On Mon, 16 Aug 2010 09:51:53 +0800, Ming Lei wrote: > 2010/8/15 Dennis Nezic > <dennisn@xxxxxxxxxxxxxxxxxx>: > > On Sun, 15 Aug 2010 12:10:30 +0800, Ming Lei wrote: > >> 2010/8/15 Dennis Nezic > >> <dennisn@xxxxxxxxxxxxxxxxxx>: > >> > It looks like that patch made things more consistently > >> > *worse* :b -- which I guess is some kind of progress? :b In the > >> > "detected XactErr" messages, absolutely nothing is transferred? > >> > (ie. the ratios are always 0/some-low-number ... whereas before > >> > there was a lot more variety like 3584/4096, etc.). > >> > >> Sorry, I don't have any ideas about it. > >> > >> 'detected XactErr' means below (from ehci 1.0 pg. 54) > >> > >> Transaction Error (XactErr). Set to a one by the Host Controller > >> during status update in the case where the host did not receive a > >> valid response from the device (Timeout, CRC, Bad PID, etc.). Refer > >> to Section 4.15.1.1 for summary of the conditions that affect this > >> bit. If the host controller sets this bit to a one, then > >> it remains a one for the duration of the transfer. > >> > >> Generally speaking, it means a hardware level issue. > >> > >> The patch only adds 125us delay before transmitting data with > >> device or receiving CSW. I don't know why it make more worse than > >> before, seems it should not put any side effect on normal device. > > > > Well, it does :P -- and IMHO it's an important clue. If adding a > > 120us delay degrades performance* considerably, perhaps (pure > > handwaving) it is some kind of timing or interrupt problem? > > > > * (Before the patch, it made a little progress -- it was able to > > detect the capacity of the memory card in it -- with the patch it > > gets almost nowhere... although in both cases, the first 3 or 4 > > usb_stor_bulk_transfer's consistently work perfectly -- the > > problematic transfers only seem to happen a bit after the scsi > > multiple lun detection thingy finishes.) > > > > > >> BTW: You try the card reader on Windows OK, is the windows > >> installed on the same machine with linux? Are you sure for the > >> same machine and same card reader(includes cable), only windows > >> can work and linux does not work? > > > > I tested it with another MSWindows machine, same cable. > > As I said, your issue is very possible a hardware related, so > recommend to do the compared(windows vs. linux) tests on > same machine for ruling out hardware problem. I can't install MSWindows on my machine (for many reasons, money being a big one), and it will take a little time before I get linux installed on that other machine. However, I don't quite understand how you figure it's a hardware issue. The device *clearly* works. Unless you think my laptop's USB controller is broken? Which I *really* don't think it is, since I use USB (2.0) storage devices on here all the time. Moreover, we just saw how a simple /software/ modification yielded a quite drastic change (reduction) in performance. -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
