summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
* | | | | Merge pull request #137 from ↵efriedrich2023-01-051-18/+2
|\ \ \ \ \ | |_|/ / / |/| | | | | | | | | | | | | | khangtb1/fix_crash_and_hang_on_simple_weston_client_example simple-weston-client: fix crash and hang when exiting
| * | | | simple-weston-client: fix crash and hang when exitingTran Ba Khang(MS/EMC31-XC)2022-12-011-18/+2
|/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | wlcontext->debug_iface didn't bind when weston has been running with non debug option. So, we need to check it before call weston_debug_v1_destroy. Need to remove the while loop after send SIGINT, the stream->ojb always available in this case, then it makes a hang here. Signed-off-by: Tran Ba Khang(MS/EMC31-XC) <Khang.TranBa@vn.bosch.com>
* | | | Merge remote-tracking branch 'upstream/pull/124'Jeka Friedrich2022-02-202-4/+7
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * upstream/pull/124 EGLWLInputEventExample: Repair array out of bounds access simple-weston-client: Repair array out of bounds access Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com> Tested-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * | | | EGLWLInputEventExample: Repair array out of bounds accessMarek Vasut2022-02-171-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The problem is in the `fprintf(..., left_ptrs[j])`, where if this code is ever triggered, the variable `j=4` always, while the `left_ptrs` array only has four entries instead of five, so the code would access one entry past the array. In case this code is triggered, it does cause segmentation fault. Signed-off-by: Marek Vasut <marex@denx.de>
| * | | | simple-weston-client: Repair array out of bounds accessMarek Vasut2022-02-171-2/+4
| |/ / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix array out of bounds access detected by gcc format fortification check: ``` In file included from /test/recipe-sysroot/usr/include/stdio.h:867, from /test/git/ivi-layermanagement-examples/simple-weston-client/src/simple-weston-client.c:23: In function 'fprintf', inlined from 'create_cursors' at /test/git/ivi-layermanagement-examples/simple-weston-client/src/simple-weston-client.c:120:9, inlined from 'seat_handle_capabilities' at /test/git/ivi-layermanagement-examples/simple-weston-client/src/simple-weston-client.c:247:9: /test/recipe-sysroot/usr/include/bits/stdio2.h:100:10: error: '%s' directive argument is null [-Werror=format-overflow=] 100 | return __fprintf_chk (__stream, __USE_FORTIFY_LEVEL - 1, __fmt, | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 101 | __va_arg_pack ()); | ~~~~~~~~~~~~~~~~~ ``` The problem is in the `fprintf(..., left_ptrs[j])`, where if this code is ever triggered, the variable `j=4` always, while the `left_ptrs` array only has four entries instead of five, so the code would access one entry past the array. In case this code is triggered, it does cause segmentation fault. Signed-off-by: Marek Vasut <marex@denx.de>
* | | | Merge remote-tracking branch 'upstream/pull/107'Jeka Friedrich2022-02-202-10/+23
|\ \ \ \ | |/ / / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | * upstream/pull/107 simple-weston-client: remove unused variable simple-weston-client: use correct dlt header simple-weston-client: make dlt optional Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * | | simple-weston-client: remove unused variableJeka2021-01-031-1/+0
| | | | | | | | | | | | | | | | Signed-off-by: Jeka <friedrix@gmail.com>
| * | | simple-weston-client: use correct dlt headerJeka2021-01-031-2/+1
| | | | | | | | | | | | | | | | | | | | | | | | dlt.h can be used instead of dlt_common.h and dlt_user.h Signed-off-by: Jeka <friedrix@gmail.com>
| * | | simple-weston-client: make dlt optionalJeka2021-01-032-10/+23
| |/ / | | | | | | | | | | | | | | | | | | | | | if weston-debug protocol is available we should use dlt if it is available so only optional otherwise we will just output to stderr Signed-off-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
* | | Merge remote-tracking branch 'upstream/pull/122'Jeka Friedrich2022-02-131-12/+0
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | * upstream/pull/122 ivi-input-controller: Prevent use-after-free when tearing down seat Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * | | ivi-input-controller: Prevent use-after-free when tearing down seatMarek Vasut2022-01-231-12/+0
| | |/ | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The seat destroy callback is triggered from weston_seat_release(), that means the only way the code can enter destroy_seat() is with all input devices associted with seat already released and all the pointers already invalid. Avoid touching any of these input device pointers in the destroy_seat() callback, otherwise this triggers use-after-free condition. ASan detects the use-after-free condition as follows: ``` ==89420==ERROR: AddressSanitizer: heap-use-after-free on address 0x61100028e6a0 at pc 0x7f50deacb6fd bp 0x7ffed8504fe0 sp 0x7ffed8504fd8 WRITE of size 8 at 0x61100028e6a0 thread T0 #0 0x7f50deacb6fc in weston_keyboard_end_grab ../libweston/input.c:1604 #1 0x7f50bc6aac97 in destroy_seat (/tmp/wt/lib/x86_64-linux-gnu/weston/ivi-input-controller.so+0x7c97) #2 0x7f50bc6ab02f in handle_seat_destroy (/tmp/wt/lib/x86_64-linux-gnu/weston/ivi-input-controller.so+0x802f) #3 0x7f50deac477d in wl_signal_emit /usr/include/wayland-server-core.h:478 #4 0x7f50dead5a93 in weston_seat_release ../libweston/input.c:3490 #5 0x7f50bc6973d0 in ss_seat_destroy ../compositor/screen-share.c:395 #6 0x7f50bc69922c in registry_handle_global_remove ../compositor/screen-share.c:751 #7 0x7f50de8b89d9 (/lib/x86_64-linux-gnu/libffi.so.8+0x79d9) #8 0x7f50de8b7b20 (/lib/x86_64-linux-gnu/libffi.so.8+0x6b20) #9 0x7f50dea582d1 (/lib/x86_64-linux-gnu/libwayland-client.so.0+0x92d1) #10 0x7f50dea54979 (/lib/x86_64-linux-gnu/libwayland-client.so.0+0x5979) #11 0x7f50dea5603b in wl_display_dispatch_queue_pending (/lib/x86_64-linux-gnu/libwayland-client.so.0+0x703b) #12 0x7f50bc69931a in shared_output_handle_event ../compositor/screen-share.c:771 #13 0x7f50dea434b1 in wl_event_loop_dispatch (/lib/x86_64-linux-gnu/libwayland-server.so.0+0xb4b1) #14 0x7f50dea41134 in wl_display_run (/lib/x86_64-linux-gnu/libwayland-server.so.0+0x9134) #15 0x7f50deed8489 in wet_main ../compositor/main.c:3564 #16 0x55961496217d in main ../compositor/executable.c:33 #17 0x7f50dece97ec in __libc_start_main ../csu/libc-start.c:332 #18 0x559614962099 in _start (/tmp/wt/bin/weston+0x1099) ``` Signed-off-by: Marek Vasut <marex@denx.de>
* | | Merge remote-tracking branch 'upstream/pull/121'Jeka Friedrich2022-02-061-11/+13
|\ \ \ | |/ / |/| | | | | | | | | | | | | | * upstream/pull/121 Fix NULL pointer exception in case of no input device Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * | Fix NULL pointer exception in case of no input deviceNaoto Yamaguchi2021-10-081-11/+13
|/ / | | | | | | | | | | | | | | | | | | | | When weston with wayland-ivi-extension run without input device, that crash by NULL access at setup_input_focus function. In this case, st_focus->seat_ctx is NULL pointer, but it use without NULL check in this function. This patch add NULL check for st_focus->seat_ctx. Signed-off-by: Naoto Yamaguchi <naoto.yamaguchi@aisin.co.jp>
* | Merge remote-tracking branch 'upstream/pull/120'Eugen Friedrich2021-01-184-13/+22
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | * upstream/pull/120 test : Fix compiler warnings EGLWLMockNavigation : Fix compiler warnings simple-weston-client : Fix compiler warnings ivi-controller : Fix compiler warnings Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * | test : Fix compiler warningsShivakumar Halagatti2020-12-211-7/+14
| | | | | | | | | | | | | | | | | | Solve indentation warnings for if-else case and for loop. Signed-off-by: Shivakumar Halagatti <Shivakumar.Halagatti@in.bosch.com>
| * | EGLWLMockNavigation : Fix compiler warningsShivakumar Halagatti2020-12-211-1/+2
| | | | | | | | | | | | | | | | | | Solve "ignoring return value" warning of fread function. Signed-off-by: Shivakumar Halagatti <Shivakumar.Halagatti@in.bosch.com>
| * | simple-weston-client : Fix compiler warningsShivakumar Halagatti2020-12-211-4/+5
| | | | | | | | | | | | | | | | | | | | | Solve "ignoring return value" of read and pipe functions and remove unused variables temp and offset. Signed-off-by: Shivakumar Halagatti <Shivakumar.Halagatti@in.bosch.com>
| * | ivi-controller : Fix compiler warningsShivakumar Halagatti2020-12-211-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | API weston_config_section_get_bool needs third argument of type int. Since enable_cursor is of uint8_t type change it to int to solve the "incompatible pointer type" warning. Signed-off-by: Shivakumar Halagatti <Shivakumar.Halagatti@in.bosch.com>
* | | Merge remote-tracking branch 'upstream/pull/115'Eugen Friedrich2021-01-036-0/+130
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * upstream/pull/115 layer-add-surfaces: Register for ilm shutdown notification ilmCommon: Implement a shutdown notification for ilm ilmControl: Implement a shutdown notification for ilm Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * | | layer-add-surfaces: Register for ilm shutdown notificationRajendraprasad K J2020-07-081-0/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Register for ilm shutdown notification so that we would be notified when some unrecoverable error occurs in ilmControl. Signed-off-by: Rajendraprasad K J <KarammelJayakumar.Rajendraprasad@in.bosch.com>
| * | | ilmCommon: Implement a shutdown notification for ilmRajendraprasad K J2020-07-083-0/+39
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Added new ilm API using which user could register for the shutdown notification. Signed-off-by: Rajendraprasad K J <KarammelJayakumar.Rajendraprasad@in.bosch.com>
| * | | ilmControl: Implement a shutdown notification for ilmRajendraprasad K J2020-07-082-0/+60
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We should gracefully exit when we encounter some unrecoverable errors in ilmControl, e.g.: OOM or wayland errors. But before we exit, we have to send a notification to the application side that we are shutting down and releasing all resources. Signed-off-by: Rajendraprasad K J <KarammelJayakumar.Rajendraprasad@in.bosch.com>
* | | | Merge pull request #116 from rajendraprasadkj/alpha_blendingefriedrich2021-01-030-0/+0
|\ \ \ \ | |_|/ / |/| | | Avoid unnecessary scaling of the view transformation matrix
| * | | ivi-controller: Avoid unnecessary scaling of the view transformation matrixRajendraprasad K J2020-07-291-3/+7
| |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The opaque region of a weston view is updated only if the alpha value is 1 and the transform matrix is of type WESTON_MATRIX_TRANSFORM_TRANSLATE. Here opaque region is never updated, as we are performing scaling operations to the view transform matrix, even when the scaling factor is 1, thus changing the matrix type to WESTON_MATRIX_TRANSFORM_SCALE. Perform scaling of the view transformation matrix only when the scaling factor is non-zero. Signed-off-by: Rajendraprasad K J <KarammelJayakumar.Rajendraprasad@in.bosch.com>
* | | Merge remote-tracking branch 'upstream/pull/116'Eugen Friedrich2020-09-101-3/+7
|\ \ \ | |/ / |/| | | | | | | | | | | | | | * upstream/pull/116 ivi-controller: Avoid unnecessary scaling of the view transformation matrix Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * | ivi-controller: Avoid unnecessary scaling of the view transformation matrixRajendraprasad K J2020-03-261-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The opaque region of a weston view is updated only if the alpha value is 1 and the transform matrix is of type WESTON_MATRIX_TRANSFORM_TRANSLATE. Here opaque region is never updated, as we are performing scaling operations to the view transform matrix(even when the scaling factor is 1), thus changing the matrix type to WESTON_MATRIX_TRANSFORM_SCALE. As a result, views are never opaque and won't be able to place it in HW planes(forced to use renderer). Perform scaling of the view transformation matrix only when the scaling factor is non-zero. Signed-off-by: Rajendraprasad K J <KarammelJayakumar.Rajendraprasad@in.bosch.com>
* | | Merge remote-tracking branch 'upstream/pull/117'Eugen Friedrich2020-07-076-78/+364
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * upstream/pull/117 ilmControl: add support for saving screenshot as png file Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com> Reviewed-by: Michael Rodin <mrodin@de.adit-jv.com> Tested-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * | | ilmControl: add support for saving screenshot as png fileRamzi BEN MEFTAH2020-06-166-78/+364
| |/ / | | | | | | | | | | | | | | | To save as png file, the file name should have ".png" as extension. Signed-off-by: Ramzi BEN MEFTAH <rbmeftah@de.adit-jv.com>
* | | Merge remote-tracking branch 'upstream/pull/118'Eugen Friedrich2020-07-071-1/+4
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | * upstream/pull/118 ilmCommon: Fix segmentation fault when ilm_isInitialized is called before ilm_init Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com> Tested-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * | | ilmCommon: Fix segmentation fault when ilm_isInitialized is called before ↵Shivakumar Halagatti2020-05-111-1/+4
| |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ilm_init Segmentation fault is observed when ilm_isInitialized is called and if ilm_init was not called before. Here gIlmComonPlatformFunc is initialized in ilm_init using init_ilmCommonPlatformTable, so without calling ilm_init, gIlmCommonPlatformFunc.isInitialized is NULL. Hence calling ilm_isInitialized results in segfault. Check whether gIlmComonPlatformFunc is initialized. ilm_isInitialized should be called only when gIlmComonPlatformFunc is initialized. Signed-off-by: Shivakumar Halagatti <Shivakumar.Halagatti@in.bosch.com>
* | | Merge remote-tracking branch 'upstream/pull/109'Eugen Friedrich2020-06-232-3/+26
|\ \ \ | |/ / |/| | | | | | | | | | | | | | | | | * upstream/pull/109 Fix segmentation fault observed at compositor termination Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com> Tested-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * | Fix segmentation fault observed at compositor terminationRajendraprasad K J2019-09-172-3/+26
| |/ | | | | | | | | | | | | | | A segmentation fault is observed on compositor de-init, if the 'simple-weston-client' is already killed. Here in termination sequence 'set_bkgnd_surface_prop()' is called without active 'simple-weston-client' resulting in segfault. 'set_bkgnd_surface_prop()' should be called only when the client is active. Signed-off-by: Rajendraprasad K J <KarammelJayakumar.Rajendraprasad@in.bosch.com>
* | Merge pull request #113 from csaavedra/gnu-install-dirsefriedrich2020-03-091-0/+2
|\ \ | | | | | | cmake: add missing include of GNUInstallDirs
| * | cmake: add missing include of GNUInstallDirsClaudio Saavedra2020-02-261-0/+2
| | | | | | | | | | | | | | | CMAKE_INSTALL_LIBDIR (at least) is used, so GNUInstallDirs needs to be included. Otherwise it might expand to empty.
* | | Merge pull request #112 from vkadasani/ivi_application_fix_typoefriedrich2020-03-011-7/+7
|\ \ \ | |/ / |/| | ivi-application: fix typo
| * | ivi-application: fix typos and grammarVeeresh Kadasani2020-02-041-7/+7
|/ / | | | | | | | | | | fix all typos and grammar. Signed-off-by: Veeresh Kadasani <external.vkadasani@jp.adit-jv.com>
* | Merge remote-tracking branch 'upstream/pull/111'Eugen Friedrich2020-01-093-5/+25
|\ \ | |/ |/| | | | | | | | | | | | | * upstream/pull/111 simple-weston-client: add cursor enable option ivi-controller: add cursor enable option Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com> Tested-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * simple-weston-client: add cursor enable optionRajendraprasad K J2020-01-081-5/+15
| | | | | | | | | | | | | | | | | | Create cursor only if 'IVI_CLIENT_ENABLE_CURSOR' environment variable is set to true. Signed-off-by: Karthiraman Ramaiah <external.kramaiah@de.adit-jv.com> Signed-off-by: Maniraj Devadoss <Maniraj.Devadoss@in.bosch.com> Signed-off-by: Rajendraprasad K J <KarammelJayakumar.Rajendraprasad@in.bosch.com>
| * ivi-controller: add cursor enable optionRajendraprasad K J2020-01-082-0/+10
|/ | | | | | | | | | | | | | | | | By default the cursor wont be created, If the user wants then the "enable-cursor" option can be added as true in the ivi-shell section of weston config file, as shown below: [ivi-shell] ... enable-cursor=true Set 'IVI_CLIENT_ENABLE_CURSOR' environment variable if user has added "enable-cursor" option in weston config file. Signed-off-by: Karthiraman Ramaiah <external.kramaiah@de.adit-jv.com> Signed-off-by: Maniraj Devadoss <Maniraj.Devadoss@in.bosch.com> Signed-off-by: Rajendraprasad K J <KarammelJayakumar.Rajendraprasad@in.bosch.com>
* Merge remote-tracking branch 'upstream/pull/106'Eugen Friedrich2019-05-314-3/+447
|\ | | | | | | | | | | | | | | | | | | | | | | * upstream/pull/106 Integrate DLT support with weston-debug protocol simple-weston-client: enable debug-proto based on pkgconfig simple-weston-client: route the weston logs to dlt simple-weston-client: use the weston-debug protocol ivi-controller: new config option for debug-scopes Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com> Tested-by: Eugen Friedrich <efriedrich@de.adit-jv.com>
| * Integrate DLT support with weston-debug protocolRajendraprasad K J2019-05-082-7/+53
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PFB details of the changes made with the commit: 1) Modified CMakeLists to make it work with Weston 6. 2) Weston debug protocol is slightly changed. The list of debug scopes available to subscribe will be advertised to the client after binding with weston debug interface. Created a listener for the advertise event. Call weston_debug_subscribe only if the debug scope is available. 3) A hang was observed while stopping the weston service. To resolve the issue, modified the termination sequence such that after destroying the weston debug interface we should wait till all the stream objects get destroyed. Signed-off-by: Rajendraprasad K J <KarammelJayakumar.Rajendraprasad@in.bosch.com>
| * simple-weston-client: enable debug-proto based on pkgconfigManiraj Devadoss2019-04-302-17/+48
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Based on the libweston-2-protocols pkgconfig file availability, decide whether to have the debug-protocol support or not. Conflicts: ivi-layermanagement-examples/simple-weston-client/src/simple-weston-client.c Changes to be committed: modified: ivi-layermanagement-examples/simple-weston-client/CMakeLists.txt modified: ivi-layermanagement-examples/simple-weston-client/src/simple-weston-client.c Cherry-picked the commit 3ad4523f8edcee4326cb8ad51a6b51bb9dae3159 (manirajd:weston_debug_support_v1) to GENIVI:master. Resolved the merge conflict present in ivi-layermanagement-examples/simple-weston-client/src/simple-weston-client.c. Signed-off-by: Rajendraprasad K J <KarammelJayakumar.Rajendraprasad@in.bosch.com>
| * simple-weston-client: route the weston logs to dltManiraj Devadoss2019-04-302-2/+168
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | create the pipe b/w stdout and stdin stdout - write end stdin - read end weston will write to stdout through weston-debug protocol and the dlt_ctx_thread will read from stdin Conflicts: ivi-layermanagement-examples/simple-weston-client/src/simple-weston-client.c Changes to be committed: modified: ivi-layermanagement-examples/simple-weston-client/CMakeLists.txt modified: ivi-layermanagement-examples/simple-weston-client/src/simple-weston-client.c Cherry-picked the commit 9203e0d7f7996f5549ce4b0782b1535b2283ddd6 (manirajd:weston_debug_support_v1) to GENIVI:master. Resolved the merge conflict present in ivi-layermanagement-examples/simple-weston-client/src/simple-weston-client.c. Signed-off-by: Rajendraprasad K J <KarammelJayakumar.Rajendraprasad@in.bosch.com>
| * simple-weston-client: use the weston-debug protocolManiraj Devadoss2019-04-302-1/+192
| | | | | | | | | | | | | | to print the weston-debug logs to the stdout, user desired debug-scopes are read from weston-ini thorugh ivi-controller. Signed-off-by: Maniraj Devadoss <Maniraj.Devadoss@in.bosch.com>
| * ivi-controller: new config option for debug-scopesManiraj Devadoss2019-04-302-0/+10
|/ | | | | | | read the debug scopes from weston.ini and export it to the weston-client Signed-off-by: Maniraj Devadoss <Maniraj.Devadoss@in.bosch.com>
* Merge remote-tracking branch 'upstream/pull/105'Emre Ucan2019-04-095-0/+507
|\ | | | | | | | | | | | | | | | | | | | | * upstream/pull/105 ivi-controller: load id-agent module ivi-id-agent: added ivi-id-agent Added ivi-id-agent to CMake Reviewed-by: Eugen Friedrich <efriedrich@de.adit-jv.com> Reviewed-by: Emre Ucan <eucan@de.adit-jv.com> Tested-by: Emre Ucan <eucan@de.adit-jv.com>
| * ivi-controller: load id-agent moduleMichael Teyfel2019-04-081-0/+36
| | | | | | | | Signed-off-by: Michael Teyfel <mteyfel@de.adit-jv.com>
| * ivi-id-agent: added ivi-id-agentMichael Teyfel2019-04-083-0/+470
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a reference implementation of an ivi-id-agent plugin. It creates surface-ids for desktop-surfaces depending on the configuration provided in "weston.ini". For more please refer to the reference implementation in this commit. The only public interface available is the "id_agent_module_init" function. It is responsible for initialization of structs and reading the configuration. In the reference "weston.ini" two types of configurations can be found: [desktop-app] is used to configure a particular application. Therefore the desired surface-id must be provided. Moreover "app-title" is the title that is provided by the desktop application (xdg-protocol). "app-id" behaves accordingly. Although both parameters can be set, it is not mandatory. Finally at least one has to be set. [desktop-app-default] enables the id-agent to generate generic surface-ids for unconfigured applications, e.g. for development. This tag is optional. To generate the id an interval starting from "default-surface-id" to "default-surface-id-max" is used. The id is incremented until the interval is exceeded. In the function "get_id" a surface-id is assigned by means of the configuration. It can be adjusted, if another behavior is desired. In this plugin the parameters, that are described above, are evaluated. To use these patches please also apply the dedicated patches for weston: https://github.com/mtey/weston/tree/xdg_support_ivi_id_agent Since libweston-desktop is used to introduce desktop-surface support, xdg-protocol is supported. Signed-off-by: Michael Teyfel <mteyfel@de.adit-jv.com>
| * Added ivi-id-agent to CMakeMichael Teyfel2019-01-151-0/+1
| | | | | | | | Signed-off-by: Michael Teyfel <mteyfel@de.adit-jv.com>
* | Merge remote-tracking branch 'upstream/pull/104'2.2.0_weston_5_compartibleEmre Ucan2019-04-092-1/+6
|\ \ | |/ |/| | | | | | | | | | | * upstream/pull/104 LayerManagerControl: Prevent conversion of filenames into lowercase Reviewed-by: Emre Ucan <eucan@de.adit-jv.com> Tested-by: Emre Ucan <eucan@de.adit-jv.com>