https://bugzilla.kernel.org/show_bug.cgi?id=60824 --- Comment #85 from Sergey Kondakov (virtuousfox@xxxxxxxxx) --- Created attachment 289871 --> https://bugzilla.kernel.org/attachment.cgi?id=289871&action=edit CSR_ds4_fail.btsnoop (In reply to Swyter from comment #80) > Hi, Andrés. Thanks for testing v1! I just submitted a revised v2 after > following the recommendations of Marcel Holtmann and finding a better way of > classifying them by type: > > https://patchwork.kernel.org/patch/11622161/ > > > In your case, and as it says after iProduct in your lsusb -v dump, yours > should be a legit Cambridge Silicon Radio CSR8510 A10 chip and detected as > such. > > Those have the HCI_QUIRK_BROKEN_STORED_LINK_KEY bug. > > Same as my dongle. Give v2 a go. :) Tried it for my DS4. Unfortunately the result is the same as when I tried to hardcode my ID: gamepad thinks that it's paired but system does not register it. I've fiddled about with `btmon` and it seems that everything goes wrong at: HCI Event: Role Change (0x12) plen 8 Status: Role Switch Failed (0x35) Role: Slave (0x01) You may read full dump with `btmon -r <file>` `dmesg --color=always | grep -i CSR` says: [ 11.516062] Bluetooth: hci0: CSR: New controller detected; bcdDevice=0x8891, HCI manufacturer=10, HCI rev=0x811, LMP subver=0x811 [ 11.516065] Bluetooth: hci0: CSR: Modern CSR controller type detected -- You are receiving this mail because: You are the assignee for the bug.
