Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=56291

When playing a loop, and if no end of loop point has been defined
(meaning = play whole segment), don't rely on segment's length for
wave tracks as their length is always 1.  Compute segment length in
this case by grabbing each track's playback length in segment state's
internal graph.  This feature is only activated for messages generated
by wave tracks, but could be extended if needed to other types of
messages.
Signed-off-by: Eric Pouech <epouech@codeweavers.com>

Store track_flags inside segment state so that they can be reused in
subsequent calls.  The start/seek/loop flags are now properly set when
calling IDirectMusicTrack_Play().
Signed-off-by: Eric Pouech <epouech@codeweavers.com>

Signed-off-by: Eric Pouech <epouech@codeweavers.com>

Signed-off-by: Eric Pouech <epouech@codeweavers.com>

Signed-off-by: Eric Pouech <epouech@codeweavers.com>

Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>

Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>

Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>

It's possible that a state object pointer not in the topology node collection gets passed to
session_get_node_object(). Instead of returning the last node when the object is not found, we
should return a NULL so that the state of the last node is not changed by mistake.
Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>

Doing so triggers a permissions prompt for input monitoring.

Patch originally by Rémi Bernon.

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=56291

Matches what's done in the GLSL shader backend.

The original intent here seems to have been to avoid disabling
GL_FRAGMENT_PROGRAM_ARB only to reënable it again. There is not actually any
harm in this, however.

It is possible (if somewhat artificial) to use lighting without using ambient
lighting. In this case ffp_light_ambient may not be an active uniform.

We could be more sophisticated here, and add lighting to the mask if we use any
lighting uniform, but this is simpler and probably good enough.

Show that it affects both the vertex and fragment FFP pipelines.

autoconf 2.72 disabled the time size increase when LARGEFILE macro is being
used: https://git.savannah.gnu.org/cgit/autoconf.git/commit/?id=cf09f48841b66fe76f606dd6018bb3a93242a7c9

That change can cause 32-bit Wine to return EOVERFLOW when running
stat() on a file in certain edge cases (which causes some files to
be skipped leading to applications failing to run properly).

This change was tested with both autoconf 2.71 and 2.72 to make sure
nothing broke there.

Signed-off-by: Eric Pouech <epouech@codeweavers.com>