After USB WLAN interface is connected for the first time, interface doesn't respond: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [ 5085.096197] usbcore: registered new interface driver rtl8xxxu [ 5104.105708] usb 1-2: new high-speed USB device number 3 using xhci_hcd [ 5104.357890] usb 1-2: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [ 5104.357901] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [ 5104.357905] usb 1-2: Product: 802.11n NIC [ 5104.357908] usb 1-2: Manufacturer: Realtek [ 5107.412769] usb 1-2: Vendor: Realtek [ 5107.412777] usb 1-2: Product: \x03802.11n NI [ 5107.412781] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [ 5107.412785] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [ 5107.412793] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 [ 5107.412797] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [ 5107.421690] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [ 5111.638510] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! After reconnecting the interface , everything is fine: [ 5162.855223] usb 1-2: new high-speed USB device number 4 using xhci_hcd [ 5163.106485] usb 1-2: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [ 5163.106496] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [ 5163.106500] usb 1-2: Product: 802.11n NIC [ 5163.106503] usb 1-2: Manufacturer: Realtek [ 5166.161384] usb 1-2: Vendor: Realtek [ 5166.161399] usb 1-2: Product: \x03802.11n NI [ 5166.161402] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [ 5166.161404] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [ 5166.161410] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 [ 5166.161414] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [ 5166.164736] usb 1-2: Firmware revision 11.1 (signature 0x88e1) I've read the wireless mailing list and noticed the patches, e.g.: https://lore.kernel.org/lkml/a0c14bfd-a502-6b19-de75-491ea9af3816@gmail.com/T/ Now I would like to inform you that rtl8xxxu_8188e.c is also affected.
Thanks for the report. When you get the message „Path A RX IQK failed!” can you still use the device? Does it scan? Does it connect? IQ calibration is attempted twice, so if you only saw one of these messages it means the second attempt was successful and the device should be working.
The device doesn't respond. After reconnecting it or after an USB reset everything is fine.
Unfortunately my TP-Link TL-WN725N doesn't have that problem. What are you trying to do with the device when it doesn't respond?
Reset the device: static void resetusbif(char *sysfsusbdevpath) { static int fd_usb; if((fd_usb = open(sysfsusbdevpath, O_WRONLY)) < 0) { fprintf(stderr, "ERROR openening USB device\n"); return; } if(ioctl(fd_usb, USBDEVFS_RESET, 0) < 0) fprintf(stderr, "ioctl failed\n"); else fprintf(stdout, "reset successful\n"); close(fd_usb); return; } This always work. BTW: That happens on this driver, too: https://github.com/kimocoder/realtek_rtwifi/issues/34#issue-1524625740 and let me assume, it is related to the driver. Confirmed by another user of that driver (not the stock kernel driver) https://github.com/kimocoder/realtek_rtwifi/issues/35#issuecomment-1444676522
This will not happen if I connect e.g. a rtl8192 device: ID 0bda:8178 Realtek Semiconductor Corp. RTL8192CU 802.11n WLAN Adapter [36029.791983] usb 1-2: new high-speed USB device number 11 using xhci_hcd [36029.935014] usb 1-2: New USB device found, idVendor=0bda, idProduct=8178, bcdDevice= 2.00 [36029.935025] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [36029.935030] usb 1-2: Product: USB WLAN [36029.935033] usb 1-2: Manufacturer: 802.11n [36029.935036] usb 1-2: SerialNumber: 00e04c000001 [36030.397257] usb 1-2: Vendor: 802.11n [36030.397267] usb 1-2: Product: USB WLAN [36030.397270] usb 1-2: RTL8192CU rev A (TSMC) romver 0, 2T2R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [36030.397277] usb 1-2: RTL8192CU MAC: f8:1a:67:0d:c5:27 [36030.397282] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8192cufw_TMSC.bin [36030.401276] usb 1-2: Firmware revision 88.2 (signature 0x88c1) [36031.451440] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0
compared to a rtl8188 device. This time a Logilink 0150 & efuse information: sudo insmod rtl8xxxu.ko debug=0x2000 [36544.627558] usbcore: registered new interface driver rtl8xxxu [36549.949836] usb 1-2: new high-speed USB device number 13 using xhci_hcd [36550.091883] usb 1-2: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [36550.091895] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [36550.091900] usb 1-2: Product: 802.11n NIC [36550.091903] usb 1-2: Manufacturer: Realtek [36550.091906] usb 1-2: SerialNumber: 70F11C27AEEC [36550.481162] usb 1-2: Vendor: Realtek [36550.481173] usb 1-2: Product: \x03802.11n NI [36550.481176] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [36550.481179] usb 1-2: Dumping efuse for RTL8188EU (0x200 bytes): [36550.481184] 00000000: 29 81 00 6c 0b 00 00 00 00 0c 00 00 00 00 00 00 )..l............ [36550.481188] 00000010: 2a 2a 2a 2a 2a 2a 2a 2a 2a 2a 2a 02 ff ff ff ff ***********..... [36550.481191] 00000020: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481193] 00000030: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481195] 00000040: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481198] 00000050: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481200] 00000060: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481202] 00000070: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481204] 00000080: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481206] 00000090: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481209] 000000a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481211] 000000b0: ff ff ff ff ff ff ff ff 20 27 1a 00 00 00 00 00 ........ '...... [36550.481213] 000000c0: 00 00 00 10 00 00 00 00 00 01 ff ff ff ff ff ff ................ [36550.481215] 000000d0: da 0b 79 81 42 66 00 70 f1 1c 27 ae ec 0a 03 52 ..y.Bf.p..'....R [36550.481218] 000000e0: 65 61 6c 74 65 6b 0d 0d 03 38 30 32 2e 31 31 6e ealtek...802.11n [36550.481220] 000000f0: 20 4e 49 43 ff ff ff ff ff ff ff ff ff ff ff ff NIC............ [36550.481222] 00000100: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481224] 00000110: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481226] 00000120: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481228] 00000130: 81 ae 96 2d 03 93 96 11 fc 8c 00 11 9b ff ff ff ...-............ [36550.481231] 00000140: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481233] 00000150: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481235] 00000160: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481237] 00000170: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481239] 00000180: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481241] 00000190: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481243] 000001a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481246] 000001b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481248] 000001c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481250] 000001d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481252] 000001e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481254] 000001f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [36550.481257] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [36550.481264] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [36550.481268] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [36550.484894] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [36
Please notice: On first connect: [ 5111.638510] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! This message does not appear any longer after an USB reset or dis/reconnecting the device. At them moment, I have no idea why this happens.
Can you try firmware v28? https://github.com/lwfinger/rtl8188eu/blob/1201a7a51fcef5228cc5aa6df749e6c94c91c70b/hal8188e_t_fw.c#L1957 What are you trying to do with the device when it doesn't respond? In other words, what command are you running when you notice it doesn't respond?
Created attachment 304103 [details] quick and dirty code to reset a "dead" USB interface gcc resetif.c -o resetif
I'm not sure if my commands to reset an USB device will helpful. But here they are: get USB ID: $ lsusb Bus 001 Device 015: ID 0bda:8179 Realtek Semiconductor Corp. RTL8188EUS 802.11n Wireless Network Adapter reset interface: $ sudo resetif /dev/bus/usb/001/015 reset successful resetif is a quick and dirty code to reset an USB device if it crashed ag.g during hcxdumptool tests. The entire code is attached above.
resetif is exactly doing what I expect it to do: Interface connected for the first time: [43110.568424] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [43110.568428] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [43110.572456] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [43111.400716] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [43111.435086] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [43111.498085] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [43111.607823] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! ... sudo ./resetif /dev/bus/usb/001/016 ... 43344.960509] usb 1-2: disconnecting [43344.960529] usb 1-2: Device still attached, trying to reset [43345.081609] usb 1-2: reset high-speed USB device number 16 using xhci_hcd [43345.609136] usb 1-2: Vendor: Realtek [43345.609155] usb 1-2: Product: \x03802.11n NI [43345.609158] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [43345.609162] usb 1-2: Dumping efuse for RTL8188EU (0x200 bytes): [43345.609168] 00000000: 29 81 00 6c 0b 00 00 00 00 0c 00 00 00 00 00 00 )..l............ [43345.609173] 00000010: 2a 2a 2a 2a 2a 2a 2a 2a 2a 2a 2a 02 ff ff ff ff ***********..... [43345.609176] 00000020: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609178] 00000030: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609181] 00000040: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609183] 00000050: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609185] 00000060: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609188] 00000070: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609190] 00000080: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609192] 00000090: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609194] 000000a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609196] 000000b0: ff ff ff ff ff ff ff ff 20 27 1a 00 00 00 00 00 ........ '...... [43345.609199] 000000c0: 00 00 00 10 00 00 00 00 00 01 ff ff ff ff ff ff ................ [43345.609201] 000000d0: da 0b 79 81 42 66 00 70 f1 1c 27 ae ec 0a 03 52 ..y.Bf.p..'....R [43345.609203] 000000e0: 65 61 6c 74 65 6b 0d 0d 03 38 30 32 2e 31 31 6e ealtek...802.11n [43345.609206] 000000f0: 20 4e 49 43 ff ff ff ff ff ff ff ff ff ff ff ff NIC............ [43345.609208] 00000100: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609210] 00000110: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609212] 00000120: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609215] 00000130: 81 ae 96 2d 03 93 96 11 fc 8c 00 11 9b ff ff ff ...-............ [43345.609217] 00000140: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609220] 00000150: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609222] 00000160: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609225] 00000170: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609227] 00000180: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609229] 00000190: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609231] 000001a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609233] 000001b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609236] 000001c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609238] 000001d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609240] 000001e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609243] 000001f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [43345.609245] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [43345.609252] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [43345.609257] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [43345.614682] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [43346.553557] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 and everything is fine.
At a first step, we should make sure that the firmware is evaluated correctly. That is not the case, right now: Product: \x03802.11n NI Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff
Edit: comment 12: I meant interface information - firmware is loaded correctly At a first step, we should make sure that the interface information is evaluated correctly. That is not the case, right now: Product: \x03802.11n NI Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff
The problem with the product and serial is completely harmless and not related to the IQ calibration failures.
I fully agree - however, fixing that, will keep dmesg log clean. At the moment a take a look at the calibration function: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c?h=v6.3-rc6#n664
Changed mdelay from 10 to 20 https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8188e.c?h=v6.3-rc6#n703 [46234.153161] usb 1-2: new high-speed USB device number 19 using xhci_hcd [46234.295693] usb 1-2: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [46234.295706] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [46234.295710] usb 1-2: Product: 802.11n NIC [46234.295713] usb 1-2: Manufacturer: Realtek [46234.295716] usb 1-2: SerialNumber: 70F11C27AEEC [46234.684314] usb 1-2: Vendor: Realtek [46234.684323] usb 1-2: Product: \x03802.11n NI [46234.684327] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [46234.684330] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [46234.684337] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [46234.684341] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [46234.688100] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [46235.659570] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0
$ sudo rmmod rtl8xxxu $ sudo insmod rtl8xxxu.ko device number 19 no warning device number 20 no warning device number 21 [46446.874622] usb 1-2: new high-speed USB device number 21 using xhci_hcd [46447.017115] usb 1-2: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [46447.017127] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [46447.017132] usb 1-2: Product: 802.11n NIC [46447.017135] usb 1-2: Manufacturer: Realtek [46447.017138] usb 1-2: SerialNumber: 70F11C27AEEC [46447.406615] usb 1-2: Vendor: Realtek [46447.406624] usb 1-2: Product: \x03802.11n NI [46447.406629] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [46447.406632] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [46447.406639] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [46447.406643] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [46447.410510] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [46448.380953] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 dmesg doesn't show a warning any longer.
rejoiced too soon try 22: [46663.828964] usb 1-2: new high-speed USB device number 22 using xhci_hcd [46663.971043] usb 1-2: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [46663.971055] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [46663.971059] usb 1-2: Product: 802.11n NIC [46663.971063] usb 1-2: Manufacturer: Realtek [46663.971066] usb 1-2: SerialNumber: 70F11C27AEEC [46664.359041] usb 1-2: Vendor: Realtek [46664.359051] usb 1-2: Product: \x03802.11n NI [46664.359055] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [46664.359058] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [46664.359066] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [46664.359071] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [46664.363132] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [46665.377920] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [46665.412318] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [46665.457378] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0
Regardless if we give the interface some more time do to the calibration, the error appears randomly Now changed the interface from a cheap Logilink 0150 to an ALFA AWUS0136NHV https://www.alfa.com.tw/products/awus036nhv Bus 001 Device 023: ID 0bda:8179 Realtek Semiconductor Corp. RTL8188EUS 802.11n Wireless Network Adapter [49276.814153] usb 1-2: new high-speed USB device number 23 using xhci_hcd [49276.958540] usb 1-2: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [49276.958552] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [49276.958557] usb 1-2: Product: 802.11n NIC [49276.958560] usb 1-2: Manufacturer: Realtek [49277.340800] usb 1-2: Vendor: Realtek [49277.340810] usb 1-2: Product: \x03802.11n NI [49277.340814] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [49277.340816] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [49277.340824] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 [49277.340828] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [49277.345761] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [49278.189693] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [49278.302947] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [49278.390829] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 $ sudo ./resetif /dev/bus/usb/001/023 reset successful [49714.455396] usb 1-2: Device still attached, trying to reset [49714.576361] usb 1-2: reset high-speed USB device number 23 using xhci_hcd [49715.098521] usb 1-2: Vendor: Realtek [49715.098530] usb 1-2: Product: \x03802.11n NI [49715.098534] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [49715.098536] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [49715.098543] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 [49715.098547] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [49715.103048] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [49716.067057] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0
Can you try this patch and tell me what it prints, please? I'm not sure it will work, it's inspired by the RTL8192FU driver. diff --git a/rtl8xxxu_8188e.c b/rtl8xxxu_8188e.c index af84360..c27c97b 100644 --- a/rtl8xxxu_8188e.c +++ b/rtl8xxxu_8188e.c @@ -665,6 +665,7 @@ static int rtl8188eu_rx_iqk_path_a(struct rtl8xxxu_priv *priv) { u32 reg_ea4, reg_eac, reg_e94, reg_e9c, val32; int result = 0; + int ktime; /* Leave IQK mode */ val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK); @@ -752,6 +753,13 @@ static int rtl8188eu_rx_iqk_path_a(struct rtl8xxxu_priv *priv) mdelay(10); + ktime = 0; + while (rtl8xxxu_read32(priv, 0xea8) == 0 && ktime < 101) { + mdelay(5); + ktime += 5; + } + pr_warn("%s: waited an extra %d ms.\n", __func__, ktime); + reg_eac = rtl8xxxu_read32(priv, REG_RX_POWER_AFTER_IQK_A_2); reg_ea4 = rtl8xxxu_read32(priv, REG_RX_POWER_BEFORE_IQK_A_2);
Thanks. Applied to kernel 6.3-rc6 Pretty good idea to make mdelay dynamic: [87092.697524] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [87093.533462] rtl8188eu_rx_iqk_path_a: waited an extra 0 ms. [87093.597921] rtl8188eu_rx_iqk_path_a: waited an extra 0 ms. [87093.645301] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 Interesting Did another test. This time I removed mdelay(10) from the applied patch and we enter the while loop directly: rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf9000000); rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf8000000); ktime = 0; while (rtl8xxxu_read32(priv, 0xea8) == 0 && ktime < 101) { mdelay(5); ktime += 5; } pr_warn("%s: waited an extra %d ms.\n", __func__, ktime); [87993.201415] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [87994.056759] rtl8188eu_rx_iqk_path_a: waited an extra 0 ms. [87994.057512] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [87994.110260] rtl8188eu_rx_iqk_path_a: waited an extra 0 ms. [87994.177262] rtl8188eu_rx_iqk_path_a: waited an extra 0 ms. [87994.229111] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 expected ktime at least 10 measured ktime = 0 Shouldn't the ktime increase in that case?
For me it looks that mdelay(10) is completely useless. Modified the patch to: pr_warn("val32: %s %04x\n", __func__, rtl8xxxu_read32(priv, 0xea8)); for(int x = 0; x < 100; x++) { pr_warn("val32: %s %04x\n", __func__, rtl8xxxu_read32(priv, 0xea8)); mdelay(5); } [ 5085.740565] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [ 5086.563554] val32: rtl8188eu_rx_iqk_path_a 3a7ea4e8 [ 5086.563929] val32: rtl8188eu_rx_iqk_path_a 3a7ea4e8 ... [ 5088.529686] val32: rtl8188eu_rx_iqk_path_a 16ea090 [ 5088.534939] val32: rtl8188eu_rx_iqk_path_a 16ea090 [ 5088.540202] val32: rtl8188eu_rx_iqk_path_a 16ea090
Created attachment 304106 [details] entire results of the for next loop entire results of the for next loop
Okay. If even 500 ms is not enough then adjusting the delay is not the way to go. Let's see what is happening: diff --git a/rtl8xxxu_8188e.c b/rtl8xxxu_8188e.c index af84360..417c85e 100644 --- a/rtl8xxxu_8188e.c +++ b/rtl8xxxu_8188e.c @@ -757,9 +757,10 @@ static int rtl8188eu_rx_iqk_path_a(struct rtl8xxxu_priv *priv) if (!(reg_eac & BIT(27)) && ((reg_ea4 & 0x03ff0000) != 0x01320000) && - ((reg_eac & 0x03ff0000) != 0x00360000)) + ((reg_eac & 0x03ff0000) != 0x00360000)) { result |= 0x02; - else + pr_warn("%s: Path A RX IQK success!\n", __func__); + } else dev_warn(&priv->udev->dev, "%s: Path A RX IQK failed!\n", __func__); @@ -985,8 +986,8 @@ static void rtl8188eu_phy_iq_calibrate(struct rtl8xxxu_priv *priv) priv->regebc = reg_ebc; reg_ec4 = result[candidate][6]; reg_ecc = result[candidate][7]; - dev_dbg(dev, "%s: candidate is %x\n", __func__, candidate); - dev_dbg(dev, + dev_warn(dev, "%s: candidate is %x\n", __func__, candidate); + dev_warn(dev, "%s: e94=%x e9c=%x ea4=%x eac=%x eb4=%x ebc=%x ec4=%x ecc=%x\n", __func__, reg_e94, reg_e9c, reg_ea4, reg_eac, reg_eb4, reg_ebc, reg_ec4, reg_ecc);
ALFA: [19636.536498] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [19637.370919] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [19637.434871] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [19637.448624] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [19637.448634] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fc e9c=2 ea4=fb eac=3fb eb4=0 ebc=0 ec4=0 ecc=0 [19637.478068] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 after USB reset: [19796.781609] usb 1-2: Device still attached, trying to reset [19796.902489] usb 1-2: reset high-speed USB device number 12 using xhci_hcd [19797.423876] usb 1-2: Vendor: Realtek [19797.423884] usb 1-2: Product: \x03802.11n NI [19797.423887] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [19797.423889] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [19797.423895] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 [19797.423898] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [19797.428054] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [19798.256273] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [19798.320140] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [19798.333640] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [19798.333644] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=df e9c=7 ea4=fd eac=3fd eb4=0 ebc=0 ec4=0 ecc=0 [19798.355158] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 after re-connect: [19827.045898] usb 1-2: new high-speed USB device number 13 using xhci_hcd [19827.189861] usb 1-2: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [19827.189872] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [19827.189876] usb 1-2: Product: 802.11n NIC [19827.189880] usb 1-2: Manufacturer: Realtek [19827.572610] usb 1-2: Vendor: Realtek [19827.572619] usb 1-2: Product: \x03802.11n NI [19827.572623] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [19827.572626] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [19827.572633] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 [19827.572638] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [19827.576491] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [19828.407015] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [19828.471488] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [19828.484988] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [19828.484997] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fc e9c=2 ea4=fd eac=3fd eb4=0 ebc=0 ec4=0 ecc=0 [19828.517877] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 looking pretty good so far. Interface respond and it as able to receive and transmit packets $ hcxdumptool -I wlp5s0f3u2 Requesting interface capabilities. This may take some time. Please be patient... interface information: phy idx hw-mac virtual-mac m ifname driver (protocol) --------------------------------------------------------------------------------------------- 10 13 00c0cab067e9 52dedbee35fc + wlp5s0f3u2 rtl8xxxu (NETLINK & WIRELESS EXTENSIONS) available frequencies: frequency [channel] tx-power 2412 [ 1] 20.0 dBm 2417 [ 2] 20.0 dBm 2422 [ 3] 20.0 dBm 2427 [ 4] 20.0 dBm 2432 [ 5] 20.0 dBm 2437 [ 6] 20.0 dBm 2442 [ 7] 20.0 dBm 2447 [ 8] 20.0 dBm 2452 [ 9] 20.0 dBm 2457 [ 10] 20.0 dBm 2462 [ 11] 20.0 dBm 2467 [ 12] 20.0 dBm 2472 [ 13] 20.0 dBm 2484 [ 14] disabled bye-bye [19828.407015] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [19828.471488] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [19828.484988] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [19828.484997] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fc e9c=2 ea4=fd eac=3fd eb4=0 ebc=0 ec4=0 ecc=0 [19828.517877] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 [19963.358671] device wlp5s0f3u2 entered promiscuous mode [19999.414727] device wlp5s0f3u2 left promiscuous mode Logilink: [20133.526541] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [20134.358414] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [20134.422657] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [20134.436629] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [20134.436639] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fc e9c=5 ea4=fd eac=3fd eb4=0 ebc=0 ec4=0 ecc=0 [20134.470647] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 after USB-reset: [20197.755799] usb 1-2: Device still attached, trying to reset [20197.877209] usb 1-2: reset high-speed USB device number 14 using xhci_hcd [20198.406289] usb 1-2: Vendor: Realtek [20198.406299] usb 1-2: Product: \x03802.11n NI [20198.406302] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [20198.406305] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [20198.406312] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [20198.406316] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [20198.410362] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [20199.245487] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [20199.327323] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [20199.403356] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [20199.416846] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 1 [20199.416851] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fe e9c=5 ea4=fd eac=3fd eb4=0 ebc=0 ec4=0 ecc=0 [20199.438913] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 after re-connect: [20234.653431] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [20235.487872] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [20235.551613] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [20235.565467] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [20235.565477] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fe e9c=5 ea4=fd eac=3fd eb4=0 ebc=0 ec4=0 ecc=0 [20235.599000] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 looking pretty good so far. Interface respond and it as able to receive and transmit packets. $ hcxdumptool -I wlp5s0f3u2 $ sudo hcxdumptool -I wlp5s0f3u2 Requesting interface capabilities. This may take some time. Please be patient... interface information: phy idx hw-mac virtual-mac m ifname driver (protocol) --------------------------------------------------------------------------------------------- 13 16 70f11c27aeec b025aac41fd3 + wlp5s0f3u2 rtl8xxxu (NETLINK & WIRELESS EXTENSIONS) available frequencies: frequency [channel] tx-power 2412 [ 1] 20.0 dBm 2417 [ 2] 20.0 dBm 2422 [ 3] 20.0 dBm 2427 [ 4] 20.0 dBm 2432 [ 5] 20.0 dBm 2437 [ 6] 20.0 dBm 2442 [ 7] 20.0 dBm 2447 [ 8] 20.0 dBm 2452 [ 9] 20.0 dBm 2457 [ 10] 20.0 dBm 2462 [ 11] 20.0 dBm 2467 [ 12] 20.0 dBm 2472 [ 13] 20.0 dBm 2484 [ 14] disabled bye-bye Good job. Thanks. I think we can mark it as fixed.
I don't get it. That last patch only prints some more messages. It shouldn't fix anything.
You're right. Epical fail of me. I applied your patch to mine. Now I clean up the source.
Looks like my entire USB subsystem crashed due to the last test series: [26418.746420] INFO: task rmmod:35303 blocked for more than 614 seconds. [26418.746430] Tainted: P OE 6.2.8-arch1-1 #1 [26418.746434] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [26418.746436] task:rmmod state:D stack:0 pid:35303 ppid:35302 flags:0x00000006 [26418.746445] Call Trace: [26418.746448] <TASK> [26418.746455] __schedule+0x3c8/0x12e0 [26418.746467] ? __slab_free+0xe0/0x310 [26418.746474] ? prb_read_valid+0x11/0x30 [26418.746484] schedule+0x5e/0xd0 [26418.746489] klist_remove+0xcc/0x110 [26418.746499] bus_remove_driver+0x4f/0xe0 [26418.746508] usb_deregister+0x7d/0xf0 [26418.746516] __do_sys_delete_module+0x1d5/0x320 [26418.746527] do_syscall_64+0x5f/0x90 [26418.746535] ? ksys_read+0x6f/0xf0 [26418.746540] ? syscall_exit_to_user_mode+0x1b/0x40 [26418.746547] ? do_syscall_64+0x6b/0x90 [26418.746552] ? ksys_read+0x6f/0xf0 [26418.746557] ? syscall_exit_to_user_mode+0x1b/0x40 [26418.746563] ? do_syscall_64+0x6b/0x90 [26418.746569] entry_SYSCALL_64_after_hwframe+0x72/0xdc [26418.746578] RIP: 0033:0x7f1cba921d6b [26418.746604] RSP: 002b:00007ffd6febecd8 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0 [26418.746610] RAX: ffffffffffffffda RBX: 00005557376aa6f0 RCX: 00007f1cba921d6b [26418.746613] RDX: 0000000000000000 RSI: 0000000000000800 RDI: 00005557376aa758 [26418.746616] RBP: 0000000000000000 R08: 1999999999999999 R09: 0000000000000000 [26418.746619] R10: 00007f1cba996ac0 R11: 0000000000000206 R12: 0000000000000000 [26418.746622] R13: 00007ffd6febef40 R14: 00007ffd6febfdf8 R15: 00005557376aa2a0 [26418.746628] </TASK> Ok, I'll go back to the roots and start again at ground lever.
This is the reason why I closed the report: I replaced mdelay(10) by if(rtl8xxxu_read32(priv, 0xea8) == 0) mdelay(25); that worked: [ 50.428908] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [ 50.428915] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [ 50.428919] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [ 50.436225] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [ 51.265642] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 51.319891] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 51.333520] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [ 51.333529] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fe e9c=5 ea4=fd eac=3fd eb4=0 ebc=0 ec4=0 ecc=0 [ 51.365568] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 Than i tried to make it dynamic: rtl8xxxu_read32(priv, 0xea8) as long as it doesn't change. That doesn't work and caused the USB subsystem to crash and I had to reboot. Now testing "if(rtl8xxxu_read32(priv, 0xea8) == 0) mdelay(25);" again. This time I do an USB reset every 10 seconds to see if the warning will appear. I dmesg it looks like this: [ 774.538763] usb 1-2: disconnecting [ 778.066423] usb 1-2: new high-speed USB device number 6 using xhci_hcd [ 778.209043] usb 1-2: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [ 778.209055] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [ 778.209059] usb 1-2: Product: 802.11n NIC [ 778.209063] usb 1-2: Manufacturer: Realtek [ 778.209065] usb 1-2: SerialNumber: 70F11C27AEEC [ 778.599164] usb 1-2: Vendor: Realtek [ 778.599173] usb 1-2: Product: \x03802.11n NI [ 778.599176] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [ 778.599179] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [ 778.599185] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [ 778.599189] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [ 778.603208] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [ 779.428162] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 779.482788] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 779.496331] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [ 779.496338] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fd e9c=4 ea4=fd eac=3fd eb4=0 ebc=0 ec4=0 ecc=0 [ 779.526613] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 ... [ 821.478449] usb 1-2: disconnecting [ 821.478461] usb 1-2: Device still attached, trying to reset [ 821.599603] usb 1-2: reset high-speed USB device number 6 using xhci_hcd [ 822.130450] usb 1-2: Vendor: Realtek [ 822.130460] usb 1-2: Product: \x03802.11n NI [ 822.130463] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [ 822.130466] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [ 822.130472] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [ 822.130476] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [ 822.134128] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [ 822.971867] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 823.044074] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 823.058111] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [ 823.058121] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fe e9c=6 ea4=fd eac=3fd eb4=0 ebc=0 ec4=0 ecc=0 every 5 second I set interface to monitor mode to receive and to transmit some packets. [ 823.080235] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 [ 875.652907] device wlp5s0f3u2 entered promiscuous mode [ 884.265597] device wlp5s0f3u2 left promiscuous mode That is still working, too. [ 986.024814] usb 1-2: disconnecting [ 986.024824] usb 1-2: Device still attached, trying to reset [ 986.148513] usb 1-2: reset high-speed USB device number 6 using xhci_hcd [ 986.681818] usb 1-2: Vendor: Realtek [ 986.681827] usb 1-2: Product: \x03802.11n NI [ 986.681831] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [ 986.681833] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [ 986.681840] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [ 986.681844] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [ 986.685439] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [ 987.510064] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 987.578312] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 987.645313] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 987.659074] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 3 [ 987.659085] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=100 e9c=0 ea4=100 eac=0 eb4=100 ebc=0 ec4=100 ecc=0 [ 987.681555] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0
I see. I was hoping you would apply my last patch without any other changes.
I feel sorry about that, but I'm a bit scattered at the moment, beacause I'm working on hcxdumptool in parallel https://github.com/ZerBea/hcxdumptool/commits/master that include to hunt for some problems on mt76 driver. Unfortunately everything is linked: drivers, USB system, NETLINK, RTNETLINK, hcxdumptool whereas one cannot work without the other.
Register values without my patch: [ 7622.337212] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 [ 7622.337217] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [ 7622.341419] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [ 7623.161187] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 7623.215061] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 7623.298312] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [ 7623.311807] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 1 [ 7623.311811] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fc e9c=2 ea4=fb eac=3fc eb4=0 ebc=0 ec4=0 ecc=0 [ 7623.340445] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0
BTW: Do you see a chance to add active monitor mode to this driver. Feature is part of the MAC80211 stack. I guess it is possible to add this to all stock drivers (like mt76).
On 10/04/2023 20:04, bugzilla-daemon@kernel.org wrote: > https://bugzilla.kernel.org/show_bug.cgi?id=217205 > > --- Comment #32 from Michael (ZeroBeat@gmx.de) --- > Register values without my patch: > [ 7622.337212] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 > [ 7622.337217] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin > [ 7622.341419] usb 1-2: Firmware revision 11.1 (signature 0x88e1) > [ 7623.161187] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! > [ 7623.215061] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! > [ 7623.298312] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! > [ 7623.311807] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 1 > [ 7623.311811] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fc e9c=2 ea4=fb > eac=3fc > eb4=0 ebc=0 ec4=0 ecc=0 > [ 7623.340445] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 > What about when it fails?
You make a good point. Here we go: [11874.160379] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 [11874.160384] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [11874.165137] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [11874.984977] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [11875.038854] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [11875.063350] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [11875.129227] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [11875.142724] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [11875.142727] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fc e9c=2 ea4=fc eac=3fd eb4=0 ebc=0 ec4=0 ecc=0 $ sudo ./resetif /dev/bus/usb/001/013 12049.631253] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 [12049.631258] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [12049.635082] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [12050.453602] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [12050.507478] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! [12050.520975] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [12050.520979] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fc e9c=2 ea4=fc eac=3fc eb4=0 ebc=0 ec4=0 ecc=0 [12050.545294] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0
On 10/04/2023 20:16, bugzilla-daemon@kernel.org wrote: > https://bugzilla.kernel.org/show_bug.cgi?id=217205 > > --- Comment #33 from Michael (ZeroBeat@gmx.de) --- > BTW: > Do you see a chance to add active monitor mode to this driver. > > Feature is part of the MAC80211 stack. I guess it is possible to add this to > all stock drivers (like mt76). > Probably yes. You should open another bug about that so we can discuss it there.
Thanks for the answer. I'll open a bug report/feature request.
On 10/04/2023 21:16, bugzilla-daemon@kernel.org wrote: > https://bugzilla.kernel.org/show_bug.cgi?id=217205 > > --- Comment #35 from Michael (ZeroBeat@gmx.de) --- > You make a good point. Here we go: > > [11874.160379] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 > [11874.160384] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin > [11874.165137] usb 1-2: Firmware revision 11.1 (signature 0x88e1) > [11874.984977] rtl8188eu_rx_iqk_path_a: Path A RX IQK success!> > [11875.038854] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! > [11875.063350] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! > [11875.129227] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! > [11875.142724] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 > [11875.142727] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fc e9c=2 ea4=fc > eac=3fd > eb4=0 ebc=0 ec4=0 ecc=0 So at this point the message "renamed from wlan0" doesn't appear? I assume "iw dev" doesn't show it? The IQ calibration looks fine. One stage of it fails once, it's tried again and it succeeds. > > $ sudo ./resetif /dev/bus/usb/001/013 > > 12049.631253] usb 1-2: RTL8188EU MAC: 00:c0:ca:b0:67:e9 > [12049.631258] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin > [12049.635082] usb 1-2: Firmware revision 11.1 (signature 0x88e1) > [12050.453602] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! > [12050.507478] rtl8188eu_rx_iqk_path_a: Path A RX IQK success! > [12050.520975] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 > [12050.520979] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fc e9c=2 ea4=fc > eac=3fc > eb4=0 ebc=0 ec4=0 ecc=0 > [12050.545294] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 >
That was a copy and paste error: [11875.169574] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 As I mentioned above. The device sometimes need a "second chance". But if this message appeared twice: [11875.038854] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! device will neither receive notr transmit: iw show it hcxdumptool show it but it does not receive packets and it does not transmit packets [15164.437895] usb 1-2: new high-speed USB device number 20 using xhci_hcd [15164.580591] usb 1-2: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [15164.580600] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [15164.580604] usb 1-2: Product: 802.11n NIC [15164.580607] usb 1-2: Manufacturer: Realtek [15164.580609] usb 1-2: SerialNumber: 70F11C27AEEC [15164.967957] usb 1-2: Vendor: Realtek [15164.967964] usb 1-2: Product: \x03802.11n NI [15164.967968] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [15164.967970] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [15164.967977] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec [15164.967981] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [15164.971864] usb 1-2: Firmware revision 11.1 (signature 0x88e1) [15165.809079] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [15165.862204] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [15165.892329] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 [15165.892334] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fe e9c=5 ea4=0 eac=0 eb4=0 ebc=0 ec4=0 ecc=0 [15165.923215] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 It is really hard to reproduce because happen by accident. It is not a problem on a notebook or a desktop. But it is a big problem on a Raspberry Pi that is running headless.
On 10/04/2023 22:11, bugzilla-daemon@kernel.org wrote: > https://bugzilla.kernel.org/show_bug.cgi?id=217205 > > --- Comment #39 from Michael (ZeroBeat@gmx.de) --- > That was a copy and paste error: > [11875.169574] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 > > As I mentioned above. The device sometimes need a "second chance". > > But if this message appeared twice: > [11875.038854] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! > device will neither receive notr transmit: > iw show it > hcxdumptool show it > but it does not receive packets > and it does not transmit packets I see. Are you testing that in station mode or monitor mode? > > [15164.437895] usb 1-2: new high-speed USB device number 20 using xhci_hcd > [15164.580591] usb 1-2: New USB device found, idVendor=0bda, idProduct=8179, > bcdDevice= 0.00 > [15164.580600] usb 1-2: New USB device strings: Mfr=1, Product=2, > SerialNumber=3 > [15164.580604] usb 1-2: Product: 802.11n NIC > [15164.580607] usb 1-2: Manufacturer: Realtek > [15164.580609] usb 1-2: SerialNumber: 70F11C27AEEC > [15164.967957] usb 1-2: Vendor: Realtek > [15164.967964] usb 1-2: Product: \x03802.11n NI > [15164.967968] usb 1-2: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff > [15164.967970] usb 1-2: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, > WiFi=1, BT=0, GPS=0, HI PA=0 > [15164.967977] usb 1-2: RTL8188EU MAC: 70:f1:1c:27:ae:ec > [15164.967981] usb 1-2: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin > [15164.971864] usb 1-2: Firmware revision 11.1 (signature 0x88e1) > [15165.809079] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! > [15165.862204] usb 1-2: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! I don't understand how this part here is possible. There should be more messages, either with "failed" or "success". > [15165.892329] usb 1-2: rtl8188eu_phy_iq_calibrate: candidate is 0 > [15165.892334] usb 1-2: rtl8188eu_phy_iq_calibrate: e94=fe e9c=5 ea4=0 eac=0 > eb4=0 ebc=0 ec4=0 ecc=0 > [15165.923215] rtl8xxxu 1-2:1.0 wlp5s0f3u2: renamed from wlan0 > > It is really hard to reproduce because happen by accident. > It is not a problem on a notebook or a desktop. But it is a big problem on a > Raspberry Pi that is running headless. >
Nevermind, I see how that can happen.
"I don't understand how this part here is possible. There should be more messages, either with "failed" or "success"." Me too, so you are in good company. The tests (connect device, re-connect, USB reset) are done in managed mode, only. "rtl8188eu_rx_iqk_path_a: Path A RX IQK failed!" is show before I set monitor mode. The tests if the device transmit/receive are done in monitor mode, only. I monitor the traffic directly via hcxdumptool and in parallel with Wireshark on the same interface an on a second reference interface. I coded hcxdumptool that it use RTNETLINK to set interface up and down and NL80211 to set monitor mode, to set bandwidth, bit rate and frequency. I set bandwidth, bit rate and frequency via a single NL80211 message: https://github.com/ZerBea/hcxdumptool/blob/master/hcxdumptool.c#L3047 I do not use libnl or libnl-tiny to create and send NETLINK messages, due to performance reasons. Realtek and iwlwifi drivers "do not like" WIRELESS EXTENSIONS so I had to refactor hcxdumptool. Now WIRELESS EXTENSIONS are retired in favor of NL80211. This is still in progress. Next release will be v 6.3.0 on 1th of May. My reference interface/driver is a mt76 interface/driver. Other drivers (Realtek, Atheros, iwlwifi) are compared to this reference.
This driver: https://github.com/kimocoder/realtek_rtwifi is working after merge of this commits: https://github.com/kimocoder/realtek_rtwifi/commit/143cf45c4af15a47b5287b1f989e567190cb8348 https://github.com/kimocoder/realtek_rtwifi/commit/cd2d02fdaf46f62f8303c2ac87ad34302a1f82ed https://github.com/kimocoder/realtek_rtwifi/commit/81005acf7fa4eb651fb7d3c0d00080e680238a14 https://github.com/kimocoder/realtek_rtwifi/commit/c58f779e75a2cd75d2a80547b5b92b07b274e34c https://github.com/kimocoder/realtek_rtwifi/commit/bf22bdb706ba53f044b1232c262f632ad7534d1a "rtl8188eu_rx_iqk_path_a: Path A RX IQK failed!" is gone: https://github.com/kimocoder/realtek_rtwifi/issues/36#issuecomment-1513644758
I think it would be best if we discuss only the problems you have with the code obtained from official sources, like your distro's kernel package, or: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/ or: https://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next.git/ kimocoder's repository has a lot of confusing commits. I don't know what's going on there. You're having this intermittent problem with the monitor mode and IQ calibration even with the driver from kernel 6.3, right? I think you said so. If that's the case, then the problem should still be there.
Created attachment 304160 [details] RTL8188EU firmware v28
I fully agree. The problems still exist on 6.3-rc7. But the driver mentioned above may help me to figure out what's wrong. On the bright side, while diving into kimocoder's driver code, I guess that cause it is not located here: https://bugzilla.kernel.org/show_bug.cgi?id=217205#c20
Do you think it is related to the firmware?
I still use this one: https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/tree/rtlwifi/rtl8188eufw.bin
I think it's worth trying the newer firmware.
Do you plan to commit v28 to linux-firmware.git?
Only if it fixes a bug.
I think v28 fixed a bug on ID 0bda:8179 Realtek Semiconductor Corp. RTL8188EUS 802.11n Wireless Network Adapter [ 8426.899088] usb 5-1.1.3: new high-speed USB device number 12 using xhci_hcd [ 8427.090428] usb 5-1.1.3: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [ 8427.090437] usb 5-1.1.3: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [ 8427.090441] usb 5-1.1.3: Product: 802.11n NIC [ 8427.090443] usb 5-1.1.3: Manufacturer: Realtek [ 8427.162049] usb 5-1.1.3: Vendor: Realtek [ 8427.162054] usb 5-1.1.3: Product: \x03802.11n NI [ 8427.162057] usb 5-1.1.3: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff [ 8427.162059] usb 5-1.1.3: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [ 8427.162065] usb 5-1.1.3: RTL8188EU MAC: 00:c0:ca:b0:67:e9 [ 8427.162068] usb 5-1.1.3: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [ 8427.162124] usb 5-1.1.3: Firmware revision 28.0 (signature 0x88e1) [ 8427.674562] rtl8xxxu 5-1.1.3:1.0 wlp39s0f3u1u1u3: renamed from wlan0 [ 8463.714876] usb 5-1.1.1: USB disconnect, device number 9 [ 8471.791443] device wlp39s0f3u1u1u3 entered promiscuous mode [ 8491.684973] device wlp39s0f3u1u1u3 left promiscuous mode [ 8471.791443] device wlp39s0f3u1u1u3 entered promiscuous mode $ hcxdumptool -I wlp39s0f3u1u1u3 Requesting interface capabilities. This may take some time. Please be patient... interface information: phy idx hw-mac virtual-mac m ifname driver (protocol) --------------------------------------------------------------------------------------------- 4 5 00c0cab063a9 c8aacc5f63a9 + wlp39s0f3u1u1u3 rtl8xxxu (NETLINK & WIRELESS EXTENSIONS) available frequencies: frequency [channel] tx-power of Regulatory Domain: DE 2412 [ 1] 20.0 dBm 2417 [ 2] 20.0 dBm 2422 [ 3] 20.0 dBm 2427 [ 4] 20.0 dBm 2432 [ 5] 20.0 dBm 2437 [ 6] 20.0 dBm 2442 [ 7] 20.0 dBm 2447 [ 8] 20.0 dBm 2452 [ 9] 20.0 dBm 2457 [ 10] 20.0 dBm 2462 [ 11] 20.0 dBm 2467 [ 12] 20.0 dBm 2472 [ 13] 20.0 dBm 2484 [ 14] disabled bye-bye [ 8491.684973] device wlp39s0f3u1u1u3 left promiscuous mode Thanks for sharing the new firmware.
rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! is gone on this driver: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/net/wireless/realtek/rtl8xxxu?h=v6.3-rc7
Do you know what changed from: Firmware revision 11.1 (signature 0x88e1) to Firmware revision 28.0 (signature 0x88e1)
No, I don't know. The manufacturer doesn't say much about the changes in the firmware.
Ok, thanks. Tested this three adapters: Bus 005 Device 030: ID 2357:010c TP-Link TL-WN722N v2/v3 [Realtek RTL8188EUS] Bus 005 Device 031: ID 0bda:8179 Realtek Semiconductor Corp. RTL8188EUS 802.11n Wireless Network Adapter Bus 005 Device 029: ID 7392:b811 Edimax Technology Co., Ltd Edimax N150 Adapter And they are working as expected running v28. The problem, reading device informatin is related to the driver and still exists: [24872.589992] usb 5-1.1.3: Product: \x03802.11n NI [24872.589995] usb 5-1.1.3: Serial: \xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff I think we can close this report.
I sent a patch to „fix” the other thing: https://lore.kernel.org/linux-wireless/a2a7d9df-0529-7890-3522-48dce613753f@gmail.com/
Great. Thanks for your effort.
Fixed by fw v28.
Still fixed? Just making sure.
Yes, working as expected. I'm still running fw v28. Arch Linux is on kernel: $ uname -r 6.3.1-arch1-1 And I was really surprised by this dmesg warning: [ 2770.939021] warning: hcxdumptool uses wireless extensions which will stop working for Wi-Fi 7 hardware; use nl80211 Took me a couple of hours to remove entire WEXT stuff and to replace it by NL80211. In hindsight, I am grateful for the warning. RTNETLINK and NL80211 are working smooth (I don't use libnl). Additional NETLINK offer some nice new (driver) features.
On 10/05/2023 12:41, bugzilla-daemon@kernel.org wrote: > https://bugzilla.kernel.org/show_bug.cgi?id=217205 > > --- Comment #61 from Michael (ZeroBeat@gmx.de) --- > Yes, working as expected. I'm still running fw v28. > That's good. I'll send the v28 firmware upstream soon-ish. > Arch Linux is on kernel: > $ uname -r > 6.3.1-arch1-1 > > And I was really surprised by this dmesg warning: > [ 2770.939021] warning: hcxdumptool uses wireless extensions which will stop > working for Wi-Fi 7 hardware; use nl80211 > > Took me a couple of hours to remove entire WEXT stuff and to replace it by > NL80211. > > In hindsight, I am grateful for the warning. RTNETLINK and NL80211 are > working > smooth (I don't use libnl). Additional NETLINK offer some nice new (driver) > features. > Yes, Wifi 7 introduces some things which are incompatible with the Wireless Extensions: https://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next.git/commit/?id=4ca69027691a
There is a major advantage of the new firmware: I noticed some calibration problems on USB cables > 1m in combination with a high TX power device. This is gone with fw 28, too.
Looks like this problem is back on kernel 6.6.1 Once a new kernel has been released, I do some driver tests regarding monitor mode an packet injection: https://github.com/ZerBea/hcxdumptool/discussions/361 This test failed: https://github.com/ZerBea/hcxdumptool/discussions/361#discussioncomment-7550830 ALFA AWUS036NHV ID 0bda:8179 Realtek Semiconductor Corp. RTL8188EUS 802.11n Wireless Network Adapter driver: rtl8xxxu (Linux stock kernel driver) 213 packet(s) captured 1 SHB written to pcapng dumpfile 1 IDB written to pcapng dumpfile 1 ECB written to pcapng dumpfile 17 EPB written to pcapng dumpfile exit on EAPOL M1M2 bye-bye real 0m6,546s user 0m0,000s sys 0m0,009s Important notice: The first attack was successful but the driver died after the first test and device has to be reconnect to the USB port! Neither tshark nor Wireshark nor hcxdumptool received packers any longer. run hcxdumptool twice: Warning: no packets received (frame injection may not work as expected) Possible reasons: no AP in range frames are filtered out by BPF driver is broken driver does not support frame injection 1 SHB written to pcapng dumpfile 1 IDB written to pcapng dumpfile 1 ECB written to pcapng dumpfile exit on sigterm bye-bye confirm that device is in monitor mode: $ iw dev phy#6 Interface wlp22s0f0u9u4 ifindex 9 wdev 0x600000001 addr 00:c0:ca:b0:67:e9 type monitor channel 10 (2457 MHz), width: 20 MHz (no HT), center1: 2457 MHz txpower 20.00 dBm multicast TXQ: qsz-byt qsz-pkt flows drops marks overlmt hashcol tx-bytes tx-packets 0 0 0 0 0 0 0 0 0 run tshark to capture packets: $ tshark -i wlp22s0f0u9u3 Capturing on 'wlp22s0f0u9u3' ** (tshark:5627) 08:24:47.585105 [Main MESSAGE] -- Capture started. ** (tshark:5627) 08:24:47.585157 [Main MESSAGE] -- File: "/tmp/wireshark_wlp22s0f0u9u3ZKCRE2.pcapng" ^Ctshark: 0 packets captured sometimes dmesg print a warning: $ dmesg [ 1697.283905] usb 1-9.3: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [ 1697.418898] usb 1-9.3: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! [ 1697.712900] usb 1-9.3: rtl8188eu_rx_iqk_path_a: Path A RX IQK failed! but mostly not: $ dmesg [ 1748.913402] usb 1-9.3: New USB device found, idVendor=0bda, idProduct=8179, bcdDevice= 0.00 [ 1748.913407] usb 1-9.3: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [ 1748.913410] usb 1-9.3: Product: 802.11n NIC [ 1748.913411] usb 1-9.3: Manufacturer: Realtek [ 1751.979371] usb 1-9.3: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [ 1751.979377] usb 1-9.3: RTL8188EU MAC: 00:c0:ca:b0:67:e9 [ 1751.979380] usb 1-9.3: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [ 1751.979560] usb 1-9.3: Firmware revision 28.0 (signature 0x88e1) [ 1756.744526] rtl8xxxu 1-9.3:1.0 wlp22s0f0u9u3: renamed from wlan0 [ 1758.045644] rtl8xxxu 1-9.3:1.0 wlp22s0f0u9u3: entered promiscuous mode The problem is not related to the device itself. All tested RTL8188EUS devices running this driver show the same behavior https://github.com/ZerBea/hcxdumptool/discussions/361#discussioncomment-7551692 https://github.com/ZerBea/hcxdumptool/discussions/361#discussioncomment-7554097 https://github.com/ZerBea/hcxdumptool/discussions/361#discussioncomment-7554193 while RTL8192CU devices are not affected: https://github.com/ZerBea/hcxdumptool/discussions/361#discussioncomment-7554247
Maybe a regression (selected check box regression), but I'm not sure.
Same behavior on kernel 6.5.0.
What is the exact hcxdumptool command I should run to reproduce this?
That should do it: $ hcxdumptool -i INTERFACENAME --rds=1 Please make sure that all services that take access to the device are stopped before you run hcxdumptool. It is not mandatory to set monitor mode by iw before, because hcxdumptool will do this, as well as set the device up. After a while stop hcxdumptool than restart it. Packets are no longer received. You can confirm this running: $ tshark -i SAME_INTERFACE BTW: rtl8xxxu (RTL8192CU) is not affected https://github.com/ZerBea/hcxdumptool/discussions/361#discussioncomment-7554247 I'll say that we can rule out an error in the test environment.
I'm not sure if this test can be done with e.g. aireplay-ng too because hcxdumptool produce a much higher work load.
Also we can rule out distribution dependent issues. I compiled kernels on Debian and kernels on Arch. Same behavior on both.
Also we can rule out hardware dependent issues. Tests are done on a RYZEN machine and on an Intel machine.
I can reproduce it with my TP-Link TL-WN725N. I'll have to see what is going on in the driver.
Thanks. I'm glad to hear that you can reproduce it. BTW: You can monitor the entire traffic between hcxdumptool and the kernel: Create a NETLINK monitor interface (nlmon) and run Wireshark on nlmon. Than start hcxdumptool in parallel.
The interface is completely controlled using RTNETLINK and NETLINK (NL80211). 802.11 frames will be read via a raw packet socket and they will be sent via raw packet socket. I don't think that the problem is related to the NETLINK control interface.
Please take care when running hcxdumptool, because it is an extreme aggressive penetration testing tool.
(In reply to rtl8821cerfe2 from comment #72) > I can reproduce it with my TP-Link TL-WN725N. I'll have to see what is going > on in the driver. This device is on my test list, too: https://github.com/ZerBea/hcxdumptool/discussions/361#discussioncomment-7554097
Does this fix it? diff --git a/rtl8xxxu_8188e.c b/rtl8xxxu_8188e.c index 6d0f975..07ddbc6 100644 --- a/rtl8xxxu_8188e.c +++ b/rtl8xxxu_8188e.c @@ -1298,7 +1298,7 @@ static void rtl8188e_disable_rf(struct rtl8xxxu_priv *priv) rtl8xxxu_write32(priv, REG_OFDM0_TRX_PATH_ENABLE, val32); /* Power down RF module */ - rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_AC, 0); + // rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_AC, 0); rtl8188eu_active_to_emu(priv); }
Looking good: TP-Link TL-WN722N v2 is working (receive and send packets) as expected: ID 2357:010c TP-Link TL-WN722N v2/v3 [Realtek RTL8188EUS] tried that for several times and everything is fine, dmesg does not show a warning any longer: [46850.077821] rtl8xxxu 5-2.4:1.0 wlp48s0f4u2u4: renamed from wlan0 [46874.418349] rtl8xxxu 5-2.4:1.0 wlp48s0f4u2u4: entered promiscuous mode [46882.492930] rtl8xxxu 5-2.4:1.0 wlp48s0f4u2u4: left promiscuous mode [46885.975786] rtl8xxxu 5-2.4:1.0 wlp48s0f4u2u4: entered promiscuous mode [46893.746354] rtl8xxxu 5-2.4:1.0 wlp48s0f4u2u4: left promiscuous mode [46895.860573] rtl8xxxu 5-2.4:1.0 wlp48s0f4u2u4: entered promiscuous mode [46902.563128] rtl8xxxu 5-2.4:1.0 wlp48s0f4u2u4: left promiscuous mode Also tested TP-Link TL-WN725N and everything is fine, too: ID 0bda:8179 Realtek Semiconductor Corp. RTL8188EUS 802.11n Wireless Network Adapter Great job, thanks. BTW: I think we can remove this information "This Realtek USB WiFi dongle (0x2357:0x010c) is untested!", too: TP-Link TL-WN722N v2: ID 2357:010c TP-Link TL-WN722N v2/v3 [Realtek RTL8188EUS] [46848.552865] usb 5-2.4: new high-speed USB device number 24 using xhci_hcd [46848.647172] usb 5-2.4: New USB device found, idVendor=2357, idProduct=010c, bcdDevice= 0.00 [46848.647183] usb 5-2.4: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [46848.647186] usb 5-2.4: Product: 802.11n NIC [46848.647188] usb 5-2.4: Manufacturer: Realtek [46848.647190] usb 5-2.4: SerialNumber: 00E04C0001 [46848.647954] usb 5-2.4: This Realtek USB WiFi dongle (0x2357:0x010c) is untested! [46848.647957] usb 5-2.4: Please report results to Jes.Sorensen@gmail.com [46849.137278] usb 5-2.4: Dumping efuse for RTL8188EU (0x200 bytes): [46849.137282] 00000000: 29 81 00 6c 0b 00 00 00 00 0c 00 00 00 00 00 00 )..l............ [46849.137285] 00000010: 2b 2b 2b 2b 2b 2b 2f 2f 2f 2e 2e f1 ff ff ff ff ++++++///....... [46849.137286] 00000020: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137288] 00000030: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137289] 00000040: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137290] 00000050: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137291] 00000060: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137293] 00000070: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137294] 00000080: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137295] 00000090: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137297] 000000a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137298] 000000b0: ff ff ff ff ff ff ff ff a1 2f 17 00 00 00 00 00 ........./...... [46849.137299] 000000c0: 00 01 00 10 00 00 00 00 00 03 ff ff ff ff ff ff ................ [46849.137301] 000000d0: 57 23 0c 01 43 66 00 9c a2 f4 09 3a f0 09 03 52 W#..Cf.....:...R [46849.137302] 000000e0: 65 61 6c 74 65 6b 0d 03 38 30 32 2e 31 31 6e 20 ealtek..802.11n [46849.137303] 000000f0: 4e 49 43 0c 03 30 30 45 30 34 43 30 30 30 31 00 NIC..00E04C0001. [46849.137304] 00000100: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137306] 00000110: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137307] 00000120: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137308] 00000130: 81 ae 96 2d 03 93 96 11 fc 8c 00 11 9b ff ff ff ...-............ [46849.137310] 00000140: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137311] 00000150: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137312] 00000160: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137314] 00000170: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137315] 00000180: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137316] 00000190: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137317] 000001a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137319] 000001b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137320] 000001c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137321] 000001d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137323] 000001e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137324] 000001f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ [46849.137325] usb 5-2.4: RTL8188EU rev D (TSMC) romver 0, 1T1R, TX queues 2, WiFi=1, BT=0, GPS=0, HI PA=0 [46849.137329] usb 5-2.4: RTL8188EU MAC: 9c:a2:f4:09:3a:f0 [46849.137331] usb 5-2.4: rtl8xxxu: Loading firmware rtlwifi/rtl8188eufw.bin [46849.137515] usb 5-2.4: Firmware revision 28.0 (signature 0x88e1)
How have you discovered this so fast?
Experience, I guess. :)
Good to hear that. I'm going to close this report, because the patch is working like a charm. Also I'll remove the driver from "hcxdumptool's not suitable driver list". Thanks again. Cheers Mike
Reopened this bug report, because the patch mentioned here: https://bugzilla.kernel.org/show_bug.cgi?id=217205#c77 has not been applied to latest kernel 6.8 rc2
On 29/01/2024 16:02, bugzilla-daemon@kernel.org wrote: > https://bugzilla.kernel.org/show_bug.cgi?id=217205 > > Michael (ZeroBeat@gmx.de) changed: > > What |Removed |Added > ---------------------------------------------------------------------------- > Status|RESOLVED |REOPENED > Resolution|PATCH_ALREADY_AVAILABLE |--- > > --- Comment #82 from Michael (ZeroBeat@gmx.de) --- > Reopened this bug report, because the patch mentioned here: > https://bugzilla.kernel.org/show_bug.cgi?id=217205#c77 > has not been applied to latest kernel 6.8 rc2 > I didn't do anything with the patch because it's not a good fix. It revealed where the problem is (probably), but not what it is. I need to do more digging.
I fully agree, because the driver needs a lot more attention. While I tried to add a new device ID: TP-Link TL-WN8200ND v3 ID 2357:0126 TP-Link 802.11n NIC I noticed that the patch has not been applied and I thought it might be a good idea to reopen the report until we have a final solution.