Commit Graph

172 Commits

Author SHA1 Message Date
Armin Novak
e009a62f06 Fixed sign-compare warnings 2019-04-05 09:14:35 +02:00
Armin Novak
02f68c35b9 Fixed -Wextra sign warnings 2018-10-24 13:25:31 +02:00
Armin Novak
1f7d33a2f2 Fixed read/write of surface bits command.
The optional field exBitmapDataHeader of TS_ BITMAP_DATA_EX was ignored.
Read and expose the data (currently unused)
2018-03-01 11:38:59 +01:00
Armin Novak
62607e3b02 Fixed accidental free of XImage data. 2018-02-12 12:44:51 +01:00
Armin Novak
7c0a33632b Fixed #4412: XImage must be freed with XDestroyImage 2018-02-12 10:02:35 +01:00
Armin Novak
d81e2ec351 Fixed X11 stride on screen update. 2017-11-21 11:09:16 +01:00
Armin Novak
24233b299c Fix #4111: 15bpp color handling for brush. 2017-11-20 15:00:09 +01:00
David Fort
9ac54ede6d Fix displaying of raw surfaces 2017-09-26 13:56:08 +02:00
Ondrej Holy
3137caebf2 Fix colors on big endian (#4135)
* client/x11: Fix colors on big endian

The bitmaps are recieved in little endian byte order (LSBFirst in terms
of X11). This is a problem on systems, where bitmaps are expected in big
endian byte order (MSBFirst). X11 client tries to handle this situation
by different color formats (e.g. RGBX vs. BGRX), but this doesn't work.
BGRX is not MSBFirst variant of RGBX, it should be XRGB. But this would
work only for 32/24 depths, where color channels matches bytes. Let's
say to the XServer that all the bitmaps are in LSBFirst encoding instead
of changing the color formats on big endian systems.

https://github.com/FreeRDP/FreeRDP/issues/2520

* client/x11: Fix cursor color on big endian

The cursor color is wrong on big endian systems. It is not probably
possible to force bitmap_order for XcursorImage as it is possible for
XImage. Fortunately, cursors are always 32 bit so we can use ABGR
instead of BGRA to deal with.

https://github.com/FreeRDP/FreeRDP/issues/2520

* client/x11: Fix comment indentation

The comment has wrong indentation for some reason, let's fix it.

* client/x11: Fix BGR vs. RGB detection

The BGR vs. RGB detection code is leftover from history and I am conviced
that it is wrong with the current color handling, where invert flag is TRUE
by default. However, I suppose that the detection still makes sense and
XServer may use the inverted formats in some cases. Maybe we can force XServer
to use our masks somehow, but let's just fix the value to FALSE now.

* client/x11: Remove unused color shifts

The color shifts are lefover from history and are not used in current
code. Let's remove them.
2017-09-18 10:47:56 +02:00
Armin Novak
0490aeb018 Fixed clang malloc integer overflow warnings. 2017-07-20 09:29:48 +02:00
Norbert Federa
fa2086686b some performance fixes
- draw only the updated region in the gdi and x11 surface bits implementation
- don't repeatedly call IsProcessorFeaturePresentEx in rfx rlgr decoder
- fix ugly and unaligned profiler print layout and remove an unnecessary value
2017-03-27 20:14:13 +02:00
Armin Novak
b2c29158be Scanbuild warning, argument checks and leak fixes.
* Added Stream_GetRemainingCapacity to check remaining stream size
  before writes.
* Fixed shadow server memory leak.
* Fixed lots of scanbuild warnings
* Added missing argument checks in many functions
* Added missing static function declarations
2017-03-02 18:13:43 +01:00
Armin Novak
198bc6d9e1 Fixed compiler warnings. 2017-02-16 13:17:49 +01:00
Armin Novak
052736a825 Fixed conversion to XColor. 2017-01-25 08:34:23 +01:00
Armin Novak
0acc54a4f7 Fixed scanline for RFX. 2017-01-23 11:30:56 +01:00
Armin Novak
085ee3ac0f Fixed patblt for mono color. #3678 2017-01-09 15:25:47 +01:00
Norbert Federa
f71b6b46e8 fix string format specifiers
- fixed invalid, missing or additional arguments
- removed all type casts from arguments
- added missing (void*) typecasts for %p arguments
- use inttypes defines where appropriate
2016-12-16 13:48:43 +01:00
Armin Novak
514a33d435 Respecting color depth with RDP_CODEC_ID_NONE. 2016-12-07 13:22:52 +01:00
Martin Fleisz
e059e3dea3 codec: Remove src format parameter from rfx_process_message
This parameter is not required. Default param is BGRX32
and if an other format is required rfx_codec_set_pixel_format
can be used.
2016-12-01 17:21:06 +01:00
Martin Fleisz
f5a6645832 Merge pull request #3571 from akallabeth/clang_fixes
Clang fixes
2016-10-19 15:45:17 +02:00
Armin Novak
4ceda8a097 Fixed uninitialized return. 2016-10-19 11:11:35 +02:00
Armin Novak
9c50e1a635 Replaced _VF formats with copy flags. 2016-10-14 10:36:52 +02:00
Armin Novak
25c7372fe6 Fixed offset and stride for X11 surface bits. 2016-10-12 09:32:05 +02:00
Armin Novak
8286fa4b96 Fixed format flip and offsets for GDI. 2016-10-11 18:50:01 +02:00
Armin Novak
404ae7d83c Fixed Windows 7 RFX issues. 2016-10-11 14:34:07 +02:00
zihao.jiang
34b4bca119 Fix nsc client after after color conversion huge fix
nsc codec client doesn't work:
1. We should mark invalid region for software gdi
2. Checked the code before color conversion fix, the correct color format should be PIXEL_FORMAT_BGRX32_VF (corresponds to old PIXEL_FORMAT_XRGB32_VF)
3. For gdi:hw: xfc->bitmap_buffer is never used/initialized. However gdi->primary_buffer is always maintained. So use primary_buffer to hold the decoded bitmap data
2016-10-11 02:18:31 +08:00
Armin Novak
c242467346 Fixed some X11 functions. 2016-10-06 13:43:10 +02:00
Armin Novak
b765374d32 Refactored BitBlt, implementing everything now. 2016-10-06 13:43:08 +02:00
Armin Novak
c8a7065f93 Refactored xfContext. 2016-10-06 13:43:06 +02:00
Armin Novak
9af71b9878 Fixed color conversion from order. 2016-10-06 13:43:06 +02:00
Armin Novak
7dbeeb2df6 Fixed hardware GDI. 2016-10-06 13:43:04 +02:00
Armin Novak
9d82c4701b Removed duplicate code. 2016-10-06 13:43:03 +02:00
Armin Novak
85bbe2a908 API refactoring. 2016-10-06 13:43:03 +02:00
Armin Novak
e650fdb2b6 Bugfixes. 2016-10-06 13:43:02 +02:00
Armin Novak
9d5ca34d0d Fixed pointer initialisation for X11. 2016-10-06 13:43:02 +02:00
Armin Novak
04fb4e4e4c Fixed rectangle clipping and warnings. 2016-10-06 13:43:02 +02:00
Armin Novak
c229a1939d Updated codec API, unified drawing order color decoding. 2016-10-06 13:43:02 +02:00
Armin Novak
bc8217ed42 Removed palette, using GDI one. 2016-10-06 13:43:01 +02:00
Armin Novak
8fffda5740 Fixed clearcodec and codecs reset. 2016-10-06 13:43:01 +02:00
Armin Novak
e860fde4bc Fixed function arguments. 2016-10-06 13:43:00 +02:00
Armin Novak
d98677094e Fixed warnings. 2016-10-06 13:43:00 +02:00
Armin Novak
716c5c53e9 256 color support. 2016-10-06 13:42:59 +02:00
Armin Novak
ac8c72238d Api update. 2016-10-06 13:42:59 +02:00
Armin Novak
24599af991 More color fixes. 2016-10-06 13:42:58 +02:00
Armin Novak
df35c135d1 Fixed color conversion, unified GFX and updated API. 2016-10-06 13:42:58 +02:00
Nathan Kidd
24aa71e6cb client/X11: Only upconvert < 24 bit RFX tiles
Fixing this logical typo gives a slight performance boost /
CPU usage reduction on heavy loads.
2016-09-23 21:26:56 -04:00
zihao.jiang
e7d9e91864 channels/rdpgfx: Make freerdp_client_codecs_prepare calls codec reset. It fix broken h264 client from #3328. Also it doesn't make sense that we don't need width and height for codec initialization while we need them for codec reset 2016-05-30 02:00:05 +08:00
Armin Novak
d83af81295 Fixed uninitialized value. 2015-06-23 12:12:16 +02:00
Norbert Federa
4d661fdb35 client/x11: fix + simplify invalidate_poly_region 2015-05-19 01:13:40 +02:00
Norbert Federa
ef1fd12b15 Fix unchecked CreateEvent calls and misc fixes
1)
Added missing checks for CreateEvent which also required the
following related changes:

- changed freerdp_context_new API to BOOL
- changed freerdp_peer_context_new API to BOOL
- changed pRdpClientNew callback to BOOL
- changed pContextNew callback to BOOL
- changed psPeerAccepted callback to BOOL
- changed psPeerContextNew callback to BOOL

2)
Fixed lots of missing alloc and error checks in the
changed code's neighbourhood.

3)
Check freerdp_client_codecs_prepare result to avoid segfaults
caused by using non-initialized codecs.

4)
Fixed deadlocks in x11 caused by missing xf_unlock_x11() calls
in some error handlers

5)
Some fixes in thread pool:
- DEFAULT_POOL assignment did not match TP_POOL definition
- don't free the pool pointer if it points to the static DEFAULT_POOL
- added error handling and cleanup in InitializeThreadpool
2015-04-29 18:18:39 +02:00