. I've pretty much dropped the per-cpu pages patches which Martin and
I developed.
These patches gave a 1-2% benefit in kernel compiles, up to 4% in
Randy Hron's testing of the autoconf build tests. 2.2% in specweb.
And a 60% speedup in a little app which just looped writing 80k to a
file and truncating it off again.
All the above came from the cache-warmth effect. The patches would
give an overall 15% speedup due to reduced lock contention in Anton's
testing on the big PPC64 machines, but he fixed that anyway by
getting NUMA working properly.
In my opinion, all the above is just too thin to justify throwing a
bunch of new stuff into the page allocator.
I shall continue to distribute these patches. Maybe someone will
find them sufficiently beneficial for something at a future time.
But it simply seems to be the case that no interesting workloads
repeatedly allocate and free small numbers of pages (in the 10-50
range).
. The shared pagetable code is not in the main diff - it has a few
problems at present. The patches are over in the experimental/
directory. order of application is:
shpte.patch
shpte-lock-ranking-fix.patch
shmmap.patch
handle-mm-fault-locking.patch
mremap-shared-pagetable-fix.patch
shpte-unmap_all_pages_fix.patch
unmap_page_range-fix.patch
. The slab rework is stable now and Manfred has some good
microbenchmark numbers from that. But we're still not quite ready
with that code because the hotplug CPU APIs have, shall we say, a few
shortcomings.
Since 2.5.43-mm1:
-disable-ppc-lbd.patch
-fs-inlines.patch
-md-fix.patch
-mpparse-fix.patch
-no-reclaim-throttle.patch
-refill-inactive-lockup-fix.patch
-reiserfs-kmap-fix.patch
-simple_rename-link-count.patch
-static-filemap_sync.patch
-uninline-highmem.patch
-vmalloc-overalloc.patch
Merged
-meminfo-numa.patch
Dropped. To be moved from procfs into the NUMA preentation in
driverfs.
+3c59x-udp-csum.patch
Make the 3c59x driver work with UDP in Linus's current tree
+dmi-warning.patch
+ide-warnings.patch
Stomp some compilation warnings
+dhowells-readahead.patch
+file_ra_state_init.patch
Expose some finer-grained readahead facilities.
+less-unlikelies.patch
Less buslocked traffic in the page allocator
+running-iowait.patch
Expose nr_running and nr_iowait task counts in /proc
+uaccess-uninline.patch
Uninline the ia32 copy_*_user functions. (Now showing a 33kbyte
shrink from this work)
+slab-cleanup.patch
Less typedefs and macros in slab.c
+mm1-incr1.patch
+mm1-incr2.patch
POSIX ACL and EA updates
All patches:
linus.patch
cset-1.781.24.13-to-1.793.txt.gz
3c59x-udp-csum.patch
Enable UDP checksums in 3c59x
ide-warnings.patch
Fix some IDE compile warnings
dmi-warning.patch
fix a compile warning in dmi_scan.c
kgdb.patch
ramfs-aops.patch
Move ramfs address_space ops into libfs
ramfs-prepare-write-speedup.patch
correctness fixes in libfs address_space ops
dio-fine-alignment.patch
Allow O_DIRECT to use 512-byte alignment
dhowells-readahead.patch
readahead generalisations
file_ra_state_init.patch
Add a function to initialise file readahead state
less-unlikelies.patch
reduced buslocked traffic in the page allocator
running-iowait.patch
expose nr_running and nr_iowait task counts in /proc
intel-user-copy-taka.patch
Faster copy_*_user for Intel ia32 CPUs
uaccess-uninline.patch
ingo-oom-kill.patch
oom-killer changes for threaded apps
add_timer_on.patch
add_timer_on(): function to start a timer on a particular CPU
slab-split-01-rename.patch
slab cleanup: rename static functions
slab-split-02-SMP.patch
slab: enable the cpu arrays on uniprocessor
slab-split-03-tail.patch
slab: reduced internal fragmentation
slab-split-04-drain.patch
slab: take the spinlock in the drain function.
slab-split-05-name.patch
slab: remove spaces from /proc identifiers
slab-split-06-mand-cpuarray.patch
slab: cleanups and speedups
slab-split-07-inline.patch
slab: uninline poisoning checks
slab-split-08-reap.patch
slab: reap timers
slab-timer.patch
slab-use-sem.patch
slab-cleanup.patch
Slab cleanup
ingo-mmap-speedup.patch
Ingo's mmap speedup
mm-inlines.patch
remove some inlines from mm/*
o_streaming.patch
O_STREAMING support
page_reserved-accounting.patch
Global PageReserved accounting
use-page_reserved_accounting.patch
Use PG_reserved accounting in the VM
mod_timer-race.patch
net-loopback.patch
Disable second copy in the network loopback driver
blkdev-o_direct-short-read.patch
Fix O_DIRECT blockdev reads at end-of-device
orlov-allocator.patch
blk-queue-bounce.patch
inline blk_queue_bounce
lseek-ext2_readdir.patch
remove lock_kernel() from ext2_readdir()
write-deadlock.patch
Fix the generic_file_write-from-same-mmapped-page deadlock
rd-cleanup.patch
Cleanup and fix the ramdisk driver (doesn't work right yet)
spin-lock-check.patch
spinlock/rwlock checking infrastructure
hugetlb-prefault.patch
hugetlbpages: factor out some code for hugetlbfs
hugetlb-header-split.patch
Move hugetlb declarations into their own header
hugetlbfs.patch
hugetlbfs file system
hugetlb-shm.patch
hugetlbfs backing for SYSV shared memory
truncate-bkl.patch
don't take the BKL in inode_setattr
akpm-deadline.patch
deadline scheduler tweaks
xattr-01-metablock-cache.patch
EA: meta block cache
xattr-02-ext3.patch
EA: ext3 support
xattr-03-ext2.patch
EA: ext2 support
fix-xattr.patch
EA: compile warning fix
posix-acl-01-core.patch
posixacl: core support
posix-acl-02-umask.patch
posixacl: umask support
posix-acl-03-user-api.patch
posixacl: user API
posix-acl-04-ext3.patch
posixacl: ext3 support
acl-ext3-fix-tree.patch
acl-ext3-inode.patch
posix-acl-05-ext2.patch
posixacl: ext2 support
mm1-incr1.patch
mm1-incr2.patch
posixacl: use getxattr in nfsd_get_posix_acl()
ext23-mount-options.patch
ext2/3 mount option processing cleanup
read_barrier_depends.patch
extended barrier primitives
dcache_rcu.patch
Use RCU for dcache
mpopulate.patch
remap_file_pages
rmqueue_bulk.patch
bulk page allocator
free_pages_bulk.patch
Bulk page freeing function
hot_cold_pages.patch
Hot/Cold pages and zone->lock amortisation
readahead-cold-pages.patch
Use cache-cold pages for pagecache reads.
pagevec-hot-cold-hint.patch
hot/cold hints for truncate and page reclaim
page-reservation.patch
Page reservation API
wli-show_free_areas.patch
show_free_areas extensions
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/