Remove a faulty assumption that a long running BAU request has
encountered a hardware problem and will never finish.

Numalink congestion can make a request appear to have
encountered such a problem, but it is not safe to cancel the
request.  If such a cancel is done but a reply is later received
we can miss a TLB shootdown.

We depend upon the max_bau_concurrent 'throttle' to prevent the
stay-busy case from happening.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: gregkh@suse.de
LKML-Reference: <E1OJvNy-0004ad-BV@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Correct the initialization-time assumption of contigous blade
numbers and of sockets numbered from zero.

There may be hubs present with no cpu's enabled.
There may be disabled sockets such that the active socket is not
number zero.

And assign a 'socket master' by assuming that a socket is a
node. (it is not safe to extract socket number from an apicid)
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: gregkh@suse.de
LKML-Reference: <E1OJvNy-0004aW-9S@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Correct the acknowledgment and the reset of a BAU
software-acknowledged message.

A retry message should be testing only for timed-out resources
(mask << 8). (And we delete a log message that might cause
unnecessary concern) The acknowledge MMR is
|--timed-out--|---pending--|,  each is 8 bits.

The IPI-driven reset of software acknowledge resources frees
both timed out and pending resources.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: gregkh@suse.de
LKML-Reference: <E1OJvNy-0004aP-7O@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Move some structure definitions from the C code to the BAU
header file, and change the organization of that header file a
little.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: gregkh@suse.de
LKML-Reference: <E1OJvNy-0004aI-54@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Use a pointer from the per-cpu BAU control structure to the
per-cpu BAU statistics structure.
We nearly always know the first before needing the second.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: gregkh@suse.de
LKML-Reference: <E1OJvNy-0004aB-2k@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The numalink network can become so congested that TLB shootdown
using the Broadcast Assist Unit becomes slower than using IPI's.

In that case, disable the use of the BAU for a period of time.
The period is tunable.  When the period expires the use of the
BAU is re-enabled. A count of these actions is added to the
statistics file.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: gregkh@suse.de
LKML-Reference: <E1OJvNy-0004a4-0a@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Make the Broadcast Assist Unit driver's nine tuning values variable by
making them accessible through a read/write debugfs file.

The file will normally be mounted as
/sys/kernel/debug/sgi_uv/bau_tunables. The tunables are kept in each
cpu's per-cpu BAU structure.

The patch also does a little name improvement, and corrects the reset of
two destination timeout counters.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: gregkh@suse.de
LKML-Reference: <E1OJvNx-0004Zx-Uo@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Calculate the Broadcast Assist Unit's destination timeout period from the
values in the relevant MMR's.

Store it in each cpu's per-cpu BAU structure so that a destination
timeout can be differentiated from a 'plugged' situation in which all
software ack resources are already allocated and a timeout is pending.
That case returns an immediate destination error.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: gregkh@suse.de
LKML-Reference: <E1OJvNx-0004Zq-RK@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

- increase performance of the interrupt handler

- release timed-out software acknowledge resources

- recover from continuous-busy status due to a hardware issue

- add a 'throttle' to keep a uvhub from sending more than a
  specified number of broadcasts concurrently (work around the hardware issue)

- provide a 'nobau' boot command line option

- rename 'pnode' and 'node' to 'uvhub' (the 'node' terminology
  is ambiguous)

- add some new statistics about the scope of broadcasts, retries, the
  hardware issue and the 'throttle'

- split off new function uv_bau_retry_msg() from
  uv_bau_process_message() per community coding style feedback.

- simplify the argument list to uv_bau_process_message(), per
  community coding style feedback.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: linux-mm@kvack.org
Cc: Jack Steiner <steiner@sgi.com>
Cc: Russ Anderson <rja@sgi.com>
Cc: Mike Travis <travis@sgi.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <E1O25Z4-0004Ur-PB@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h

percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files.  percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.

percpu.h -> slab.h dependency is about to be removed.  Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability.  As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.

  http://userweb.kernel.org/~tj/misc/slabh-sweep.py

The script does the followings.

* Scan files for gfp and slab usages and update includes such that
  only the necessary includes are there.  ie. if only gfp is used,
  gfp.h, if slab is used, slab.h.

* When the script inserts a new include, it looks at the include
  blocks and try to put the new include such that its order conforms
  to its surrounding.  It's put in the include block which contains
  core kernel includes, in the same order that the rest are ordered -
  alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
  doesn't seem to be any matching order.

* If the script can't find a place to put a new include (mostly
  because the file doesn't have fitting include block), it prints out
  an error message indicating which .h file needs to be added to the
  file.

The conversion was done in the following steps.

1. The initial automatic conversion of all .c files updated slightly
   over 4000 files, deleting around 700 includes and adding ~480 gfp.h
   and ~3000 slab.h inclusions.  The script emitted errors for ~400
   files.

2. Each error was manually checked.  Some didn't need the inclusion,
   some needed manual addition while adding it to implementation .h or
   embedding .c file was more appropriate for others.  This step added
   inclusions to around 150 files.

3. The script was run again and the output was compared to the edits
   from #2 to make sure no file was left behind.

4. Several build tests were done and a couple of problems were fixed.
   e.g. lib/decompress_*.c used malloc/free() wrappers around slab
   APIs requiring slab.h to be added manually.

5. The script was run on all .h files but without automatically
   editing them as sprinkling gfp.h and slab.h inclusions around .h
   files could easily lead to inclusion dependency hell.  Most gfp.h
   inclusion directives were ignored as stuff from gfp.h was usually
   wildly available and often used in preprocessor macros.  Each
   slab.h inclusion directive was examined and added manually as
   necessary.

6. percpu.h was updated not to include slab.h.

7. Build test were done on the following configurations and failures
   were fixed.  CONFIG_GCOV_KERNEL was turned off for all tests (as my
   distributed build env didn't work with gcov compiles) and a few
   more options had to be turned off depending on archs to make things
   build (like ipr on powerpc/64 which failed due to missing writeq).

   * x86 and x86_64 UP and SMP allmodconfig and a custom test config.
   * powerpc and powerpc64 SMP allmodconfig
   * sparc and sparc64 SMP allmodconfig
   * ia64 SMP allmodconfig
   * s390 SMP allmodconfig
   * alpha SMP allmodconfig
   * um on x86_64 SMP allmodconfig

8. percpu.h modifications were reverted so that it could be applied as
   a separate patch and serve as bisection point.

Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: NTejun Heo <tj@kernel.org>
Guess-its-ok-by: NChristoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>

Update UV mmr definitions header file. Eliminate definitions no
longer needed. Move 2 definitions from tlb_uv.c into the header
file where they belong.
Signed-off-by: NJack Steiner <steiner@sgi.com>
LKML-Reference: <20100310204458.GA28835@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

A memory mapped register that affects the SGI UV Broadcast
Assist Unit's interrupt handling may sometimes be unintialized.

Remove the condition on its initialization, as that condition
can be randomly satisfied by a hardware reset.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: <stable@kernel.org>
LKML-Reference: <E1NBGB9-0005nU-Dp@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Create an inline function to extract the pnode from a global
physical address and then convert the broadcast assist unit to
use the newly created uv_gpa_to_pnode function.

The open-coded code was wrong as well - it might explain a
few of our unexplained bau hangs.
Signed-off-by: NRobin Holt <holt@sgi.com>
Acked-by: NCliff Whickman <cpw@sgi.com>
Cc: linux-mm@kvack.org
Cc: Jack Steiner <steiner@sgi.com>
LKML-Reference: <20091016112920.GZ8903@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

A few parts of the uv_hub_info structure are initialized
incorrectly.

 - n_val is being loaded with m_val.
 - gpa_mask is initialized with a bytes instead of an unsigned long.
 - Handle the case where none of the alias registers are used.

Lastly I converted the bau over to using the uv_hub_info->m_val
which is the correct value.

Without this patch, booting a large configuration hits a
problem where the upper bits of the gnode affect the pnode
and the bau will not operate.
Signed-off-by: NRobin Holt <holt@sgi.com>
Acked-by: NJack Steiner <steiner@sgi.com>
Cc: Cliff Whickman <cpw@sgi.com>
Cc: stable@kernel.org
LKML-Reference: <20091015224946.396355000@alcatraz.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

create_proc_entry() is getting duhprecated.
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Cc: cpw@sgi.com
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The SGI UV Broadcast Assist Unit is used to send TLB shootdown
messages to remote nodes of the system.  The header of the
message must contain the subnode id of the block in the
receiving hub that handles such messages.  It should always be
0x10, the id of the "LB" block.

It had previously been documented as a "must be zero" field.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Acked-by: NJack Steiner <steiner@sgi.com>
LKML-Reference: <E1Mc1x7-0005Ce-6t@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The initialization of the UV Broadcast Assist Unit's sending
buffers was making an invalid assumption about the
initialization of an MMR that defines its address.

The BIOS will not be providing that MMR.  So
uv_activation_descriptor_init() should unconditionally set it.

Tested on UV simulator.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: <stable@kernel.org> # for v2.6.30.x
LKML-Reference: <E1MJTfj-0005i1-W8@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

These are defined as static cpumask_var_t so if MAXSMP is not used,
they are cleared already.  Avoid surprises when MAXSMP is enabled.
Signed-off-by: NYinghai Lu <yinghai.lu@kernel.org>
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>

The UV tlb shootdown code has a serious initialization error.

An array of structures [32*8] is initialized as if it were [32].
The array is indexed by (cpu number on the blade)*8, so the short
initialization works for up to 4 cpus on a blade.
But above that, we provide an invalid opcode to the hub's
broadcast assist unit.

This patch changes the allocation of the array to use its symbolic
dimensions for better clarity. And initializes all 32*8 entries.

Shortened 'UV_ACTIVATION_DESCRIPTOR_SIZE' to 'UV_ADP_SIZE' per Ingo's
recommendation.

Tested on the UV simulator.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: <stable@kernel.org>
LKML-Reference: <E1M6lZR-0007kV-Aq@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This patch correctly sets BAU memory mapped registers to point
to the sending activation descriptor table and target payload table.

The "Broadcast Assist Unit" is used for TLB shootdown in UV.

The memory mapped registers that point to sending and receiving
memory structures contain node numbers.

In one case the __pa() function did not provide the node id of
memory on blade zero in configurations where that id is nonzero.
In another case, it was assumed that memory was allocated on
the local node.  That assumption is not true in a configuration
in which the node has no memory.

Tested on the UV hardware simulator.

[ Impact: fix possible runtime crash due to incorrect TLB logic ]
Signed-off-by: NCliff Wickman <cpw@sgi.com>
LKML-Reference: <E1LuR5Z-0007An-B8@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This patch enables each partition's BAU distribution bit map
to be partition-relative.

The distribution bitmap had been constructed assuming 0 as the base
node number.  That construct would not have allowed a total system of
greater than 256 nodes.
It also corrects an error that occurred when the first blade's nasid
was not zero.  That nasid was stored as the base node.
The base node number gets added by hardware to the node numbers implied
in the distribution bitmap, resulting in invalid target nasids.

Tested on the UV hardware simulator.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
LKML-Reference: <E1Ltl0C-0004Ob-37@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This patch replaces a 'nop' uv_enable_timeouts() in the
UV TLB shootdown code. (somehow, long ago that function got
eviscerated)

If any cpu in the destination node does not get interrupted by the
message and post completion in a reasonable time the hardware
should respond to the sender with an error.  This function
enables such timeouts.

Tested on the UV hardware simulator.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
LKML-Reference: <E1LpjXU-00007e-Qh@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This patch fixes BAU initialization for systems containing
nodes with no memory and for systems with non-consecutive
node numbers.

Fixes and clarifies situations where pnode should be used instead
of node id.

Tested on the UV hardware simulator.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
LKML-Reference: <E1LpjX3-00007N-12@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: fix boot crash on UV systems

Commit 76ba0ecd "cpumask: use
cpumask_var_t in uv_flush_tlb_others" used cur_cpu as an iterator;
it was supposed to be zero for the code below it.
Reported-by: NCliff Wickman <cpw@sgi.com>
Original-From: Cliff Wickman <cpw@sgi.com>
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Acked-by: NMike Travis <travis@sgi.com>
Cc: steiner@sgi.com
Cc: <stable@kernel.org>
LKML-Reference: <200903180822.31196.rusty@rustcorp.com.au>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: remove cpumask_t, reduce per-cpu size for CONFIG_CPUMASK_OFFSTACK=y
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>

In uv_flush_tlb_others() (arch/x86/kernel/tlb_uv.c),
the "WARN_ON(!in_atomic())" fails if CONFIG_PREEMPT is not enabled.

And CONFIG_PREEMPT is not enabled by default in the distribution that
most UV owners will use.

We could #ifdef CONFIG_PREEMPT the warning, but that is not good form.
And there seems to be no suitable fix to in_atomic() when CONFIG_PREMPT
is not on.

As Ingo commented:

  > and we have no proper primitive to test for atomicity. (mainly
  > because we dont know about atomicity on a non-preempt kernel)

So we drop the WARN_ON.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: cleanup
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: cleanup

Remove genapic.h and remove all references to it.
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: fix possible tlb mis-flushing on UV

uv_flush_send_and_wait() should return a pointer if the broadcast
remote tlb shootdown requests fail. That causes the conventional IPI
method of shootdown to be used.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Signed-off-by: NTejun Heo <tj@kernel.org>

Spread mach_apic.h definitions into genapic.h. (with some knock-on effects
on smp.h and apic.h.)
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: cleanup

Make the following uv related cleanups.

* collect visible uv related definitions and interfaces into uv/uv.h
  and use it.  this cleans up the messy situation where on 64bit, uv
  is defined properly, on 32bit generic it's dummy and on the rest
  undefined.  after this clean up, uv is defined on 64 and dummy on
  32.

* update uv_flush_tlb_others() such that it takes cpumask of
  to-be-flushed cpus as argument, instead of that minus self, and
  returns yet-to-be-flushed cpumask, instead of modifying the passed
  in parameter.  this interface change will ease dummy implementation
  of uv_flush_tlb_others() and makes uv tlb flush related stuff
  defined in tlb_uv proper.
Signed-off-by: NTejun Heo <tj@kernel.org>

The function uv_wait_completion() spins on reads of a memory-mapped
register, waiting for completion of BAU hardware replies.

It should call "cpu_relax()" between those reads to improve performance
on hyperthreaded configurations.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Acked-by: NJack Steiner <steiner@sgi.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: reduce stack usage, use new cpumask API.

This is made a little more tricky by uv_flush_tlb_others which
actually alters its argument, for an IPI to be sent to the remaining
cpus in the mask.

I solve this by allocating a cpumask_var_t for this case and falling back
to IPI should this fail.

To eliminate temporaries in the caller, all flush_tlb_others implementations
now do the this-cpu-elimination step themselves.

Note also the curious "cpus_or(f->flush_cpumask, cpumask, f->flush_cpumask)"
which has been there since pre-git and yet f->flush_cpumask is always zero
at this point.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NMike Travis <travis@sgi.com>

Impact: fix crash on x86/UV

UV is the SGI "UltraViolet" machine, which is x86_64 based.
BAU is the "Broadcast Assist Unit", used for TLB shootdown in UV.

This patch removes the allocation and initialization of an unused table.

This table is left over from a development test mode.  It is unused in
the present code.

And it was incorrectly initialized: 8 entries allocated but 17 initialized,
causing slab corruption.

This patch should go into 2.6.27 and 2.6.28 as well as the current tree.

Diffed against 2.6.28 (linux-next, 12/30/08)
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: <stable@kernel.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: fix double entry creation in /proc

There is a collision between two UV functions:
  both uv_ptc_init() and gru_proc_init() try to make /proc/sgi_uv

So move it's creation to a single place: uv_system_init()
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

For some reason tlb_uv was including linux/mc146818rtc.h.  It really
just needs linux/seq_file.h
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citix.com>
Cc: Cliff Wickman <cpw@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The UV TLB shootdown mechanism needs a system interrupt vector.

Its vector had been hardcoded as 200, but needs to moved to the reserved
system vector range so that it does not collide with some device vector.

This is still temporary until dynamic system IRQ allocation is provided.
But it will be needed when real UV hardware becomes available and runs 2.6.27.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Someone could write 0 bytes to /proc/sgi_uv/ptc_statistics,
causing
  optstr[count - 1] = '\0';
to write to who-knows-where.

(Andi Kleen noticed this need from a patch I sent for
 similar code in the ia64 world (sn2_ptc_proc_write()).)

(count less than zero is not possible here, as count is unsigned)
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

v6: 6/19 close the security hole in uv_ptc_proc_write())

  > Found a potential security hole while doing that:
  > static ssize_t uv_ptc_proc_write(struct file *file, const char __user *user,
  >                              size_t count, loff_t *data)
  >     if (copy_from_user(optstr, user, count))
  >             return -EFAULT;
  >
  > is count guaranteed to never be larger than 64?

is fixed below.

It adds tlb_uv.o to the Makefile.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: mingo@elte.hu
Signed-off-by: NIngo Molnar <mingo@elte.hu>

fix:

arch/x86/kernel/tlb_uv.c: In function ‘uv_table_bases_init':
arch/x86/kernel/tlb_uv.c:612: error: ‘bau_tabsp' undeclared (first use in this function)
arch/x86/kernel/tlb_uv.c:612: error: (Each undeclared identifier is reported only once
arch/x86/kernel/tlb_uv.c:612: error: for each function it appears in.)
Signed-off-by: NIngo Molnar <mingo@elte.hu>