Re: [Nano-devel] solving a history bug in nano [patch]

[Benno Schulenberg]

On Tue, May 31, 2016, at 14:42, Tito wrote:
> On 05/31/2016 10:20 AM, Benno Schulenberg wrote:
> > Tito, did you try using ((~(size_t)0)>>1) instead of (size_t)-1?
> 
> Yes works.

Okay, then we have a fallback.

> > Did you try running nano with gdb?  I would still like to know why and
> > where it segfaults for you.
> 
> Program received signal SIGABRT, Aborted.

Oh...  But an Abort is not the same as a Segmentation fault.
When you run nano outside of gdb, you get the same error
message?  "Aborted"?

> 0x400c6548 in raise () from /lib/libc.so.6
> (gdb) backtrace
> #0  0x400c6548 in raise () from /lib/libc.so.6
> #1  0x400c7b6c in abort () from /lib/libc.so.6
> #2  0x400fe344 in __libc_message () from /lib/libc.so.6
> #3  0x4010423c in malloc_printerr () from /lib/libc.so.6
> #4  0x401043b8 in malloc_consolidate () from /lib/libc.so.6
> #5  0x40106218 in _int_malloc () from /lib/libc.so.6
> #6  0x40107878 in malloc () from /lib/libc.so.6
> #7  0x00033d30 in nmalloc (howmuch=714) at utils.c:388
> #8  0x0003847c in statusbar (msg=0x1 <Address 0x1 out of bounds>) at 
> winio.c:2126
> #9  0x00028e9c in findnextstr (whole_word_only=false, begin=0xc91a0, 
> begin_x=0, needle=0x4e048 "prova", needle_len=0x0)  at search.c:364

Strange.  Why would it go wrong in statusbar()?  There are only two calls
of statusbar() in findnextstr() -- I am looking at 2.5.3 now.  The first
shouldn't happen, because you don't cancel the search, so it must be the
second.  Anyway -- can you comment out both those calls (in search.c, line
290 and 364) and try the run in gdb again?

Can you also paste the output of 'locale' and 'stty size'?
And can you try a run with 'LC_ALL=C LANGUAGE=en src/nano'?

Benno

-- 
http://www.fastmail.com - Same, same, but different...