From: Stefan Richter Found and debugged by Jay Fenlason . The bug was especially noticeable with direct I/O over fw-sbp2. Same as commit 9c9bdf4d50730fd04b06077e22d7a83b585f26b5. Signed-off-by: Stefan Richter Signed-off-by: Kristian Høgsberg Signed-off-by: Greg Kroah-Hartman --- drivers/firewire/fw-transaction.c | 4 +++- drivers/firewire/fw-transaction.h | 4 ++++ 2 files changed, 7 insertions(+), 1 deletion(-) --- a/drivers/firewire/fw-transaction.c +++ b/drivers/firewire/fw-transaction.c @@ -605,8 +605,10 @@ fw_send_response(struct fw_card *card, s * check is sufficient to ensure we don't send response to * broadcast packets or posted writes. */ - if (request->ack != ACK_PENDING) + if (request->ack != ACK_PENDING) { + kfree(request); return; + } if (rcode == RCODE_COMPLETE) fw_fill_response(&request->response, request->request_header, --- a/drivers/firewire/fw-transaction.h +++ b/drivers/firewire/fw-transaction.h @@ -124,6 +124,10 @@ typedef void (*fw_transaction_callback_t size_t length, void *callback_data); +/* + * Important note: The callback must guarantee that either fw_send_response() + * or kfree() is called on the @request. + */ typedef void (*fw_address_callback_t)(struct fw_card *card, struct fw_request *request, int tcode, int destination, int source, -- - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/