<html> <head> <base href="https://bugs.freedesktop.org/" /> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - NULL dereference in weston_pointer_send_frame with RDP backend" href="https://bugs.freedesktop.org/show_bug.cgi?id=93833#c5">Comment # 5</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - NULL dereference in weston_pointer_send_frame with RDP backend" href="https://bugs.freedesktop.org/show_bug.cgi?id=93833">bug 93833</a> from <a class="email" href="mailto:lnicola@dend.ro" title="Laurentiu Nicola <lnicola@dend.ro>"> Laurentiu Nicola</a> <pre>(In reply to Jonas Ådahl from <a href="show_bug.cgi?id=93833#c3">comment #3</a>) > (In reply to Laurentiu Nicola from <a href="show_bug.cgi?id=93833#c2">comment #2</a>) > > Connecting with freerdp seems to work, at least with my patch applied. > > Do you have that patch somewhere? Sorry, what I meant is that even after fixing the crash here (see attached patch), weston still crashes with mstsc, while it works with wfreerdp. When I made that comment I didn't know whether connecting with wfreerdp worked without my initial patch. I tested it in the meanwhile and: 1. the original code crashes with both wfreerdp and mstsc 2. with my fix, wfreerdp can connect, while mstsc can't: [08:12:30.778] kbd_layout:0x409 kbd_type:0x7 kbd_subType:0x0 kbd_functionKeys:0xc [08:12:30.778] xf_peer_activate: matching layout=us variant=(null) [08:12:30.886] unable to checkDescriptor for 0x731010 [Thread 0x7fffe860a700 (LWP 18653) exited] [Thread 0x7fffe7e09700 (LWP 18652) exited] [Thread 0x7fffe7608700 (LWP 18651) exited] [Thread 0x7fffe6e07700 (LWP 18650) exited] [08:12:30.889] input_method disconnected, respawning... [08:12:30.889] launching '/usr/lib/weston/weston-keyboard' wl_registry@2: error 0: invalid global wl_seat (14) [08:12:30.893] Error: /usr/lib/weston/weston-desktop-shell apparently cannot run at all. Quitting...wl_registry@2: error 0: invalid global wl_seat (14) [Inferior 1 (process 18640) exited normally] 3. there are a few other places in the same file that lack the same NULL check I'm not familiar with the code, so I don't know whether focus_client should be NULL. PS: while toying with it, I also got another crash: Program received signal SIGSEGV, Segmentation fault. weston_compositor_wake (compositor=compositor@entry=0x0) at ../src/compositor.c:3894 3894 uint32_t old_state = compositor->state; (gdb) bt #0 weston_compositor_wake (compositor=compositor@entry=0x0) at ../src/compositor.c:3894 #1 0x00000000004131c3 in notify_motion_absolute (seat=seat@entry=0x71c608, time=2089892085, x=x@entry=143872, y=y@entry=84992) at ../src/input.c:1281 #2 0x00007ffff61c4e3b in xf_mouseEvent (input=<optimized out>, flags=<optimized out>, x=<optimized out>, y=<optimized out>) at ../src/compositor-rdp.c:956 #3 0x00007ffff5f0f61c in fastpath_recv_inputs () from /usr/lib/libfreerdp.so.2.0 #4 0x00007ffff5f1f1b7 in ?? () from /usr/lib/libfreerdp.so.2.0 #5 0x00007ffff5f1f7a8 in ?? () from /usr/lib/libfreerdp.so.2.0 #6 0x00007ffff5f124f0 in transport_check_fds () from /usr/lib/libfreerdp.so.2.0 #7 0x00007ffff5f0b338 in rdp_check_fds () from /usr/lib/libfreerdp.so.2.0 #8 0x00007ffff5f1ef23 in ?? () from /usr/lib/libfreerdp.so.2.0 #9 0x00007ffff61c4c88 in rdp_client_activity (fd=<optimized out>, mask=<optimized out>, data=0x730ec0) at ../src/compositor-rdp.c:658 #10 0x00007ffff7bd2462 in wl_event_loop_dispatch () from /usr/lib/libwayland-server.so.0 #11 0x00007ffff7bd0cc5 in wl_display_run () from /usr/lib/libwayland-server.so.0 #12 0x000000000041c22b in main (argc=1, argv=0x7fffffffe4d8) at ../src/main.c:859 (gdb) l 3889 * Restarts the idle timer. 3890 */ 3891 WL_EXPORT void 3892 weston_compositor_wake(struct weston_compositor *compositor) 3893 { 3894 uint32_t old_state = compositor->state; 3895 3896 /* The state needs to be changed before emitting the wake 3897 * signal because that may try to schedule a repaint which 3898 * will not work if the compositor is still sleeping */ (gdb) p compositor $1 = (struct weston_compositor *) 0x0</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the assignee for the bug.</li> </ul> </body> </html>