The tcp_ehash hash table gets too big on systems with really big memory.
It is worse on systems with pages larger than 4KB. It wastes memory that
could be better used. It also makes the netstat command slow because reading
/proc/net/tcp and /proc/net/tcp6 needs to go through the full hash table.
The default value should not be larger for larger page sizes. It seems
that the effect of page size is an unintended error dating back a long
time. I also wonder if the default value really should be a larger
fraction of memory for systems with more memory. While systems with
really big ram can afford more space for hash tables, it is not clear to
me that they benefit from increasing the allocation ratio for this table.
The amount of memory allocated is determined by net/ipv4/tcp.c:tcp_init and
mm/page_alloc.c:alloc_large_system_hash.
tcp_init calls alloc_large_system_hash passing parameters-
bucketsize=sizeof(struct tcp_ehash_bucket)
numentries=thash_entries
scale=(num_physpages >= 128 * 1024) ? (25-PAGE_SHIFT) : (27-PAGE_SHIFT)
limit=0
On i386, PAGE_SHIFT is 12 for a page size of 4K
On ia64, PAGE_SHIFT defaults to 14 for a page size of 16K
The num_physpages test above makes the allocation take a larger fraction
of the total memory on systems with larger memory. The threshold size
for a i386 system is 512MB. For an ia64 system with 16KB pages the
threshold is 2GB.
For smaller memory systems-
On i386, scale = (27 - 12) = 15
On ia64, scale = (27 - 14) = 13
For larger memory systems-
On i386, scale = (25 - 12) = 13
On ia64, scale = (25 - 14) = 11
For the rest of this discussion, I'll just track the larger memory case.
The default behavior has numentries=thash_entries=0, so the allocated
size is determined by either scale or by the default limit of 1/16 of
total memory.
In alloc_large_system_hash-
| numentries = (flags & HASH_HIGHMEM) ? nr_all_pages : nr_kernel_pages;
| numentries += (1UL << (20 - PAGE_SHIFT)) - 1;
| numentries >>= 20 - PAGE_SHIFT;
| numentries <<= 20 - PAGE_SHIFT;
At this point, numentries is pages for all of memory, rounded up to the
nearest megabyte boundary.
| /* limit to 1 bucket per 2^scale bytes of low memory */
| if (scale > PAGE_SHIFT)
| numentries >>= (scale - PAGE_SHIFT);
| else
| numentries <<= (PAGE_SHIFT - scale);
On i386, numentries >>= (13 - 12), so numentries is 1/8196 of
bytes of total memory.
On ia64, numentries <<= (14 - 11), so numentries is 1/2048 of
bytes of total memory.
| log2qty = long_log2(numentries);
|
| do {
| size = bucketsize << log2qty;
bucketsize is 16, so size is 16 times numentries, rounded
down to a power of two.
On i386, size is 1/512 of bytes of total memory.
On ia64, size is 1/128 of bytes of total memory.
For smaller systems the results are
On i386, size is 1/2048 of bytes of total memory.
On ia64, size is 1/512 of bytes of total memory.
The large page effect can be removed by just replacing
the use of PAGE_SHIFT with a constant of 12 in the calls to
alloc_large_system_hash. That makes them more like the other uses of
that function from fs/inode.c and fs/dcache.c
Signed-off-by: David S. Miller <davem@davemloft.net>
This reverts commit 252ac86553.
It impacts the LSI customers using the mptstm target mode drivers
(source tar-ball at
ftp://ftp.lsil.com/HostAdapterDrivers/linux/Fusion-MPT/mptstm-1.00.13-src.tar.gz
for those who care).
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Use copy_user_highpage directly instead of cow_user_page in do_no_page:
in the immediately following page_cache_release, and elsewhere, it is
assuming that new_page is normal. If any VM_PFNMAP driver can get to
do_no_page, it's just a BUG (but not in the case of do_anonymous_page).
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
High Dickins points out that the user virtual address passed to the page
fault handler isn't necessarily page-aligned.
Also, add a comment on why the copy could fail for the user address case.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Increment the PFN field of the PTE so that the tests
on vm_pfn in mm/memory.c match up. The TLB ignores these
lower bits for larger page sizes, so it's OK to set things
like this.
Signed-off-by: David S. Miller <davem@davemloft.net>
This fixes locking in megaraid.c, namely:
(1) make sure megaraid_queue release the adapter lock by changing the
code to have a single return
(2) remove the errornous scsi_assign_lock call
Testing by Burton Windle.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Burton Windle <bwindle@fint.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
These get created by some drivers that don't generally even want a pfn
remapping at all, but would really mostly prefer to just map pages
they've allocated individually instead.
For now, create a helper function that turns such an incomplete PFN
remapping call into a loop that does that explicit mapping. In the long
run we almost certainly want to export a totally different interface for
that, though.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I image this showed up because of "unused var..." when the changes
occured, because flush_cache_page() is a noop in most places. This
showed up for me on parisc however, where flush_cache_page() is a real
function.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch contains the following possible cleanups:
- every file should #include the headers containing the prototypes for
it's global functions
- make needlessly global functions static
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Update the pre-CFI Sharp driver sharps.c so it compiles. map_read32 /
map_write32 no longer exist in the kernel so the driver is totally broken
as it stands. The replacement functions use different parameters resulting
in the other changes.
Change collie to use this driver until someone works out why the cfi driver
fails on that machine.
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Tested-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The patch below fixes the following sparse warning:
drivers/mtd/maps/nettel.c:482:27: warning: Using plain integer as NULL pointer
Signed-off-by: Luiz Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
arjan: drivers/mtd/maps/sc520cdp.c:167: warning: par_table is never written to and should be declared 'const'
arjan: drivers/mtd/maps/pci.c:105: warning: mtd_pci_map is never written to and should be declared 'const'
arjan: mind fixing those up ?
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
ixp4xx updates:
- Handle reads that don't start on a half-word boundary.
- Make it work when CPU is in little-endian mode.
Signed-off-by: John Bowler <jbowler@acm.org>
Signed-off-by: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: David Vrabel <dvrabel@arcom.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Recent models of Intel/Sharp and Spansion CFI flash now have significant
bits in the upper byte of device ID codes, read via what Spansion calls
"autoselect" and Intel calls "read device identifier". Currently these
values are truncated to the low 8 bits in the mtd data structures, as
all CFI read query info has previously been read one byte at a time.
Add a new method for reading 16-bit info, currently just manufacturer
and device codes; datasheets hint at future uses for upper bytes in
other fields.
Signed-off-by: Todd Poynor <tpoynor@mvista.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Merge from linux-mips:
Use physical addresses at the interface level, letting drivers remap
them as appropriate.
Signed-off-by: Maciej W. Rozycki <macro@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
break.b always sets cr.iim to 0 and the current code tries to
get the break_num by decoding instruction. However, their
seems to be a race condition while reading the regs->cr_iip,
as on other cpu the break.b at regs->cr_iip might have been
replaced with the original instruction as a result of
unregister_kprobe() and hence decoding instruction to
obtain break_num will result in wrong value in this case.
Also includes changes to kprobes.c which now has to handle
break number zero.
Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
A single SGI Altix system can be divided into multiple partitions,
each running their own instance of the Linux kernel. pfn_valid()
is currently not optimal for any but the first partition, since it
does not compare the pfn with min_low_pfn before calling the more
costly ia64_pfn_valid().
Signed-off-by: Dean Roe <roe@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
JFFS2 initialize f->sem mutex as "locked" in the slab constructor which is a
bug. Objects are freed with unlocked f->sem mutex. So, when they allocated
again, f->sem is unlocked because the slab cache constructor is not called for
them. The constructor is called only once when memory pages are allocated for
objects (namely, when the slab layer allocates new slabs). So, sometimes
'struct jffs2_inode_info' are allocated with unlocked f->sem, sometimes with
locked. This is a bug. Instead, initialize f->sem as unlocked in the
constructor. I.e., in the "constructed" state f->sem must be unlocked.
From: Keijiro Yano <keijiro_yano@yahoo.co.jp>
Acked-by: Artem B. Bityutskiy <dedekind@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The system call gate area handling called vm_normal_page() with the
wrong vma (which was always NULL, and caused an oops).
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Ensure to update hiscore.rule in dummy rule 4 in ipv6_dev_get_saddr().
Pointed out by Yan Zheng <yanzheng@21cn.com>.
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Both 32-bit and 64-bit use the same inline flush_icache_range definition
now, so both need to export __flush_icache_range, not just 64-bit.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Export symbol needed to allow MOL to run. This was changed to be inline
in past and forgot to be change here.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Check for invalid node ID values in the new atomic create+open method.
Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Check the created directory inode for aliases in the mkdir() method.
Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
With Andrew Morton <akpm@osdl.org>
The slab scanning code tries to balance the scanning rate of slabs versus the
scanning rate of LRU pages. To do this, it retains state concerning how many
slabs have been scanned - if a particular slab shrinker didn't scan enough
objects, we remember that for next time, and scan more objects on the next
pass.
The problem with this is that with (say) a huge number of GFP_NOIO
direct-reclaim attempts, the number of objects which are to be scanned when we
finally get a GFP_KERNEL request can be huge. Because some shrinker handlers
just bail out if !__GFP_FS.
So the patch clamps the number of objects-to-be-scanned to 2* the total number
of objects in the slab cache.
Signed-off-by: Andrea Arcangeli <andrea@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When quota file specified in mount options did not exist, we tried to
dereference NULL pointer later. Fix it.
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
If you have an array with a write-intent-bitmap, and you remove a device, then
re-add it, a full recovery isn't needed. We detect a re-add by looking at
saved_raid_disk. For raid1, it doesn't matter which disk it was, only whether
or not it was an active device. The old code being removed set a value of
'mirror' which was then ignored, so it can go. The changed code performs the
correct check.
For raid6, if there are two missing devices, make sure we chose the right slot
on --re-add rather than always the first slot.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
If an array is created using set_array_info, default_bitmap_offset isn't set
properly meaning that an internal bitmap cannot be hot-added until the array
is stopped and re-assembled.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When doing a recovery, we need to know whether the array will still be
degraded after the recovery has finished, so we can know whether bits can be
clearred yet or not. This patch performs the required check.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
bitmap_unplug actually writes data (bits) to storage, so we shouldn't be
holding a spinlock...
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
raid10 has two different layouts. One uses near-copies (so multiple
copies of a block are at the same or similar offsets of different
devices) and the other uses far-copies (so multiple copies of a block
are stored a greatly different offsets on different devices). The point
of far-copies is that it allows the first section (normally first half)
to be layed out in normal raid0 style, and thus provide raid0 sequential
read performance.
Unfortunately, the read balancing in raid10 makes some poor decisions
for far-copies arrays and you don't get the desired performance. So
turn off that bad bit of read_balance for far-copies arrays.
With this patch, read speed of an 'f2' array is comparable with a raid0
with the same number of devices, though write speed is ofcourse still
very slow.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Some users (hi Zwane) have seen a problem when running a workload that
eats nearly all of physical memory - th system does an OOM kill, even
when there is still a lot of swap free.
The problem appears to be a very big task that is holding the swap
token, and the VM has a very hard time finding any other page in the
system that is swappable.
Instead of ignoring the swap token when sc->priority reaches 0, we could
simply take the swap token away from the memory hog and make sure we
don't give it back to the memory hog for a few seconds.
This patch resolves the problem Zwane ran into.
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Assign the appropriate dentry operations to the dentry. Fixes memory leak.
Signed-off-by: Latchesar Ionkov <lucho@ionkov.net>
Cc: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Move the cpuset_fork() call below the write_unlock_irq call in
kernel/fork.c copy_process().
Since the cpuset-dual-semaphore-locking-overhaul.patch, the cpuset_fork()
routine acquires task_lock(), so cannot be called while holding the
tasklist_lock for write.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I believe this patch is required to fix breakage in the asynch reclaim
watermark logic introduced by this patch:
http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=7fb1d9fca5c6e3b06773b69165a73f3fb786b8ee
Just some background of the watermark logic in case it isn't clear...
Basically what we have is this:
--- pages_high
|
| (a)
|
--- pages_low
|
| (b)
|
--- pages_min
|
| (c)
|
--- 0
Now when pages_low is reached, we want to kick asynch reclaim, which gives us
an interval of "b" before we must start synch reclaim, and gives kswapd an
interval of "a" before it need go back to sleep.
When pages_min is reached, normal allocators must enter synch reclaim, but
PF_MEMALLOC, ALLOC_HARDER, and ALLOC_HIGH (ie. atomic allocations, recursive
allocations, etc.) get access to varying amounts of the reserve "c".
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: "Seth, Rohit" <rohit.seth@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
