Be warned (Windows user):
I get a crash after some c&p
with floats and ERTs. But I
couldn't reproduce it.
Peter
> "Joost" == Joost Verburg <[EMAIL PROTECTED]> writes:
Joost> Jean-Marc Lasgouttes wrote:
>> For 1.4.4, we will have to get people to prepare test builds much
>> earlier.
Joost> I agree. Because of all the regressions that were caused by the
Joost> wrong bugfix in 1.4.3, I was not able to cre
Jean-Marc Lasgouttes wrote:
For 1.4.4, we will have to get people to prepare test builds much earlier.
I agree. Because of all the regressions that were caused by the wrong
bugfix in 1.4.3, I was not able to create Windows test builds earlier.
Joost
> "Joost" == Joost Verburg <[EMAIL PROTECTED]> writes:
Joost> Jean-Marc Lasgouttes wrote:
>>> "Georg" == Georg Baum <[EMAIL PROTECTED]> writes:
>>
Georg> But if it is OK I'll put it in 1.4.4svn once the version number
Georg> in configure.ac is up.
>> This is done now. So what is the final
Joost Verburg wrote:
> Jean-Marc Lasgouttes wrote:
>>> "Georg" == Georg Baum
>>> <[EMAIL PROTECTED]>
>>> writes:
>>
>> Georg> But if it is OK I'll put it in 1.4.4svn once the version number
>> Georg> in configure.ac is up.
>>
>> This is done now. So what is the final patch? I am lost
Jean-Marc Lasgouttes wrote:
"Georg" == Georg Baum <[EMAIL PROTECTED]> writes:
Georg> But if it is OK I'll put it in 1.4.4svn once the version number
Georg> in configure.ac is up.
This is done now. So what is the final patch? I am lost in the thread.
This one, created by Georg.
Joost
Index:
> "Joost" == Joost Verburg <[EMAIL PROTECTED]> writes:
Joost> Peter Kümmel wrote:
>> It would not be a 1.4.3 release, but without this patch it makes no
>> sense to release on Windows. So we could make it 1.4.3.1 or
>> something like that.
Joost> Windows releases already have an additional nu
> "Georg" == Georg Baum <[EMAIL PROTECTED]> writes:
Georg> But if it is OK I'll put it in 1.4.4svn once the version number
Georg> in configure.ac is up.
This is done now. So what is the final patch? I am lost in the thread.
JMarc
Georg Baum wrote:
> Peter Kümmel wrote:
>
>> It's already a file.
>> JMarc could we commit this patch to 1.4.3?
>
> Not possible :-(
>
Yes 1.4.3 is frozen by the release.
> But if it is OK I'll put it in 1.4.4svn once the version number in
> configure.ac is up.
>
>
> Georg
>
>
--
Peter
Joost Verburg wrote:
> Peter Kümmel wrote:
>> It would not be a 1.4.3 release, but without this patch it makes no sense
>> to release on Windows. So we could make it 1.4.3.1 or something like
>> that.
>
> Windows releases already have an additional number. The first one will
> be called 1.4.3-1.
>
Joost Verburg wrote:
> Georg Baum wrote:
>> Not possible :-(
>>
>> But if it is OK I'll put it in 1.4.4svn once the version number in
>> configure.ac is up.
>
> For the LyX 1.4.3/Windows release I can just get 1.4.3 from SVN and add
> this patch before compiling.
>
> Joost
>
>
Maybe you have f
Peter Kümmel wrote:
It would not be a 1.4.3 release, but without this patch it makes no sense
to release on Windows. So we could make it 1.4.3.1 or something like that.
Windows releases already have an additional number. The first one will
be called 1.4.3-1.
Joost
Joost Verburg wrote:
> Georg Baum wrote:
>> Indeed, the signal is also static. Maybe this patch fixes the problem? I
>> don't like the quitting variable at all, but if it works it would be the
>> most simple solution.
>
> Yes, it works! Jean-Marc, can I release 1.4.3/Windows with this patch?
>
>
Georg Baum wrote:
Not possible :-(
But if it is OK I'll put it in 1.4.4svn once the version number in
configure.ac is up.
For the LyX 1.4.3/Windows release I can just get 1.4.3 from SVN and add
this patch before compiling.
Joost
Georg Baum wrote:
>
> Indeed, the signal is also static. Maybe this patch fixes the problem? I
> don't like the quitting variable at all, but if it works it would be the
> most simple solution.
>
Maybe GCC destructs the signaling first.
We should really have a singleton with all the global stu
Peter Kümmel wrote:
> It's already a file.
> JMarc could we commit this patch to 1.4.3?
Not possible :-(
But if it is OK I'll put it in 1.4.4svn once the version number in
configure.ac is up.
Georg
Peter Kümmel wrote:
> Peter Kümmel wrote:
>> Georg Baum wrote:
>>>
>>>
>>>
>>> Index: src/frontends/Dialogs.C
>>> ===
>>> --- src/frontends/Dialogs.C (Revision 15
Peter Kümmel wrote:
> Georg Baum wrote:
>>
>>
>>
>>
>> Index: src/frontends/Dialogs.C
>> ===
>> --- src/frontends/Dialogs.C (Revision 15111)
>> +++ src/frontends
Georg Baum wrote:
Indeed, the signal is also static. Maybe this patch fixes the problem? I
don't like the quitting variable at all, but if it works it would be the
most simple solution.
Yes, it works! Jean-Marc, can I release 1.4.3/Windows with this patch?
Joost
Jean-Marc Lasgouttes wrote:
>> "Joost" == Joost Verburg <[EMAIL PROTECTED]> writes:
>
> Joost> Jean-Marc Lasgouttes wrote:
>>> Can you make a mingw-based release for now? If we find a simple
>>> enough fix for 1.4.3, you could release an updated version later.
>
> Joost> I have not tested whe
Georg Baum wrote:
> Peter Kümmel wrote:
>
>> disconnection also doesn't help, there is still a signaling:
>>
>>
> lyx-qt3.exe!boost::signals::detail::call_notification::call_notification(const
>> boost::shared_ptr & b={...})
>> Line 158 + 0x42 bytes C++
>>
>> and it crashed while sending the sign
> "Joost" == Joost Verburg <[EMAIL PROTECTED]> writes:
Joost> Jean-Marc Lasgouttes wrote:
>> Can you make a mingw-based release for now? If we find a simple
>> enough fix for 1.4.3, you could release an updated version later.
Joost> I have not tested whether using MinGW hides the problem or n
Jean-Marc Lasgouttes wrote:
Can you make a mingw-based release for now? If we find a simple enough
fix for 1.4.3, you could release an updated version later.
I have not tested whether using MinGW hides the problem or not. However
I do not have the time anyway to switch everything back.
Joost
Peter Kümmel wrote:
But I've found a hack to fix the bug:
we could replace the crash with a memory leak ;)
Index: CutAndPaste.C
===
--- CutAndPaste.C (revision 15109)
+++ CutAndPaste.C (working copy)
@@ -67,7 +67,7 @@
Peter Kümmel wrote:
> disconnection also doesn't help, there is still a signaling:
>
>
lyx-qt3.exe!boost::signals::detail::call_notification::call_notification(const
> boost::shared_ptr & b={...})
> Line 158 + 0x42 bytes C++
>
> and it crashed while sending the signal, could this be blocked?
I
> "Joost" == Joost Verburg <[EMAIL PROTECTED]> writes:
Joost> What are we going to do with this bug? LyX 1.4.3 has been
Joost> released, but I get this crash all the time when using cut &
Joost> paste on Windows.
Joost> Do you want me to create a 1.4.3 Windows release with this bug?
Joost> It
Peter Kümmel wrote:
>
> Doesn't help, but I think the idea is right, but the Dialog is not
> destructed before theCuts, so no disconnect happens, I try it with a
> static function in Dialogs, which I call before calling exit.
>
disconnection also doesn't help, there is still a signaling:
lyx-qt
Peter Kümmel wrote:
> Georg Baum wrote:
>> Index: src/frontends/Dialogs.h
>> ===
>> --- src/frontends/Dialogs.h (Revision 15111)
>> +++ src/frontends/Dialogs.h (Arbeitskopie)
>> @@ -33,6 +33,8 @@ class Dialogs : boost::noncopyable {
Georg Baum wrote:
> Peter Kümmel wrote:
>
>> Georg Baum wrote:
>>> Peter Kümmel wrote:
>>>
I think the reason for the crash is :
lyx-qt3.exe!`anonymous namespace'::`dynamic atexit destructor for
'theCuts''() + 0xd bytes C++ msvcr80d.dll!102020c6()
lyx-qt3.exe!LyXServe
Georg Baum wrote:
>
> Index: src/frontends/Dialogs.C
> ===
> --- src/frontends/Dialogs.C (Revision 15111)
> +++ src/frontends/Dialogs.C (Arbeitskopie)
> @@ -73,6 +73,14 @@ Dialogs::Dialogs(LyXView & lyxview)
> }
>
>
> +Dialog
Georg Baum wrote:
> Peter Kümmel wrote:
>
>> Georg Baum wrote:
>>> Peter Kümmel wrote:
>>>
I think the reason for the crash is :
lyx-qt3.exe!`anonymous namespace'::`dynamic atexit destructor for
'theCuts''() + 0xd bytes C++ msvcr80d.dll!102020c6()
lyx-qt3.exe!LyXServe
Peter Kümmel wrote:
> Georg Baum wrote:
>> Peter Kümmel wrote:
>>
>>> I think the reason for the crash is :
>>>
>>> lyx-qt3.exe!`anonymous namespace'::`dynamic atexit destructor for
>>> 'theCuts''() + 0xd bytes C++ msvcr80d.dll!102020c6()
>>> lyx-qt3.exe!LyXServer::`scalar deleting destructo
Georg Baum wrote:
> Peter Kümmel wrote:
>
>> I've replaced the complete boost tree (.cpp included) with the version
>> from trunk. But it also makes no difference.
>>
>> BTW should we update the 1.4.x tree to boost 1.34?
>
> Not before it is released. We might want to update to 1.33.1, we had tha
Peter Kümmel wrote:
> I've replaced the complete boost tree (.cpp included) with the version
> from trunk. But it also makes no difference.
>
> BTW should we update the 1.4.x tree to boost 1.34?
Not before it is released. We might want to update to 1.33.1, we had that
for some time without probl
The same code in 1.5:
CutAndPAst.C:
namespace {
typedef std::pair PitPosPair;
typedef limited_stack > CutStack;
CutStack theCuts(10);
// store whether the tabular stack is newer than the normal copy stack
// FIXME: this is a workaround for bug 1919. Should be removed for 1.5,
// when we (hope
Joost Verburg wrote:
> Georg Baum wrote:
>>> Looks like it. I get a crash in boost/boost/detail/shared_count.hpp
>>> line 124.
>>
>> Then try to replace the smart pointers with a newer version, e.g. the
>> one from trunk. Does it then work?
>
> I copied boost\boost\detail from trunk. It doesn't ma
Georg Baum wrote:
Looks like it. I get a crash in boost/boost/detail/shared_count.hpp line
124.
Then try to replace the smart pointers with a newer version, e.g. the one
from trunk. Does it then work?
I copied boost\boost\detail from trunk. It doesn't make any difference.
Joost
Am Donnerstag, 21. September 2006 21:02 schrieb Joost Verburg:
> Michael Gerz wrote:
> > Could it be another boost problem?
If that is tzhe case it is not similar to the last ones, since I can't
reprocduce it.
> Looks like it. I get a crash in boost/boost/detail/shared_count.hpp line
> 124.
Th
Michael Gerz wrote:
Could it be another boost problem?
Looks like it. I get a crash in boost/boost/detail/shared_count.hpp line
124.
Joost
Peter Kümmel wrote:
Ahh...
This bug is somehow related to c&p because just deleting
the TOC inset makes no problems.
CTRL-x moves the TOC into a buffer and then on exiting
lyx the destruction fails or happens to often, but I'm
only speculating.
What are we going to do with this bug? LyX 1.4.3
Enrico Forestieri schrieb:
Uhm, I think that it is more correct to say that it crashes when
compiling with MSVC. I have not tried mingw but I would bet that it
will not crash...
Could it be another boost problem?
Michael
Enrico Forestieri wrote:
> On Thu, Sep 21, 2006 at 12:25:08AM +0200, Joost Verburg wrote:
>> Enrico Forestieri wrote:
>>> On Wed, Sep 20, 2006 at 11:32:28PM +0200, Joost Verburg wrote:
I can also reproduce this. LyX crashed a few times last week when I
closed it, but I was never able to
Jean-Marc Lasgouttes wrote:
> Peter> Valgrin on Windows? Does this mean cygwin/LyX?
>
> Sorry, I meant 'on linux'.
Ahh...
This bug is somehow related to c&p because just deleting
the TOC inset makes no problems.
CTRL-x moves the TOC into a buffer and then on exiting
lyx the destruction fails o
On Thu, Sep 21, 2006 at 03:10:22PM +0200, Peter Kümmel wrote:
> Jean-Marc Lasgouttes wrote:
> >> "Peter" == Peter Kümmel <[EMAIL PROTECTED]> writes:
> >
> > Peter> It reduces to removing the TOC inset, maybe something is
> > Peter> destructed twice.
> >
> > Valgrind does not notice anything o
On Thu, Sep 21, 2006 at 12:25:08AM +0200, Joost Verburg wrote:
> Enrico Forestieri wrote:
> >On Wed, Sep 20, 2006 at 11:32:28PM +0200, Joost Verburg wrote:
> >>I can also reproduce this. LyX crashed a few times last week when I
> >>closed it, but I was never able to find the cause.
> >
> >I tried
Enrico Forestieri wrote:
On Wed, Sep 20, 2006 at 11:32:28PM +0200, Joost Verburg wrote:
I can also reproduce this. LyX crashed a few times last week when I
closed it, but I was never able to find the cause.
I tried hard but cannot reproduce this neither on cygwin nor on solaris.
Maybe it cra
> "Peter" == Peter Kümmel <[EMAIL PROTECTED]> writes:
Peter> Valgrin on Windows? Does this mean cygwin/LyX?
Sorry, I meant 'on linux'.
Peter> And currently I use a TT dll, so it's not a Q../Free problem.
OK.
JMarc
Jean-Marc Lasgouttes wrote:
>> "Peter" == Peter Kümmel <[EMAIL PROTECTED]> writes:
>
> Peter> It reduces to removing the TOC inset, maybe something is
> Peter> destructed twice.
>
> Valgrind does not notice anything on windows. It may be a problem with
> Q../Free.
>
>
> JMarc
>
>
Valgrin
> "Peter" == Peter Kümmel <[EMAIL PROTECTED]> writes:
Peter> It reduces to removing the TOC inset, maybe something is
Peter> destructed twice.
Valgrind does not notice anything on windows. It may be a problem with
Q../Free.
JMarc
Joost Verburg wrote:
> Peter Kümmel wrote:
>> Only Joost could reproduce it, maybe a msvc specific
>> problem.
>
> Or Windows specific?
>
>> The backtrace is not very informative:
>
> I'll have a look at it later today. I get the crash quite often, so it
> must be fixed in 1.4.3/Windows.
>
> Jo
Peter Kümmel wrote:
Only Joost could reproduce it, maybe a msvc specific
problem.
Or Windows specific?
The backtrace is not very informative:
I'll have a look at it later today. I get the crash quite often, so it
must be fixed in 1.4.3/Windows.
Joost
Juergen Spitzmueller wrote:
> Peter Kümmel wrote:
>> Next crash:
>>
>> - open Introduction
>> - select much text
>> - cut (CTRL-x)
>> - close lyx
>> - discard -> CRASH
>
> I cannot reproduce (onlinux).
> Do you have a backtrace?
>
> Jürgen
>
>
Only Joost could reproduce it, maybe a msvc specif
Peter Kümmel wrote:
> Next crash:
>
> - open Introduction
> - select much text
> - cut (CTRL-x)
> - close lyx
> - discard -> CRASH
I cannot reproduce (onlinux).
Do you have a backtrace?
Jürgen
On Wed, Sep 20, 2006 at 11:32:28PM +0200, Joost Verburg wrote:
> Peter Kümmel wrote:
> >Next crash:
> >
> >- open Introduction
> >- select much text
> >- cut (CTRL-x)
> >- close lyx
> >- discard -> CRASH
>
> I can also reproduce this. LyX crashed a few times last week when I
> closed it, but I wa
Peter Kümmel wrote:
Next crash:
- open Introduction
- select much text
- cut (CTRL-x)
- close lyx
- discard -> CRASH
I can also reproduce this. LyX crashed a few times last week when I
closed it, but I was never able to find the cause.
Joost
55 matches
Mail list logo
