Re: 2.6.24-rc5 the LED does not come on after a switch-off switch-on

[Ehud Gavron]

Michael Buesch wrote:

On Tuesday 18 December 2007 18:33:43 Michael Buesch wrote:

On Tuesday 18 December 2007 18:29:34 Michael Buesch wrote:

And here's the code:

        if (unlikely(report_change)) {
                b43info(wl,"EHUD: sleeping\n");
                msleep(1);
                b43info(wl,"EHUD: LED coming on\n);
                input_report_key(poll_dev->input, KEY_WLAN, 1);
                input_report_key(poll_dev->input, KEY_WLAN, 0);
        }



Ehud

So my question (other than why do I need two messages and one msleep to get my LED lit) is... what happened to the "EHUD: sleeping" debug message? It never showed up on the console. I did this several times. The full dmesg is attached.

This smells like a compiler bug.
Can you try an older compiler version?
(Most distros ship several versions)

Actually I do remember a gcc bug related to strict-aliasing.
What happens is that about two lines of source code are
skipped. So this might also apply here. We skip two lines.
But I don't remember the gcc bug #. :(

I think this is the one I was talking about:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32328

From Bugzilla:
Known to Work: 4.1.2 4.3.0
gcc --version
gcc (GCC) 4.1.2 20070925 (Red Hat 4.1.2-33)

smime.p7s
Description: S/MIME Cryptographic Signature

_______________________________________________
Bcm43xx-dev mailing list
Bcm43xx-dev@lists.berlios.de
https://lists.berlios.de/mailman/listinfo/bcm43xx-dev