I think the following patch introduced a memory leak into the cm. (commit 110cf374a809817d5c080c0ac82d65d029820a66, committed on July 27):
http://git.kernel.org/?p=linux/kernel/git/roland/infiniband.git;a=commitdiff;h=110cf374a809817d5c080c0ac82d65d029820a66;hp=d4c4196f24ade5f336882587480652efde2c739c Now, no one seems responsible for freeing the memory allocated by kzalloc in procedure cm_add_one (file drivers/infiniband/core/cm.c): /**** jpm: Who deallocates this kzalloc when the cm device is removed??? */ cm_dev = kzalloc(sizeof(*cm_dev) + sizeof(*port) * ib_device->phys_port_cnt, GFP_KERNEL); if (!cm_dev) return; cm_dev->ib_device = ib_device; cm_get_ack_delay(cm_dev); cm_dev->device = device_create_drvdata(&cm_class, &ib_device->dev, MKDEV(0, 0), NULL, "%s", ib_device->name); Am I correct? - Jack _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
