kernel/linux.git/drivers/thunderbolt/ctl.c, branch v5.15.209

thunderbolt: Do not double dequeue a configuration request

2025-06-27T10:05:09+00:00

commit 0f73628e9da1ee39daf5f188190cdbaee5e0c98c upstream. Some of our devices crash in tb_cfg_request_dequeue(): general protection fault, probably for non-canonical address 0xdead000000000122 CPU: 6 PID: 91007 Comm: kworker/6:2 Tainted: G U W 6.6.65 RIP: 0010:tb_cfg_request_dequeue+0x2d/0xa0 Call Trace: ? tb_cfg_request_dequeue+0x2d/0xa0 tb_cfg_request_work+0x33/0x80 worker_thread+0x386/0x8f0 kthread+0xed/0x110 ret_from_fork+0x38/0x50 ret_from_fork_asm+0x1b/0x30 The circumstances are unclear, however, the theory is that tb_cfg_request_work() can be scheduled twice for a request: first time via frame.callback from ring_work() and second time from tb_cfg_request(). Both times kworkers will execute tb_cfg_request_dequeue(), which results in double list_del() from the ctl->request_queue (the list poison deference hints at it: 0xdead000000000122). Do not dequeue requests that don't have TB_CFG_REQUEST_ACTIVE bit set. Signed-off-by: Sergey Senozhatsky Cc: stable@vger.kernel.org Signed-off-by: Mika Westerberg Signed-off-by: Greg Kroah-Hartman

thunderbolt: Use the actual buffer in tb_async_error()

2022-09-08T10:28:05+00:00

commit eb100b8fa8e8b59eb3e5fc7a5fd4a1e3c5950f64 upstream. The received notification packet is held in pkg->buffer and not in pkg itself. Fix this by using the correct buffer. Fixes: 81a54b5e1986 ("thunderbolt: Let the connection manager handle all notifications") Cc: stable@vger.kernel.org Signed-off-by: Mika Westerberg Signed-off-by: Greg Kroah-Hartman

thunderbolt: Decrease control channel timeout for software connection manager

2021-03-18T15:25:30+00:00

When the firmware connection manager is not proxying between the software and the hardware we can decrease the timeout for control packets significantly. The USB4 spec recommends 10 ms +- 1 ms but we use slightly larger value (100 ms) which is recommendation from Intel Thunderbolt firmware folks. When firmware connection manager is running then we keep using the existing 5000 ms. To implement this we move the control channel allocation to tb_domain_alloc(), and pass the timeout from that function to the tb_ctl_alloc(). Then make both connection manager implementations pass the timeout when they alloc the domain structure. While there update kernel-doc of struct tb_ctl to match the reality. Signed-off-by: Mika Westerberg

thunderbolt: Do not pass timeout for tb_cfg_reset()

2021-03-18T15:25:30+00:00

There is only one user for this function and it passes the default timeout to it anyway, so remove the parameter completely. This is also needed in the subsequent patch where we allow connection manager implementations to use different timeout for non-raw control channel messages. Signed-off-by: Mika Westerberg

thunderbolt: Disable retry logic for intra-domain control packets

2021-03-18T15:25:30+00:00

In most cases the response packet is lost because the router in question was disconnected by the user. Resending the control packet in that case just adds unnecessary delays, so disable that for intra-domain control packets. For inter-domain (XDomain) packets we continue retrying. This also aligns the driver better what the Intel connection manager firmware is doing. Signed-off-by: Mika Westerberg

thunderbolt: ctl: Fix kernel-doc descriptions of non-static functions

2021-02-04T07:38:37+00:00

Fix kernel-doc descriptions of all non-static functions and struct tb_cfg. Gets rid of several warnings on W=1 builds too. Reported-by: Lee Jones Signed-off-by: Mika Westerberg Reviewed-by: Lee Jones

thunderbolt: ctl: Demote non-conformant kernel-doc headers

2021-01-28T09:55:28+00:00

Fixes the following W=1 kernel build warning(s): drivers/thunderbolt/ctl.c:38: warning: expecting prototype for struct tb_cfg. Prototype was for struct tb_ctl instead drivers/thunderbolt/ctl.c:350: warning: Function parameter or member 'ctl' not described in 'tb_ctl_tx' drivers/thunderbolt/ctl.c:350: warning: Function parameter or member 'data' not described in 'tb_ctl_tx' drivers/thunderbolt/ctl.c:350: warning: Function parameter or member 'len' not described in 'tb_ctl_tx' drivers/thunderbolt/ctl.c:350: warning: Function parameter or member 'type' not described in 'tb_ctl_tx' drivers/thunderbolt/ctl.c:350: warning: expecting prototype for tb_cfg_tx(). Prototype was for tb_ctl_tx() instead drivers/thunderbolt/ctl.c:383: warning: Function parameter or member 'ctl' not described in 'tb_ctl_handle_event' drivers/thunderbolt/ctl.c:383: warning: Function parameter or member 'type' not described in 'tb_ctl_handle_event' drivers/thunderbolt/ctl.c:383: warning: Function parameter or member 'pkg' not described in 'tb_ctl_handle_event' drivers/thunderbolt/ctl.c:383: warning: Function parameter or member 'size' not described in 'tb_ctl_handle_event' Cc: Andreas Noever Cc: Michael Jamet Cc: Yehezkel Bernat Cc: linux-usb@vger.kernel.org Signed-off-by: Lee Jones [ mw: Demote only static functions ] Signed-off-by: Mika Westerberg

thunderbolt: Return -ENOTCONN when ERR_CONN is received

2020-11-30T11:39:24+00:00

This allows the calling code to distinguish if the error was due to ERR_CONN (adapter is disconneced or disabled) or something else. Will be needed in USB4 router NVM update in the following patch. Signed-off-by: Mika Westerberg

thunderbolt: Add support for end-to-end flow control

2020-11-11T07:20:16+00:00

USB4 spec defines end-to-end (E2E) flow control that can be used between hosts to prevent overflow of a RX ring. We previously had this partially implemented but that code was removed with commit 53f13319d131 ("thunderbolt: Get rid of E2E workaround") with the idea that we add it back properly if there ever is need. Now that we are going to add DMA traffic test driver (in subsequent patches) this can be useful. For this reason we modify tb_ring_alloc_rx/tx() so that they accept RING_FLAG_E2E and configure the hardware ring accordingly. The RX side also requires passing TX HopID (e2e_tx_hop) used in the credit grant packets. Signed-off-by: Mika Westerberg Acked-by: Yehezkel Bernat Reviewed-by: Greg Kroah-Hartman

thunderbolt: Log correct zeroX entries in decode_error()

2020-09-16T11:57:46+00:00

There was copy & paste error so it always printed value of pkg->zero1. Also use tb_ctl_warn() here, no need to print backtrace. Signed-off-by: Mika Westerberg