4.4-stable review patch. If anyone has any objections, please let me know.
------------------ From: Alex Vesker <va...@mellanox.com> [ Upstream commit d412c31dae053bf30a1bc15582a9990df297a660 ] The command interface can work in two modes: Events and Polling. In the general case, each time we invoke a command, a work is queued to handle it. When working in events, the interrupt handler completes the command execution. On the other hand, when working in polling mode, the work itself completes it. Due to a bug in the work handler, a command could have been completed by the interrupt handler, while the work handler hasn't finished yet, causing the it to complete once again if the command interface mode was changed from Events to polling after the interrupt handler was called. mlx5_unload_one() mlx5_stop_eqs() // Destroy the EQ before cmd EQ ...cmd_work_handler() write_doorbell() --> EVENT_TYPE_CMD mlx5_cmd_comp_handler() // First free free_ent(cmd, ent->idx) complete(&ent->done) <-- mlx5_stop_eqs //cmd was complete // move to polling before destroying the last cmd EQ mlx5_cmd_use_polling() cmd->mode = POLL; --> cmd_work_handler (continues) if (cmd->mode == POLL) mlx5_cmd_comp_handler() // Double free The solution is to store the cmd->mode before writing the doorbell. Fixes: e126ba97dba9 ("mlx5: Add driver for Mellanox Connect-IB adapters") Signed-off-by: Alex Vesker <va...@mellanox.com> Signed-off-by: Saeed Mahameed <sae...@mellanox.com> Signed-off-by: Greg Kroah-Hartman <gre...@linuxfoundation.org> --- drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -643,6 +643,7 @@ static void cmd_work_handler(struct work struct semaphore *sem; unsigned long flags; int alloc_ret; + int cmd_mode; sem = ent->page_queue ? &cmd->pages_sem : &cmd->sem; down(sem); @@ -688,6 +689,7 @@ static void cmd_work_handler(struct work set_signature(ent, !cmd->checksum_disabled); dump_command(dev, ent, 1); ent->ts1 = ktime_get_ns(); + cmd_mode = cmd->mode; /* ring doorbell after the descriptor is valid */ mlx5_core_dbg(dev, "writing 0x%x to command doorbell\n", 1 << ent->idx); @@ -695,7 +697,7 @@ static void cmd_work_handler(struct work iowrite32be(1 << ent->idx, &dev->iseg->cmd_dbell); mmiowb(); /* if not in polling don't use ent after this point */ - if (cmd->mode == CMD_MODE_POLLING) { + if (cmd_mode == CMD_MODE_POLLING) { poll_timeout(ent); /* make sure we read the descriptor after ownership is SW */ rmb();