6:36 a.m.
We are hopefully going to get rid of OBD_ALLOC_LARGE() as well,
though.
It's simple enough to write a function:
void *obd_zalloc(size_t size)
{
if (size > 4 * PAGE_CACHE_SIZE)
return vzalloc(size);
else
return kmalloc(size, GFP_NOFS);
}
Except, huh? Shouldn't we be using GFP_NOFS for the vzalloc() side?
There was some discussion of that GFP_NOFS was a bit buggy back in 2010
(http://marc.info/?l=linux-mm&m=128942194520631&w=4) but the current
lustre code doesn't try to pass GFP_NOFS.
The version in the upstream client is out of date. The current macro in the Intel master
Branch is:
#define __OBD_VMALLOC_VERBOSE(ptr, cptab, cpt, size) \
do { \
(ptr) = cptab == NULL ? \
__vmalloc(size, GFP_NOFS | __GFP_HIGHMEM | __GFP_ZERO, \
PAGE_KERNEL) : \
cfs_cpt_vzalloc(cptab, cpt, size); \
if (unlikely((ptr) == NULL)) { \
CERROR("vmalloc of '" #ptr "' (%d bytes)
failed\n", \
(int)(size)); \
CERROR(LPU64" total bytes allocated by Lustre, %d by LNET\n", \
obd_memory_sum(), atomic_read(&libcfs_kmemory)); \
} else { \
OBD_ALLOC_POST(ptr, size, "vmalloced"); \
} \
} while(0)
Then it's simple enough to change OBD_FREE_LARGE() to kvfree().
Also it's weird that only the lustre people have thought of this trick
to allocate big chunks of RAM and no one else has. What would happen if
we just change vmalloc() so it worked this way for everyone?
Do we really want to encourage vmalloc usages?