Re: [PATCH 1/2] drivers: caam: add RNG software self-test

[Roland Hieber <r.hieber@pengutronix.de>]

On Mon, Nov 26, 2018 at 11:53:47AM +0100, Roland Hieber wrote:
> > > +	pr_vdebug("Result\n");
> > > +	for (i = 0; i < RESULT_SIZE; i++) {
> > > +		pr_vdebug("%02X\n", result[i]);
> > > +	}
> > > +
> > > +	pr_vdebug("Expected Result:\n");
> > > +	for (i = 0; i < RESULT_SIZE; i++) {
> > > +		pr_vdebug("%02X\n", exp_result[i]);
> > > +	}
> > 
> > Use memory_display to display this. Also this is only interesting if
> > both differ, so better print it in the failure path.
> 
> I thought about that too, but I didn't see a way to make memory_display
> use pr_vdebug, or otherwise make its output depend on the debug level.

I have several solutions in mind: one would be to introduce a function
log_level() to get the current log level and decide on that whether to
call memory_display. However then memory dump won't pick up the pr_*
prefixes and use plain printf nevertheless. So my second thought is to
add a callback function to memory_display to pass pr_debug or else to it
which it can use to print stuff. However then we need to remodel
memory_display so it always prints full lines (or uses pr_cont which it
would need to know about, and you wanted to get rid of pr_cont anyway).
We could also use a third way and implement a custom printf escape, say
%md, that gets replaced with memory_display output when calling
pr_printk, and move the core formatting of memory_display into a helper.

I'm not sure which one is best.

-  Roland

-- 
Roland Hieber                     | r.hieber@pengutronix.de     |
Pengutronix e.K.                  | https://www.pengutronix.de/ |
Peiner Str. 6-8, 31137 Hildesheim | Phone: +49-5121-206917-5086 |
Amtsgericht Hildesheim, HRA 2686  | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox