View previous topic :: View next topic |
Author |
Message |
an7h Tux's lil' helper
Joined: 10 Jan 2018 Posts: 90
|
Posted: Fri Feb 26, 2021 8:22 pm Post subject: [solved] PCSX2 - File not found (Path: IsoFileSystem) |
|
|
How do I get rid of this error? For whatever reason PCSX cannot read any ISO correctly. I compiled pcsx2-9999, so I'm using the latest dev release.
Here's a screenshot with more information: https://i.imgur.com/pzlcSNX.png
Clearly I'm doing something wrong, but I don't understand what. Used to work years ago on an older version of PCSX2.
Can somebody please give some pointers? Thanks.
solution: https://forums.gentoo.org/viewtopic-p-8581380.html#8581380 _________________ If dreams were horses, beggars would ride.
Last edited by an7h on Mon Mar 01, 2021 12:37 pm; edited 1 time in total |
|
Back to top |
|
|
CaptainBlood Advocate
Joined: 24 Jan 2010 Posts: 3661
|
Posted: Sat Feb 27, 2021 7:37 am Post subject: |
|
|
Do you have stable running fine?
Thks 4 ur attention, interest & support. _________________ USE="-* ..." in /etc/portage/make.conf here, i.e. a countermeasure to portage implicit braces, belt & diaper paradigm
LT: "I've been doing a passable imitation of the Fontana di Trevi, except my medium is mucus. Sooo much mucus. " |
|
Back to top |
|
|
an7h Tux's lil' helper
Joined: 10 Jan 2018 Posts: 90
|
Posted: Sat Feb 27, 2021 8:30 am Post subject: |
|
|
Hmm, I'll see if there's any difference. _________________ If dreams were horses, beggars would ride. |
|
Back to top |
|
|
an7h Tux's lil' helper
Joined: 10 Jan 2018 Posts: 90
|
Posted: Sat Feb 27, 2021 9:22 pm Post subject: |
|
|
unfortunately the stable release has the same problem. I don't understand why? _________________ If dreams were horses, beggars would ride. |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54421 Location: 56N 3W
|
Posted: Sat Feb 27, 2021 10:14 pm Post subject: |
|
|
Moved from Gamers & Players to Unsupported Software.
It's not in portage so it fits here.
an7h,
It looks as if PCSX2 thinks its found a CDDA image, not an ISO, as the file name suggests.
CDDA does not have a filesystem, so cannot be mounted.
The original disk may have been a split CDDA and data disk.
That has the audio part first, followed by data. This lets any random CD player play the audio.
Can you try a real CD, or a different image file?
Does your normal user belong to the cdrom group?
Does your kernel have loopback support?
Do you have ISO 9960 filesystem support in the kernel?
If you poind a CD player app at the image file, does it find anything?
Is so, how many minutes of music, total are reported?
What size is the ISO ? _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
an7h Tux's lil' helper
Joined: 10 Jan 2018 Posts: 90
|
Posted: Sun Feb 28, 2021 3:47 pm Post subject: |
|
|
NeddySeagoon wrote: | Can you try a real CD, or a different image file? | This laptop doesn't have a CD/DVD drive, but trying a different image doesn't change things.
NeddySeagoon wrote: | Does your normal user belong to the cdrom group? | yes
NeddySeagoon wrote: | Does your kernel have loopback support? | Yes, enabled under Device Drivers ---> Block devices ---> Loopback device support
and
Number of loop devices to pre-create at init time is set to 8
NeddySeagoon wrote: | Do you have ISO 9960 filesystem support in the kernel? | Yes, this one is enabled under File systems ---> CD-ROM/DVD Filesystems ---> ISO 9660 CDROM file system support
You probably meant 9660, not 9960, right?
NeddySeagoon wrote: | If you poind a CD player app at the image file, does it find anything?
Is so, how many minutes of music, total are reported?
What size is the ISO ? | Which app would you suggest I try? Quodlibet doesn't detect the image at all, for instance. _________________ If dreams were horses, beggars would ride. |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54421 Location: 56N 3W
|
Posted: Sun Feb 28, 2021 3:58 pm Post subject: |
|
|
an7h,
All the right answers so far.
mplayer, vlan, grip. ...
grip is a ripper rather than a player. It may want a real optical drive to control though.
What does Code: | file </path/to/iso> | tell? _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
an7h Tux's lil' helper
Joined: 10 Jan 2018 Posts: 90
|
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54421 Location: 56N 3W
|
Posted: Sun Feb 28, 2021 4:29 pm Post subject: |
|
|
an7h,
Ahh. The content uses the UDF file system, not iso9660.
You need kernel support for UDF.
You can paste things into posts, or use wgetpaste to put things on the web, then post links.
That allows helpers to copy/paste back to you.
That helps us help you. It also preserves the important stuff for those that find this topic in years to come. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
an7h Tux's lil' helper
Joined: 10 Jan 2018 Posts: 90
|
Posted: Sun Feb 28, 2021 11:53 pm Post subject: |
|
|
@NeddySeagoon
No, unfortunately UDF was already enabled. With this laptop, even though there's no disk drive, I still enabled support for it, inluding UDF in case I really need to write a DVD or something.
In other words the same problem persists even with all the right setting it seems. Any other suggestions perhaps?
Also this forum supports attachments or something? I would just attach images to my posts, of course, but I didn't notice any of that fuctionality really. I'm not really sure what's the point of posting links to images on a pastebin-like website either? _________________ If dreams were horses, beggars would ride. |
|
Back to top |
|
|
Hu Administrator
Joined: 06 Mar 2007 Posts: 21849
|
Posted: Mon Mar 01, 2021 1:16 am Post subject: |
|
|
an7h wrote: | Also this forum supports attachments or something? I would just attach images to my posts, of course, but I didn't notice any of that fuctionality really. I'm not really sure what's the point of posting links to images on a pastebin-like website either? | Attachments are not supported. Inline images are not supported. Links to images are discouraged, for the reasons that Neddy cited. We strongly prefer that text from your system become either text in a post, or text in a pastebin, depending on the size of what you need to post. If you post to a pastebin, it is common that someone who responds will quote inline the fragment of the pastebin that is considered important, so that the fragment is preserved after the pastebin expires. You use the pastebin because the blob is too big to post inline, and you do not know yet what is important, so you cannot trim the raw output to just what we need. |
|
Back to top |
|
|
an7h Tux's lil' helper
Joined: 10 Jan 2018 Posts: 90
|
Posted: Mon Mar 01, 2021 4:50 am Post subject: |
|
|
@Hu
Ah, sorry. should have thought about that really.
Either way hope somebody can please explain why would PCSX2 have this problem under Gentoo. I just checked on Windows with the same images, and it works fine there. Of course, I wouldn't want use anything by Bill Gates, so I'd rather learn what the heck I did wrong with my system config to have an issue like this. As a sidenote I'd like to point out that other emulators like Dolphin and Mednafen work just fine with *.iso images. _________________ If dreams were horses, beggars would ride. |
|
Back to top |
|
|
CaptainBlood Advocate
Joined: 24 Jan 2010 Posts: 3661
|
Posted: Mon Mar 01, 2021 7:01 am Post subject: |
|
|
It may sound a burden of a work for some, still: Code: | [ebuild R ] gnome-base/nautilus-3.36.3::gentoo USE="-gnome -gstreamer -gtk-doc -introspection -previewer (-selinux) -sendto -test" 0 KiB
[ebuild R ] app-cdr/gcdemu-3.2.4::gentoo PYTHON_SINGLE_TARGET="python3_7 -python3_8" 0 KiB
[ebuild R ] app-cdr/cdemu-3.2.4::gentoo USE="cdemu-daemon" PYTHON_SINGLE_TARGET="python3_7 -python3_8" 0 KiB
[ebuild R ] sys-fs/udisks-2.9.1:2::gentoo USE="-acl daemon -debug elogind -introspection -lvm -nls (-selinux) -systemd -vdo -zram" 0 KiB
[ebuild R ~] x11-misc/pcmanfm-1.3.1::local-repo USE="-debug" 0 KiB
[ebuild R ] app-cdr/cdemu-daemon-3.2.4:0/7::gentoo 0 KiB
[ebuild R ] sys-fs/vhba-20200106-r1::gentoo USE="(-dist-kernel)" 0 KiB
[ebuild R ] x11-libs/libfm-1.3.1:0/5.2.1::gentoo USE="automount -debug -doc -examples exif gtk udisks vala" 0 KiB | That app stack allows to right click mount iso in pcmanfm file manager via CDemu client entry.
Mounts will appears in nautilus as well, and can be mounted via 'Open with' right click option if selecting CDemu client on combo box.
vhba needs to be builded with proper kernel eselected, if multiple kernels are installed to boot.
It also needs to be build with same gcc as kernel; IIRC. Cf Gentoo forum
Having such a stack built could be a step to validate high level app properly handle ISO file.
Stack has been build long time ago here, and never failed since: Hence some critical failure points may have been forgotten, sorry.
Thks 4 ur attention, interest & support. _________________ USE="-* ..." in /etc/portage/make.conf here, i.e. a countermeasure to portage implicit braces, belt & diaper paradigm
LT: "I've been doing a passable imitation of the Fontana di Trevi, except my medium is mucus. Sooo much mucus. " |
|
Back to top |
|
|
CaptainBlood Advocate
Joined: 24 Jan 2010 Posts: 3661
|
Posted: Mon Mar 01, 2021 7:12 am Post subject: |
|
|
pcmanfm + libfm & nautilus are local preferences, which build could be escaped.
Possibly, any file browser requiring gnome-base/gvfs might fit.
vhba module needs manual load in /etc/conf.d/modules: Code: | modules_5_11_2_gentoo_iptables="${modules_5_11_2_gentoo_iptables}vhba " |
No systemd experience to share, sorry.
Thks 4 ur attention, interest & support. _________________ USE="-* ..." in /etc/portage/make.conf here, i.e. a countermeasure to portage implicit braces, belt & diaper paradigm
LT: "I've been doing a passable imitation of the Fontana di Trevi, except my medium is mucus. Sooo much mucus. " |
|
Back to top |
|
|
Ionen Developer
Joined: 06 Dec 2018 Posts: 2732
|
Posted: Mon Mar 01, 2021 7:36 am Post subject: |
|
|
Kernel's udf isn't used with PCSX2 either way (works even without it), uses its own cdvd plugin to read directly -- just to rule that one out (but I guess it's not impossible there's some other kernel settings that could be affecting this (somehow), if in doubt try gentoo-kernel-bin).
pcsx2-9999 did go through a major change recently (switched to experimental 64bit build), but if stable does the same then I don't really know. You aren't using any strange C(XX)FLAGS optimizations, right?
It works for me™ for what it's worth. |
|
Back to top |
|
|
Ionen Developer
Joined: 06 Dec 2018 Posts: 2732
|
Posted: Mon Mar 01, 2021 8:16 am Post subject: |
|
|
NeddySeagoon wrote: | Moved from Gamers & Players to Unsupported Software.
It's not in portage so it fits here. | Also, it is, it's games-emulation/pcsx2::gentoo, and well maintained by its proxy maintainer I might add that keeps in touch with upstream and keeps the -9999 ebuild in order (slow release cycle so it's often used). |
|
Back to top |
|
|
CaptainBlood Advocate
Joined: 24 Jan 2010 Posts: 3661
|
Posted: Mon Mar 01, 2021 8:46 am Post subject: |
|
|
Ionen wrote: | It works for me™ for what it's worth. | Maybe USE flags configuration may help some.
Sidenote: ebuilds got portage update early today.
Thks 4 ur attention, interest & support. _________________ USE="-* ..." in /etc/portage/make.conf here, i.e. a countermeasure to portage implicit braces, belt & diaper paradigm
LT: "I've been doing a passable imitation of the Fontana di Trevi, except my medium is mucus. Sooo much mucus. " |
|
Back to top |
|
|
Ionen Developer
Joined: 06 Dec 2018 Posts: 2732
|
Posted: Mon Mar 01, 2021 8:53 am Post subject: |
|
|
CaptainBlood wrote: | Maybe USE flags configuration may help some. | pcsx2 doesn't really have USE, test is only for tests, and filecaps (added by fcaps.eclass) is to allow dev9 support without being root (not related and causes no code changes). |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54421 Location: 56N 3W
|
Posted: Mon Mar 01, 2021 9:24 am Post subject: |
|
|
Moved from Unsupported Software to Gamers & Players.
Ionen,
Thank you, I must have messed up my eix command. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
an7h Tux's lil' helper
Joined: 10 Jan 2018 Posts: 90
|
Posted: Mon Mar 01, 2021 12:34 pm Post subject: |
|
|
Ionen wrote: | f in doubt try gentoo-kernel-bin). |
Thank you. This helped me pinpoint the CONFIG option within the kernel that was missing for PCSX2 to work correctly.
This is what was missing:
General setup ---> Configure standard kernel features (expert users) ---> Enable AIO support
I disabled that during the initial kernel configuration. Usually I don't have problems with the stuff I disable, but as it turns out this functionality is required for PCSX2 to actually read ISO images. Btw, before this I was actually able to start up the PS2 bios via PCSX2, but not ISO images. _________________ If dreams were horses, beggars would ride. |
|
Back to top |
|
|
Hu Administrator
Joined: 06 Mar 2007 Posts: 21849
|
Posted: Mon Mar 01, 2021 4:16 pm Post subject: |
|
|
an7h wrote: | Ah, sorry. should have thought about that really. | You're not the first to ask this. an7h wrote: | This is what was missing:
General setup ---> Configure standard kernel features (expert users) ---> Enable AIO support
I disabled that during the initial kernel configuration. Usually I don't have problems with the stuff I disable, but as it turns out this functionality is required for PCSX2 to actually read ISO images. Btw, before this I was actually able to start up the PS2 bios via PCSX2, but not ISO images. | Considering Ionen's comment that upstream is active, it might be worth a problem report to upstream to request a more helpful error message when the application is run on an AIO=n kernel, so that the next person to hit this can jump straight to fixing the kernel configuration, and not need to reproduce the debugging you did here. |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54421 Location: 56N 3W
|
Posted: Mon Mar 01, 2021 4:33 pm Post subject: |
|
|
an7h,
Ebuilds can check for kernel options too. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
Ionen Developer
Joined: 06 Dec 2018 Posts: 2732
|
Posted: Mon Mar 01, 2021 4:38 pm Post subject: |
|
|
Hu wrote: | Considering Ionen's comment that upstream is active, it might be worth a problem report to upstream to request a more helpful error message when the application is run on an AIO=n kernel, so that the next person to hit this can jump straight to fixing the kernel configuration, and not need to reproduce the debugging you did here. | Normally I'd agree but we're talking about these: Code: | │ This option allows certain base kernel options and settings │
│ to be disabled or tweaked. This is for specialized │
│ environments which can tolerate a "non-standard" kernel. │
│ Only use this if you really know what you are doing. | These include things eventfd, user caps (setuid), futex, etc... It's not things you normally check for and instead just expect to exist (that is, if you're expecting semi-modern Linux and not some other kernel), unless it's some specific embedded system.
Even with uring around, AIO is still the POSIX standard. I don't think even ebuilds should have to check for this.
That aside, great managed to figure it out |
|
Back to top |
|
|
Hu Administrator
Joined: 06 Mar 2007 Posts: 21849
|
Posted: Mon Mar 01, 2021 5:13 pm Post subject: |
|
|
I know it is an "expert" option, and I don't think it is unreasonable for pcsx2 to require it. However, I think "Failed to initialize AIO" or similar would be a better error than reporting "File not found" as apparently happened here. If the error had explicitly mentioned AIO, that might have led to the solution sooner. |
|
Back to top |
|
|
|