[Cin] iris square transition crash (termux/arm)

Andrew Randrianasulu randrianasulu at gmail.com
Sun Aug 29 11:11:03 CEST 2021


On Sunday, August 29, 2021, Phyllis Smith via Cin <
cin at lists.cinelerra-gg.org> wrote:

> Andrew,
> WOW, highly reproducible -- I can not believe you found this as it is
> always hard to remember what was done to cause the problem.
> The fix works Great! to stop the crash and I will check it into GIT next
> time I power up the desktop.
> What was the purpose of the previous email: "illustration about 'iris
> square; transition"?  Was it just to show that it does not work correctly
> in the case of the steps outlined here?
>

it was illustration for Andrea, and while making this illustration I
discovered this crash :-} thankfully, it was  easy to fix. Hopefully fix
have no side effects.. (we do not need to actually process 0.0000 seconds
long  transition?)



>
>
> Thanks, Phyllis
> Inbox
>
> On Fri, Aug 27, 2021 at 12:31 PM Andrew Randrianasulu via Cin <
> cin at lists.cinelerra-gg.org> wrote:
>
>> for me attached patch seems to fix crash (because 'add silence'
>> apparently  add second transition with duration exactly 0.000 sec - so we
>> crash trying to process this)
>>
>>
>>
>> On Friday, August 27, 2021, Andrew Randrianasulu <randrianasulu at gmail.com>
>> wrote:
>>
>>> for future history:
>>>
>>> [New LWP 26812]
>>> --Type <RET> for more, q to quit, c to continue without paging--
>>>
>>> Thread 156 "cin" received signal SIGFPE, Arithmetic exception.
>>> [Switching to LWP 26812]
>>> 0xf7316cf8 in tgkill ()
>>>    from /apex/com.android.runtime/lib/bionic/libc.so
>>> (gdb) bt full
>>> #0 0xf7316cf8 in tgkill ()
>>>    from /apex/com.android.runtime/lib/bionic/libc.so
>>> No symbol table info available.
>>> #1 0xf5e3aa40 in __aeabi_ldiv0 ()
>>>     at /buildbot/src/android/gcc/toolchain/build/../gcc/gcc-4.
>>> 9/libgcc/config/arm/lib1funcs.S:1337
>>> No locals.
>>> #2 0xf5e36740 in IrisSquareMain::process_realtime (this=0xf6044000,
>>>     incoming=0xf1959740, outgoing=0xf1959900) at irissquare.C:312
>>>         h = <optimized out>
>>>         w = <optimized out>
>>> #3 0x00b36dc4 in PluginServer::process_transition(VFrame*, VFrame*,
>>> long long, long long) ()
>>> No symbol table info available.
>>> #4 0x00bea320 in VModule::render(VFrame*, long long, int, double, int,
>>> int, int) ()
>>> No symbol table info available.
>>> #5 0x00be7fd8 in VirtualVNode::read_data(VFrame*, long long, double,
>>> int) ()
>>> No symbol table info available.
>>> #6 0x00be824c in VirtualVNode::render_as_module(VFrame*, VFrame*, long
>>> long, double, int) ()
>>> No symbol table info available.
>>> #7 0x00be8070 in VirtualVNode::render(VFrame*, long long, double, int) ()
>>> No symbol table info available.
>>> #8 0x00be7aa0 in VirtualVConsole::process_buffer(long long, int) ()
>>> No symbol table info available.
>>> #9 0x00bef280 in VRender::process_buffer(long long, int) ()
>>> No symbol table info available.
>>> #10 0x00bef804 in VRender::run() ()
>>> --Type <RET> for more, q to quit, c to continue without paging--
>>> No symbol table info available.
>>> #11 0x00cc04d4 in Thread::entrypoint(void*) ()
>>> No symbol table info available.
>>> #12 0xf7325544 in __pthread_start(void*) ()
>>>    from /apex/com.android.runtime/lib/bionic/libc.so
>>> No symbol table info available.
>>> #13 0xf72dfc44 in __start_thread ()
>>>    from /apex/com.android.runtime/lib/bionic/libc.so
>>> No symbol table info available.
>>> #14 0x00000000 in ?? ()
>>> No symbol table info available.
>>> Backtrace stopped: previous frame identical to this frame (corrupt
>>> stack?)
>>>
>>> ---
>>>
>>> part of gdb session...
>>>
>>> for reproduction.
>>>
>>> 1) load video
>>> 2) seek to some frame by setting there playhead/cursor on timeline
>>> 3) apply edit > cut (x) {blade cut}
>>> 4) attach iris square transition from video > attach transition menu.
>>> 5) set in point at very same frame
>>> 6) move cursor a bit after transition bar ended..
>>> 7) set out point
>>> 8) use edit>add silence..
>>>
>>> in my case it crashes....
>>>
>>> --
>> Cin mailing list
>> Cin at lists.cinelerra-gg.org
>> https://lists.cinelerra-gg.org/mailman/listinfo/cin
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cinelerra-gg.org/pipermail/cin/attachments/20210829/373f0e9e/attachment.htm>


More information about the Cin mailing list