For the x2apic cluster mode, vector for an interrupt is
currently reserved on all the cpu's that are part of the x2apic
cluster. But the interrupts will be routed only to the cluster
(derived from the first cpu in the mask) members specified in
the mask. So there is no need to reserve the vector in the
unused cluster members.
Modify __assign_irq_vector() to reserve the vectors based on the
user specified irq destination mask. If the new mask is a proper
subset of the currently used mask, cleanup the vector allocation
on the unused cpu members.
Also, allow the apic driver to tune the vector domain based on
the affinity mask (which in most cases is the user-specified
mask).
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Acked-by: Yinghai Lu <yinghai@kernel.org>
Acked-by: Alexander Gordeev <agordeev@redhat.com>
Acked-by: Cyrill Gorcunov <gorcunov@openvz.org>
Link: http://lkml.kernel.org/r/1340656709-11423-3-git-send-email-suresh.b.siddha@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Currently __assign_irq_vector() goes through each cpu in the
specified mask until it finds a free vector in all the cpu's
that are part of the same interrupt domain. We visit all the
interrupt domain sibling cpus to reserve the free vector. So,
when we fail to find a free vector in an interrupt domain, it is
safe to continue our search with a cpu belonging to a new
interrupt domain. No need to go through each cpu, if the domain
containing that cpu is already visited.
Use the irq_cfg's old_domain to track the visited domains and
optimize the cpu traversal while finding a free vector in the
given cpumask.
NOTE: We can also optimize the search by using for_each_cpu() and
skip the current cpu, if it is not the first cpu in the mask
returned by the vector_allocation_domain(). But re-using the
cfg->old_domain to track the visited domains will be slightly
faster.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Acked-by: Yinghai Lu <yinghai@kernel.org>
Acked-by: Alexander Gordeev <agordeev@redhat.com>
Acked-by: Cyrill Gorcunov <gorcunov@openvz.org>
Link: http://lkml.kernel.org/r/1340656709-11423-2-git-send-email-suresh.b.siddha@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since there are only two locations where cpu_mask_to_apicid() is
called from, remove the operation and use only
cpu_mask_to_apicid_and() instead.
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Suggested-and-acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Acked-by: Yinghai Lu <yinghai@kernel.org>
Link: http://lkml.kernel.org/r/20120614074935.GE3383@dhcp-26-207.brq.redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since commit 8637e38 ("x86/apic: Avoid useless scanning thru a
cpumask in assign_irq_vector()") vector_allocation_domain()
operation indicates if a cpumask is dynamic or static. This
update fixes the oversight and makes the operation to return a
value.
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Link: http://lkml.kernel.org/r/20120614103933.GJ3383@dhcp-26-207.brq.redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Currently cpu_mask_to_apicid() should not get a offline CPU with
the cpumask. Otherwise some apic drivers might try to access
non-existent per-cpu variables (i.e. x2apic). In that regard
cpu_mask_to_apicid() and cpu_mask_to_apicid_and() operations are
inconsistent.
This fix makes the two operations do not rely on calling
functions and always return the apicid for only online CPUs. As
result, the meaning and implementations of cpu_mask_to_apicid()
and cpu_mask_to_apicid_and() operations become straight.
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Link: http://lkml.kernel.org/r/20120607131624.GG4759@dhcp-26-207.brq.redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Current cpu_mask_to_apicid() and cpu_mask_to_apicid_and()
implementations have few shortcomings:
1. A value returned by cpu_mask_to_apicid() is written to
hardware registers unconditionally. Should BAD_APICID get ever
returned it will be written to a hardware too. But the value of
BAD_APICID is not universal across all hardware in all modes and
might cause unexpected results, i.e. interrupts might get routed
to CPUs that are not configured to receive it.
2. Because the value of BAD_APICID is not universal it is
counter- intuitive to return it for a hardware where it does not
make sense (i.e. x2apic).
3. cpu_mask_to_apicid_and() operation is thought as an
complement to cpu_mask_to_apicid() that only applies a AND mask
on top of a cpumask being passed. Yet, as consequence of 18374d8
commit the two operations are inconsistent in that of:
cpu_mask_to_apicid() should not get a offline CPU with the cpumask
cpu_mask_to_apicid_and() should not fail and return BAD_APICID
These limitations are impossible to realize just from looking at
the operations prototypes.
Most of these shortcomings are resolved by returning a error
code instead of BAD_APICID. As the result, faults are reported
back early rather than possibilities to cause a unexpected
behaviour exist (in case of [1]).
The only exception is setup_timer_IRQ0_pin() routine. Although
obviously controversial to this fix, its existing behaviour is
preserved to not break the fragile check_timer() and would
better addressed in a separate fix.
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Link: http://lkml.kernel.org/r/20120607131559.GF4759@dhcp-26-207.brq.redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
If the HW implements round-robin interrupt delivery, this
enables multiple cpu's (which are part of the user specified
interrupt smp_affinity mask and belong to the same x2apic
cluster) to service the interrupt.
Also if the platform supports Power Aware Interrupt Routing,
then this enables the interrupt to be routed to an idle cpu or a
busy cpu depending on the perf/power bias tunable.
We are now grouping all the cpu's in a cluster to one vector
domain. So that will limit the total number of interrupt sources
handled by Linux. Previously we support "cpu-count *
available-vectors-per-cpu" interrupt sources but this will now
reduce to "cpu-count/16 * available-vectors-per-cpu".
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: yinghai@kernel.org
Cc: gorcunov@openvz.org
Cc: agordeev@redhat.com
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1337644682-19854-2-git-send-email-suresh.b.siddha@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We know both register and value for eoi beforehand,
so there's no need to check it and no need to do math
to calculate the msr. Saves instructions/branches
on each EOI when using x2apic.
I looked at the objdump output to verify that the
generated code looks right and actually is shorter.
The real improvemements will be on the KVM guest side
though, those come in a later patch.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: gleb@redhat.com
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/e019d1a125316f10d3e3a4b2f6bda41473f4fb72.1337184153.git.mst@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add eoi_write callback so that kvm can override
eoi accesses without touching the rest of the apic.
As a side-effect, this will enable a micro-optimization
for apics using msr.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: gleb@redhat.com
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/0df425d746c49ac2ecc405174df87752869629d2.1337184153.git.mst@redhat.com
[ tidied it up a bit ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
As suggested by Suresh Siddha and Yinghai Lu:
For x2apic pre-enabled systems, apic driver is set already early
through early_acpi_boot_init()/early_acpi_process_madt()/
acpi_parse_madt()/default_acpi_madt_oem_check() path so that
apic_id_valid() checking will be sufficient during MADT and SRAT
parsing.
For non-x2apic pre-enabled systems, all apic ids should be less
than 255.
This allows us to substitute the checks in
arch/x86/kernel/acpi/boot.c::acpi_parse_x2apic() and
arch/x86/mm/srat.c::acpi_numa_x2apic_affinity_init() with
apic->apic_id_valid().
In addition we can avoid feigning the x2apic cpu feature in the
NumaChip apic code.
The following apic drivers have separate apic_id_valid()
functions which will accept x2apic type IDs :
x2apic_phys
x2apic_cluster
x2apic_uv_x
apic_numachip
Signed-off-by: Steffen Persvold <sp@numascale.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Daniel J Blueman <daniel@numascale-asia.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Jack Steiner <steiner@sgi.com>
Link: http://lkml.kernel.org/r/1331925935-13372-1-git-send-email-sp@numascale.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Move APIC ID validity check into platform APIC code, so it can
be overridden when needed. For NumaChip systems, always trust
MADT, as it's constructed with high APIC IDs.
Behaviour verifies on standard x86 systems and on NumaChip
systems with this, and compile-tested with allyesconfig.
Signed-off-by: Daniel J Blueman <daniel@numascale-asia.com>
Reviewed-by: Steffen Persvold <sp@numascale.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: H. Peter Anvin <hpa@linux.intel.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Link: http://lkml.kernel.org/r/1331709454-27966-1-git-send-email-daniel@numascale-asia.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This will pave the way for each apic driver to be self-contained
and eliminate the need for apic_probe[].
Order in which apic drivers are listed in the .apicdrivers
section is important, as this determines the apic probe order.
And this is enforced by the ordering of apic driver files in the
Makefile and the macros apic_driver()/apic_drivers().
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Tested-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: steiner@sgi.com
Cc: gorcunov@openvz.org
Cc: yinghai@kernel.org
Link: http://lkml.kernel.org/r/20110521005526.068775085@sbsiddha-MOBL3.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
In the case of x2apic cluster mode we can group IPI register
writes based on the cluster group instead of individual per-cpu
destination messages.
This reduces the apic register writes and reduces the amount of
IPI messages (in the best case we can reduce it by a factor of
16).
With this change, the cost of flush_tlb_others(), with the flush
tlb IPI being sent from a cpu in the socket-1 to all the logical
cpus in socket-2 (on a Westmere-EX system that has 20 logical
cpus in a socket) is 3x times better now (compared to the former
'send one-by-one' algorithm).
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: steiner@sgi.com
Cc: yinghai@kernel.org
Link: http://lkml.kernel.org/r/20110519234637.512271057@sbsiddha-MOBL3.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
In the case of x2apic cluster mode, we can group IPI register
writes based on the cluster group instead of individual per-cpu
destination messages.
For this purpose, track the cpu's that belong to the same x2apic
cluster.
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: steiner@sgi.com
Cc: yinghai@kernel.org
Link: http://lkml.kernel.org/r/20110519234637.421800999@sbsiddha-MOBL3.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Use the unused probe routine in the apic driver to finalize the
apic model selection. This cleans up the
default_setup_apic_routing() and this probe routine in future
can also be used for doing any apic model specific
initialisation.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Acked-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: steiner@sgi.com
Cc: yinghai@kernel.org
Link: http://lkml.kernel.org/r/20110519234637.247458931@sbsiddha-MOBL3.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
apic->apicid_to_node() is 32bit specific apic operation which
determines NUMA node for a CPU. Depending on the APIC
implementation, it can be easier to determine NUMA node from
either physical or logical apicid. Currently,
->apicid_to_node() takes @logical_apicid and calls
hard_smp_processor_id() if the physical apicid is needed.
This prevents NUMA mapping from being queried from a different
CPU, which in turn makes it impossible to initialize NUMA
mapping before SMP bringup.
This patch replaces apic->apicid_to_node() with
->x86_32_numa_cpu_node() which takes @cpu, from which both
logical and physical apicids can easily be determined. While at
it, drop duplicate implementations from bigsmp_32 and summit_32,
and use the default one.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Pekka Enberg <penberg@kernel.org>
Cc: eric.dumazet@gmail.com
Cc: yinghai@kernel.org
Cc: brgerst@gmail.com
Cc: gorcunov@gmail.com
Cc: shaohui.zheng@intel.com
Cc: rientjes@google.com
LKML-Reference: <1295789862-25482-13-git-send-email-tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
After the previous patch, apic->cpu_to_logical_apicid() is no
longer used. Kill it.
For apic types with custom cpu_to_logical_apicid() which is also
used for other purposes, remove the function and modify its
users to do the mapping directly.
#ifdef's on CONFIG_SMP in es7000_32 and summit_32 are ignored
during conversion as they are not used for UP kernels.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: eric.dumazet@gmail.com
Cc: yinghai@kernel.org
Cc: brgerst@gmail.com
Cc: gorcunov@gmail.com
Cc: penberg@kernel.org
Cc: shaohui.zheng@intel.com
Cc: rientjes@google.com
LKML-Reference: <1295789862-25482-7-git-send-email-tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
John Blackwood reported:
> on an older Dell PowerEdge 6650 system with 8 cpus (4 are hyper-threaded),
> and 32 bit (x86) kernel, once you change the irq smp_affinity of an irq
> to be less than all cpus in the system, you can never change really the
> irq smp_affinity back to be all cpus in the system (0xff) again,
> even though no error status is returned on the "/bin/echo ff >
> /proc/irq/[n]/smp_affinity" operation.
>
> This is due to that fact that BAD_APICID has the same value as
> all cpus (0xff) on 32bit kernels, and thus the value returned from
> set_desc_affinity() via the cpu_mask_to_apicid_and() function is treated
> as a failure in set_ioapic_affinity_irq_desc(), and no affinity changes
> are made.
set_desc_affinity() is already checking if the incoming cpu mask
intersects with the cpu online mask or not. So there is no need
for the apic op cpu_mask_to_apicid_and() to check again
and return BAD_APICID.
Remove the BAD_APICID return value from cpu_mask_to_apicid_and()
and also fix set_desc_affinity() to return -1 instead of using BAD_APICID
to represent error conditions (as cpu_mask_to_apicid_and() can return
logical or physical apicid values and BAD_APICID is really to represent
bad physical apic id).
Reported-by: John Blackwood <john.blackwood@ccur.com>
Root-caused-by: John Blackwood <john.blackwood@ccur.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1261103386.2535.409.camel@sbs-t61>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: reduce kernel size a bit, address sparse warning
Addresses the problem pointed out by this sparse warning:
arch/x86/kernel/apic/x2apic_cluster.c:13:1: warning: symbol 'per_cpu__x86_cpu_to_logical_apicid' was not declared. Should it be static?
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1239434726.4418.24.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: optimize APIC IPI related barriers
Uncached MMIO accesses for xapic are inherently serializing and hence
we don't need explicit barriers for xapic IPI paths.
x2apic MSR writes/reads don't have serializing semantics and hence need
a serializing instruction or mfence, to make all the previous memory
stores globally visisble before the x2apic msr write for IPI.
Add x2apic_wrmsr_fence() in flush tlb path to x2apic specific paths.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: "steiner@sgi.com" <steiner@sgi.com>
Cc: Nick Piggin <npiggin@suse.de>
LKML-Reference: <1237313814.27006.203.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
- rename apic->wakeup_cpu to apic->wakeup_secondary_cpu, to
make it apparent that this is an SMP-only method
- handle NULL ->wakeup_secondary_cpus to mean the default INIT
wakeup sequence - this allows simplification of the APIC
driver templates.
Cc: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
x86_quirks->update_apic() calling looks crazy. so try to remove it:
1. every apic take wakeup_cpu member directly
2. separate es7000_apic to es7000_apic_cluster
3. use uv_wakeup_cpu directly
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
If BIOS hands over the control to OS in legacy xapic mode, select
legacy xapic related ops in the early apic probe and shift to x2apic
ops later in the boot sequence, only after enabling x2apic mode.
If BIOS hands over the control in x2apic mode, select x2apic related
ops in the early apic probe.
This fixes the early boot panic, where we were selecting x2apic ops,
while the cpu is still in legacy xapic mode.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
arch/x86/kernel/ is getting a bit crowded, and the APIC
drivers are scattered into various different files.
Move them to arch/x86/kernel/apic/*, and also remove
the 'gen' prefix from those which had it.
Also move APIC related functionality: the IO-APIC driver,
the NMI and the IPI code.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
