Bug#872375: lirc: irrecord segfaults when recording a button

[Alec Leamas]

On Tue, 13 Feb 2018 21:54:19 +0100 Alec Leamas 
wrote:

> 
> This message really says it all: using irrecord  with the devinput
> driver is a useless and dangerous idea. See my other replies in the bug
> for more.
> 
> 

This is bug is outdated and partly off-topic. Closing. That is not to
say that irrecord works for all input devices -- it doesn't. It also
occasionally segfaults, I have not been able to find the reason.

Bug#872375: lirc: irrecord segfaults when recording a button

[Alec Leamas]

On Sun, 11 Feb 2018 10:07:02 -0800 James Bottomley
 wrote:
> I'm getting this same segfault as well.  My old remote fell apart, so I
> was trying to train the IR receiver on a new one (well, the unused VCR
> section of the current TV remote.

> This is what I get:

[cut]

> Usually you should not create a new config file for devinput
> devices. LIRC is installed with a devinput.lircd.conf file which 
> is built for the current system which works with all remotes 
> supported by the kernel. There might be a need to update 
> this file so it matches the current kernel. For this, use the 
> lirc-make-devinput(1) script.
>
> Press RETURN to continue.


This message really says it all: using irrecord  with the devinput
driver is a useless and dangerous idea. See my other replies in the bug
for more.


Cheers!

--alec

Bug#872375: lirc: irrecord segfaults when recording a button

[James Bottomley]

I'm getting this same segfault as well.  My old remote fell apart, so I
was trying to train the IR receiver on a new one (well, the unused VCR
section of the current TV remote.

This is what I get:

Starting program: /usr/bin/irrecord -d /dev/lirc0 -f
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-
gnu/libthread_db.so.1".
Warning: Running as root.
Using raw access on device /dev/lirc0

irrecord -  application for recording IR-codes for usage with lirc
Copyright (C) 1998,1999 Christoph Bartelmus(l...@bartelmus.de)

This program will record the signals from your remote control
and create a config file for lircd.

A proper config file for lircd is maybe the most vital part of this
package, so you should invest some time to create a working config
file. Although I put a good deal of effort in this program it is often
not possible to automatically recognize all features of a remote
control. Often short-comings of the receiver hardware make it nearly
impossible. If you have problems to create a config file READ THE
DOCUMENTATION at https://sf.net/p/lirc-remotes/wiki

If there already is a remote control of the same brand available at
http://sf.net/p/lirc-remotes you might want to try using such a
remote as a template. The config files already contains all
parameters of the protocol used by remotes of a certain brand and
knowing these parameters makes the job of this program much
easier. There are also template files for the most common protocols
available. Templates can be downloaded using irdb-get(1). You use a
template file by providing the path of the file as a command line
parameter.

Please take the time to finish the file as described in
https://sourceforge.net/p/lirc-remotes/wiki/Checklist/ an send it
to   so it can be made available to others.

Press RETURN to continue.




Usually you should not create a new config file for devinput
devices. LIRC is installed with a devinput.lircd.conf file which 
is built for the current system which works with all remotes 
supported by the kernel. There might be a need to update 
this file so it matches the current kernel. For this, use the 
lirc-make-devinput(1) script.

Press RETURN to continue.

Checking for ambient light  creating too much disturbances.
Please don't press any buttons, just wait a few seconds...

No significant noise (received 0 bytes)

Enter name of remote (only ascii, no spaces) :c
Using c.lircd.conf as output filename

Hold down an arbitrary key
...
.
Found gap (7941 us)

Please enter the name for the next button (press  to finish
recording)
KEY_A

Now hold down button "KEY_A".

Program received signal SIGSEGV, Segmentation fault.
0x in ?? ()
(gdb) bt
#0  0x in ?? ()
#1  0x779af1eb in record_buttons (btn_state=0x7fffe560, 
last_status=, state=, opts=)
at irrecord.c:1966
#2  0x814d in ?? ()
#3  0x6e4c in ?? ()
#4  0x77613f2a in __libc_start_main (main=0x64c0,
argc=4, 
argv=0x7fffeb58, init=, fini=, 
rtld_fini=, stack_end=0x7fffeb48)
at ../csu/libc-start.c:310
#5  0x74da in ?? ()
(gdb) 

James

Bug#872375: lirc: irrecord segfaults when recording a button

[Francois Gouget]

On Sun, 24 Sep 2017, Alec Leamas wrote:
[...]
> I note that you seemingly havn't used mode2 (?) This is the tool to determine
> what kind of data the kernel makes available to lirc which in the end
> determines what driver to use. This is described in the configuration guide.

I ran it at some point and ir-keytable too but It did not really 
enlighten me.

# mode2 --driver devinput --list-devices
/dev/input/event13 [147a:e03e] Formosa21 eHome Infrared Transceiver version:  
2.00 serial: 000100B6
/dev/input/by-id/usb-Formosa21_eHome_Infrared_Transceiver_000100B6-event-if00 
-> ../event13
/dev/input/by-path/pci-:03:00.0-usb-0:1.4:1.0-event -> ../event13

# mode2 --driver default --list-devices
/dev/lirc0 [147a:e03e] Formosa21 eHome Infrared Transceiver version:  2.00 
serial: 000100B6


# ir-keytable 
Found /sys/class/rc/rc0/ (/dev/input/event13) with:
Driver mceusb, table rc-rc6-mce
Supported protocols: lirc rc-5 rc-5-sz jvc sony nec sanyo mce_kbd rc-6 
sharp xmp 
Enabled protocols: lirc 
Name: Media Center Ed. eHome Infrared 
bus: 3, vendor/product: 147a:e03e, version: 0x1201
Repeat delay = 500 ms, repeat period = 125 ms



> > but I had to recreate
> >the remote configuration file from scratch as the one I used with lirc
> >0.9.4c wouldn't work. The official guidelines don't talk about that
> >incompatibility either.
> 
> Actually, there shouldn't be anything incompatible between 0.9.4c and 0.10.0.
> This seems strange...

My configuration files were actually a bit older so maybe that's why. Or 
it's yet again a devinput issue. I don't know if that will offer a clue 
but here are the three generations:

 * http://fgouget.free.fr/tmp/lirc/Hauppauge_PVR350.conf.0.9.0-pre1.0
   I think this is the oldest one, created with lirc 0.9.0-pre1 and 
   apparently using devinput. This one has two sets of numbers for each 
   key but the second one is always 0.

   E.g.: KEY_MENU 0x0400041F0D 0x00

 * http://fgouget.free.fr/tmp/lirc/Hauppauge_PVR350.conf.0.9.0-pre1
   This is the same as the previous one except I removed the extra 0x0 
   values. I think I had to do so to keep the configuration file 
   working.
   (I also removed flags and tweaked the gap at some point)

   E.g.: KEY_MENU 0x0400041F0D

 * http://fgouget.free.fr/tmp/lirc/Hauppauge_PVR350.conf.0.10.0
   This is the new configuration, recreated from scratch with Lirc 
   0.10.0-2 using the default driver.
   The values are completely different.

   E.g.: KEY_MENU 0x17CD

Unfortunately I don't remember exactly when the second generation 
stopped working, but it was probably 6 to 12 months ago.


-- 
Francois Gouget   http://fgouget.free.fr/
 "Only wimps use tape backup: _real_ men just upload their important stuff on
   ftp, and let the rest of the world mirror it ;)" -- Linus Torvalds

Bug#872375: lirc: irrecord segfaults when recording a button

[Alec Leamas]

On 24/09/17 16:12, Francois Gouget wrote:


Now as to why I have driver=devinput in lirc_options.conf, it's because
it's the only way I found to get my remote to work. Setting
driver=default driver does not work and if I only provide
devinput.lircd.conf and no Hauppauge_PVR350.conf file then it does not
work either.


I think you really need to take a step back here. The way you use lirc does
not match the official guidelines in [1], which you probably need to check.


Right. The default configuration and the official guidelines don't work.

* I used the devinput.lircd.conf provided by the package, together with
   driver = devinput and my remote wouldn't work.


The default configuration is, well, the default one. It's not guaranteed 
to work, but do work in a lot of cases (i. e., when the remote is 
supported by the kernel). Your statement that the official guidelines 
doesn't work needs a citation, so to speak.


I note that you seemingly havn't used mode2 (?) This is the tool to 
determine what kind of data the kernel makes available to lirc which in 
the end determines what driver to use. This is described in the 
configuration guide.




* I ran lirc-make-devinput as recommended by irrecord to generate a new
   devinput.lircd.conf file, still using driver=devinput, and the remote
   still wouldn't work.

* Seconds after I regenerated the devinput.lircd.conf with
   lirc-make-devinput, lircd complained that it is outdated:

Sep 24 01:33:47 habsheim lircd[1039]: lircd-0.10.0[1039]: Warning: Obsolete 
devinput config file used
Sep 24 01:33:47 habsheim lircd[1039]: lircd-0.10.0[1039]: Notice: Use the 
distributed devinput.lircd.dist by renaming it.
Sep 24 01:33:47 habsheim lircd[1039]: lircd-0.10.0[1039]: Notice: Or use irdb-get to 
search for "devinput" and download it.


These messages are actually misleading. I will look into this. That 
said, they are not blocking.



* With both versions of the devinput.lircd.conf lircd complained that
   all buttons were duplicated, which makes sense since they appear in
   both the devinput-32 and devinput-64 remotes.


Yes, but these are just "notice" messages. They can be ignored in this case.

* Now I finally got the remote to work with the default driver (i.e. not
   devinput (despite it being the default driver)) 


Happy to hear that it works for you!


but I had to recreate
   the remote configuration file from scratch as the one I used with lirc
   0.9.4c wouldn't work. The official guidelines don't talk about that
   incompatibility either.


Actually, there shouldn't be anything incompatible between 0.9.4c and 
0.10.0. This seems strange...



* Oh, and for the remote configuration files comment lines must have a
   '#' in the first column. Put spaces before and the configuration file
   breaks. If that's documented somewhere then I did not find where.


It's in the lircd.conf(4) manpage...


So crashes in irrecord? No one will notice that bit of brittleness.


I definitely agree that the crash is a bug. If I could reproduce it (or 
get a proper stacktrace) we might make a try to fix it. At least, we 
could augment the message with the fact that irrecord might crash. 
However, besides this the state of this bug is basically "wontfix".


--alec

Bug#872375: lirc: irrecord segfaults when recording a button

[Francois Gouget]

On Sun, 24 Sep 2017, Alec Leamas wrote:
[...]
> > But it's not 100% reproducible now, particularly when I try to use
> > my own libirrecord.so so I have debugging symbols. What I can give you
> > though is the irrecord backtrace:
> > 
> > #0  0x in ?? ()
> > #1  0x7f5644aaae73 in record_buttons ()
> > from /usr/lib/x86_64-linux-gnu/libirrecord.so.0
> > #2  0x802d in get_button_data (opts=0x7fffe2d0,
> > state=0x7fffe290,
> >  btn_state=0x7fffe110) at irrecord-ui.cpp:759
> > #3  do_record_buttons (state=0x7fffe290, opts=0x7fffe2d0) at
> > irrecord-ui.cpp:814
> > #4  0x6ddf in main (argc=, argv=)
> >  at irrecord-ui.cpp:1179
> > 
> > Compiled from git commit be62884be227.
> > So the crash happens somewhere in record_buttons() in irrecord.c.
> > I suspect the curr_driver methods are not getting initialized...
> > sometimes. 
> 
> Thanks. That said, if this from a version you compiled yourself, the
> libirrecord.so seems to be the wrong one (i. e., the one installed by the
> package).

True. It might not crash in the same way as when I use the package's 
irrecord with the package's libirrecord.so library. I'm done chasing it 
anyway.


> > Now as to why I have driver=devinput in lirc_options.conf, it's because
> > it's the only way I found to get my remote to work. Setting
> > driver=default driver does not work and if I only provide
> > devinput.lircd.conf and no Hauppauge_PVR350.conf file then it does not
> > work either.
> 
> I think you really need to take a step back here. The way you use lirc does
> not match the official guidelines in [1], which you probably need to check.

Right. The default configuration and the official guidelines don't work.

* I used the devinput.lircd.conf provided by the package, together with 
  driver = devinput and my remote wouldn't work.

* I ran lirc-make-devinput as recommended by irrecord to generate a new 
  devinput.lircd.conf file, still using driver=devinput, and the remote 
  still wouldn't work.

* Seconds after I regenerated the devinput.lircd.conf with 
  lirc-make-devinput, lircd complained that it is outdated:

Sep 24 01:33:47 habsheim lircd[1039]: lircd-0.10.0[1039]: Warning: Obsolete 
devinput config file used
Sep 24 01:33:47 habsheim lircd[1039]: lircd-0.10.0[1039]: Notice: Use the 
distributed devinput.lircd.dist by renaming it.
Sep 24 01:33:47 habsheim lircd[1039]: lircd-0.10.0[1039]: Notice: Or use 
irdb-get to search for "devinput" and download it.

* With both versions of the devinput.lircd.conf lircd complained that 
  all buttons were duplicated, which makes sense since they appear in 
  both the devinput-32 and devinput-64 remotes.

Sep 24 01:11:43 habsheim lircd[1047]: lircd-0.10.0[1047]: Notice: 
/etc/lirc/lircd.conf.d/devinput.lircd.conf: devinput-64: Multiple values for 
same code: BTN_MISC
Sep 24 01:11:43 habsheim lircd[1047]: lircd-0.10.0[1047]: Notice: 
/etc/lirc/lircd.conf.d/devinput.lircd.conf: devinput-64: Multiple values for 
same code: BTN_MOUSE

* Now I finally got the remote to work with the default driver (i.e. not 
  devinput (despite it being the default driver)) but I had to recreate 
  the remote configuration file from scratch as the one I used with lirc 
  0.9.4c wouldn't work. The official guidelines don't talk about that 
  incompatibility either.

* Oh, and for the remote configuration files comment lines must have a 
  '#' in the first column. Put spaces before and the configuration file 
  breaks. If that's documented somewhere then I did not find where.

So crashes in irrecord? No one will notice that bit of brittleness.

-- 
Francois Gouget   http://fgouget.free.fr/
 You can have my guns when you pry them from my kids cold, dead hands.

Bug#872375: lirc: irrecord segfaults when recording a button

[Alec Leamas]

On 24/09/17 03:50, Francois Gouget wrote:

Actually it's the -f option that causes the crash. I can reproduce it
with:

# irrecord -f /tmp/foo.conf


This would just indicate that foo.conf is broken somehow. Given the 
state of lirc, I don't think it's feasible to make it handle any kind of 
broken input files - it's just too fragile.



But it's not 100% reproducible now, particularly when I try to use
my own libirrecord.so so I have debugging symbols. What I can give you
though is the irrecord backtrace:

#0  0x in ?? ()
#1  0x7f5644aaae73 in record_buttons ()
from /usr/lib/x86_64-linux-gnu/libirrecord.so.0
#2  0x802d in get_button_data (opts=0x7fffe2d0, 
state=0x7fffe290,
 btn_state=0x7fffe110) at irrecord-ui.cpp:759
#3  do_record_buttons (state=0x7fffe290, opts=0x7fffe2d0) at 
irrecord-ui.cpp:814
#4  0x6ddf in main (argc=, argv=)
 at irrecord-ui.cpp:1179

Compiled from git commit be62884be227.
So the crash happens somewhere in record_buttons() in irrecord.c.
I suspect the curr_driver methods are not getting initialized...
sometimes. 


Thanks. That said, if this from a version you compiled yourself, the 
libirrecord.so seems to be the wrong one (i. e., the one installed by 
the package). This will by definition be unstable - the libirrecord ABI 
being part of lirc_private.h is not supposed to be the same between 
versions.



Now as to why I have driver=devinput in lirc_options.conf, it's because
it's the only way I found to get my remote to work. Setting
driver=default driver does not work and if I only provide
devinput.lircd.conf and no Hauppauge_PVR350.conf file then it does not
work either.


I think you really need to take a step back here. The way you use lirc 
does not match the official guidelines in [1], which you probably need 
to check. In short:


  - As mentioned in the irrecord message the way to update 
devinput.lircd.conf is the lirc-make-devinput(1) script, not irrecord.
  - Use mode2(1) to check if you get something sane using the devinput 
and/or the default driver.
  - If you go with the devinput driver, use devinput.lircd.conf. In a 
off-the-shelf setup, just copy it into /etc/lirc/lircd.conf.d. Check 
that /etc/lircd.conf does nothing but include the files in lircd.conf.d.
  - If you go with the default driver, you might very well need  to use 
irrecord with this driver. Copy the generated file into lircd.conf.d.
  - In any case, don't use irrecord with the devinput driver since this 
is not a supported usecase.
  - When running into problems, raise the loglevel and submit the 
console logs.


IMHO, this is now more of a support issue than a bug report.


Cheers!

--alec


[1] http://lirc.org/html/configuration-guide.html

Bug#872375: lirc: irrecord segfaults when recording a button

[Francois Gouget]

On Fri, 15 Sep 2017, Alec Leamas wrote:
[...]
> > As before I get:
> > # irrecord -H devinput -f -d /dev/input/event13 /tmp/foo.conf
> 
> This is actually not sane, and a prominent note is displayed when you start
> irrecord like this [1]

Actually it's the -f option that causes the crash. I can reproduce it 
with:

# irrecord -f /tmp/foo.conf

But it's not 100% reproducible now, particularly when I try to use 
my own libirrecord.so so I have debugging symbols. What I can give you 
though is the irrecord backtrace:

#0  0x in ?? ()
#1  0x7f5644aaae73 in record_buttons ()
   from /usr/lib/x86_64-linux-gnu/libirrecord.so.0
#2  0x802d in get_button_data (opts=0x7fffe2d0, 
state=0x7fffe290, 
btn_state=0x7fffe110) at irrecord-ui.cpp:759
#3  do_record_buttons (state=0x7fffe290, opts=0x7fffe2d0) at 
irrecord-ui.cpp:814
#4  0x6ddf in main (argc=, argv=)
at irrecord-ui.cpp:1179

Compiled from git commit be62884be227.
So the crash happens somewhere in record_buttons() in irrecord.c.
I suspect the curr_driver methods are not getting initialized... 
sometimes.


Now as to why I have driver=devinput in lirc_options.conf, it's because 
it's the only way I found to get my remote to work. Setting 
driver=default driver does not work and if I only provide 
devinput.lircd.conf and no Hauppauge_PVR350.conf file then it does not 
work either. And I don't see any other driver that looks like it matches 
my receiver.

IR receiver  : IR605Q
Universal Remote : Logitech Harmony 650
Emulated Remote  : Hauppauge 350


That said, now I've found the right way to recreate the 
Hauppauge_PVR350.conf file is with this command (I think):

# irrecord -H default /tmp/foo.conf


-- 
Francois Gouget   http://fgouget.free.fr/
 f u kn rd ts, ur wy 2 gky 4 ur wn gd.

Bug#872375: lirc: irrecord segfaults when recording a button

[Alec Leamas]

Hi!

On Tue, 22 Aug 2017 17:06:29 +0200 (CEST) Francois Gouget 
 wrote:

> On Fri, 18 Aug 2017, Gianfranco Costamagna wrote:
>
> > Hello,
> >
> > > Dear Maintainer,
> > >
> > > When trying to create a configuration file for my remote, irrecord
> > > crashes when it gets to the 'Now hold down button Xxx' step:
> > can you please try 0.10.0 from unstable?
>
> I tried with 0.10.0-2 but I'm still getting a segfault. Note that the
> segfault happens *before* I press the button on the remote.
>
> # dpkg -l | grep lirc
> ii liblirc-client0:amd64 0.10.0-2 amd64 infra-red remote control 
support - client library
> ii liblirc0:amd64 0.10.0-2 amd64 Infra-red remote control support - 
Run-time libraries
> ii lirc 0.10.0-2 amd64 Infra-red remote control support - daemons and 
utils

>
> As before I get:
> # irrecord -H devinput -f -d /dev/input/event13 /tmp/foo.conf

This is actually not sane, and a prominent note is displayed when you 
start irrecord like this [1]


> [...]
> Hold down an arbitrary key
> 


> Found gap (115997 us)
>
> Please enter the name for the next button (press  to finish
> recording)
> KEY_1
>
> Now hold down button "KEY_1".
> Segmentation fault

A segmentation fault is obviously bad. However, I cannot reproduce this. 
For me the gap detection sort of works after a looong time, and the 
recording actually results in a file.  Havn't bothered to test it, this 
is just not the correct way to create a config file for the devinput driver.


I'm leaning to close this as not-a-bug. After all, the usecase is not 
supported. On top of that, the crash is not reproducible.


Perhaps the message [1] should be augmented with info on possible 
crashes, dunno


Cheers!
--alec


[1]
Usually you should not create a new config file for devinput
devices. LIRC is installed with a devinput.lircd.conf file which
is built for the current system which works with all remotes
supported by the kernel. There might be a need to update
this file so it matches the current kernel. For this, use the
lirc-make-devinput(1) script.

Press RETURN to continue.

Bug#872375: lirc: irrecord segfaults when recording a button

[Francois Gouget]

On Fri, 18 Aug 2017, Gianfranco Costamagna wrote:

> Hello,
> 
> > Dear Maintainer,
> > 
> > When trying to create a configuration file for my remote, irrecord
> > crashes when it gets to the 'Now hold down button Xxx' step:
> can you please try 0.10.0 from unstable?

I tried with 0.10.0-2 but I'm still getting a segfault. Note that the 
segfault happens *before* I press the button on the remote.

# dpkg -l | grep lirc
ii  liblirc-client0:amd64 0.10.0-2  
  amd64infra-red remote control support - client library
ii  liblirc0:amd640.10.0-2  
  amd64Infra-red remote control support - Run-time libraries
ii  lirc  0.10.0-2  
  amd64Infra-red remote control support - daemons and utils

As before I get:
# irrecord -H devinput -f -d /dev/input/event13 /tmp/foo.conf  
[...]
Hold down an arbitrary key

Found gap (115997 us)

Please enter the name for the next button (press  to finish 
recording)
KEY_1

Now hold down button "KEY_1".
Segmentation fault


Running that command in gdb does not give me much more data due to the 
lack of debugging symbols:

Program received signal SIGSEGV, Segmentation fault.
0x in ?? ()
(gdb) bt
#0  0x in ?? ()
#1  0x779afe73 in record_buttons ()
   from /usr/lib/x86_64-linux-gnu/libirrecord.so.0
#2  0x814d in ?? ()
#3  0x6e4c in ?? ()
#4  0x7762b2b1 in __libc_start_main (main=0x64c0, argc=7, 
argv=0x7fffe748, init=, fini=, 
rtld_fini=, stack_end=0x7fffe738) at 
../csu/libc-start.c:291
#5  0x74da in ?? ()



-- 
Francois Gouget   http://fgouget.free.fr/
  Good judgment comes from experience, and experience comes from bad judgment
   -- Barry LePatner

Bug#872375: lirc: irrecord segfaults when recording a button

[Alec Leamas]

On Fri, 18 Aug 2017 22:38:15 +0200 Gianfranco Costamagna 
 wrote:

> Hello,
>
> > Dear Maintainer,
> >
> > When trying to create a configuration file for my remote, irrecord
> > crashes when it gets to the 'Now hold down button Xxx' step:
> can you please try 0.10.0 from unstable?
>
> thanks


The usage here actually does not look  sane. We have

> # irrecord -H devinput -f -d /dev/input/event13 /tmp/tmp.conf

According to the sources, when presented with the devinput driver 
irrecord displays a message like this:


"Usually you should not create a new config file for devinput\n"
"devices. LIRC is installed with a devinput.lircd.conf file 
which \n"

"is built for the current system which works with all remotes \n"
"supported by the kernel. There might be a need to update \n"
"this file so it matches the current kernel. For this, use the \n"
"lirc-make-devinput(1) script.";

Perhaps you missed this message, irrecord does indeed spit out a lot of 
text. That said, this usecase is not really supported.


OTOH, a segfault is never acceptable. Filed upstream bug [1]


Cheers!

--alec

[1] https://sourceforge.net/p/lirc/tickets/299/

Bug#872375: lirc: irrecord segfaults when recording a button

[Gianfranco Costamagna]

Hello,

> Dear Maintainer,
> 
> When trying to create a configuration file for my remote, irrecord
> crashes when it gets to the 'Now hold down button Xxx' step:
can you please try 0.10.0 from unstable?

thanks



signature.asc
Description: OpenPGP digital signature

Bug#872375: lirc: irrecord segfaults when recording a button

[Francois Gouget]

Package: lirc
Version: 0.9.4c-9
Severity: important

Dear Maintainer,

When trying to create a configuration file for my remote, irrecord
crashes when it gets to the 'Now hold down button Xxx' step:

# /etc/init.d/lircd stop
[] Stopping lircd (via systemctl): lircd.serviceWarning: Stopping
lircd.service, but it can still be activated by:
lircd.socket
. ok 

# cat /sys/class/rc/rc0/protocols
[rc-5] nec rc-6 jvc sony rc-5-sz sanyo sharp mce_kbd xmp [lirc]

# cat /sys/class/rc/rc0/input14/name 
Media Center Ed. eHome Infrared Remote Transceiver (147a:e03e)

# ll -d /sys/class/rc/rc0/input14/event13
drwxr-xr-x 3 root root 0 août  15 20:07 /sys/class/rc/rc0/input14/event13

# irrecord -H devinput -f -d /dev/input/event13 /tmp/tmp.conf
[...]
Checking for ambient light  creating too much disturbances.
Please don't press any buttons, just wait a few seconds...

No significant noise (received 0 bytes)

Enter name of remote (only ascii, no spaces) :Test
Using Test.lircd.conf as output filename

Hold down an arbitrary key

Found gap (227049 us)

Please enter the name for the next button (press  to finish
recording)
KEY_1

Now hold down button "KEY_1".
Segmentation fault


This was working fine in the Debian Jessie package, lirc 0.9.0~pre1-1.2.


-- System Information:
Debian Release: buster/sid
  APT prefers testing
  APT policy: (500, 'testing'), (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.11.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=fr_FR.utf8, LC_CTYPE=fr_FR.utf8 (charmap=UTF-8), 
LANGUAGE=fr_FR.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages lirc depends on:
ii  init-system-helpers  1.49
ii  libasound2   1.1.3-5
ii  libc62.24-12
ii  libftdi1-2   1.3-2+b3
ii  libgcc1  1:7.1.0-10
ii  liblirc-client0  0.9.4c-9
ii  liblirc0 0.9.4c-9
ii  libportaudio219.6.0-1
ii  libstdc++6   7.1.0-10
ii  libsystemd0  234-2
ii  libudev1 234-2
ii  libusb-0.1-4 2:0.1.12-31
ii  libusb-1.0-0 2:1.0.21-2
ii  lsb-base 9.20161125
ii  python3  3.5.3-3

Versions of packages lirc recommends:
pn  gir1.2-vte
ii  python3-gi3.22.0-2+b1
ii  python3-yaml  3.12-1+b1

Versions of packages lirc suggests:
pn  ir-keytable  
pn  lirc-compat-remotes  
pn  lirc-doc 
pn  lirc-drv-irman   
pn  lirc-x   
pn  setserial

-- Configuration Files:
/etc/lirc/lircd.conf.d/devinput.lircd.conf [Errno 2] Aucun fichier ou dossier 
de ce type: '/etc/lirc/lircd.conf.d/devinput.lircd.conf'

-- no debconf information