mailing list of musl libc
 help / color / mirror / code / Atom feed
From: Sebastian Kemper <sebastian_ml@gmx.net>
To: musl@lists.openwall.com
Subject: [musl] freeswitch and musl 1.2.x (time64 most likely)
Date: Thu, 28 Oct 2021 00:16:56 +0200	[thread overview]
Message-ID: <YXnP2OyicBdPLRKM@darth.lan> (raw)

Hi all,

I put OpenWrt with musl 1.2.x on my router. freeswitch was updated a bit
in preparation for time64 libc. This snippet was added:

#if __USE_TIME_BITS64
#define TIME_T_FMT SWITCH_INT64_T_FMT
#else
#define TIME_T_FMT "ld"
#endif

And the few places where during compilation there were new warnings
(time_t related) I replaced some '%ld' with TIME_T_FMT and they
disappeared.

But now I try to start freeswitch on the router and it doesn't work. It
either just exits somehow or it segfaults. The logs on the router show
that freeswitch puts the wrong time stamps. So there must still be some
time64 problem hidden somewhere.

Wed Oct 27 23:44:35 2021 daemon.info freeswitch[10181]: 1970-01-01 00:24:12.-055227 0.00% [NOTICE] switch_loadable_module.c:223 Adding Codec SPEEX 99 Speex 8000hz 20ms 1ch 24600bps
Wed Oct 27 23:44:35 2021 daemon.info freeswitch[10181]: 1970-01-01 00:24:12.-066240 0.00% [CONSOLE] switch_loadable_module.c:2191 open of pre_load_modules.conf failed
Wed Oct 27 23:44:35 2021 daemon.info freeswitch[10181]: 1970-01-01 00:24:12.-066240 0.00% [INFO] switch_core_sqldb.c:3593 Opening DB
Wed Oct 27 23:44:35 2021 daemon.info freeswitch[10181]: 1970-01-01 00:24:12.-054973 0.00% [INFO] switch_core_sqldb.c:1842 CORE Starting SQL thread.
Wed Oct 27 23:44:35 2021 daemon.info freeswitch[10181]: 1970-01-01 00:24:12.-062042 0.00% [CONSOLE] switch_loadable_module.c:1768 Successfully Loaded [mod_logfile]

This is the end of the log. And then freeswitch just disappeared.

I'm not a programmer, so looking at the code is a bit of a hurdle :D I
think that the time basics are set up in src/switch_time.c ([1]) and
src/include/switch_apr.h ([2]). I'm wondering if anybody could maybe
give these files a look and check if you see something obvious? *
(pretty please) * Or maybe give me a hint how to narrow it down? I guess
I should do a backtrace?

Kind regards,
Sebastian

[1] https://github.com/signalwire/freeswitch/blob/master/src/switch_time.c
[2] https://github.com/signalwire/freeswitch/blob/master/src/include/switch_apr.h

             reply	other threads:[~2021-10-27 22:17 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-27 22:16 Sebastian Kemper [this message]
2021-10-27 23:24 ` Érico Nogueira
2021-10-28 20:21 ` [musl] " Sebastian Kemper
2021-10-28 21:34   ` Sebastian Kemper
2021-10-29  3:10     ` Érico Nogueira
2021-10-29 14:40       ` Sebastian Kemper
2021-10-29 21:04         ` Szabolcs Nagy
2021-10-29 22:44           ` Sebastian Kemper
2021-10-30  1:00             ` Rich Felker
2021-10-30 10:57               ` Sebastian Kemper
2021-10-30 17:56                 ` Rich Felker
2021-10-30 19:07                   ` Sebastian Kemper

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=YXnP2OyicBdPLRKM@darth.lan \
    --to=sebastian_ml@gmx.net \
    --cc=musl@lists.openwall.com \
    --subject='Re: [musl] freeswitch and musl 1.2.x (time64 most likely)' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Code repositories for project(s) associated with this inbox:

	https://git.vuxu.org/mirror/musl/

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).