From mboxrd@z Thu Jan 1 00:00:00 1970 From: Oliver Neukum Subject: Re: JMS56x not working reliably with uas driver Date: Wed, 21 Dec 2016 13:12:12 +0100 Message-ID: <1482322332.7638.12.camel@suse.com> References: <585A69E6.6040009@caviumnetworks.com> <1482320547.7638.7.camel@suse.com> <585A7073.3060904@caviumnetworks.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Return-path: Received: from mx2.suse.de ([195.135.220.15]:59910 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752138AbcLUMSg (ORCPT ); Wed, 21 Dec 2016 07:18:36 -0500 In-Reply-To: <585A7073.3060904@caviumnetworks.com> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: George Cherian Cc: hdegoede@redhat.com, stern@rowland.harvard.edu, linux-scsi@vger.kernel.org, linux-usb@vger.kernel.org On Wed, 2016-12-21 at 17:37 +0530, George Cherian wrote: > > On 12/21/2016 05:12 PM, Oliver Neukum wrote: > > On Wed, 2016-12-21 at 17:09 +0530, George Cherian wrote: > >> Hi Oliver, > >> > >> I was working with this JMicron device and using the uas driver. > >> I am seeing the following 2 issues. > >> > >> 1) On connect I see the following messages. > > > > Thanks. Do you want to submit it to Greg? > > The patch is fine. > Yes please!!! So you want me to submit it? It would be your chance to get a patch upstream. > >> 2) On disconnect I am seeing the following issue > >> > >> scsi host4: uas_post_reset: alloc streams error -19 after reset > >> sd 4:0:0:0: [sdb] Synchronizing SCSI cache > >> > >> This is more fatal because after these messages the USB port becomes > >> unusable. Even an lsusb invocation hangs for ever. > > > > Ouch. That points to a logic error. We should not reset if > > a device is gone. > > Could you send dmesg of such a case? > here is the dmesg!! > [ 203.475382] usb 4-1.3: new SuperSpeed USB device number 3 using xhci_hcd > [ 203.496172] usb 4-1.3: New USB device found, idVendor=152d, > idProduct=9561 > [ 203.503037] usb 4-1.3: New USB device strings: Mfr=1, Product=2, > SerialNumber=5 > [ 203.510352] usb 4-1.3: Product: JMS56x Series > [ 203.514698] usb 4-1.3: Manufacturer: JMicron > [ 203.518966] usb 4-1.3: SerialNumber: 00000000000000000000 > [ 203.594383] usbcore: registered new interface driver usb-storage > [ 203.612425] scsi host4: uas > [ 203.615418] usbcore: registered new interface driver uas > [ 203.620979] scsi 4:0:0:0: Direct-Access ST4000NM 0033-9ZM170 > 0001 PQ: 0 ANSI: 6 > [ 203.630240] sd 4:0:0:0: Attached scsi generic sg1 type 0 > [ 203.630382] sd 4:0:0:0: [sdb] 7814037168 512-byte logical blocks: > (4.00 TB/3.63 TiB) > [ 203.631338] sd 4:0:0:0: [sdb] Write Protect is off > [ 203.631342] sd 4:0:0:0: [sdb] Mode Sense: 67 00 10 08 > [ 203.631734] sd 4:0:0:0: [sdb] Write cache: enabled, read cache: > enabled, supports DPO and FUA > [ 203.631899] xhci_hcd 0000:00:11.0: ERROR Transfer event for disabled > endpoint or incorrect stream ring > [ 203.631904] xhci_hcd 0000:00:11.0: @0000001f610a1c10 00000000 > 00000000 1b000000 03078001 state 14 ep_info 9403 > [ 203.631906] xhci_hcd 0000:00:11.0: No epring > [ 203.674546] sdb: sdb1 > [ 203.676639] sd 4:0:0:0: [sdb] Attached SCSI disk > [ 213.222913] scsi host4: uas_post_reset: alloc streams error -19 after > reset > [ 213.230548] sd 4:0:0:0: [sdb] Synchronizing SCSI cache > > Above is the dmesg without the unusual_uas patch applied. > Do you need me to enable any specific dev_dbg and then the dmesg output? Damn, that is a strange error. Do you know whether it happens on other XHCI controllers? Which controller are you using and can you please also post "lsusb -v"? Regards Oliver