On Thu, Sep 21, 2023 at 02:22:13PM +0200, Massimo Burcheri wrote: > On Tue, 2023-09-19 at 11:13 -0400, Alan Stern wrote: > > > > coming from https://lore.kernel.org/all/20200818041324.GA3173@Susan/ That email refers to vendor & product 357d:7788. But your device, as shown in the earlier email message, is 152d:0567. So it's puzzling why you referred back to this bug report, even though the devices do have similar names (JMicron J566 and J567). > > > as I understand UAS was working for JMicron JMS567 in the past, then was > > > disabled in the kernel, now using usb-storage. > > > How did you get that idea? After looking through the email archives, I > > found this bug report from 2015: > > > > https://bugzilla.redhat.com/show_bug.cgi?id=1260207 > > I got the idea from my original thread. > I did not test it on old kernel. In fact, I don't see anything in the current kernel that would prevent your device from working with uas. The entry in unusual_uas.h merely has the BROKEN_FUA and NO_REPORT_OPCODES flags set, neither of which should have a bad effect. Am I missing something? > > That bug report indicates that the device wasn't working properly with a > > 4.1.4 kernel. Of course, it's possible that the problem had more to do > > with the drive inside the enclosure than the enclosure itself. > > From the report the idProduct is different from my product, might be a different > bridge. Indeed. > So from what I understand it is blacklisted currently because of issues. > > Is there a way I can bypass the blacklisting, like forcing to use uas similar to > the workaround by forcing usb-storage? I would like to know if there are still > issues in uas mode with this bridge. > > The (chinese?) product is sold with UAS support. I know this is no warranty. The > JMicron bridge itself should support UASP: > https://pcper.com/wp-content/uploads/2014/12/0813-jms567.pdf > > I found someone that fixed it by flashing an older firmware: > https://forum.openmediavault.org/index.php?user-post-list/36559-sanjager/ > (Version 20.06.00.01, before 138.01.00.01) > > It seems a similar device Icybox 4HDD enclosure is also using the JMS567: > https://answers.launchpad.net/ubuntu/+source/linux/+question/688498 > > This all sounds like UAS is not entirely missing but somehow faulty in this > bridge. > > I'm going to test with a Windows machine soon, if UAS is working stable there. > But that won't help me on Linux, just for saying the enclosure hardware is UAS > capable. > > How can I help exploring / debugging this? If you can, start by figuring out why your kernel isn't using the uas driver for the JMS567. Is the driver itself blacklisted on your computer? Alan Stern
