On 06/01/2012 11:53 AM, Darrell Anderson wrote:

...

I'm looking for breakage with a patch set for bug report 892. In that testing I tried to play twin4, a Connect Four type game.

The game is broken here.

All I'm asking is for somebody who is using GIT packages to test whether the game works.

Select one of the arrows in the top row to insert a disk in a column. That's all. :-)

The game is listed as kwin4 in the TDE menu (Games/Board games). I have a patch for those branding issues.

Anybody?

The game plays fine in 3.5.13 (named kwin4). The breakage occured thereafter. I only want confirmation of the breakage.

Thanks.

Darrell

Darrell,

Summer has started with a BANG! Shipping 13 yo. son off to camp tomorrow. After I return, I'll build and test and see if twin4 is behaving and report back. Any word on the kwrite/kate gcc 47 bug yet? I'm basically in a holding pattern. I'm also trying to determine what arch moving to systemd is going to do to trinity on Arch. I'll ask under a separate post....

On 06/03/2012 09:07 AM, Darrell Anderson wrote:

...

After I return, I'll build and test and see if twin4 is behaving and report back.

Thanks!

...

Any word on the kwrite/kate gcc 47 bug yet? I'm basically in a holding pattern. I'm also trying to determine what arch moving to systemd is going to do to trinity on Arch. I'll ask under a separate post....

I haven't tested in several weeks. :-(

Is this the same bug?:

http://bugs.pearsoncomputing.net/show_bug.cgi?id=979

Darrell

That's the one - - added the following:

OK, if this is a Not Our Problem issue -- what next? If you build on gcc 4.7 - then you have no TDE. That's something we need to get elevated quickly to whoever we think needs to look at it.

I'm dead in the water building on Arch until this is resolved. If I need to file it with gcc, then I need to know what code is causing it. I've filed gcc bugs before and they are going to need an example that is short-concise-and-to-the-point. Literally they deal is 'a+b=c' primitives and right now I don't know anything other than something in tdelibs is getting built into a runaway loop.

How is this going to get fixed?

I can start the game, and File->New Game, but that is about it. Then the game just sits there. If I misconfigure the game (i.e. player 2 plays with mouse, player 1 plays with computer), then when I (player 1) click on the game -- I don't get all the messages like "Hold your horses.." "The other player hasn't gone" "etc..."

It kind of seems like the player 1/player 2 roles are backwards in the game. Like somebody started a loop at 1 instead of 0 :p

Okay. Thanks. Exactly the same here. Would you please add a confirmation comment to bug report 1014?

http://bugs.pearsoncomputing.net/show_bug.cgi?id=1014

I also have all packages built and current now - so I can test again. (kwrite still craters, but the remainder seems to work well)

The GNU gcc web site indicates gcc 4.7.1 is frozen and in release candidate mode:

http://gcc.gnu.org/ http://gcc.gnu.org/ml/gcc/2012-06/msg00060.html

I presume that the 4.7.1 release will contain the fixes from Fedora and hopefully resolves the kate/kwrite mess.

If you don't want to wait for 4.7.1 then looks like you'll have to sift through the Arch and Fedora gcc 4.7.0-x patches to identify differences. Ugh. But Francois' report makes this sounds like a gcc issue. Slackware rarely updates packages in such a fashion. I have to wait until Slackware updates gcc, which is at version 4.7.0 original. I'm holding my breath that 4.7.1 is released before Slackware Current goes into release candidate status....

Darrell