AGPv3 command parsing
PATCH 1/1 AGPv3 command parsing
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi,
there's a trivial but fatal typo in agp/generic.c:agp_v3_parse_one() that completely messes up the command generation.
I'll let the attached patch against 2.6.0 explain the rest...
cu,
Nicolai
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)
iD8DBQE/MvsBsxPozBga0lwRAgsQAJ9AHqqG0uBXDKzE9pHxe1m67aZV2wCggBFH p+BUB22EdrDIgvNs+kZHyVY=
=8khY
-----END PGP SIGNATURE-----
char/agp/generic.c renamed from char/agp/generic.c~ +2 -2
@@ -453,9 +453,9 @@/* Clear out unwanted bits. */ if (*cmd &AGPSTAT3_8X ) - *cmd = ~(AGPSTAT3_4X |AGPSTAT3_RSVD ); + *cmd &= ~(AGPSTAT3_4X |AGPSTAT3_RSVD );if (*cmd &AGPSTAT3_4X ) - *cmd = ~(AGPSTAT3_8X |AGPSTAT3_RSVD ); + *cmd &= ~(AGPSTAT3_8X |AGPSTAT3_RSVD ); }//FIXME: This doesn't smell right.
Quoted 4 lines · Nicolai Haehnle
there's a trivial but fatal typo in agp/generic.c:agp_v3_parse_one() that completely messes up the command generation.
I'll let the attached patch against 2.6.0 explain the rest...
Whoops. Thanks, applied.
Dave
Quoted 4 lines · Nicolai Haehnle
there's a trivial but fatal typo in agp/generic.c:agp_v3_parse_one() that completely messes up the command generation.
I'll let the attached patch against 2.6.0 explain the rest...
Already applied to agpgart repository last week, and Linus pulled a few hours ago. Thanks again.
Dave
Signature · 1 line
Dave Jones http://www.codemonkey.org.uk