On Thursday 14 January 2010, Chris Cannam wrote:
> Well, it's just very peculiar. I don't like not being able to understand
> it.
I don't like not being able to understand it either, but I can confirm that
doing a clean rebuild cures the crash. I re-applied the patch, did a make
clean, and a
On Thu, Jan 14, 2010 at 2:37 PM, Julie S wrote:
> I've attached a .diff file and a gdb backtrace along with a couple lines from
> RG right before the crash.
Mm, very strange.
Have you built from clean since making the change (in whatever working
copy that diff came from)? Shouldn't make any di
Hello Chris,
Concerning the crash:
> Can you get a stack trace?
I've attached a .diff file and a gdb backtrace along with a couple lines from
RG right before the crash.
I didn't like what I saw, but since it was deep inside a Qt Library, but since
RG is threaded that may not mean anything. I'
On Wed, Jan 13, 2010 at 6:32 PM, Julie S wrote:
> I've attached a .diff based off the revision 11636. Bsically it is just two
> lines of code.
>
> This makes RG crash. Change DeviceEventMap to DeviceEventMap * and all is
> well.
Hm, it doesn't crash for me -- I've tried it several times -- and
On Wed, 13 Jan 2010 20:57:45 -0500
"D. Michael McIntyre" wrote:
> On Wednesday 13 January 2010, Julie S wrote:
> > The .diff I submitted was based off of 11636 -- just those two
> > lines in the .h crash RG.
> >
> > I have no code in the .cpp.
>
> That's bizarre. I get a crash about 3/5 of the
With the pointer version, you're not calling a constructor, but with
this version:
typedef std::map > DeviceEventMap;
DeviceEventMap m_pendSysExcMap;
I believe you're calling the map constructor. That
appears to be the only difference. Doesn't seem like it would cause a
core dump, but perhaps
Dear Yves,
You wrote:
> What is the size of this map ?
> Maybe RG is running out of stack space...
This should not be an issue. It is just an empty container. I have not put
anything in it...there is not code written in this test to add and contents to
the map.
That is why I'm so perplexed.
On Wednesday 13 January 2010, Julie S wrote:
> The .diff I submitted was based off of 11636 -- just those two lines in the
> .h crash RG.
>
> I have no code in the .cpp.
That's bizarre. I get a crash about 3/5 of the time myself. The crashes are
all bizarre too.
All you do is declare the thi
Le mercredi 13 janvier 2010 19:32:14, Julie S a écrit :
> This makes RG crash. Change DeviceEventMap to DeviceEventMap * and all is
> well.
What is the size of this map ?
Maybe RG is running out of stack space...
Yves
---
Hello Michael,
Concerning:
> this. How you're using
> the thing might well change that though.
The .diff I submitted was based off of 11636 -- just those two lines in the .h
crash RG.
I have no code in the .cpp.
I submitted a complete .diff.
This is what I did:
I copied my files-in-progress
On Wednesday 13 January 2010, Julie S wrote:
> I've attached a .diff based off the revision 11636. Bsically it is just
> two lines of code.
>
> This makes RG crash. Change DeviceEventMap to DeviceEventMap * and all is
> well.
Just adding a new typedef and declaring a member variable of that t
Hello Chris,
I've attached a .diff based off the revision 11636. Bsically it is just two
lines of code.
This makes RG crash. Change DeviceEventMap to DeviceEventMap * and all is well.
At this point this is more of a learning experience thing. I have code up and
running using the pointer and
On Wed, Jan 13, 2010 at 4:47 PM, Julie S wrote:
> I've been working on pooling the ALSA system exclusive messages while
> recording. To properly manage these I had created a map so I could track
> incomplete messages. Here is what I used in the AlsaDriver.h file (these are
> private members):
Hello All,
I've been working on pooling the ALSA system exclusive messages while
recording. To properly manage these I had created a map so I could track
incomplete messages. Here is what I used in the AlsaDriver.h file (these are
private members):
> typedef std::map std::pa
14 matches
Mail list logo
