On Tue, 2004-05-25 at 10:58, Randy.Dunlap wrote:
> On Thu, 20 May 2004 20:50:44 -0600 (MDT) MJK wrote:
>
> |
> | I made sure that this wasn't one that Mr. Walter Harms didn't
> | graciously already do or wasn't in the new source tree.
> |
> | --- old.linux-2.6.6/drivers/scsi/aacraid/rx.c 2004-05-09
> 20:33:22.000000000 -0600
> | +++ new.linux-2.6.6/drivers/scsi/aacraid/rx.c 2004-05-20
> 20:40:32.000000000 -0600
> | @@ -435,6 +435,11 @@
> | * Start any kernel threads needed
> | */
> | dev->thread_pid = kernel_thread((int (*)(void *))aac_command_thread,
> dev, 0);
> | + if (dev->thread_pid < 0) {
> | + printk(KERN_ERR "aacraid: unable to create i960 command
> thread.\n");
> | + return -1;
> | + }
> | +
> | /*
> | * Tell the adapter that all is configured, and it can start
> | * accepting requests
>
> There are a few things that need to be undone if kernel_thread()
> fails, before returning here:
>
> a. release the irq that request_irq() claimed;
> b. free the memory that aac_init_adapter() allocated (dev->queues);
Also would a iounmap of aac->regs.rx need to be done?
Mark.
--
Mark Haverkamp <markh@xxxxxxxx>
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@xxxxxxxxxxxxxx
http://lists.osdl.org/mailman/listinfo/kernel-janitors
|
