Messages in this thread | | | Date | Fri, 18 Apr 2008 15:20:43 -0700 | From | Andrew Morton <> | Subject | Re: kgdb: add kgdb internal test suite |
| |
On Fri, 18 Apr 2008 19:00:48 GMT Linux Kernel Mailing List <linux-kernel@vger.kernel.org> wrote:
> Gitweb: http://git.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=e8d31c204e36e019b9134f2a11926cac0fcf9b19 > Commit: e8d31c204e36e019b9134f2a11926cac0fcf9b19 > Parent: 737a460f21febe551ff1d2299b63bae9b154078f > Author: Jason Wessel <jason.wessel@windriver.com> > AuthorDate: Fri Mar 7 16:34:17 2008 -0600 > Committer: Ingo Molnar <mingo@elte.hu> > CommitDate: Thu Apr 17 20:05:42 2008 +0200 > > kgdb: add kgdb internal test suite > > ... > > +static const char hexchars[] = "0123456789abcdef";
42! That's a great place to stop.
> +static char get_buf[BUFMAX];
This really needs to be renamed.
> + int (*run_test) (int, int); > + int (*validate_put) (char *); > +}; > + > +static struct test_state ts; > + > +static int kgdbts_unreg_thread(void *ptr) > +{ > + /* Wait until the tests are complete and then ungresiter the I/O > + * driver. > + */ > + while (!final_ack) > + msleep_interruptible(1500);
Why 1.5 seconds? Seems like a pointless wats of time - can we use just 1 here?
> +#ifdef CONFIG_X86
btw, is the selftest code known to work on any other architectures?
afacit pretty much all of this code could/should be __init, but isn't?
| |