This bug was assigned to Mian Yousaf who is no longer working with Linaro. https://bugs.launchpad.net/linux-linaro/+bug/709245 - USB Disk IO Slow
Per Andy's last comment: "It seems the issue may be to do with the Panda's hub / ethernet chip idle / wakeup characteristics but since there are no hub registers in there settable from Panda side AFAIK (there are some settable in the missing EEPROM that can configure the hub / Ethernet bridge)"
Is there anyone that is willing to handle? Or provide additional insight? Thanks, Mounir
On 07/11/2011 09:52 PM, Somebody in the thread at some point said:
This bug was assigned to Mian Yousaf who is no longer working with Linaro.
https://bugs.launchpad.net/linux-linaro/+bug/709245 - USB Disk IO Slow
Per Andy's last comment: "It seems the issue may be to do with the Panda's hub / ethernet chip idle / wakeup characteristics but since there are no hub registers in there settable from Panda side AFAIK (there are some settable in the missing EEPROM that can configure the hub / Ethernet bridge)"
Is there anyone that is willing to handle? Or provide additional insight?
Actually I don't know where to go with debugging it any further. There was some talk it might be to do with Transaction Translators and to meddle with kernel config around those, however there are no non-Highspeed devices connected and it still acts the same.
I guess we could try to contact the hub / ethernet bridge chip vendor for comment.
-Andy
On Tue, Jul 12, 2011 at 12:58 AM, Andy Green andy.green@linaro.org wrote:
On 07/11/2011 09:52 PM, Somebody in the thread at some point said:
This bug was assigned to Mian Yousaf who is no longer working with Linaro.
https://bugs.launchpad.net/linux-linaro/+bug/709245 - USB Disk IO Slow
Per Andy's last comment: "It seems the issue may be to do with the Panda's hub / ethernet chip idle / wakeup characteristics but since there are no hub registers in there settable from Panda side AFAIK (there are some settable in the missing EEPROM that can configure the hub / Ethernet bridge)"
Is there anyone that is willing to handle? Or provide additional insight?
Actually I don't know where to go with debugging it any further. There was some talk it might be to do with Transaction Translators and to meddle with kernel config around those, however there are no non-Highspeed devices connected and it still acts the same.
I've observed many times that MPU frequency is directly related device throughput (USB, MMC, WLAN) on OMAP platforms. On both OMAP3 and OMAP4 some customer devices would indeed hold a minimum MPU frequency constraint during use cases. In the OMAP PM layer in the past this API took the form of "omap_pm_min_mpu_freq" and was the MPU analogy of the common L3 bus throughput constraint.
If pinging the device over ethernet caused an increase in performance, I wonder what performance delta for USB throughput could be achieved by first using userspace CPUfreq governor at 300MHz, and then doing the same test at 600MHz. If the delta is meaningful then it looks like the MPU frequency dependency is there as described above.
Regards, Mike
I guess we could try to contact the hub / ethernet bridge chip vendor for comment.
-Andy
-- Andy Green | TI Landing Team Leader Linaro.org │ Open source software for ARM SoCs | Follow Linaro http://facebook.com/pages/Linaro/155974581091106 - http://twitter.com/#%21/linaroorg - http://linaro.org/linaro-blog
linaro-kernel mailing list linaro-kernel@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-kernel
linaro-kernel@lists.linaro.org