0.3.17 testing

[EmuandCo]

Rapps Links are fixed, (I hope) ^^

[Pi_User5]

Has anyone gotten tried updating a video driver from device manager? I tried and I keep getting a message about an unspecified error. I am trying this because I managed to get the video card driver for Virtual PC 2007 and I was going to try it out. Is this a bug or is updating the video driver via Device Manager a work in progress?

Video driver in case anyone wants to try: http://www.mediafire.com/download/bsqaw ... Driver.vfd
I included the Windows 2000 and NT driver. (That's what the INF file said)

[Jedi-to-be]

New RC incorporating offbyone browser fix, keyboard lights fix, and an update for uniata up to 45b1. I'd like to ask people to help check and make sure the new uniata version doesn't break anything.

http://sourceforge.net/projects/reactos ... p/download

There is NO update for uniata up to 45b1 curently 0.3.17 RC4
https://code.reactos.org/browse/reactos ... .h?r=64751

[Z98]

Because I didn't commit it, I wanted verification that it didn't break anything beforehand. And it's starting to look like it does.

[Webunny]

Because I didn't commit it, I wanted verification that it didn't break anything beforehand. And it's starting to look like it does.

Anyone knows when there was made a change concerning (audio)drivers, between build 61357 and RC1? Because, in all honesty, last time I started 'in the wild' with regression testing, it took me 20+ burns on a CD-RW, and more than half a day, just to pinpoint it.

I would like to proceed a bit less haphazardly, this time. If there was some change made in that area, I'll do better to try the one before and after that change.

[Black_Fox]

That makes it 61357-64751 which is 10 months apart. Unless there was a single miraculous commit, I'm afraid nobody will remember that much to be able to tell (commit log for the drivers folder to illustrate). Using binary division you should be able to pinpoint the precise revision using no more than 12 steps. IMO in the r60000+ range you have very low chance to find faulty ISO (that won't boot or whatever), so it shouldn't get higher than the 12 burns - you'll also usually be able to tell which of 4 commits did the breakage, so that leaves you even additional attempt to try the "next" or "previous" revision to be sure within the 12 attempts.
I regresstested 4 or 5 tickets recently on a VM, so I know it's quite annoying, but it's also helpful.

[Webunny]

That makes it 61357-64751 which is 10 months apart. Unless there was a single miraculous commit, I'm afraid nobody will remember that much to be able to tell (commit log for the drivers folder to illustrate). Using binary division you should be able to pinpoint the precise revision using no more than 12 steps. IMO in the r60000+ range you have very low chance to find faulty ISO (that won't boot or whatever), so it shouldn't get higher than the 12 burns - you'll also usually be able to tell which of 4 commits did the breakage, so that leaves you even additional attempt to try the "next" or "previous" revision to be sure within the 12 attempts.
I regresstested 4 or 5 tickets recently on a VM, so I know it's quite annoying, but it's also helpful.

*sigh*

I'm really not looking forward to it...

I'll guess I'll start with 6300...


Come to think of it, why isn't it sufficient if I take the RC4, and make a complete bugreport with my serial cable when I try to install the driver and it doesn't work? That way, devs should have enough info too, no?

[Black_Fox]

I try to burn a CD-R bootcd every 500-1000 revisions so that it's easier to track changes, but I recently also dug out a CD-RW to do "quick" out-of-order checks if needed.

As for the RC4 + serial vs regresstesting, it depends. I had an issue on RealHW and knew it appeared within last 400 revisions, but it was a specific change so it was enough to provide one line of log and the devs knew what was going on. On the other hand, there are other issues whose log may not help that much, and pinpointing the guilty revision made the fix easier. Can't say which one of these your issue will be. If you have some time to invest, finding the guilty revision has never hurt anybody, maybe except in the process

[Webunny]

I try to burn a CD-R bootcd every 500-1000 revisions so that it's easier to track changes, but I recently also dug out a CD-RW to do "quick" out-of-order checks if needed.

As for the RC4 + serial vs regresstesting, it depends. I had an issue on RealHW and knew it appeared within last 400 revisions, but it was a specific change so it was enough to provide one line of log and the devs knew what was going on. On the other hand, there are other issues whose log may not help that much, and pinpointing the guilty revision made the fix easier. Can't say which one of these your issue will be. If you have some time to invest, finding the guilty revision has never hurt anybody, maybe except in the process

Well, that's strange. Even with the 64000 it works.

Is 64751 the very last before the RC1?

Wait...64751 doesn't show anything.

64750, I guess?

[Jedi-to-be]

https://jira.reactos.org/browse/CORE-8667

May it be backported to 0.3.17?

[Webunny]

I try to burn a CD-R bootcd every 500-1000 revisions so that it's easier to track changes, but I recently also dug out a CD-RW to do "quick" out-of-order checks if needed.

As for the RC4 + serial vs regresstesting, it depends. I had an issue on RealHW and knew it appeared within last 400 revisions, but it was a specific change so it was enough to provide one line of log and the devs knew what was going on. On the other hand, there are other issues whose log may not help that much, and pinpointing the guilty revision made the fix easier. Can't say which one of these your issue will be. If you have some time to invest, finding the guilty revision has never hurt anybody, maybe except in the process

Well, that's strange. Even with the 64000 it works.

Is 64751 the very last before the RC1?

Wait...64751 doesn't show anything.

64750, I guess?

Well, there you have it. With 64000 it works fine, with 65000, it doesn't anymore. Range: 64000-65000! And now I have to sleep..

[middings]

I walked the tree /trunk/reactos/drivers and picked revisions that (by my untrained eye) might affect the audio driver. (The file(s) modified column lists only the files modified in the paths that I walked.)

Code: Select all

rev.  file(s) modified
===== ============================================
64795 /trunk/reactos/drivers/wdm/audio/backpln/audio_test/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/drivers/CMIDriver/cmicontrol/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/drivers/CMIDriver/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/drm/drmk/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/filters/kmixer/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/CMakeLists.txt
63247 /trunk/reactos/drivers/wdm/audio/sysaudio/deviface.c
      /trunk/reactos/drivers/wdm/audio/sysaudio/pin.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/control.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/deviface.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/mmixer.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/sup.c
62124 /trunk/reactos/drivers/wdm/audio/drivers/CMIDriver/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/drivers/CMIDriver/guid.cpp
      /trunk/reactos/drivers/wdm/audio/drivers/CMIDriver/precomp.h
62077 /trunk/reactos/drivers/wdm/audio/sysaudio/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/filters/kmixer/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/CMakeLists.txt
62075 /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/CMakeLists.txt
62006 /trunk/reactos/drivers/wdm/audio/sysaudio/deviface.c
      /trunk/reactos/drivers/wdm/audio/sysaudio/sysaudio.h
      /trunk/reactos/drivers/wdm/audio/filters/kmixer/kmixer.h
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/deviface.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/mmixer.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/wdmaud.h
61848 /trunk/reactos/drivers/base/beep/beep.c
61501 /trunk/reactos/drivers/wdm/audio/sysaudio/sysaudio.h
61500 /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/mmixer.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/sup.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/wdmaud.h
61499 /trunk/reactos/drivers/wdm/audio/filters/kmixer/filter.c
      /trunk/reactos/drivers/wdm/audio/filters/kmixer/kmixer.c
      /trunk/reactos/drivers/wdm/audio/filters/kmixer/kmixer.h
      /trunk/reactos/drivers/wdm/audio/filters/kmixer/pin.c
61498 /trunk/reactos/drivers/wdm/audio/backpln/portcls/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/adapter.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/api.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/connection.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/dispatcher.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/dll.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/dma_slave.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/drm.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/drm_port.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/filter_dmus.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/filter_topology.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/filter_wavecyclic.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/filter_wavepci.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/filter_wavert.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/guid.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/interrupt.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/irp.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/irpstream.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/miniport.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/miniport_dmus.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/miniport_fmsynth.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/pin_dmus.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/pin_wavecyclic.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/pin_wavepci.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/pin_wavert.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/pool.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/port.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/port_dmus.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/port_topology.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/port_wavecyclic.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/port_wavepci.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/port_wavert.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/port_wavertstream.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/power.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/private.hpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/propertyhandler.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/purecall.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/registry.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/resource.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/service_group.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/undoc.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/unregister.cpp
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/version.cpp
61334 /trunk/reactos/drivers/wdm/audio/drivers/CMIDriver/CMakeLists.txt
      /trunk/reactos/drivers/wdm/audio/drivers/CMIDriver/debug.hpp
61331 /trunk/reactos/drivers/wdm/audio/drivers/CMIDriver/adapter.cpp
      /trunk/reactos/drivers/wdm/audio/drivers/CMIDriver/adapter.hpp
61305 /trunk/reactos/drivers/wdm/audio/sysaudio/control.c
      /trunk/reactos/drivers/wdm/audio/sysaudio/deviface.c
      /trunk/reactos/drivers/wdm/audio/sysaudio/dispatcher.c
      /trunk/reactos/drivers/wdm/audio/sysaudio/main.c
      /trunk/reactos/drivers/wdm/audio/sysaudio/pin.c
      /trunk/reactos/drivers/wdm/audio/sysaudio/sysaudio.h
61304 /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/control.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/deviface.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/entry.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/mmixer.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/sup.c
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/wdmaud.h
61132 /trunk/reactos/drivers/wdm/audio/sysaudio/sysaudio.rc
      /trunk/reactos/drivers/wdm/audio/backpln/portcls/portcls.rc
      /trunk/reactos/drivers/wdm/audio/drivers/mpu401/mpu401.rc
      /trunk/reactos/drivers/wdm/audio/drm/drmk/drmk.rc
      /trunk/reactos/drivers/wdm/audio/legacy/stream/stream.rc
      /trunk/reactos/drivers/wdm/audio/legacy/wdmaud/wdmaud.rc
      /trunk/reactos/drivers/base/beep/beep.rc
60650 /trunk/reactos/drivers/wdm/audio/drivers/CMIDriver/cmicontrol/window.rc
      /trunk/reactos/drivers/wdm/audio/drivers/CMIDriver/cpl/cmicpl.rc
60129 /trunk/reactos/drivers/wdm/audio/filters/kmixer/kmixer.c
      /trunk/reactos/drivers/wdm/audio/filters/kmixer/pin.c
60126 /trunk/reactos/drivers/wdm/audio/backpln/portcls/irpstream.cpp

You appear to have discovered the revision with the regression before I made this post, Webunny. Good work.

[Black_Fox]

Is 64751 the very last before the RC1?

It's code-wise identical to 64750 and it's the branching commit, basis of RC1 - http://svn.reactos.org/svn/reactos?revi ... w=revision

https://jira.reactos.org/browse/CORE-8667

May it be backported to 0.3.17?

It already is http://svn.reactos.org/svn/reactos?view ... sion=65047

[Webunny]

Been regressing testing the whole evening, but I'm getting a bit tired. Already burned half a dozen times the same RW by now.

The current range is 64187 to 64251.

Could someone tell me, if, within that range, something has changed in how (sound)drivers are handled? Or possible how an exe is handled (it's been installed as an exe, after all)?


Edit: tried out some guesswork/gamble which was pointed out to me on irc, but it didn't pan out to be right.

anyway, range is now 64217 to 64251

[Black_Fox]

There's no sound or drivers related commit in there. Since you have 34 revisions remaining without any suspect, the best idea would be to go in power-of-2 steps. You're almost there, should know in 4 attempts.