On Sat, 13 Dec 2014, Udo van den Heuvel wrote: > Hello, > > I noticed an usb related issue when recording from my airspy device; > so I created a bug in the bugzilla: > https://bugzilla.kernel.org/show_bug.cgi?id=89671 > > Please let me know what info I should add to help find the root cause. What are we supposed to see? The lsusb output in the bug report lists a bunch of devices, but it's not clear which one is the airspy. The bug report says there should be 10 M samples per second. How many bytes are in a sample? 2? The usbmon trace you attached to the bug report shows a bunch of bulk data being transferred in from the device, at an average rate close to 262144 bytes every 10.8 ms. This works out to a little more than 24 MB/s, or 12 MS/s assuming 2 bytes per sample. If the data transfer really was running at 20 MB/s, then 262144 bytes would require 13.11 ms. Since the usbmon trace shows that many bytes coming in every 10.5 - 11 ms (with slight variations due to other loads but almost always less than 11 ms), it's hard to see what could be wrong. Now, maybe the transfer isn't just raw data. Protocol overhead would reduce the total throughput. But since I don't know what the protocol is, all I can tell you is the raw data rate. Alan Stern -- 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
