On Wed, 01 Oct 2014 11:45:47 -0400, Jeff Moyer said:
> This sounds an awful lot like posix_fadvise' POSIX_FADV_NOREUSE flag.
Gaah. Premature click. man posix_fadvise says this:
In kernels before 2.6.18, POSIX_FADV_NOREUSE had the same semantics as
POSIX_FADV_WILLNEED. This was probably a bug; since kernel 2.6.18,
this flag is a no-op.
and mm/fadvise.c says this:
switch (advice) {
case POSIX_FADV_NORMAL:
f.file->f_ra.ra_pages = bdi->ra_pages;
spin_lock(&f.file->f_lock);
f.file->f_mode &= ~FMODE_RANDOM;
spin_unlock(&f.file->f_lock);
...
*/
force_page_cache_readahead(mapping, f.file, start_index,
nrpages);
break;
case POSIX_FADV_NOREUSE:
break;
case POSIX_FADV_DONTNEED:
if (!bdi_write_congested(mapping->backing_dev_info))
__filemap_fdatawrite_range(mapping, offset, endbyte,
WB_SYNC_NONE);
/* First and last FULL page! */
So... not much interface there, actually. One wonders if removing the 'break;'
and allowing a fall-through would actually be an improvement....
Attachment:
pgpV7OOftQncp.pgp
Description: PGP signature