Hmm. I'm not sure how to reproduce this, because I was just moving
around between the dives, checking out how the globe moved, and
occasionally pressing the dive site quick-edit button.
And suddenly I get
*** Error in `/home/torvalds/src/install-root/bin/subsurface':
malloc(): memory corruption (
On Sat, Jul 11, 2015 at 7:00 PM, Linus Torvalds <
torva...@linux-foundation.org> wrote:
> Hmm. I'm not sure how to reproduce this, because I was just moving
> around between the dives, checking out how the globe moved, and
> occasionally pressing the dive site quick-edit button.
>
> And suddenly I
Compile with no Facebook, no user manual which stops the crash. You still get
two bazillion errors from marble which mostly seem bogus to me but occasionally
one can spot an actual bug in the midst of all that crap.
Or compile with no marble as well to see if you can make something interesting
s
On Sat, Jul 11, 2015 at 3:45 PM, Dirk Hohndel wrote:
> Compile with no Facebook, no user manual which stops the crash. You still get
> two bazillion errors from marble which mostly seem bogus to me but
> occasionally one can spot an actual bug in the midst of all that crap.
> Or compile with no
On Sat, Jul 11, 2015 at 4:29 PM, Linus Torvalds
wrote:
>
> Anyway, that makes things runnable with valgrind, although very slow.
> I'm not seeing anything suspicious. I'll play with it some more.
.. and immediately after sending that, I played some more with the
quick-edit button (but no actual e
On Saturday 11 July 2015 19:28:42 Tomaz Canabrava wrote:
> On Sat, Jul 11, 2015 at 7:00 PM, Linus Torvalds <
>
> torva...@linux-foundation.org> wrote:
> > #0 0x7fffefa67a98 in raise () at /lib64/libc.so.6
> > #1 0x7fffefa6972a in abort () at /lib64/libc.so.6
> > #2 0x7fffefaaaea2 in
On Sat, Jul 11, 2015 at 4:39 PM, Linus Torvalds
wrote:
>
> which is just the taxonomy.category writes. Looks like it writes past
> the end of the allocation.
Ok, looking at this, the problem seems to be pretty clear.
When the code does the taxonomy lookup in
ReverseGeoLookupThread::run(), it kee
On Mon, Jul 13, 2015 at 3:46 PM, Linus Torvalds <
torva...@linux-foundation.org> wrote:
> On Sat, Jul 11, 2015 at 4:39 PM, Linus Torvalds
> wrote:
> >
> > which is just the taxonomy.category writes. Looks like it writes past
> > the end of the allocation.
>
> Ok, looking at this, the problem seem
On Mon, Jul 13, 2015 at 11:46:28AM -0700, Linus Torvalds wrote:
> On Sat, Jul 11, 2015 at 4:39 PM, Linus Torvalds
> wrote:
> >
> > which is just the taxonomy.category writes. Looks like it writes past
> > the end of the allocation.
>
> Ok, looking at this, the problem seems to be pretty clear.
>
On Mon, Jul 13, 2015 at 1:10 PM, Dirk Hohndel wrote:
>
> I still don't see how this can overflow. If you put a printf or qDebug()
> or something at this line
>
> ds->taxonomy.nr = ri;
You missed the part where I pointed out that that store is conditional.
It's inside the
if (geo
On Mon, Jul 13, 2015 at 01:27:54PM -0700, Linus Torvalds wrote:
> On Mon, Jul 13, 2015 at 1:10 PM, Dirk Hohndel wrote:
> >
> > I still don't see how this can overflow. If you put a printf or qDebug()
> > or something at this line
> >
> > ds->taxonomy.nr = ri;
>
> You missed the part where
On Mon, Jul 13, 2015 at 1:39 PM, Dirk Hohndel wrote:
>
> The idea is that the taxomony should always be consistent. NULL pointer
> and nothing there, or .nr in sync with the number of valid pointers.
Yes, yes. But if that initialization never happens, then you may have
several *old* taxonomy valu
On Mon, Jul 13, 2015 at 1:58 PM, Linus Torvalds
wrote:
>
> The problem is that "nr" is not necessarily 0 or 1, because we may
> have *previous* taxonomy data, and we keep potentially growing it.
So I'm testing the attached patch, but so far I haven't seen anything.
As mentioned, I'm not sure wha
On Mon, Jul 13, 2015 at 01:58:25PM -0700, Linus Torvalds wrote:
> On Mon, Jul 13, 2015 at 1:39 PM, Dirk Hohndel wrote:
> >
> > The idea is that the taxomony should always be consistent. NULL pointer
> > and nothing there, or .nr in sync with the number of valid pointers.
>
> Yes, yes. But if that
On Mon, Jul 13, 2015 at 2:37 PM, Linus Torvalds
wrote:
>
> I'm wondering if maybe your "Fix memory handling for taxonomy data"
> commit fixed it. I ended up recompiling due to the debug patch, so I'm
> now running a different version of subsurface than I was when I saw it
> originally.
That said,
On Mon, Jul 13, 2015 at 2:25 PM, Linus Torvalds
wrote:
>
> So I'm testing the attached patch, but so far I haven't seen anything.
>
> As mentioned, I'm not sure what triggers it.
I've been clicking on the dive edit thing and the dive list like a
deranged monkey on amphetamines for several minutes
On Mon, Jul 13, 2015 at 02:40:03PM -0700, Linus Torvalds wrote:
> On Mon, Jul 13, 2015 at 2:37 PM, Linus Torvalds
> wrote:
> >
> > I'm wondering if maybe your "Fix memory handling for taxonomy data"
> > commit fixed it. I ended up recompiling due to the debug patch, so I'm
> > now running a differ
On Mon, Jul 13, 2015 at 02:51:24PM -0700, Dirk Hohndel wrote:
> That said, I would appreciate a code review of that patch once it's
> pushed.
I went out of my way to try to capture all corner cases here.
Pushed.
/D
___
subsurface mailing list
subsurface
18 matches
Mail list logo
