[Bug 54575] New: False positive detection of mmc reader as hard drive since kernel 6.1

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=54575

Bug ID: 54575
           Summary: False positive detection of mmc reader as hard drive
                    since kernel 6.1
           Product: Wine
           Version: 8.2
          Hardware: x86-64
                OS: Linux
            Status: UNCONFIRMED
          Severity: major
          Priority: P2
         Component: -unknown
          Assignee: [e-mail]
          Reporter: [e-mail]
      Distribution: ---

Since kernel 6.1 wine interprets mmc reader as valid /dev/sda device.

Problem is that kernel instead of /dev/mmc created /dev/sda... However its easy
to fix in wine. Please just detect device class type and problem never appear
again.

Reproduction steps install fusion360
https://github.com/str0g/Autodesk-Fusion360-Linux
or
https://github.com/cryinkfly/Autodesk-Fusion-360-for-Linux

1.Open project
2.Modify something
3.leave it for 30minuts (don't know if screen saver is important or not)
4.add additional modification
5.try to save.

Expected outcome
Error:
D:\Stage\workspace\protein\release\2021.10.R2023\protein\src\libs\protein4\archiveadaptor.cpp(210)stream
can not open!!

Additional information:
winprefixes/
d:: -> /dev/sda

Actual mmc device...
❯ dmesg |grep sda
[    3.455870] sd 0:0:0:0: [sda] Media removed, stopped polling
[    3.456045] sd 0:0:0:0: [sda] Attached SCSI removable disk
[99954.410598] sd 0:0:0:0: [sda] 250085376 512-byte logical blocks: (128 GB/119
GiB)
[99954.411058] sda: detected capacity change from 0 to 250085376
[99954.411675]  sda: sda1

when sd card is in slot
winprefixes/
d:: -> /dev/sda1
e:: -> /dev/sda

[Bug 54575] False positive detection of mmc reader as hard drive since kernel 6.1

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=54575

--- Comment #1 from Rafał Mużyło  ---
While I'm not quite sure about it, this sounds more like a local configuration
problem, best solved by adjusting the udev rule that's handling the device.

Also, if I'm reading your post correctly, 'expected outcome' has a different
meaning then the one you used here.

Please clarify what happens and what you think is supposed to happen.

[Bug 54575] False positive detection of mmc reader as hard drive since kernel 6.1

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=54575

--- Comment #2 from Łukasz Buśko  ---
Rafał thank You for fast response.

Expected outcome was referring to reproduction steps.

From user standpoint d:: should not be mapped to /dev/sda which is card reader
device so there is not possibility to read it this way. /dev/sda1 is whole
different story.

Regarding to configuration I have checked developer board running arch with
kernel 6.1.2 emmc is also reported as SCSI removable so question is what have
change in kernel.

However still change in detection method would be ideal solution which will
bypass problems with further kernel refactoring.

I've started this topic and there is one more person reporting same problem so
there be more reporters in the future.
https://github.com/cryinkfly/Autodesk-Fusion-360-for-Linux/issues/299

Tomorrow will play a little with udev and update topic if find solution or more
details information what had happen.

[Bug 54575] False positive detection of mmc reader as hard drive since kernel 6.1

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=54575

--- Comment #3 from Łukasz Buśko  ---
Created attachment 74121
  --> https://bugs.winehq.org/attachment.cgi?id=74121
udev dump, mmc no sd, mmc with sd, ssd with partitions

ID_PART_TABLE_TYPE looks like unique value for correct filtering. Unfortunately
from udev it can be clearly seen that ID_TYPE=disk is being set to mmc device.

[Bug 54575] False positive detection of mmc reader as hard drive since kernel 6.1

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=54575

--- Comment #4 from Łukasz Buśko  ---
Easiest temporary fix put sd card to reader, card do not need to be mounted.

[Bug 54575] False positive detection of mmc reader as hard drive since kernel 6.1

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=54575

Łukasz Buśko  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|UNCONFIRMED                 |RESOLVED

--- Comment #5 from Łukasz Buśko  ---
Fusion error

[Bug 54575] False positive detection of mmc reader as hard drive since kernel 6.1

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=54575

[e-mail] changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

--- Comment #6 from [e-mail] ---
was anything fixed in wine source code? if not, this isn't fixed.

[Bug 54575] False positive detection of mmc reader as hard drive since kernel 6.1

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=54575

--- Comment #7 from Łukasz Buśko  ---
Created attachment 76740
  --> https://bugs.winehq.org/attachment.cgi?id=76740
fusion settings

[Bug 54575] False positive detection of mmc reader as hard drive since kernel 6.1

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=54575

--- Comment #8 from Łukasz Buśko  ---
To be more specific fusion has something messed up with network protocol. When
I change to settings attached as image, problem goes away. Solution for this
was found on fusion forum from windows users who had network problems with ISP.
Symptoms where similar to what I had so gave it a try.

[Bug 54575] False positive detection of mmc reader as hard drive since kernel 6.1

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=54575

Alexandre Julliard  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED

--- Comment #9 from Alexandre Julliard  ---
Closing bugs fixed in 9.13.