On 12/6/18 4:49 AM, Nicolas Boichat wrote: >> So it would be fine even unchanged. The check would anyway need some >> more love to catch the same with __GFP_DMA to be consistent and cover >> all corner cases. > Yes, the test is not complete. If we really wanted this to be > accurate, we'd need to check that GFP_* exactly matches SLAB_CACHE_*. > > The only problem with dropping this is test that we should restore > GFP_DMA32 warning/errors somewhere else (as Christopher pointed out > here: https://lkml.org/lkml/2018/11/22/430), especially for kmalloc > case. I meant just dropping that patch hunk, not the whole test. Then the test stays as it is and will keep warning anyone calling kmalloc(GFP_DMA32). It would also warn anyone calling kmem_cache_alloc(GFP_DMA32) on SLAB_CACHE_DMA32 cache, but since the gfp can be just dropped, and you as the only user of this so far will do that, it's fine? > Maybe this can be done in kmalloc_slab.
