This reverts commit 8eba1842.

uv_trace() is not used by anything, nor is uv_trace_nmi_func, nor
uv_trace_func.

That's not how we do instrumentation code in the kernel: we add
tracepoints, printk()s, etc. so that everyone not just those with
magic kernel modules can debug a system.

So remove this unused (and misguied) piece of code.
Signed-off-by: NIngo Molnar <mingo@kernel.org>
Cc: Mike Travis <travis@sgi.com>
Cc: Dimitri Sivanich <sivanich@sgi.com>
Cc: Hedi Berriche <hedi@sgi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Cc: Jason Wessel <jason.wessel@windriver.com>
Link: http://lkml.kernel.org/n/tip-tumfBffmr4jmnt8Gyxanoblg@git.kernel.org

This patch adds support for the uvtrace module by providing a
skeleton call to the registered trace function.  It also
provides another separate 'NMI' tracer that is triggered by the
system wide 'power nmi' command.
Signed-off-by: NMike Travis <travis@sgi.com>
Reviewed-by: NDimitri Sivanich <sivanich@sgi.com>
Reviewed-by: NHedi Berriche <hedi@sgi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Cc: Jason Wessel <jason.wessel@windriver.com>
Link: http://lkml.kernel.org/r/20130923212501.185052551@asylum.americas.sgi.comSigned-off-by: NIngo Molnar <mingo@kernel.org>

The current UV NMI handler has not been updated for the changes
in the system NMI handler and the perf operations.  The UV NMI
handler reads an MMR in the UV Hub to check to see if the NMI
event was caused by the external 'system NMI' that the operator
can initiate on the System Mgmt Controller.

The problem arises when the perf tools are running, causing
millions of perf events per second on very large CPU count
systems.  Previously this was okay because the perf NMI handler
ran at a higher priority on the NMI call chain and if the NMI
was a perf event, it would stop calling other NMI handlers
remaining on the NMI call chain.

Now the system NMI handler calls all the handlers on the NMI
call chain including the UV NMI handler.  This causes the UV NMI
handler to read the MMRs at the same millions per second rate.
This can lead to significant performance loss and possible
system failures.  It also can cause thousands of 'Dazed and
Confused' messages being sent to the system console.  This
effectively makes perf tools unusable on UV systems.

To avoid this excessive overhead when perf tools are running,
this code has been optimized to minimize reading of the MMRs as
much as possible, by moving to the NMI_UNKNOWN notifier chain.
This chain is called only when all the users on the standard
NMI_LOCAL call chain have been called and none of them have
claimed this NMI.

There is an exception where the NMI_LOCAL notifier chain is
used.  When the perf tools are in use, it's possible that the UV
NMI was captured by some other NMI handler and then either
ignored or mistakenly processed as a perf event.  We set a
per_cpu ('ping') flag for those CPUs that ignored the initial
NMI, and then send them an IPI NMI signal.  The NMI_LOCAL
handler on each cpu does not need to read the MMR, but instead
checks the in memory flag indicating it was pinged.  There are
two module variables, 'ping_count' indicating how many requested
NMI events occurred, and 'ping_misses' indicating how many stray
NMI events.  These most likely are perf events so it shows the
overhead of the perf NMI interrupts and how many MMR reads were avoided.

This patch also minimizes the reads of the MMRs by having the
first cpu entering the NMI handler on each node set a per HUB
in-memory atomic value.  (Having a per HUB value avoids sending
lock traffic over NumaLink.)  Both types of UV NMIs from the SMI
layer are supported.
Signed-off-by: NMike Travis <travis@sgi.com>
Reviewed-by: NDimitri Sivanich <sivanich@sgi.com>
Reviewed-by: NHedi Berriche <hedi@sgi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Cc: Jason Wessel <jason.wessel@windriver.com>
Link: http://lkml.kernel.org/r/20130923212500.353547733@asylum.americas.sgi.comSigned-off-by: NIngo Molnar <mingo@kernel.org>

This patch moves the UV NMI support from the x2apic file to a
new separate uv_nmi.c file in preparation for the next sequence
of patches. It prevents upcoming bloat of the x2apic file, and
has the added benefit of putting the upcoming /sys/module
parameters under the name 'uv_nmi' instead of 'x2apic_uv_x',
which was obscure.
Signed-off-by: NMike Travis <travis@sgi.com>
Reviewed-by: NDimitri Sivanich <sivanich@sgi.com>
Reviewed-by: NHedi Berriche <hedi@sgi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Cc: Jason Wessel <jason.wessel@windriver.com>
Link: http://lkml.kernel.org/r/20130923212500.183295611@asylum.americas.sgi.comSigned-off-by: NIngo Molnar <mingo@kernel.org>

[Purpose of this patch]

As Vaibhav explained in the thread below, tracepoints for irq vectors
are useful.

http://www.spinics.net/lists/mm-commits/msg85707.html

<snip>
The current interrupt traces from irq_handler_entry and irq_handler_exit
provide when an interrupt is handled.  They provide good data about when
the system has switched to kernel space and how it affects the currently
running processes.

There are some IRQ vectors which trigger the system into kernel space,
which are not handled in generic IRQ handlers.  Tracing such events gives
us the information about IRQ interaction with other system events.

The trace also tells where the system is spending its time.  We want to
know which cores are handling interrupts and how they are affecting other
processes in the system.  Also, the trace provides information about when
the cores are idle and which interrupts are changing that state.
<snip>

On the other hand, my usecase is tracing just local timer event and
getting a value of instruction pointer.

I suggested to add an argument local timer event to get instruction pointer before.
But there is another way to get it with external module like systemtap.
So, I don't need to add any argument to irq vector tracepoints now.

[Patch Description]

Vaibhav's patch shared a trace point ,irq_vector_entry/irq_vector_exit, in all events.
But there is an above use case to trace specific irq_vector rather than tracing all events.
In this case, we are concerned about overhead due to unwanted events.

So, add following tracepoints instead of introducing irq_vector_entry/exit.
so that we can enable them independently.
   - local_timer_vector
   - reschedule_vector
   - call_function_vector
   - call_function_single_vector
   - irq_work_entry_vector
   - error_apic_vector
   - thermal_apic_vector
   - threshold_apic_vector
   - spurious_apic_vector
   - x86_platform_ipi_vector

Also, introduce a logic switching IDT at enabling/disabling time so that a time penalty
makes a zero when tracepoints are disabled. Detailed explanations are as follows.
 - Create trace irq handlers with entering_irq()/exiting_irq().
 - Create a new IDT, trace_idt_table, at boot time by adding a logic to
   _set_gate(). It is just a copy of original idt table.
 - Register the new handlers for tracpoints to the new IDT by introducing
   macros to alloc_intr_gate() called at registering time of irq_vector handlers.
 - Add checking, whether irq vector tracing is on/off, into load_current_idt().
   This has to be done below debug checking for these reasons.
   - Switching to debug IDT may be kicked while tracing is enabled.
   - On the other hands, switching to trace IDT is kicked only when debugging
     is disabled.

In addition, the new IDT is created only when CONFIG_TRACING is enabled to avoid being
used for other purposes.
Signed-off-by: NSeiji Aguchi <seiji.aguchi@hds.com>
Link: http://lkml.kernel.org/r/51C323ED.5050708@hds.comSigned-off-by: NH. Peter Anvin <hpa@linux.intel.com>
Cc: Steven Rostedt <rostedt@goodmis.org>

This patch trims the MMR register definitions after the updates for the
SGI UV3 system have been applied.  Note that because these definitions
are automatically generated from the RTL we cannot control the length
of the names.  Therefore there are lines that exceed 80 characters.
Signed-off-by: NMike Travis <travis@sgi.com>
Link: http://lkml.kernel.org/r/20130211194509.173026880@gulag1.americas.sgi.comAcked-by: NRuss Anderson <rja@sgi.com>
Reviewed-by: NDimitri Sivanich <sivanich@sgi.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

This patch updates the UV HUB info for UV3.  The "is_uv3_hub" and
"is_uvx_hub" (UV2 or UV3) functions are added as well as the addresses
and sizes of the MMR regions for UV3.
Signed-off-by: NMike Travis <travis@sgi.com>
Link: http://lkml.kernel.org/r/20130211194508.610723192@gulag1.americas.sgi.comAcked-by: NRuss Anderson <rja@sgi.com>
Reviewed-by: NDimitri Sivanich <sivanich@sgi.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

This patch updates the MMR register definitions for the SGI UV3 system.
Note that because these definitions are automatically generated from
the RTL we cannot control the length of the names.  Therefore there are
lines that exceed 80 characters.

All the new MMR definitions are added in this patch.  The patches that
follow then update the references. The last patch is a "trim" patch
which reduces the size of the MMR definitions file by about a third.
This keeps "bi-sectability" in place as the intermediate patches would
not compile correctly if the trimmed MMR defines were done first.
Signed-off-by: NMike Travis <travis@sgi.com>
Link: http://lkml.kernel.org/r/20130211194508.326204556@gulag1.americas.sgi.comAcked-by: NRuss Anderson <rja@sgi.com>
Reviewed-by: NDimitri Sivanich <sivanich@sgi.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

The flush tlb optimization code has logical issue on UV
platform.  It doesn't flush the full range at all, since it
simply ignores its 'end' parameter (and hence also the "all"
indicator) in uv_flush_tlb_others() function.

Cliff's notes:

 | I tested the patch on a UV.  It has the effect of either
 | clearing 1 or all TLBs in a cpu.  I added some debugging to
 | test for the cases when clearing all TLBs is overkill, and in
 | practice it happens very seldom.
Reported-by: NJan Beulich <jbeulich@suse.com>
Signed-off-by: NAlex Shi <alex.shi@intel.com>
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Tested-by: NCliff Wickman <cpw@sgi.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NIngo Molnar <mingo@kernel.org>

x86 has no flush_tlb_range support in instruction level. Currently the
flush_tlb_range just implemented by flushing all page table. That is not
the best solution for all scenarios. In fact, if we just use 'invlpg' to
flush few lines from TLB, we can get the performance gain from later
remain TLB lines accessing.

But the 'invlpg' instruction costs much of time. Its execution time can
compete with cr3 rewriting, and even a bit more on SNB CPU.

So, on a 512 4KB TLB entries CPU, the balance points is at:
	(512 - X) * 100ns(assumed TLB refill cost) =
		X(TLB flush entries) * 100ns(assumed invlpg cost)

Here, X is 256, that is 1/2 of 512 entries.

But with the mysterious CPU pre-fetcher and page miss handler Unit, the
assumed TLB refill cost is far lower then 100ns in sequential access. And
2 HT siblings in one core makes the memory access more faster if they are
accessing the same memory. So, in the patch, I just do the change when
the target entries is less than 1/16 of whole active tlb entries.
Actually, I have no data support for the percentage '1/16', so any
suggestions are welcomed.

As to hugetlb, guess due to smaller page table, and smaller active TLB
entries, I didn't see benefit via my benchmark, so no optimizing now.

My micro benchmark show in ideal scenarios, the performance improves 70
percent in reading. And in worst scenario, the reading/writing
performance is similar with unpatched 3.4-rc4 kernel.

Here is the reading data on my 2P * 4cores *HT NHM EP machine, with THP
'always':

multi thread testing, '-t' paramter is thread number:
	       	        with patch   unpatched 3.4-rc4
./mprotect -t 1           14ns		24ns
./mprotect -t 2           13ns		22ns
./mprotect -t 4           12ns		19ns
./mprotect -t 8           14ns		16ns
./mprotect -t 16          28ns		26ns
./mprotect -t 32          54ns		51ns
./mprotect -t 128         200ns		199ns

Single process with sequencial flushing and memory accessing:

		       	with patch   unpatched 3.4-rc4
./mprotect		    7ns			11ns
./mprotect -p 4096  -l 8 -n 10240
			    21ns		21ns

[ hpa: http://lkml.kernel.org/r/1B4B44D9196EFF41AE41FDA404FC0A100BFF94@SHSMSX101.ccr.corp.intel.com
  has additional performance numbers. ]
Signed-off-by: NAlex Shi <alex.shi@intel.com>
Link: http://lkml.kernel.org/r/1340845344-27557-3-git-send-email-alex.shi@intel.comSigned-off-by: NH. Peter Anvin <hpa@zytor.com>

On SGI's UV2 the BAU (Broadcast Assist Unit) driver can hang
under a heavy load. To cure this:

- Disable the UV2 extended status mode (see UV2_EXT_SHFT), as
  this mode changes BAU behavior in more ways then just delivering
  an extra bit of status.  Revert status to just two meaningful bits,
  like UV1.

- Use no IPI-style resets on UV2.  Just give up the request for
  whatever the reason it failed and let it be accomplished with
  the legacy IPI method.

- Use no alternate sending descriptor (the former UV2 workaround
  bcp->using_desc and handle_uv2_busy() stuff).  Just disable the
  use of the BAU for a period of time in favor of the legacy IPI
  method when the h/w bug leaves a descriptor busy.

  -- new tunable: giveup_limit determines the threshold at which a hub is
     so plugged that it should do all requests with the legacy IPI method for a
     period of time
  -- generalize disable_for_congestion() (renamed disable_for_period()) for
     use whenever a hub should avoid using the BAU for a period of time

Also:

 - Fix find_another_by_swack(), which is part of the UV2 bug workaround

 - Correct and clarify the statistics (new stats s_overipilimit, s_giveuplimit,
   s_enters, s_ipifordisabled, s_plugged, s_congested)
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Link: http://lkml.kernel.org/r/20120622131459.GC31884@sgi.comSigned-off-by: NIngo Molnar <mingo@kernel.org>

This patch enables the BAU to be turned on or off dynamically.

  echo "on"  > /proc/sgi_uv/ptc_statistics
  echo "off" > /proc/sgi_uv/ptc_statistics

The system may be booted with or without the nobau option.

Whether the system currently has the BAU off can be seen in
the /proc file -- normally with the baustats script.
Each cpu will have a 1 in the bauoff field if the BAU was turned
off, so baustats will give a count of cpus that have it off.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Link: http://lkml.kernel.org/r/20120622131330.GB31884@sgi.comSigned-off-by: NIngo Molnar <mingo@kernel.org>

The SGI Altix UV2 BAU (Broadcast Assist Unit) as used for
tlb-shootdown (selective broadcast mode) always uses UV2
broadcast descriptor format. There is no need to clear the
'legacy' (UV1) mode, because the hardware always uses UV2 mode
for selective broadcast.

But the BIOS uses general broadcast and legacy mode, and the
hardware pays attention to the legacy mode bit for general
broadcast. So the kernel must not clear that mode bit.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: <stable@kernel.org>
Link: http://lkml.kernel.org/r/E1SccoO-0002Lh-Cb@eag09.americas.sgi.comSigned-off-by: NIngo Molnar <mingo@kernel.org>

uv_gpa_to_soc_phys_ram() was inadvertently ignoring the
shift values.  This fix takes the shift into account.
Signed-off-by: NRuss Anderson <rja@sgi.com>
Cc: <stable@kernel.org>
Link: http://lkml.kernel.org/r/20120119020753.GA7228@sgi.comSigned-off-by: NIngo Molnar <mingo@elte.hu>

This patch adds separate accounting of UV2 message "strong
nack's" in the BAU statistics.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Link: http://lkml.kernel.org/r/20120116212238.GF5767@sgi.comSigned-off-by: NIngo Molnar <mingo@elte.hu>

This patch implements a workaround for a UV2 hardware bug.
The bug is a non-atomic update of a memory-mapped register. When
hardware message delivery and software message acknowledge occur
simultaneously the pending message acknowledge for the arriving
message may be lost.  This causes the sender's message status to
stay busy.

Part of the workaround is to not acknowledge a completed message
until it is verified that no other message is actually using the
resource that is mistakenly recorded in the completed message.

Part of the workaround is to test for long elapsed time in such
a busy condition, then handle it by using a spare sending
descriptor. The stay-busy condition is eventually timed out by
hardware, and then the original sending descriptor can be
re-used. Most of that logic change is in keeping track of the
current descriptor and the state of the spares.

The occurrences of the workaround are added to the BAU
statistics.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Link: http://lkml.kernel.org/r/20120116211947.GC5767@sgi.com
Cc: <stable@kernel.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Update the use of the Broadcast Assist Unit on SGI Altix UV2 to
the use of native UV2 mode on new hardware (not the legacy mode).

UV2 native mode has a different format for a broadcast message.
We also need quick differentiaton between UV1 and UV2.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Link: http://lkml.kernel.org/r/20120116211750.GA5767@sgi.com
Cc: <stable@kernel.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

There was a mixup when the SGI UV2 hub chip was sent to be
fabricated, and it ended up with the wrong part number in the
HRP_NODE_ID mmr. Future versions of the chip will (may) have the
correct part number. Change the UV infrastructure to recognize
both part numbers as valid IDs of a UV2 hub chip.
Signed-off-by: NJack Steiner <steiner@sgi.com>
Link: http://lkml.kernel.org/r/20111129210058.GA20452@sgi.comSigned-off-by: NIngo Molnar <mingo@elte.hu>

This is a workaround for a UV2 hub bug that affects the format of system
global addresses.

The GRU API for UV2 was inadvertently broken by a hardware change.  The
format of the physical address used for TLB dropins and for addresses used
with instructions running in unmapped mode has changed.  This change was
not documented and became apparent only when diags failed running on
system simulators.

For UV1, TLB and GRU instruction physical addresses are identical to
socket physical addresses (although high NASID bits must be OR'ed into the
address).

For UV2, socket physical addresses need to be converted.  The NODE portion
of the physical address needs to be shifted so that the low bit is in bit
39 or bit 40, depending on an MMR value.

It is not yet clear if this bug will be fixed in a silicon respin.  If it
is fixed, the hub revision will be incremented & the workaround disabled.
Signed-off-by: NJack Steiner <steiner@sgi.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

This covers the trivial cases from open-coded xadd to the xadd macros.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Link: http://lkml.kernel.org/r/4E5BCC40.3030501@goop.orgSigned-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Correct the UV2 broacast assist unit's destination timeout
period. And the activation status register in UV2 should be
tested for a destination timeout with a 4, not a 2.  The values
for Active versus Timeout were reversed.

This patch is critical for TLB shootdown on an Altix UV2 system
(i.e. the follow-on to the current Altix UV).

 Destination timeout period:
  The period is set in 4 bits of memory-mapped register MISC_CONTROL.
  The left bit toggles base period between 10us and 80us.
  The other 3 bits are the multiplier.
 Decimal 15, hex f, gives the maximum: 7 * 80us
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Link: http://lkml.kernel.org/r/20110621122243.117324443@sgi.comSigned-off-by: NIngo Molnar <mingo@elte.hu>

Remove the large stack-resident cpumask_t from
reset_with_ipi()'s stack by allocating one per uvhub.

Due to the limited size of the stack the potentially huge cpumask_t may
cause stack overrun.  We haven't seen it happen yet, but we need to make it
a practice not to push such structures onto the stack.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Reviewed-by: NPekka Enberg <penberg@kernel.org>
Link: http://lkml.kernel.org/r/20110621122242.832589130@sgi.comSigned-off-by: NIngo Molnar <mingo@elte.hu>

Rename 'bau_targ_hubmask' to 'pnmask' for clarity.

The BAU distribution bit mask is indexed by pnode number, not hub or
blade number.  This important fact is not clear while the mask is
called a 'hubmask'.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Link: http://lkml.kernel.org/r/20110621122242.630995969@sgi.comSigned-off-by: NIngo Molnar <mingo@elte.hu>

Make all the functions in uv_bau.h inline so that it can
be included in the fake prom (used in simulations).

If not inlined the unused functions will generate compiler warnings.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Reviewed-by: NPekka Enberg <penberg@kernel.org>
Link: http://lkml.kernel.org/r/20110621122242.230529678@sgi.comSigned-off-by: NIngo Molnar <mingo@elte.hu>

No code changes. Reformat definitions to make it more readable.

I fixed alignment of comments in the structure definitions.

Also aligned comments and most field definitions & values. Also
sorted the defines for the SHIFT & MASK values for each MMR.
This make the file visually much more acceptable.

Some of the symbol names are still quite long. The file is based
on post-processing of verilog definitions that are used for the
node controller chip design. Although some symbol names are not
what I would chose, I would like to maintain compatibility with
the names used by the chip designers. We have a number of
cross-reference utilities & having common names is important.
Signed-off-by: NJack Steiner <steiner@sgi.com>
Link: http://lkml.kernel.org/r/20110527145256.GA31224@sgi.comSigned-off-by: NIngo Molnar <mingo@elte.hu>
--
 arch/x86/include/asm/uv/uv_mmrs.h | 2873 +++++++++++++++++++++-----------------
 1 file changed, 1600 insertions(+), 1273 deletions(-)

SGI UV's uv_tlb.c driver has become rather hard to read, with overly large
functions, non-standard coding style and (way) too long variable, constant
and function names and non-obvious code flow sequences.

This patch improves the readability and maintainability of the driver
significantly, by doing the following strict code cleanups with no side
effects:

 - Split long functions into shorter logical functions.

 - Shortened some variable and structure member names.

 - Added special functions for reads and writes of MMR regs with
   very long names.

 - Added the 'tunables' table to shortened tunables_write().

 - Added the 'stat_description' table to shorten uv_ptc_proc_write().

 - Pass fewer 'stat' arguments where it can be derived from the 'bcp'
   argument.

 - Function definitions consistent on one line, and inline in few (short) cases.

 - Moved some small structures and an atomic inline function to the header file.

 - Moved some local variables to the blocks where they are used.

 - Updated the copyright date.

 - Shortened uv_write_global_mmr64() etc. using some aliasing; no
   line breaks. Renamed many uv_.. functions that are not exported.

 - Aligned structure fields.
    [ note that not all structures are aligned the same way though; I'd like
      to keep the extensive commenting in some of them. ]

 - Shortened some long structure names.

 - Standard pass/fail exit from init_per_cpu()

 - Vertical alignment for mass initializations.

 - More separation between blocks of code.

Tested on a 16-processor Altix UV.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: penberg@kernel.org
Link: http://lkml.kernel.org/r/E1QOw12-0004MN-Lp@eag09.americas.sgi.comSigned-off-by: NIngo Molnar <mingo@elte.hu>

This patch adds support for a new version of the SGI UV hub
chip. The hub chip is the node controller that connects multiple
blades into a larger coherent SSI.

For the most part, UV2 is compatible with UV1. The majority of
the changes are in the addresses of MMRs and in a few cases, the
contents of MMRs. These changes are the result in changes in the
system topology such as node configuration, processor types,
maximum nodes, physical address sizes, etc.
Signed-off-by: NJack Steiner <steiner@sgi.com>
Link: http://lkml.kernel.org/r/20110511175028.GA18006@sgi.comSigned-off-by: NIngo Molnar <mingo@elte.hu>

This is a fix for the SGI Altix-UV Broadcast Assist Unit code,
which is used for TLB flushing.

Certain hardware configurations (that customers are ordering)
cause nasids (numa address space id's) to be non-consecutive.
Specifically, once you have more than 4 blades in a IRU
(Individual Rack Unit - or 1/2 rack) but less than the maximum
of 16, the nasid numbering becomes non-consecutive.  This
currently results in a 'catastrophic error' (CATERR) detected by
the firmware during OS boot.  The BAU is generating an 'INTD'
request that is targeting a non-existent nasid value. Such
configurations may also occur when a blade is configured off
because of hardware errors. (There is one UV hub per blade.)

This patch is required to support such configurations.

The problem with the tlb_uv.c code is that is using the
consecutive hub numbers as indices to the BAU distribution bit
map. These are simply the ordinal position of the hub or blade
within its partition.  It should be using physical node numbers
(pnodes), which correspond to the physical nasid values. Use of
the hub number only works as long as the nasids in the partition
are consecutive and increase with a stride of 1.

This patch changes the index to be the pnode number, thus
allowing nasids to be non-consecutive.
It also provides a table in local memory for each cpu to
translate target cpu number to target pnode and nasid.
And it improves naming to properly reflect 'node' and 'uvhub'
versus 'nasid'.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: <stable@kernel.org>
Link: http://lkml.kernel.org/r/E1QJmxX-0002Mz-Fk@eag09.americas.sgi.comSigned-off-by: NIngo Molnar <mingo@elte.hu>

This fixes problems seen on UV systems handling NMIs from the
node controller.

I isolated the "dazed..." messages that I saw earlier to a bug in
the BMC on our platform. It was sending NMIs w/o properly setting
a register that indicated the source of NMI.

So rather than _assuming_ any unhandled NMI came from the UV system
maintenance console (SMC), add a check to verify that the SMC actually
sent the NMI.
Signed-off-by: NJack Steiner <steiner@sgi.com>
Cc: gorcunov@gmail.com
Cc: dzickus@redhat.com
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The BAU's initialization of the broadcast description header is
lacking the coherence domain (high bits) in the nasid.  This
causes a catastrophic system failure when running on a system
with multiple coherence domains.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
LKML-Reference: <E1PxKBB-0005F0-3U@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Fix a hard-coded limit of a maximum of 16 cpu's per socket.

The UV Broadcast Assist Unit code initializes by scanning the
cpu topology of the system and assigning a master cpu for each
socket and UV hub. That scan had an assumption of a limit of 16
cpus per socket. With Westmere we are going over that limit.
The UV hub hardware will allow up to 32.

If the scan finds the system has gone over that limit it returns
an error and we print a warning and fall back to doing TLB
shootdowns without the BAU.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: <stable@kernel.org> # .37.x
LKML-Reference: <E1PZol7-0000mM-77@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This patch for SGI UV systems addresses a problem whereby
interrupt transactions being looped back from a local IOH,
through the hub to a local CPU can (erroneously) conflict with
IO port operations and other transactions.

To workaound this we set a high bit in the APIC IDs used for
interrupts. This bit appears to be ignored by the sockets, but
it avoids the conflict in the hub.
Signed-off-by: NDimitri Sivanich <sivanich@sgi.com>
LKML-Reference: <20101116222352.GA8155@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
___

 arch/x86/include/asm/uv/uv_hub.h   |    4 ++++
 arch/x86/include/asm/uv/uv_mmrs.h  |   19 ++++++++++++++++++-
 arch/x86/kernel/apic/x2apic_uv_x.c |   25 +++++++++++++++++++++++--
 arch/x86/platform/uv/tlb_uv.c      |    2 +-
 arch/x86/platform/uv/uv_time.c     |    4 +++-
 5 files changed, 49 insertions(+), 5 deletions(-)

A new version of the SGI UV hub node controller is being
developed. A few of the MMRs (control registers) that exist on
the current hub no longer exist on the new hub. Fortunately,
there are alternate MMRs that are are functionally equivalent
and that exist on both hubs.

This patch changes the UV code to use MMRs that exist in BOTH
versions of the hub node controller.
Signed-off-by: NJack Steiner <steiner@sgi.com>
LKML-Reference: <20101106204056.GA27584@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Enable Westmere support on SGI UV.  The UV initialization code is dependent on
the APICID bits.  Westmere-EX uses different APIC bit mapping than Nehalem-EX.
This code reads the apic shift value from a UV MMR to do the proper bit
decoding to determint the pnode.
Signed-off-by: NRuss Anderson <rja@sgi.com>
LKML-Reference: <20101026212728.GB15071@sgi.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Streamline the large uv_flush_send_and_wait() function by use of
a couple of helper functions.

And remove some excess comments.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: gregkh@suse.de
LKML-Reference: <E1OJvNy-0004ay-IH@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Make the Broadcast Assist Unit driver use the BAU for TLB
shootdowns of cpu's on the local uvhub.

It was previously thought that IPI might be faster to the cpu's
on the local hub.  But the IPI operation would have to follow
the completion of the BAU broadcast anyway.  So we broadcast to
the local uvhub in all cases except when the current cpu was the
only local cpu in the mask.

This simplifies uv_flush_send_and_wait() in that it returns
either all shootdowns complete, or none.

Adjust the statistics to account for shootdowns on the local
uvhub.
Signed-off-by: NCliff Wickman <cpw@sgi.com>
Cc: gregkh@suse.de
LKML-Reference: <E1OJvNy-0004aq-G7@eag09.americas.sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

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>

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>