[Intel-gfx] [PATCH] Make I830FALLBACK debugging a runtime instead of compile-time option.

[Carl Worth]

On Wed, 2008-11-05 at 13:52 -0800, Eric Anholt wrote:
> +.BI "Option \*qFallbackDebug\*q \*q" boolean \*q
> +Enable printing of debugging information on acceleration fallbacks to the
> +server log.

A wholehearted ACK from me. This will be very helpful as we triage bug
reports of performance problems. So thanks!

As for review of the patch, should we mention the default value of False
in the documentation above? (Should be fairly obvious though.)

> +#define I830FALLBACK(s, arg...)				\
> +do {							\
> +    if (I830PTR(pScrn)->fallback_debug) {		\
> +	xf86DrvMsg(pScrn->scrnIndex, X_INFO,		\
> +		   "EXA fallback: " s "\n", ##arg);	\
> +	return FALSE;					\
> +    }							\
> +} while(0)
> +

A macro that assumes the existence of a local variable is obviously
unpleasant. But it's also not new with this patch, just moving, so not
something that must be fixed now. (Nor is this likely the only macro
using that convention). So that's just yet another cleanup we should do
sometime.

-Carl

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.freedesktop.org/archives/intel-gfx/attachments/20081105/3f0d941a/attachment.sig>