* Re: frysk-core/frysk gui/srcwin/SourceWindow.java ...
[not found] <20061212143305.11268.qmail@sourceware.org>
@ 2006-12-13 13:58 ` Andrew Cagney
2006-12-14 14:13 ` Mike Cvet
0 siblings, 1 reply; 4+ messages in thread
From: Andrew Cagney @ 2006-12-13 13:58 UTC (permalink / raw)
To: frysk
Mike,
is this null stack frames, or null source and line? The source window
should be able to assume that there is at least one stack frame.
Andrew
mcvet@sourceware.org wrote:
> CVSROOT: /cvs/frysk
> Module name: frysk-core
> Changes by: mcvet@sourceware.org 2006-12-12 14:33:05
>
> Modified files:
> frysk/gui/srcwin: SourceWindow.java ChangeLog
> frysk/rt : RunState.java
>
> Log message:
> 2006-12-12 Mike Cvet <mcvet@redhat.com>
>
> * SourceWindow.java (populateStackBrowser): Catch more null
> StackFrames. Assume some stacktrace function names could be null
> even if the objects aren't. Fixes #3696.
>
> Patches:
> http://sourceware.org/cgi-bin/cvsweb.cgi/frysk-gui/frysk/gui/srcwin/SourceWindow.java.diff?cvsroot=frysk&r1=1.153&r2=1.154
> http://sourceware.org/cgi-bin/cvsweb.cgi/frysk-gui/frysk/gui/srcwin/ChangeLog.diff?cvsroot=frysk&r1=1.353&r2=1.354
> http://sourceware.org/cgi-bin/cvsweb.cgi/frysk-core/frysk/rt/RunState.java.diff?cvsroot=frysk&r1=1.10&r2=1.11
>
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: frysk-core/frysk gui/srcwin/SourceWindow.java ...
2006-12-13 13:58 ` frysk-core/frysk gui/srcwin/SourceWindow.java Andrew Cagney
@ 2006-12-14 14:13 ` Mike Cvet
2006-12-14 15:18 ` Andrew Cagney
0 siblings, 1 reply; 4+ messages in thread
From: Mike Cvet @ 2006-12-14 14:13 UTC (permalink / raw)
To: Andrew Cagney; +Cc: frysk
On Wed, 2006-12-13 at 08:58 -0500, Andrew Cagney wrote:
> Mike,
>
> is this null stack frames, or null source and line? The source window
> should be able to assume that there is at least one stack frame.
>
> Andrew
>
Both, actually - its possible for the call to
StackFactory.createStackFrame to break completely and leave the stack
trace null. This commit also covers the situation where the StackFrame
seems to have been initialized properly, but for some reason a function
name can't be pulled out of it. I believe both of these cases have been
witnessed when testing on johntheripper with varying amounts of
optimization and debuginfo/unwindinfo
- Mike
> mcvet@sourceware.org wrote:
> > CVSROOT: /cvs/frysk
> > Module name: frysk-core
> > Changes by: mcvet@sourceware.org 2006-12-12 14:33:05
> >
> > Modified files:
> > frysk/gui/srcwin: SourceWindow.java ChangeLog
> > frysk/rt : RunState.java
> >
> > Log message:
> > 2006-12-12 Mike Cvet <mcvet@redhat.com>
> >
> > * SourceWindow.java (populateStackBrowser): Catch more null
> > StackFrames. Assume some stacktrace function names could be null
> > even if the objects aren't. Fixes #3696.
> >
> > Patches:
> > http://sourceware.org/cgi-bin/cvsweb.cgi/frysk-gui/frysk/gui/srcwin/SourceWindow.java.diff?cvsroot=frysk&r1=1.153&r2=1.154
> > http://sourceware.org/cgi-bin/cvsweb.cgi/frysk-gui/frysk/gui/srcwin/ChangeLog.diff?cvsroot=frysk&r1=1.353&r2=1.354
> > http://sourceware.org/cgi-bin/cvsweb.cgi/frysk-core/frysk/rt/RunState.java.diff?cvsroot=frysk&r1=1.10&r2=1.11
> >
> >
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: frysk-core/frysk gui/srcwin/SourceWindow.java ...
2006-12-14 14:13 ` Mike Cvet
@ 2006-12-14 15:18 ` Andrew Cagney
2006-12-14 16:13 ` Mike Cvet
0 siblings, 1 reply; 4+ messages in thread
From: Andrew Cagney @ 2006-12-14 15:18 UTC (permalink / raw)
To: Mike Cvet; +Cc: frysk
Mike Cvet wrote:
> On Wed, 2006-12-13 at 08:58 -0500, Andrew Cagney wrote:
>
>> Mike,
>>
>> is this null stack frames, or null source and line? The source window
>> should be able to assume that there is at least one stack frame.
>>
>> Andrew
>>
>>
>
> Both, actually - its possible for the call to
> StackFactory.createStackFrame to break completely and leave the stack
> trace null. This commit also covers the situation where the StackFrame
> seems to have been initialized properly, but for some reason a function
> name can't be pulled out of it. I believe both of these cases have been
> witnessed when testing on johntheripper with varying amounts of
> optimization and debuginfo/unwindinfo
>
The source window code doesn't need to handle this - instead it should
rely on there being a frame.
The place to fix this is in .createStackFrame setting things up so that
it always guarentees some sort of stack frame.
Andrew
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: frysk-core/frysk gui/srcwin/SourceWindow.java ...
2006-12-14 15:18 ` Andrew Cagney
@ 2006-12-14 16:13 ` Mike Cvet
0 siblings, 0 replies; 4+ messages in thread
From: Mike Cvet @ 2006-12-14 16:13 UTC (permalink / raw)
To: Andrew Cagney; +Cc: frysk
On Thu, 2006-12-14 at 10:18 -0500, Andrew Cagney wrote:
> Mike Cvet wrote:
> > On Wed, 2006-12-13 at 08:58 -0500, Andrew Cagney wrote:
> >
> >> Mike,
> >>
> >> is this null stack frames, or null source and line? The source window
> >> should be able to assume that there is at least one stack frame.
> >>
> >> Andrew
> >>
> >>
That makes a lot of sense - I'll make the change.
- Mike
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2006-12-14 16:13 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <20061212143305.11268.qmail@sourceware.org>
2006-12-13 13:58 ` frysk-core/frysk gui/srcwin/SourceWindow.java Andrew Cagney
2006-12-14 14:13 ` Mike Cvet
2006-12-14 15:18 ` Andrew Cagney
2006-12-14 16:13 ` Mike Cvet
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 read-only IMAP folder(s) and NNTP newsgroup(s).