X-Git-Url: https://git.cworth.org/git?a=blobdiff_plain;f=src%2Fexa%2Fi965%2Fnew_job_old_tricks.mdwn;h=03a07b25c5566773ca56695a9c3ab58a2de1bba7;hb=58e046dcdfeea6374c732d1bb5630674f3286423;hp=f320434702f52888885710d6cfd39b90813ea88d;hpb=07dfb1a6de254cbad30524dccbf73c8dd48eeae4;p=cworth.org

diff --git a/src/exa/i965/new_job_old_tricks.mdwn b/src/exa/i965/new_job_old_tricks.mdwn
index f320434..03a07b2 100644
--- a/src/exa/i965/new_job_old_tricks.mdwn
+++ b/src/exa/i965/new_job_old_tricks.mdwn
@@ -1,6 +1,6 @@
-[[meta title="A new job, but old performance fixes"]]
+[[!meta title="A new job, but old performance fixes"]]
 
-[[tag exa performance i965]]
+[[!tag exa performance intel i965]]
 
 Many readers have heard already, but it will be news to some that I
 recently changed jobs. After just short of 4 years with Red Hat, I've
@@ -84,7 +84,7 @@ So, with that in place the only thing left to decide was how large of
 a vertex buffer to allocate upfront. And that gives me an excuse to
 put in a performance plot:
 
-[[img vertex_buffers.png]]
+[[!img vertex_buffers.png]]
 
 So the more the better, (obviously), until we get to 256 composite
 operations fitting into a single buffer. Then we start losing