If a node runs out of memory, ensure that memory on nodes w/o cpus is used
before using memory on nodes with cpus.
Signed-off-by: NJack Steiner <steiner@sgi.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

)

From: Ingo Molnar <mingo@elte.hu>

This is the latest version of the scheduler cache-hot-auto-tune patch.

The first problem was that detection time scaled with O(N^2), which is
unacceptable on larger SMP and NUMA systems. To solve this:

- I've added a 'domain distance' function, which is used to cache
  measurement results. Each distance is only measured once. This means
  that e.g. on NUMA distances of 0, 1 and 2 might be measured, on HT
  distances 0 and 1, and on SMP distance 0 is measured. The code walks
  the domain tree to determine the distance, so it automatically follows
  whatever hierarchy an architecture sets up. This cuts down on the boot
  time significantly and removes the O(N^2) limit. The only assumption
  is that migration costs can be expressed as a function of domain
  distance - this covers the overwhelming majority of existing systems,
  and is a good guess even for more assymetric systems.

  [ People hacking systems that have assymetries that break this
    assumption (e.g. different CPU speeds) should experiment a bit with
    the cpu_distance() function. Adding a ->migration_distance factor to
    the domain structure would be one possible solution - but lets first
    see the problem systems, if they exist at all. Lets not overdesign. ]

Another problem was that only a single cache-size was used for measuring
the cost of migration, and most architectures didnt set that variable
up. Furthermore, a single cache-size does not fit NUMA hierarchies with
L3 caches and does not fit HT setups, where different CPUs will often
have different 'effective cache sizes'. To solve this problem:

- Instead of relying on a single cache-size provided by the platform and
  sticking to it, the code now auto-detects the 'effective migration
  cost' between two measured CPUs, via iterating through a wide range of
  cachesizes. The code searches for the maximum migration cost, which
  occurs when the working set of the test-workload falls just below the
  'effective cache size'. I.e. real-life optimized search is done for
  the maximum migration cost, between two real CPUs.

  This, amongst other things, has the positive effect hat if e.g. two
  CPUs share a L2/L3 cache, a different (and accurate) migration cost
  will be found than between two CPUs on the same system that dont share
  any caches.

(The reliable measurement of migration costs is tricky - see the source
for details.)

Furthermore i've added various boot-time options to override/tune
migration behavior.

Firstly, there's a blanket override for autodetection:

	migration_cost=1000,2000,3000

will override the depth 0/1/2 values with 1msec/2msec/3msec values.

Secondly, there's a global factor that can be used to increase (or
decrease) the autodetected values:

	migration_factor=120

will increase the autodetected values by 20%. This option is useful to
tune things in a workload-dependent way - e.g. if a workload is
cache-insensitive then CPU utilization can be maximized by specifying
migration_factor=0.

I've tested the autodetection code quite extensively on x86, on 3
P3/Xeon/2MB, and the autodetected values look pretty good:

Dual Celeron (128K L2 cache):

 ---------------------
 migration cost matrix (max_cache_size: 131072, cpu: 467 MHz):
 ---------------------
           [00]    [01]
 [00]:     -     1.7(1)
 [01]:   1.7(1)    -
 ---------------------
 cacheflush times [2]: 0.0 (0) 1.7 (1784008)
 ---------------------

Here the slow memory subsystem dominates system performance, and even
though caches are small, the migration cost is 1.7 msecs.

Dual HT P4 (512K L2 cache):

 ---------------------
 migration cost matrix (max_cache_size: 524288, cpu: 2379 MHz):
 ---------------------
           [00]    [01]    [02]    [03]
 [00]:     -     0.4(1)  0.0(0)  0.4(1)
 [01]:   0.4(1)    -     0.4(1)  0.0(0)
 [02]:   0.0(0)  0.4(1)    -     0.4(1)
 [03]:   0.4(1)  0.0(0)  0.4(1)    -
 ---------------------
 cacheflush times [2]: 0.0 (33900) 0.4 (448514)
 ---------------------

Here it can be seen that there is no migration cost between two HT
siblings (CPU#0/2 and CPU#1/3 are separate physical CPUs). A fast memory
system makes inter-physical-CPU migration pretty cheap: 0.4 msecs.

8-way P3/Xeon [2MB L2 cache]:

 ---------------------
 migration cost matrix (max_cache_size: 2097152, cpu: 700 MHz):
 ---------------------
           [00]    [01]    [02]    [03]    [04]    [05]    [06]    [07]
 [00]:     -    19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)
 [01]:  19.2(1)    -    19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)
 [02]:  19.2(1) 19.2(1)    -    19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)
 [03]:  19.2(1) 19.2(1) 19.2(1)    -    19.2(1) 19.2(1) 19.2(1) 19.2(1)
 [04]:  19.2(1) 19.2(1) 19.2(1) 19.2(1)    -    19.2(1) 19.2(1) 19.2(1)
 [05]:  19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)    -    19.2(1) 19.2(1)
 [06]:  19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)    -    19.2(1)
 [07]:  19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1) 19.2(1)    -
 ---------------------
 cacheflush times [2]: 0.0 (0) 19.2 (19281756)
 ---------------------

This one has huge caches and a relatively slow memory subsystem - so the
migration cost is 19 msecs.
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NAshok Raj <ashok.raj@intel.com>
Signed-off-by: NKen Chen <kenneth.w.chen@intel.com>
Cc: <wilder@us.ibm.com>
Signed-off-by: NJohn Hawkes <hawkes@sgi.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Fixes a compiler error in node_to_first_cpu, __ffs expects unsigned long as
a parameter; instead cpumask_t was being passed.  The macro
node_to_first_cpu was not yet used in x86_64 and ia64 arches, and so we never
hit this.  This patch replaces __ffs with first_cpu macro, similar to other
arches.
Signed-off-by: NAlok N Kataria <alokk@calsoftinc.com>
Signed-off-by: NRavikiran G Thirumalai <kiran@scalex86.org>
Signed-off-by: NShai Fultheim <shai@scalex86.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NJohn Hawkes <hawkes@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

pcibus_to_node provides a way for the Linux kernel to identify to which
node a certain pcibus connects to. Allocations of control structures
for devices can then be made on the node where the pci bus is located
to allow local access during interrupt and other device manipulation.

This patch provides a new "node" field in the the pci_controller
structure. The node field will be set based on ACPI information (thanks
to Alex Williamson  <alex.williamson@hp.com for that piece).
Signed-off-by: NChristoph Lameter <clameter@sgi.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Do some basic initial tuning.
Signed-off-by: NNick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.

Let it rip!