On Tue, Sep 28, 2010 at 02:45:18PM -0700, Andrew Morton wrote: > On Tue, 28 Sep 2010 22:41:42 +0100 > > > hm. I hope these sparse warnings are sufficiently useful to justify > > > all the gunk we're adding to support them. > > > > > > Is it actually finding any bugs? > > > > FWIW, bitwise or done in the right-hand argumet of shift looks ugly as hell; > > what the hell is that code _doing_? > > There's a nice fat comment a few lines up... [snip] Egads... IMO the cleanest way to deal with that is to add integer constants, not to be used anywhere else (e.g. ___GFP_DMA, with #define __GFP_DMA ((__force gfp_t)___GFP_DMA) and use them in that horror. As for the gfp_t warnings - yes, they'd caught a bunch of bugs at some point; considering the bitrot rates... might be worth rechecking. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxxx For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>