http://bugs.winehq.org/show_bug.cgi?id=36546
Bug ID: 36546 Summary: Please remove `tmpfs` special treatment in ntdll Product: Wine Version: 1.7.19 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: major Priority: P1 Component: ntdll Assignee: wine-bugs@winehq.org Reporter: t.artem@mailcity.com
I do love running Wine in /tmp which is mounted as `tmpfs` in my Linux.
However every other application installer whines that disk C: is not local so I have to use tricks to force applications to install in a WINEPREFIX which is located in RAM.
Please, remove the special treatment of tmpfs mount points.
http://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #1 from Bruno Jesus 00cpxxx@gmail.com --- I do that too but I don't think I've ever hit this issue. Is there a small program that can trigger it?
http://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #2 from Artem S. Tashkinov t.artem@mailcity.com --- (In reply to Bruno Jesus from comment #1)
I do that too but I don't think I've ever hit this issue. Is there a small program that can trigger it?
The most recent example is Futuremark SystemInfo which refuses to install on tmpfs'ed .wine prefix:
http://www.futuremark.com/support/systeminfo
Many other applications bundled as MSI installers have this problem.
http://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #3 from Artem S. Tashkinov t.artem@mailcity.com --- Actually I can think of a kiosk type computer where /home is mounted as tmpfs.
So treating tmpfs differently is totally unwarranted and even harmful.
http://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #4 from Bruno Jesus 00cpxxx@gmail.com --- I see it in the code now: http://source.winehq.org/git/wine.git/blob/HEAD:/dlls/ntdll/file.c#l2579
http://bugs.winehq.org/show_bug.cgi?id=36546
Nikolay Sivov bunglehead@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Severity|major |enhancement
http://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #5 from Henri Verbeet hverbeet@gmail.com --- Does overriding the drive type in winecfg not help, or are you arguing that it shouldn't be needed?
http://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #6 from Artem S. Tashkinov t.artem@mailcity.com --- (In reply to Henri Verbeet from comment #5)
are you arguing that it shouldn't be needed?
This.
Again, some people may have their home dirs mounted as tmpfs. Are you saying they are all screwed?
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #7 from Henri Verbeet hverbeet@gmail.com --- (In reply to Artem S. Tashkinov from comment #6)
Again, some people may have their home dirs mounted as tmpfs. Are you saying they are all screwed?
I'm not really saying anything on the topic since it's not quite my area of expertise, I'm just trying to clarify the bug report. If you're just asking for my opinion, the current behaviour seems reasonable to me, but I don't have the kind of setup the bug report is about, so that's probably not worth much.
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #8 from Bruno Jesus 00cpxxx@gmail.com --- I can confirm that Futuremark SystemInfo will not install inside tmpfs drive, but it will in ramfs.
AFAIK the difference between tmpfs and ramfs is that ramfs will grow if full while tmpfs will not. Is this enough for you to solve this bug?
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #9 from Artem S. Tashkinov t.artem@mailcity.com --- (In reply to Bruno Jesus from comment #8)
I can confirm that Futuremark SystemInfo will not install inside tmpfs drive, but it will in ramfs.
I don't use ramfs.
AFAIK the difference between tmpfs and ramfs is that ramfs will grow if full while tmpfs will not. Is this enough for you to solve this bug?
Solve for whom? My tmpfs disk is 8GB in size - I think that's more than enough to install casual Windows applications.
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #10 from Bruno Jesus 00cpxxx@gmail.com --- Hi, Artem. I was simply saying that if you switch from tmpfs to ramfs you will not have the problem anymore.
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #11 from Artem S. Tashkinov t.artem@mailcity.com --- (In reply to Bruno Jesus from comment #10)
Hi, Artem. I was simply saying that if you switch from tmpfs to ramfs you will not have the problem anymore.
ramfs is unsuitable for too many use cases.
I'd like Alexander Julliard to chime in. This bug can be solved by removing a single line from ntdll.c and it will save my nerves.
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #12 from Bruno Jesus 00cpxxx@gmail.com --- Removing that line would break the detection of drives as virtual disks, IMO it's better to check if the drive in question is C and always return FIXED because it could happen that C is in a network drive and I guess some installers would fail for that too.
https://bugs.winehq.org/show_bug.cgi?id=36546
lilydjwg@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |lilydjwg@gmail.com
--- Comment #13 from lilydjwg@gmail.com --- I doubt this is why I can't get QQ6.3 or similar installed, see bug 37291.
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #14 from Austin English austinenglish@gmail.com --- *** Bug 37291 has been marked as a duplicate of this bug. ***
https://bugs.winehq.org/show_bug.cgi?id=36546
Artem S. Tashkinov t.artem@mailcity.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Severity|enhancement |major
--- Comment #15 from Artem S. Tashkinov t.artem@mailcity.com --- This is not an enhancement request.
Fedora and many other modern Linux distros mount /tmp as a tmpfs and you cannot install anything on this "disk".
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #16 from Artem S. Tashkinov t.artem@mailcity.com --- (In reply to Bruno Jesus from comment #12)
Removing that line would break the detection of drives as virtual disks, IMO it's better to check if the drive in question is C and always return FIXED because it could happen that C is in a network drive and I guess some installers would fail for that too.
Now how many Windows apps use virtual disks? How many rely on this feature? I cannot think of any, actually. I doubt Linux users are even aware that Wine supports virtual disks.
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #17 from Artem S. Tashkinov t.artem@mailcity.com --- *** Bug 38346 has been marked as a duplicate of this bug. ***
https://bugs.winehq.org/show_bug.cgi?id=36546
Artem S. Tashkinov t.artem@mailcity.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Summary|Please remove `tmpfs` |Please remove `tmpfs` |special treatment in ntdll |special treatment in ntdll | |/ MSI packages cannot be | |installed from `tmpfs`
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #18 from Bruno Jesus 00cpxxx@gmail.com --- Created attachment 51367 --> https://bugs.winehq.org/attachment.cgi?id=51367 patch
The attached patch solves the issue for me.
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #19 from Artem S. Tashkinov t.artem@mailcity.com --- (In reply to Bruno Jesus from comment #18)
Created attachment 51367 [details] patch
The attached patch solves the issue for me.
Thanks! I hope it's gonna be merged soon.
I wonder if this patch also fixes MSI's behavior for drive letters other than C. Something tells me it doesn't.
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #20 from lilydjwg@gmail.com --- I've bought a more powerful computer, thus I can run more resource-consuming programs. However, when I want to try if version X of program Y runs fine under Wine, what winetricks commands it will need, I want to run it on tmpfs with a clean wine prefix, so I can get correct results. But many programs just refuse to install due to the special treatment with tmpfs.
Running wine in tmpfs make it fast to destroy and recreate wine prefixes many many times, faster and without making my hard disks fail any earlier.
https://bugs.winehq.org/show_bug.cgi?id=36546
Felix Yan felixonmars@archlinux.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |felixonmars@archlinux.org
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #21 from Bruno Jesus 00cpxxx@gmail.com --- I think this bug makes sense but only related to the drive C, which is where most applications will try to install as default. I also use ramfs a lot to test wine since it's much faster to create and operate prefixes.
https://bugs.winehq.org/show_bug.cgi?id=36546
Ken Sharp imwellcushtymelike@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download, source Status|UNCONFIRMED |NEW Ever confirmed|0 |1
--- Comment #22 from Ken Sharp imwellcushtymelike@gmail.com --- Confirming. Running bisect/testing loops on a tmpfs saves a considerable amount of time (when it works).
https://bugs.winehq.org/show_bug.cgi?id=36546
Bruno Jesus 00cpxxx@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |00cpxxx@gmail.com
--- Comment #23 from Bruno Jesus 00cpxxx@gmail.com --- *** Bug 37545 has been marked as a duplicate of this bug. ***
https://bugs.winehq.org/show_bug.cgi?id=36546
Anastasius Focht focht@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |focht@gmx.net
--- Comment #24 from Anastasius Focht focht@gmx.net --- Hello folks,
I apologize beforehand since it's unrelated to the actual tmpfs problem but since a few people here are claiming this feature to be used in conjunction with rapid testing let me offer another potential solution.
--- snip --- I also use ramfs a lot to test wine since it's much faster to create and operate prefixes. --- snip ---
I admit prefixes backed by physical disks will always have inferior performance but there might be an alternative.
I use btrfs subvolumes for prefixes for more than two years now, along with 'snapper' to ease snapshot management. It allows me to snapshot/restore prefixes to any state in an instant using btrfs builtin features. This is especially useful for incremental testing and/or when intermediate prefix states need to be captured/restored quickly. In addition you get nifty features like builtin "diff" between arbitrary snapshots. You can for example see which files changed in the prefix during install or even which registry changes have been done (since the hives are text files).
Regards
https://bugs.winehq.org/show_bug.cgi?id=36546
Michael Müller michael@fds-team.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |STAGED CC| |michael@fds-team.de, | |sebastian@fds-team.de Staged patchset| |https://github.com/wine-com | |pholio/wine-staging/tree/ma | |ster/patches/ntdll-DeviceTy | |pe_Systemroot
https://bugs.winehq.org/show_bug.cgi?id=36546
babyshakes911@hotmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |babyshakes911@hotmail.com
https://bugs.winehq.org/show_bug.cgi?id=36546
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |patch
https://bugs.winehq.org/show_bug.cgi?id=36546
Bruno Jesus 00cpxxx@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |marc-alexandre.espiaut@etu. | |u-bordeaux.fr
--- Comment #25 from Bruno Jesus 00cpxxx@gmail.com --- *** Bug 40793 has been marked as a duplicate of this bug. ***
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #26 from Artem S. Tashkinov t.artem@mailcity.com --- I'm curious why the patch for this issue is not getting merged.
https://bugs.winehq.org/show_bug.cgi?id=36546
Ken Sharp imwellcushtymelike@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |imwellcushtymelike@gmail.co | |m
https://bugs.winehq.org/show_bug.cgi?id=36546
André H. nerv@dawncrow.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |nerv@dawncrow.de Staged patchset|https://github.com/wine-com |https://github.com/wine-sta |pholio/wine-staging/tree/ma |ging/wine-staging/tree/mast |ster/patches/ntdll-DeviceTy |er/patches/ntdll-DeviceType |pe_Systemroot |_Systemroot
https://bugs.winehq.org/show_bug.cgi?id=36546
Marcel maco_cz@azet.sk changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |maco_cz@azet.sk
https://bugs.winehq.org/show_bug.cgi?id=36546
Robert Walker bob.mt.wya@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |bob.mt.wya@gmail.com
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #27 from Artem S. Tashkinov aros@gmx.com --- (In reply to Artem S. Tashkinov from comment #26)
I'm curious why the patch for this issue is not getting merged.
I still am.
https://bugs.winehq.org/show_bug.cgi?id=36546
--- Comment #28 from Zebediah Figura z.figura12@gmail.com --- Well, for one, nobody has submitted it to the mailing list. Patches aren't picked up directly from bugzilla.
https://bugs.winehq.org/show_bug.cgi?id=36546
Zebediah Figura z.figura12@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |FIXED CC| |z.figura12@gmail.com Fixed by SHA1| |fb8a44f3ee50badd85af4b50fdb | |c1db28ce76ab0 Status|STAGED |RESOLVED
--- Comment #29 from Zebediah Figura z.figura12@gmail.com --- A patch to fix this issue was committed as https://source.winehq.org/git/wine.git/commitdiff/fb8a44f3ee50badd85af4b50fdbc1db28ce76ab0.
https://bugs.winehq.org/show_bug.cgi?id=36546
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #30 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 5.13.