Re: Wine users with a CD-ROM drive, please test

[Greg Geldorp]

> From:
>
> this is my second call for participation, as only Jeff Zaroyko responded to the first one.
>
> I wrote a testsuite for mcicda (play music from CD-ROM using the MCI).
> Wine's testbot job #4330 at
> https://testbot.winehq.org/JobDetails.pl?Key=4330
> contains both a patch and binary of what I'd initially want to test about mcida.dll, 
> solely lacking compatibility fixes that testing on YOUR machine will show to be necessary.

I can attach a .iso to each VM and run your tests. Would that help? Keep in mind that almost
none of the VMs have sound support.

Ge.

Wine users with a CD-ROM drive, please test

Hi,

David c/o paulo lesgaz wrote:
> mcicda.c:425: Test failed: SEEK to 00203A43 position last + length: MCIERR_HARDWARE
> mcicda.c:436: Test failed: PLAY from 1 notify: MCIERR_HARDWARE

Same as Jeff, please see if one of you can identify which prior MCI command causes this.
> mcicda.c:275: Test failed: status position initially 00:02:33
> mcicda.c:280: Test failed: status position start 00:02:33

Interesting, thank you. Will change the tests. I once read there
must be a 2 second gap but it seems it need not be exactly 2 seconds.

Greg Geldorp wrote:
> I can attach a .iso to each VM and run your tests. Would that help?

AFAIK, a .iso contains a data CD only, not music tracks.
If that's right, then it would not be very interesting,
as data-only CDs are already skipped.
> Keep in mind that almost none of the VMs have sound support.

Thanks for reminding me to temporarily disable sound on a native
system and see how that changes test results. From what I remember,
disabling sound in the preferences (XP then wants to reboot) achieved
the same results as sound-less machines in the MCI tests.

Regards,
 Jörg Höhle

Wine users with a CD-ROM drive, please test

[Matijn Woudt]

Test results from Windows 7 x64:

I tried with original audio CD, and a burned copy, mcicda_test.exe and
mcicda_test64.exe, and they all give:

mcicda.c:270: CD length 46:20:53
mcicda.c:325: track #1 length 118107ms
mcicda.c:392: Tests skipped: Got no mixed data+audio CD.
mcicda.c:410: last track length: 04m:15s:37frames
mcicda.c:418: last track position: 42m:07s:16frames
mcicda.c:425: Test failed: SEEK to 0035162E position last + length:
MCIERR_HARDWARE
mcicda.c:436: Test failed: PLAY from 1 notify: MCIERR_HARDWARE
mcicda.c:439: Tests skipped: Cannot manage to play track 1.
mcicda: 64 tests executed (0 marked as todo, 2 failures), 2 skipped.
> David c/o paulo lesgaz wrote:
> > mcicda.c:425: Test failed: SEEK to 00203A43 position last + length: MCIERR_HARDWARE
> > mcicda.c:436: Test failed: PLAY from 1 notify: MCIERR_HARDWARE
>
> Same as Jeff, please see if one of you can identify which prior MCI command causes this.

Do you have any instructions how I can identify this?

Wine users with a CD-ROM drive, please test

Hi,

Thank you for your test results so far.
> > > mcicda.c:425: Test failed: SEEK to 00203A43 position last + length: MCIERR_HARDWARE
> > > mcicda.c:436: Test failed: PLAY from 1 notify: MCIERR_HARDWARE
> >
> > Same as Jeff, please see if one of you can identify which prior MCI command causes this.
>
> Do you have any instructions how I can identify this?

This sequence of commands worked for Jeff:
open cdaudio alias c
status c type track 1
set c time format tmsf
seek c to start
play c from 1 to 2 wait

The sequence of commands that the tests use is below.  I don't see
anything there that explains MCIERR_HARDWARE.  Perhaps it's a timing
issue and the suite would work after adding a couple of Sleep()?
Copy&Paste of this sequence into the mcishell (see bug 20232, comment #10)
works for me.  Does it generate MCIERR_HARDWARE on your machine?  Can
you try and eliminate commands until you reach a minimal set that
generates the error (start with removing capability, info and notify)?

open cdaudio alias c notify shareable
capability c has video notify
capability c can play
capability c
capability c 16385
capability c device type
status c time format
info c product wait notify
status c media present
status c mode
status c mode
status c ready
info c product identity
info c identity
info c upc
status c number of tracks
status c length
status c position
status c position start notify
status c position start track 1 notify
play c from 00:02:00 to 00:01:00 notify
resume c
seek c wait
seek c to start notify
# # Perhaps use seek c to start notify wait
set c video audio all off
set c time format ms
status c position start
status c position start track 1
status c notify wait
status c length track 1
set c door closed notify
status c type track 1
# # skipping data track tests
set c time format msf
status c length track 20
# # Replace 20 with number of tracks here
status c position track 20
# # Perhaps add status c mode (still seeking?)
seek c to 24:55:74
# # Replace with position last track start + length
seek c to 24:56:00
set c time format tmsf
play c from 1

Thank you for your help,
 Jörg Höhle

Wine users with a CD-ROM drive, please test

[Eric Pouech]

Jörg-Cyril : may be the driver doesn't support TMSF, and from/to fields must
be provided in MSF
would you mind trying that ?

A+

Wine users with a CD-ROM drive, please test

Hi,

Eric Pouech wrote:
> may be the driver doesn't support TMSF, and from/to fields must be provided in MSF
> would you mind trying that ?

Me? The others ;) The 5 machines I've had access to all pass the tests.

I somehow doubt it'll help because Jeff Zaroyko wrote:
> > set c time format tmsf
> > seek c to start
> > play c from 1 to 2 wait
>
> This works - I hear music.

So TMSF works for Jeff, even though the testsuite yields MCIERR_HARDWARE there.
Furthermore the failing Seek before Play uses MSF in the proposed testsuite.

Nevertheless worth a try. Any hint may work.

Perhaps also add the "wait" flag to the tests:
seek c to start notify WAIT

Perhaps the cause is one of the unusual commands:
set c video audio all off
set c door closed notify
# # Perhaps add status c mode after that?

On some machines, inserting an audio CD starts the media player.
Perhaps this interacts with the tests?

Thank you for your help,
 Jörg Höhle

Wine users with a CD-ROM drive, please test

Hi,

http://testbot.winehq.org/JobDetails.pl?Key=4395
contains a new patch and binary where I eliminated a few commands
that might cause the MCIERR_HARDWARE observed with the second
Seek and first Play, as well as a few more calls to "status c mode" to
verify the current state.

Thank you for testing,
 Jörg Höhle

Re : Wine users with a CD-ROM drive, please test

[paulo lesgaz]

You will attached the results of the tests on my real vista box

As the last time, I used the original CD of Police: Greatest hits

David

________________________________

De : "[e-mail]" 
À : [e-mail]
Cc : [e-mail]; [e-mail]; [e-mail]; 
[e-mail]
Envoyé le : Mar 10 août 2010, 17h 50min 39s
Objet : Wine users with a CD-ROM drive, please test

Hi,

http://testbot.winehq.org/JobDetails.pl?Key=4395
contains a new patch and binary where I eliminated a few commands
that might cause the MCIERR_HARDWARE observed with the second
Seek and first Play, as well as a few more calls to "status c mode" to
verify the current state.

Thank you for testing,
Jörg Höhle

Wine users with a CD-ROM drive, please test

Hi,

testbot.winehq.org job #4404 now contains another patch and binary. I removed the call to
set c door closed
estimating that this was the most likely among the three suspects to cause the error
(I also reduced the Sleep() time).

Please test yet once again. I've not changed anything about the 00:02:33 start position
which I'll leave in until testbot gathers more data from users.

Thank you very much,
 Jörg Höhle

Wine users with a CD-ROM drive, please test

[Erik Inge Bolsø]

On Wed, 11 Aug 2010, [e-mail] wrote:
> Hi,
>
> testbot.winehq.org job #4404 now contains another patch and binary. I removed the call to
> set c door closed
> estimating that this was the most likely among the three suspects to cause the error
> (I also reduced the Sleep() time).

Yay, no more MCIERR_HARDWARE!

Some others, though. Win7 64bit.

Z:\test>mcicda_test_4404.exe
mcicda.c:287: CD length 60:42:06
mcicda.c:351: track #1 length 245334ms
mcicda.c:423: Tests skipped: Got no mixed data+audio CD.
mcicda.c:441: last track length: 06m:46s:25frames
mcicda.c:449: last track position: 53m:57s:56frames
mcicda.c:541: Test failed: status mode after play is playing
mcicda.c:548: Test failed: status mode after play is playing
mcicda: 105 tests executed (0 marked as todo, 2 failures), 1 skipped.
> Please test yet once again. I've not changed anything about the 00:02:33 start position
> which I'll leave in until testbot gathers more data from users.

I had that one too, for one out of five test cds or so.

--
-erik	http://useofwords.blogspot.com/

Wine users with a CD-ROM drive, please test

Hi,
> mcicda.c:541: Test failed: status mode after play is playing
> mcicda.c:548: Test failed: status mode after play is playing
> mcicda: 105 tests executed (0 marked as todo, 2 failures), 1 skipped.

Here I don't know what to do. Even after the former Sleep(3900ms),
status mode reported playing. Perhaps some drives just enjoy spinning?

If you look at the source, you'll see it does
set c time format tmsf
play c from 1 to 1:00:00:21 notify
Sleep
status c mode
Within almost 4 seconds time, I'd have expect the drive to spin up,
play for 280ms and stop again.

I'm tempted to
a) leave the test as is, so that test.winehq.org will slowly
   show how often this happens among users.
Or do people think I should rather
b) silence the error and don't care.
   After all, it tells me nothing about how to implement.

BTW, in the next version, the 00:02:33 start position will not be flagged as error any more.

Regards,
 Jörg Höhle

Wine users with a CD-ROM drive, please test

Hi,

I found out that some machines ignore
set c audio all off
which was in the testsuite, while others react to it and even
update the system preferences volume mixer control if
it happens to be open.

I don't know whether that difference depends on the HW (CD-ROM
driver) or the OS. You can find it out by posting your results here.

I'm now using
set c audio all on
(which should be the default)
such that all users running the testsuite get to hear something.

https://testbot.winehq.org/JobDetails.pl?Key=4434
mcicda.c:612: Test failed: STOP all returned MCIERR_HARDWARE
(I fixed that, but not yet on testbot).

Regards,
 Jörg Höhle

Wine users with a CD-ROM drive, please test

[Jeff Zaroyko]

On Fri, Aug 13, 2010 at 11:03 PM,   wrote:
> I'm now using
> set c audio all on
> (which should be the default)
> such that all users running the testsuite get to hear something.
>
> https://testbot.winehq.org/JobDetails.pl?Key=4434
> mcicda.c:612: Test failed: STOP all returned MCIERR_HARDWARE
> (I fixed that, but not yet on testbot).

I hear the playback, no more failures here with the same media as
tested before, except for the one you mention 612 which only occurred
with a blank cd.

Nice work!

Wine users with a CD-ROM drive, please test

[Matijn Woudt]

On Fri, Aug 13, 2010 at 3:03 PM,   wrote:
> Hi,
>
> I found out that some machines ignore
> set c audio all off
> which was in the testsuite, while others react to it and even
> update the system preferences volume mixer control if
> it happens to be open.
>
> I don't know whether that difference depends on the HW (CD-ROM
> driver) or the OS. You can find it out by posting your results here.
>
> I'm now using
> set c audio all on
> (which should be the default)
> such that all users running the testsuite get to hear something.
>
> https://testbot.winehq.org/JobDetails.pl?Key=4434
> mcicda.c:612: Test failed: STOP all returned MCIERR_HARDWARE
> (I fixed that, but not yet on testbot).
>
> Regards,
>  Jörg Höhle

That fixed playback.
One thing, with my Beethoven CD:
mcicda.c:300: Test failed: status position initially 00:03:00
mcicda.c:306: Test failed: status position start 00:03:00
mcicda.c:350: Test failed: status position initially 3001ms

IIRC, it can be anything, but should atleast be 2 seconds.

- Matijn

Wine users with a CD-ROM drive, please test

[James McKenzie]

Matijn Woudt wrote:
> On Fri, Aug 13, 2010 at 3:03 PM,   wrote:
> > Hi,
> >
> > I found out that some machines ignore
> > set c audio all off
> > which was in the testsuite, while others react to it and even
> > update the system preferences volume mixer control if
> > it happens to be open.
> >
> > I don't know whether that difference depends on the HW (CD-ROM
> > driver) or the OS. You can find it out by posting your results here.
> >
> > I'm now using
> > set c audio all on
> > (which should be the default)
> > such that all users running the testsuite get to hear something.
> >
> > https://testbot.winehq.org/JobDetails.pl?Key=4434
> > mcicda.c:612: Test failed: STOP all returned MCIERR_HARDWARE
> > (I fixed that, but not yet on testbot).
> >
> > Regards,
> >  Jörg Höhle
>
> That fixed playback.
> One thing, with my Beethoven CD:
> mcicda.c:300: Test failed: status position initially 00:03:00
> mcicda.c:306: Test failed: status position start 00:03:00
> mcicda.c:350: Test failed: status position initially 3001ms
>
> IIRC, it can be anything, but should atleast be 2 seconds.

It is.  3001 ms is one ms more than three seconds....

James McKenzie