On 8 March 2012 10:35, Larry Finger <Larry.Finger@xxxxxxxxxxxx> wrote: > On 03/07/2012 08:16 PM, jerome huang wrote: >> >> log: >> [53614.238023] Compat-wireless backport release: compat-wireless-v3.2-1-s >> [53614.244729] Backport based on linux-stable.git v3.2 >> [53614.368074] alg: No test for arc4 (arc4-generic) >> [53614.402781] cfg80211: Calling CRDA to update world regulatory domain >> [53614.498603] rtl8192cu 3-3:1.0: usb_probe_interface >> [53614.503552] rtl8192cu 3-3:1.0: usb_probe_interface - got id >> [53614.509920] rtl8192cu: rtl8192cu: Chip version 0x11 >> [53615.359336] rtl8192cu: MAC address: 08:10:74:f1:a1:0f >> [53615.364575] rtl8192cu: Board Type 0 >> [53615.400583] rtlwifi: rx_max_size 15360, rx_urb_num 8, in_ep 1 >> [53615.408918] usbcore: registered new interface driver rtl8192cu >> [53652.483311] rtl8192cu: MAC auto ON okay! >> [53652.887348] rtl8192cu: Tx queue select: 0x05 >> [53652.921326] rtl8192c_common: Loading firmware file >> rtlwifi/rtl8192cufw.bin >> >> And I'm sure the function rtl_block_fw_writeN is used for firmware >> uploading. >> >> On 8 March 2012 10:03, jerome huang<jerome.syno@xxxxxxxxx> wrote: >>> >>> Hi, >>> >>> I found a problem just like in the older discussion "Question about >>> error from xhci-hcd", >>> there was a xhci error "no room on ep ring" when plugin rtl8192cu on >>> USB3. >>> (it works on USB2) >>> >>> After applying Larry's patch "rtlwifi: rtl8192cu: Change firmware >>> upload to use block writes", >>> I still get the same error "no room on ep ring". >>> >>> But under the same compat-wireless version, >>> rt8712 works on USB3. >>> >>> Does anyone still encounter this problem after applying this patch? >>> or do I miss something/patches? > > > This problem is in the USB system, not in the wireless driver. Keep me as a > Cc, but send mail to Sarah Sharp <sarah.a.sharp@xxxxxxxxxxxxxxx> and > linux-usb@xxxxxxxxxxxxxxx. They will tell you what debugging is needed. > > The patch was an attempt to minimize the number of writes needed to upload > the firmware, but it had no effect; however, it was still a good change. > > I'm confused about one statement you made. There is no rt8712 driver, but > there is one named r8712u; however, it controls a completely different > device than rtl8192cu. Do you have both devices? > > Larry Hi Larry, thanks for your help, and sorry for my typo, I have one rtl8192cu and one "r8712u". Just test 8712 to check if xhci works. I also tried to slow down between each write during firmware upload, and enlarge the ring size in xhci_endpoint_init(), but it seems that the trbs in the ring are not processed and the incoming trb is always queued till ring full. BR, Jerome -- 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
