From: Andrea V. <and...@un...> - 2018-05-14 06:30:19
|
Hi, Il giorno sab, 12/05/2018 alle 09.43 -0700, Philip Langdale ha scritto: > On Thu, 10 May 2018 12:47:39 +0200 > Andrea Vai <and...@un...> wrote: > > > Hi, > > I obtain the above error while browsing an mtp device in gnome- > > nautilus on Fedora 27. > > > > The problem seems to be triggered by the number of files in the > > directory. > > > > See https://bugzilla.redhat.com/show_bug.cgi?id=1401826 > > > > Is there any way to investigate it? > > Let me know if I can help in any way. > > > > Running libmtp 1.1.14 > > > > Thank you very much, > > Best regards > > Andrea Vai > > You can override the device timeout value, and libmtp actually does > this in a different call that retrieves the list of all files on the > device. The gvfs code that is relevant in your example is only > getting > the file list for a single directory but if the device has a silly > default timeout (likely - thank's samsung), it might still hit it. > So > I'd try setting a larger timeout and see if it helps. If that does > help, you can then look at implementing it as a device quirk. > > See get_all_metadata_fast for the current usage and then look at > LIBMTP_Get_Files_And_Folders for where you'd want to adjust the > timeout > (around the call to ptp_getobjecthandles). Thank you for your kind reply, Phil. I downloaded the current source code and found the strings you mention in libmtp.c. I guess I should increase the "60000" value at line 2620. Though, I am not skilled in linux development, so some further hints would be greatly appreciated (should I change the value, remove my libmtp installed package, build and install the hacked source code, and try if it works?). In the meantime, I will try to figure it out. Thank you, best regards Andrea |