Hi Jim,

Please don't take this personal, the code has been #iffed out not reverted.
I agree with you that we should keep good code, problem is that lately someone introduces a regression with new or modified code (even correct),
and the regression stays there forever, we should try to avoid such a case.
In this specific case we know the gdibatch code introduced that regression, so my suggestion is: do the commit anyway with a specific roadmap to fix the regression or disable it temporary until the regression is fixed.
Perception is really important, what's gonna think about ros someone that tried it before, now after a new release tries it again and sees that what was working before doesn't work anymore or works worse than it did, your references are a good example of this scenario, but see also:

4461 Regression: Notepad Lite doesn't install
4677 Regression: Can't restore minimized windows
4948 Regression: Can't close locale dialog
5276 Regression: shell about dialog doesn't display header bitmap

And there are more...

I'm not gonna blame win32k only here as regressions are not its exclusivity, it's funny that before uniata we were suggesting people to rename uniata to atapi to test in boxes with sata controllers, now we do the oposite (atapi -> uniata )to let people test in pata.
I've mentioned in ros-dev also that ros takes more than 2 minutes in my virtual machine to install when it tooks some seconds before (two hd attached to the primary channel), I'm not able to test ros so frequently now because of this.
I've also mentioned that AC97 stopped working, has regressed for several reasons several revisions away, now I'm not being prompted anymore for its driver as 46998...
We can't keep working like this, testers should work in a coordinated way with developers, both of them are important, there aren't many skilled testers to be honest just take a look at bugzilla...
Sometimes I test and see regressions and don't know what to do anymore, stop testing, tell ros-dev, bug the dev, give up...

Gabriel.



> Date: Thu, 22 Apr 2010 20:47:52 -0500
> From: jimtabor.rosdev@gmail.com
> To: ros-dev@reactos.org
> Subject: [ros-dev] [ros-diffs] [tkreuzer] 46998: Disable gdi batch for SelectObject with fonts. Fixes font regression.
>
> Not surprised!
>
> I guess ReactOS new polices are for hiding the real issues and not
> fixing them and revert all the correct code. Since the kernel rewrites
> and the new order of coding, these oddities have now surfaced. The
> TEB, is inaccessible from kernel mode as in bug 5265 and 5314, and the
> strange processes access issues in bug 5310. This revert silliness
> will result in moving the project back to post windows 95 architecture
> (arwinss). ReactOS is about moving forward and taking chances with
> innovations from learned information then moving away to make it work
> even better. Personally, I hope someone will take up where this left
> off and move on.
>
> Good luck!
>
> Reference:
>
> http://www.reactos.org/bugzilla/show_bug.cgi?id=5265
> http://www.reactos.org/bugzilla/show_bug.cgi?id=5310
> http://www.reactos.org/bugzilla/show_bug.cgi?id=5314
>
> _______________________________________________
> Ros-dev mailing list
> Ros-dev@reactos.org
> http://www.reactos.org/mailman/listinfo/ros-dev

Condividi le tue emozioni e proteggi la tua privacy. Chiacchiera su Messenger