) > (unsigned)65535) { \
                              ^ ~~~~~~~~~~~~~~~
    Xv.c:1100:20: warning: implicit conversion changes signedness: 'int' to 'unsigned int' [-Wsign-conversion]
        SetReqLen(req, len, len);
        ~~~~~~~~~~~~~~~^~~~~~~~~
    X11/Xlibint.h:532:21: note: expanded from macro 'SetReqLen'
                MakeBigReq(req,n) \
                               ^
    X11/Xlibint.h:505:35: note: expanded from macro 'MakeBigReq'
        ((CARD32 *)req)[1] = _BRlen + n + 2; \
                                    ~ ^
    
    Signed-off-by: Alan Coopersmith <alan.coopersmith at oracle.com>

commit 89dc3ff457644271b1c633efa9682ae45e55d9d8
Author: Alan Coopersmith <alan.coopersmith at oracle.com>
Date:   Sat Jun 22 11:07:13 2013 -0700

    Remove unnecessary casts from Xmalloc calls
    
    Signed-off-by: Alan Coopersmith <alan.coopersmith at oracle.com>
    Reviewed-by: Mark Kettenis <kettenis at openbsd.org>



More information about the xorg-commit mailing list