[Intel-gfx] [RFC 20/22] drm/i915: Fix MI_STORE_DWORD_IMM parser defintion

Chris Wilson chris at chris-wilson.co.uk
Tue Nov 26 19:08:48 CET 2013


On Tue, Nov 26, 2013 at 08:51:37AM -0800, bradley.d.volkin at intel.com wrote:
> From: Brad Volkin <bradley.d.volkin at intel.com>
> 
> The length mask is different for each ring and the size can vary,
> so we should duplicate the definition with the correct encoding
> for each ring.

Jumping in here since this highlights the most confusing aspect of this
series - the meta patching. Please implement the command parsing
infrastructure upfront and in a very small number of patches (most
preferably one) that avoids having to add fixes late in the series.

I think using
s/S/ALLOW/
s/R/REJECT/
s/B/BLACKLIST/
s/W/WHITELIST/
makes the action much more clear, and would rather that every unsafe
command starts off as REJECT. (With the whitelist/blacklisting being
added as separate patches with justification as they are now.)

Since we do disable the security, I would also reject all
unknown/unmatched commands and make ALLOW explicit.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre



More information about the Intel-gfx mailing list