Hmm well we should be including util.h via compat.h. Although I notice
we only do so if the platform has forkpty... maybe that is a problem,
try this:
diff --git a/compat.h b/compat.h
index e06f104d..b83e6189 100644
--- a/compat.h
+++ b/compat.h
@@ -101,17 +101,17 @@ void warnx(const char *, ...);
#include <paths.h>
#endif
-#ifdef HAVE_FORKPTY
#ifdef HAVE_LIBUTIL_H
#include <libutil.h>
#endif
+
#ifdef HAVE_PTY_H
#include <pty.h>
#endif
+
#ifdef HAVE_UTIL_H
#include <util.h>
#endif
-#endif
#ifdef HAVE_VIS
#include <vis.h>
Or you could have a look at util.h and see if fparseln is under some
stupid #ifdef?
On Sat, Oct 07, 2017 at 10:10:09AM -0500, Jason White wrote:
> Some additional information, and another mystery:
> While compiling tmux-2.6/cfg.c I get these warnings:
> cfg.c:126:16: warning: implicit declaration of function 'fparseln' is
> invalid in
> C99 [-Wimplicit-function-declaration]
> while ((buf = fparseln(f, NULL, &line, delim, 0)) != NULL) {
> ^
> cfg.c:126:14: warning: incompatible integer to pointer conversion
> assigning to
> 'char *' from 'int' [-Wint-conversion]
> while ((buf = fparseln(f, NULL, &line, delim, 0)) != NULL) {
> ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Regarding the implicit declaration warning, I noticed that the macOS man
> page for fparseln() indicates that one needs to include util.h, which is
> absent from cfg.c, so I added it. Odd thing is that the warning did not
> go away.
> However, if I take out including util.h and add the function prototype
> from util.h then that eliminates both warnings and tmux does not segfault
> when starting a server.
> char *fparseln(FILE *, size_t *, size_t *, const char[3], int);
> -Jason
>
> On Oct 5, 2017, at 3:03 PM, Jason White <[email protected]> wrote:
> That does resolve the issue for me on 2.6. I'm not sure what, precisely,
> changed or when, but:
> On macOS 10.11.6/Xcode 7.3.1 I do not have this issue with 2.5.
> On macOS 10.12.6/Xcode 9 it doesn't work out of the box with at least
> 2.5 or 2.6.
> Both environments have fparseln as part of libc.
> FPARSELN(3) BSD Library Functions Manual
> FPARSELN(3)
> NAME
> fparseln -- return the next logical line from a stream
> LIBRARY
> Standard C Library (libc, -lc)
> SYNOPSIS
> #include <stdio.h>
> #include <util.h>
> char *
> fparseln(FILE *stream, size_t *len, size_t *lineno, const char
> delim[3], int flags);
> -Jason
>
> On Oct 5, 2017, at 11:45 AM, Nicholas Marriott
> <[email protected]> wrote:
>
> This seems most likely to be a bug in fparseln on your platform, does
> it
> work if you do this (you will need to run sh autogen.sh again before
> configure):
>
> diff --git a/configure.ac b/configure.ac
> index 83c104c3..15820a14 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -457,6 +457,7 @@ fi
>
> # Look for fparseln in libutil.
> AC_SEARCH_LIBS(fparseln, util, found_fparseln=yes, found_fparseln=no)
> +found_fparseln=no
> if test "x$found_fparseln" = xyes; then
> AC_DEFINE(HAVE_FPARSELN)
> else
>
> On Thu, Oct 05, 2017 at 10:48:54AM -0500, Jason White wrote:
>
> I've had this issue on Darwin for a while now, but only recently
> started
> digging in to it.
> My environment: macOS 10.12.6, Xcode 9, libevent-2.1.8, tmux-2.6
> (same
> issue with 2.5).
> Issue: if I start tmux (as root or otherwise), I get a "lost
> server"
> message.
> If I start with "tmux -f /dev/null" it stars fine. When I "source
> ~/.tmux.conf", then "lost server".
> If I create a file with a single hash "#" character in it and use
> that for
> my config file it also crashes with "lost server", so server falls
> over
> reading any config file.
> Client/server debug logs:
> # cat tmux-server-36281.log
> 1507217486.996408 server started (36281): version 2.6, socket
> /private/tmp/tmux-0/default, protocol 8
> 1507217486.996427 on Darwin 16.7.0 Darwin Kernel Version 16.7.0:
> Thu Jun
> 15 17:36:27 PDT 2017; root:xnu-3789.70.16~2/RELEASE_X86_64;
> libevent
> 2.1.8-stable (select)
> 1507217486.997625 add peer 0x7fee9781e200: 8 (0x7fee9781d800)
> 1507217486.997642 new client 0x7fee9781d800
> 1507217486.997683 loading /etc/tmux.conf
> 1507217486.997697 loading /Users/jdwhite/.tmux.conf
> # cat tmux-client-36279.log
> 1507217486.994835 client started (36279): version 2.6, socket
> /private/tmp/tmux-0/default, protocol 8
> 1507217486.994860 on Darwin 16.7.0 Darwin Kernel Version 16.7.0:
> Thu Jun
> 15 17:36:27 PDT 2017; root:xnu-3789.70.16~2/RELEASE_X86_64;
> libevent
> 2.1.8-stable (select)
> 1507217486.994887 socket is /private/tmp/tmux-0/default
> 1507217486.994899 trying connect
> 1507217486.994915 connect failed: Connection refused
> 1507217486.995191 lock file is /private/tmp/tmux-0/default.lock
> 1507217486.995310 flock succeeded
> 1507217486.995314 got lock (6)
> 1507217486.995318 trying connect
> 1507217486.995323 connect failed: Connection refused
> 1507217486.995606 add peer 0x7fee98001800: 7 (0x0)
> 1507217486.996815 sending message 100 to peer 0x7fee98001800 (4
> bytes)
> 1507217486.996827 sending message 101 to peer 0x7fee98001800 (12
> bytes)
> 1507217486.996830 sending message 102 to peer 0x7fee98001800 (13
> bytes)
> 1507217486.996833 sending message 108 to peer 0x7fee98001800 (18
> bytes)
> 1507217486.996841 sending message 104 to peer 0x7fee98001800 (0
> bytes)
> 1507217486.996843 sending message 107 to peer 0x7fee98001800 (4
> bytes)
> 1507217486.996846 sending message 105 to peer 0x7fee98001800 (24
> bytes)
> 1507217486.996849 sending message 105 to peer 0x7fee98001800 (16
> bytes)
> 1507217486.996851 sending message 105 to peer 0x7fee98001800 (17
> bytes)
> 1507217486.996854 sending message 105 to peer 0x7fee98001800 (16
> bytes)
> 1507217486.996856 sending message 105 to peer 0x7fee98001800 (41
> bytes)
> 1507217486.996859 sending message 105 to peer 0x7fee98001800 (38
> bytes)
> 1507217486.996861 sending message 105 to peer 0x7fee98001800 (16
> bytes)
> 1507217486.996864 sending message 105 to peer 0x7fee98001800 (19
> bytes)
> 1507217486.996867 sending message 105 to peer 0x7fee98001800 (14
> bytes)
> 1507217486.996869 sending message 105 to peer 0x7fee98001800 (10
> bytes)
> 1507217486.996872 sending message 105 to peer 0x7fee98001800 (18
> bytes)
> 1507217486.996874 sending message 105 to peer 0x7fee98001800 (13
> bytes)
> 1507217486.996877 sending message 105 to peer 0x7fee98001800 (66
> bytes)
> 1507217486.996879 sending message 105 to peer 0x7fee98001800 (38
> bytes)
> 1507217486.996882 sending message 105 to peer 0x7fee98001800 (14
> bytes)
> 1507217486.996885 sending message 105 to peer 0x7fee98001800 (11
> bytes)
> 1507217486.996887 sending message 105 to peer 0x7fee98001800 (13
> bytes)
> 1507217486.996890 sending message 105 to peer 0x7fee98001800 (8
> bytes)
> 1507217486.996892 sending message 105 to peer 0x7fee98001800 (20
> bytes)
> 1507217486.996895 sending message 105 to peer 0x7fee98001800 (259
> bytes)
> 1507217486.996898 sending message 105 to peer 0x7fee98001800 (13
> bytes)
> 1507217486.996900 sending message 105 to peer 0x7fee98001800 (22
> bytes)
> 1507217486.996903 sending message 105 to peer 0x7fee98001800 (12
> bytes)
> 1507217486.996905 sending message 105 to peer 0x7fee98001800 (17
> bytes)
> 1507217486.996908 sending message 105 to peer 0x7fee98001800 (30
> bytes)
> 1507217486.996911 sending message 105 to peer 0x7fee98001800 (193
> bytes)
> 1507217486.996913 sending message 105 to peer 0x7fee98001800 (23
> bytes)
> 1507217486.996916 sending message 105 to peer 0x7fee98001800 (8
> bytes)
> 1507217486.996918 sending message 105 to peer 0x7fee98001800 (23
> bytes)
> 1507217486.996921 sending message 105 to peer 0x7fee98001800 (15
> bytes)
> 1507217486.996923 sending message 105 to peer 0x7fee98001800 (20
> bytes)
> 1507217486.996926 sending message 105 to peer 0x7fee98001800 (26
> bytes)
> 1507217486.996929 sending message 105 to peer 0x7fee98001800 (41
> bytes)
> 1507217486.996931 sending message 105 to peer 0x7fee98001800 (122
> bytes)
> 1507217486.996934 sending message 105 to peer 0x7fee98001800 (13
> bytes)
> 1507217486.996936 sending message 105 to peer 0x7fee98001800 (12
> bytes)
> 1507217486.996939 sending message 105 to peer 0x7fee98001800 (30
> bytes)
> 1507217486.996941 sending message 105 to peer 0x7fee98001800 (13
> bytes)
> 1507217486.996944 sending message 105 to peer 0x7fee98001800 (12
> bytes)
> 1507217486.996946 sending message 105 to peer 0x7fee98001800 (20
> bytes)
> 1507217486.996949 sending message 105 to peer 0x7fee98001800 (9
> bytes)
> 1507217486.996952 sending message 105 to peer 0x7fee98001800 (17
> bytes)
> 1507217486.996961 sending message 105 to peer 0x7fee98001800 (15
> bytes)
> 1507217486.996964 sending message 106 to peer 0x7fee98001800 (0
> bytes)
> 1507217486.996968 sending message 200 to peer 0x7fee98001800 (4
> bytes)
> 1507217486.996971 client loop enter
> 1507217489.162538 client loop exit
> Debugger backtrace:
> # lldb -c /cores/core.36281 ./tmux
> (lldb) target create "./tmux" --core "/cores/core.36281"
> warning: (x86_64) /cores/core.36281 load command 346 LC_SEGMENT_64
> has a
> fileoff + filesize (0x2cf6a000) that extends beyond the end of the
> file
> (0x2cf69000), the segment will be truncated to match
> warning: (x86_64) /cores/core.36281 load command 347 LC_SEGMENT_64
> has a
> fileoff (0x2cf6a000) that extends beyond the end of the file
> (0x2cf69000),
> ignoring this section
> Core file '/cores/core.36281' (x86_64) was loaded.
> (lldb) bt
> * thread #1, stop reason = signal SIGSTOP
> * frame #0: 0x00007fffcc638b52 libsystem_c.dylib`strlen + 18
> frame #1: 0x00007fffcc67e9fc libsystem_c.dylib`__vfprintf +
> 5701
> frame #2: 0x00007fffcc6a7423 libsystem_c.dylib`__v2printf +
> 699
> frame #3: 0x00007fffcc67cf46 libsystem_c.dylib`_vasprintf +
> 554
> frame #4: 0x0000000100bac587 tmux`log_vwrite + 39
> frame #5: 0x0000000100bac540 tmux`log_debug + 144
> frame #6: 0x0000000100b89097 tmux`load_cfg + 439
> frame #7: 0x0000000100b88e94 tmux`start_cfg + 164
> frame #8: 0x0000000100bbce62 tmux`server_start + 450
> frame #9: 0x0000000100b89998 tmux`client_main + 1080
> frame #10: 0x0000000100bc52d2 tmux`main + 1874
> frame #11: 0x00007fffcc602235 libdyld.dylib`start + 1
> frame #12: 0x00007fffcc602235 libdyld.dylib`start + 1
> --
> Jason White
> [email protected]
> "The single biggest problem in communication is the illusion that
> it has
> taken place." - George Bernard Shaw
>
> --
> You received this message because you are subscribed to the Google
> Groups
> "tmux-users" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an
> email to [email protected].
> To post to this group, send email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
>
> --
> Jason White
> [email protected]
> "The single biggest problem in communication is the illusion that it has
> taken place." - George Bernard Shaw
>
> --
> Jason White
> [email protected]
> "The single biggest problem in communication is the illusion that it has
> taken place." - George Bernard Shaw
--
You received this message because you are subscribed to the Google Groups
"tmux-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send an email to [email protected].
For more options, visit https://groups.google.com/d/optout.
