On 15/01/2025 17:51, Bitterblue Smith wrote: > On 10/01/2025 00:42, Bitterblue Smith wrote: >> On 15/07/2024 16:54, Jakub Kicinski wrote: >>> On Mon, 15 Jul 2024 17:47:14 +0800 WangYuli wrote: >>>> Some mt7601 devices cannot establish a connection properly. >>>> This patch fixes the issue. >>>> We do not know why, but it just works. >>> >>> Any chance we can gate this on the version of EEPROM or chip or >>> something else? It'd be good to avoid regressions on older devices. >>> Or possibly - could we issue the MCU command as a test and disable >>> the calibration if it fails? >> >> (Removed some people and lists from CC because they seemed >> unnecessary.) >> >> Hi Jakub, >> >> I have Tenda W311MI2 which works with vanilla mt7601u, and a no-name >> "free driver" device from Aliexpress which doesn't. They have the >> same USB IDs. >> >> The Tenda: >> >> Jan 09 22:39:40 ideapad2 kernel: mt7601u 1-2:1.0: ASIC revision: 76010001 MAC revision: 76010500 >> Jan 09 22:39:40 ideapad2 kernel: mt7601u 1-2:1.0: Firmware Version: 0.1.00 Build: 7640 Build time: 201302052146____ >> Jan 09 22:39:40 ideapad2 kernel: mt7601u 1-2:1.0: EEPROM ver:0d fae:00 >> >> 00000000: 01 76 00 0d 50 2b 73 e8 19 76 ff ff ff ff ff ff .v..P+s..v...... >> 00000010: 01 02 8f 14 01 76 00 00 4a 00 01 00 80 50 08 00 .....v..J....P.. >> 00000020: 03 00 02 00 00 00 ff 01 40 60 80 ff ff ff ff ff ........@`...... >> 00000030: ff ff ff ff 11 ff 04 28 ff ff 78 01 ff ff 99 99 .......(..x..... >> 00000040: 8c 88 ff 07 08 00 00 00 00 00 00 00 00 00 00 00 ................ >> 00000050: 00 00 1b 1b 1b 1b 1b 1b 1c 1c 1c 1c 1c 1c 1c 1c ................ >> 00000060: ff ff ff ff ff ff ff ff ff ff ff ff ff ff 80 00 ................ >> 00000070: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ >> 00000080: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> 00000090: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> 000000a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> 000000b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> 000000c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> 000000d0: 20 f9 00 00 00 00 00 00 00 00 00 00 00 00 05 05 ............... >> 000000e0: 03 03 04 00 00 00 02 00 00 00 02 00 00 00 02 00 ................ >> 000000f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> >> The no-name: >> >> Jan 09 21:50:34 ideapad2 kernel: mt7601u 1-4:1.0: ASIC revision: 76010001 MAC revision: 76010500 >> Jan 09 21:50:34 ideapad2 kernel: mt7601u 1-4:1.0: EEPROM ver:0d fae:00 >> >> 00000000: 01 76 00 0d 00 50 11 00 5b 9b ff ff ff ff ff ff .v...P..[....... >> 00000010: 01 02 8f 14 01 76 00 00 4a 00 01 00 80 50 08 00 .....v..J....P.. >> 00000020: 03 00 02 00 00 00 ff 01 40 60 80 ff ff ff ff ff ........@`...... >> 00000030: ff ff ff ff 11 ff 04 10 ff ff 82 01 ff ff 99 99 ................ >> 00000040: 8c 88 ff 07 08 00 00 00 00 00 00 00 00 00 00 00 ................ >> 00000050: 00 00 05 05 05 04 04 04 05 05 06 06 07 07 07 07 ................ >> 00000060: ff ff ff ff ff ff ff ff ff ff ff ff ff ff 80 00 ................ >> 00000070: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ >> 00000080: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> 00000090: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> 000000a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> 000000b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> 000000c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> 000000d0: 1e f9 00 00 00 00 00 00 00 00 00 00 00 00 05 05 ................ >> 000000e0: 03 03 04 00 00 00 02 00 00 00 02 00 00 00 02 00 ................ >> 000000f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................ >> >> The no-name device seems to die after uploading the third chunk of >> the firmware: >> >> Jan 09 16:48:43 ideapad2 kernel: mt7601u 1-4:1.0: Firmware Version: 0.1.00 Build: 7640 Build time: 201302052146____ >> Jan 09 16:48:43 ideapad2 kernel: __mt7601u_dma_fw:327 len 14336 dst_addr 64 >> Jan 09 16:48:43 ideapad2 kernel: __mt7601u_dma_fw:327 len 14336 dst_addr 14400 >> Jan 09 16:48:43 ideapad2 kernel: __mt7601u_dma_fw:327 len 14336 dst_addr 28736 >> Jan 09 16:48:43 ideapad2 kernel: mt7601u 1-4:1.0: Vendor request req:07 off:09a8 failed:-110 >> Jan 09 16:48:43 ideapad2 kernel: mt7601u 1-4:1.0: Vendor request req:02 off:09a8 failed:-110 >> Jan 09 16:48:43 ideapad2 kernel: mt7601u 1-4:1.0: Vendor request req:07 off:0734 failed:-110 >> Jan 09 16:48:43 ideapad2 kernel: mt7601u 1-4:1.0: Vendor request req:42 off:0230 failed:-110 >> Jan 09 16:48:43 ideapad2 kernel: mt7601u 1-4:1.0: Vendor request req:07 off:0080 failed:-110 >> Jan 09 16:48:43 ideapad2 kernel: mt7601u 1-4:1.0: Vendor request req:02 off:0080 failed:-110 >> Jan 09 16:48:43 ideapad2 kernel: mt7601u 1-4:1.0: Vendor request req:02 off:0080 failed:-110 >> Jan 09 16:48:43 ideapad2 kernel: mt7601u 1-4:1.0: probe with driver mt7601u failed with error -110 >> >> Removing the call to mt7601u_vendor_reset() fixes that problem. >> The firmware uploads, the device connects to my network. The >> Tenda still works. The driver included in Windows 10 doesn't do >> that reset with either device. >> >> At first I tried with len=8192 because that's what the Windows >> driver does but it didn't help. >> >> After fixing the firmware upload, I had some USB disconnects with >> the no-name device. They were kind of random. Now I'm testing it >> with those two MCU_CAL_DPD removed. The Windows driver doesn't >> seem to use MCU_CAL_DPD with the Tenda. I didn't check with the >> other one. >> >> I hope this helps. > > So after using the no-name device some more, I can say that only > the mt7601u_vendor_reset() needs to be removed to make this device > work. > > The USB disconnects only happen in one port, no idea why. They > happen with or without the MCU_CAL_DPD. I pulled it apart and disconnected pins 9, 10, and 25 (the CHIP_MODE pins) according to the advice found here: https://github.com/openwrt/mt76/issues/393#issuecomment-1092754717 Now the CDROM mode is gone and the device works with unpatched mt7601u. So I guess you can make it call mt7601u_vendor_reset() only when the device doesn't have the CDROM mode... if there is a way for mt7601u to know this.
