[ Upstream commit b82592199032bf7c778f861b936287e37ebc9f62 ]
If the number of NUMA nodes exceeds the number of MSI/MSI-X interrupts
which are allocated for a device, the interrupt affinity spreading code
fails to spread them across all nodes.
The reason is, that the spreading code starts from node 0 and continues up
to the number of interrupts requested for allocation. This leaves the nodes
past the last interrupt unused.
This results in interrupt concentration on the first nodes which violates
the assumption of the block layer that all nodes are covered evenly. As a
consequence the NUMA nodes above the number of interrupts are all assigned
to hardware queue 0 and therefore NUMA node 0, which results in bad
performance and has CPU hotplug implications, because queue 0 gets shut
down when the last CPU of node 0 is offlined.
Go over all NUMA nodes and assign them round-robin to all requested
interrupts to solve this.
[ tglx: Massaged changelog ]
Signed-off-by: Long Li <longli@microsoft.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Cc: Michael Kelley <mikelley@microsoft.com>
Link: https://lkml.kernel.org/r/20181102180248.13583-1-longli@linuxonhyperv.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 84676c1f21 ("genirq/affinity: assign vectors to all possible CPUs")
tried to spread the interrupts accross all possible CPUs to make sure that
in case of phsyical hotplug (e.g. virtualization) the CPUs which get
plugged in after the device was initialized are targeted by a hardware
queue and the corresponding interrupt.
This has a downside in cases where the ACPI tables claim that there are
more possible CPUs than present CPUs and the number of interrupts to spread
out is smaller than the number of possible CPUs. These bogus ACPI tables
are unfortunately not uncommon.
In such a case the vector spreading algorithm assigns interrupts to CPUs
which can never be utilized and as a consequence these interrupts are
unused instead of being mapped to present CPUs. As a result the performance
of the device is suboptimal.
To fix this spread the interrupt vectors in two stages:
1) Spread as many interrupts as possible among the present CPUs
2) Spread the remaining vectors among non present CPUs
On a 8 core system, where CPU 0-3 are present and CPU 4-7 are not present,
for a device with 4 queues the resulting interrupt affinity is:
1) Before 84676c1f21 ("genirq/affinity: assign vectors to all possible CPUs")
irq 39, cpu list 0
irq 40, cpu list 1
irq 41, cpu list 2
irq 42, cpu list 3
2) With 84676c1f21 ("genirq/affinity: assign vectors to all possible CPUs")
irq 39, cpu list 0-2
irq 40, cpu list 3-4,6
irq 41, cpu list 5
irq 42, cpu list 7
3) With the refined vector spread applied:
irq 39, cpu list 0,4
irq 40, cpu list 1,6
irq 41, cpu list 2,5
irq 42, cpu list 3,7
On a 8 core system, where all CPUs are present the resulting interrupt
affinity for the 4 queues is:
irq 39, cpu list 0,1
irq 40, cpu list 2,3
irq 41, cpu list 4,5
irq 42, cpu list 6,7
This is independent of the number of CPUs which are online at the point of
initialization because in such a system the offline CPUs can be easily
onlined afterwards, while in non-present CPUs need to be plugged physically
or virtually which requires external interaction.
The downside of this approach is that in case of physical hotplug the
interrupt vector spreading might be suboptimal when CPUs 4-7 are physically
plugged. Suboptimal from a NUMA point of view and due to the single target
nature of interrupt affinities the later plugged CPUs might not be targeted
by interrupts at all.
Though, physical hotplug systems are not the common case while the broken
ACPI table disease is wide spread. So it's preferred to have as many
interrupts as possible utilized at the point where the device is
initialized.
Block multi-queue devices like NVME create a hardware queue per possible
CPU, so the goal of commit 84676c1f21 to assign one interrupt vector per
possible CPU is still achieved even with physical/virtual hotplug.
[ tglx: Changed from online to present CPUs for the first spreading stage,
renamed variables for readability sake, added comments and massaged
changelog ]
Reported-by: Laurence Oberman <loberman@redhat.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Cc: Christoph Hellwig <hch@infradead.org>
Link: https://lkml.kernel.org/r/20180308105358.1506-5-ming.lei@redhat.com
To support two stage irq vector spreading, it's required to add a starting
point to the spreading function. No functional change, just preparatory
work for the actual two stage change.
[ tglx: Renamed variables, tidied up the code and massaged changelog ]
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Cc: Laurence Oberman <loberman@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
Link: https://lkml.kernel.org/r/20180308105358.1506-4-ming.lei@redhat.com
No functional change, just prepare for converting to 2-stage irq vector
spreading.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Cc: Laurence Oberman <loberman@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
Link: https://lkml.kernel.org/r/20180308105358.1506-3-ming.lei@redhat.com
The following patches will introduce two stage irq spreading for improving
irq spread on all possible CPUs.
No functional change.
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Cc: Laurence Oberman <loberman@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
Link: https://lkml.kernel.org/r/20180308105358.1506-2-ming.lei@redhat.com
When the allocation of node_to_possible_cpumask fails, then
irq_create_affinity_masks() returns with a pointer to the empty affinity
masks array, which will cause malfunction.
Reorder the allocations so the masks array allocation comes last and every
failure path returns NULL.
Fixes: 9a0ef98e18 ("genirq/affinity: Assign vectors to all present CPUs")
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Ming Lei <ming.lei@redhat.com>
Currently we assign managed interrupt vectors to all present CPUs. This
works fine for systems were we only online/offline CPUs. But in case of
systems that support physical CPU hotplug (or the virtualized version of
it) this means the additional CPUs covered for in the ACPI tables or on
the command line are not catered for. To fix this we'd either need to
introduce new hotplug CPU states just for this case, or we can start
assining vectors to possible but not present CPUs.
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Tested-by: Christian Borntraeger <borntraeger@de.ibm.com>
Tested-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Fixes: 4b855ad371 ("blk-mq: Create hctx for each present CPU")
Cc: linux-kernel@vger.kernel.org
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Many source files in the tree are missing licensing information, which
makes it harder for compliance tools to determine the correct license.
By default all files without license information are under the default
license of the kernel, which is GPL version 2.
Update the files which contain no license information with the 'GPL-2.0'
SPDX license identifier. The SPDX identifier is a legally binding
shorthand, which can be used instead of the full boiler plate text.
This patch is based on work done by Thomas Gleixner and Kate Stewart and
Philippe Ombredanne.
How this work was done:
Patches were generated and checked against linux-4.14-rc6 for a subset of
the use cases:
- file had no licensing information it it.
- file was a */uapi/* one with no licensing information in it,
- file was a */uapi/* one with existing licensing information,
Further patches will be generated in subsequent months to fix up cases
where non-standard license headers were used, and references to license
had to be inferred by heuristics based on keywords.
The analysis to determine which SPDX License Identifier to be applied to
a file was done in a spreadsheet of side by side results from of the
output of two independent scanners (ScanCode & Windriver) producing SPDX
tag:value files created by Philippe Ombredanne. Philippe prepared the
base worksheet, and did an initial spot review of a few 1000 files.
The 4.13 kernel was the starting point of the analysis with 60,537 files
assessed. Kate Stewart did a file by file comparison of the scanner
results in the spreadsheet to determine which SPDX license identifier(s)
to be applied to the file. She confirmed any determination that was not
immediately clear with lawyers working with the Linux Foundation.
Criteria used to select files for SPDX license identifier tagging was:
- Files considered eligible had to be source code files.
- Make and config files were included as candidates if they contained >5
lines of source
- File already had some variant of a license header in it (even if <5
lines).
All documentation files were explicitly excluded.
The following heuristics were used to determine which SPDX license
identifiers to apply.
- when both scanners couldn't find any license traces, file was
considered to have no license information in it, and the top level
COPYING file license applied.
For non */uapi/* files that summary was:
SPDX license identifier # files
---------------------------------------------------|-------
GPL-2.0 11139
and resulted in the first patch in this series.
If that file was a */uapi/* path one, it was "GPL-2.0 WITH
Linux-syscall-note" otherwise it was "GPL-2.0". Results of that was:
SPDX license identifier # files
---------------------------------------------------|-------
GPL-2.0 WITH Linux-syscall-note 930
and resulted in the second patch in this series.
- if a file had some form of licensing information in it, and was one
of the */uapi/* ones, it was denoted with the Linux-syscall-note if
any GPL family license was found in the file or had no licensing in
it (per prior point). Results summary:
SPDX license identifier # files
---------------------------------------------------|------
GPL-2.0 WITH Linux-syscall-note 270
GPL-2.0+ WITH Linux-syscall-note 169
((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) 21
((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) 17
LGPL-2.1+ WITH Linux-syscall-note 15
GPL-1.0+ WITH Linux-syscall-note 14
((GPL-2.0+ WITH Linux-syscall-note) OR BSD-3-Clause) 5
LGPL-2.0+ WITH Linux-syscall-note 4
LGPL-2.1 WITH Linux-syscall-note 3
((GPL-2.0 WITH Linux-syscall-note) OR MIT) 3
((GPL-2.0 WITH Linux-syscall-note) AND MIT) 1
and that resulted in the third patch in this series.
- when the two scanners agreed on the detected license(s), that became
the concluded license(s).
- when there was disagreement between the two scanners (one detected a
license but the other didn't, or they both detected different
licenses) a manual inspection of the file occurred.
- In most cases a manual inspection of the information in the file
resulted in a clear resolution of the license that should apply (and
which scanner probably needed to revisit its heuristics).
- When it was not immediately clear, the license identifier was
confirmed with lawyers working with the Linux Foundation.
- If there was any question as to the appropriate license identifier,
the file was flagged for further research and to be revisited later
in time.
In total, over 70 hours of logged manual review was done on the
spreadsheet to determine the SPDX license identifiers to apply to the
source files by Kate, Philippe, Thomas and, in some cases, confirmation
by lawyers working with the Linux Foundation.
Kate also obtained a third independent scan of the 4.13 code base from
FOSSology, and compared selected files where the other two scanners
disagreed against that SPDX file, to see if there was new insights. The
Windriver scanner is based on an older version of FOSSology in part, so
they are related.
Thomas did random spot checks in about 500 files from the spreadsheets
for the uapi headers and agreed with SPDX license identifier in the
files he inspected. For the non-uapi files Thomas did random spot checks
in about 15000 files.
In initial set of patches against 4.14-rc6, 3 files were found to have
copy/paste license identifier errors, and have been fixed to reflect the
correct identifier.
Additionally Philippe spent 10 hours this week doing a detailed manual
inspection and review of the 12,461 patched files from the initial patch
version early this week with:
- a full scancode scan run, collecting the matched texts, detected
license ids and scores
- reviewing anything where there was a license detected (about 500+
files) to ensure that the applied SPDX license was correct
- reviewing anything where there was no detection but the patch license
was not GPL-2.0 WITH Linux-syscall-note to ensure that the applied
SPDX license was correct
This produced a worksheet with 20 files needing minor correction. This
worksheet was then exported into 3 different .csv files for the
different types of files to be modified.
These .csv files were then reviewed by Greg. Thomas wrote a script to
parse the csv files and add the proper SPDX tag to the file, in the
format that the file expected. This script was further refined by Greg
based on the output to detect more types of files automatically and to
distinguish between header and source .c files (which need different
comment types.) Finally Greg ran the script using the .csv files to
generate the patches.
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Philippe Ombredanne <pombredanne@nexb.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently the irq vector spread algorithm is restricted to online CPUs,
which ties the IRQ mapping to the currently online devices and doesn't deal
nicely with the fact that CPUs could come and go rapidly due to e.g. power
management.
Instead assign vectors to all present CPUs to avoid this churn.
Build a map of all possible CPUs for a given node, as the architectures
only provide a map of all onlines CPUs. Do this dynamically on each call
for the vector assingments, which is a bit suboptimal and could be
optimized in the future by provinding a mapping from the arch code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: linux-nvme@lists.infradead.org
Cc: Keith Busch <keith.busch@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20170603140403.27379-5-hch@lst.de
min_vecs is the minimum amount of vectors needed to operate in MSI-X mode
which may just include the vectors that don't need affinity.
Disabling affinity settings causes the qla2xxx driver scsi_add_host() to fail
when blk_mq is enabled as the blk_mq_pci_map_queues() expects affinity masks
on each vector.
Fixes: dfef358bd1 ("PCI/MSI: Don't apply affinity if there aren't enough vectors left")
Signed-off-by: Michael Hernandez <michael.hernandez@cavium.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@cavium.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: stable@vger.kernel.org # v4.10+
The vectors_per_node is calculated from the remaining available vectors.
The current vector starts after pre_vectors, so we need to subtract that
from the current to properly account for the number of remaining vectors
to assign.
Fixes: 3412386b53 ("irq/affinity: Fix extra vecs calculation")
Reported-by: Andrei Vagin <avagin@virtuozzo.com>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Link: http://lkml.kernel.org/r/1492645870-13019-1-git-send-email-keith.busch@intel.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This fixes a math error calculating the extra_vecs. The error assumed
only 1 cpu per vector, but the value needs to account for the actual
number of cpus per vector in order to get the correct remainder for
extra CPU assignment.
Fixes: 7bf8222b9b ("irq/affinity: Fix CPU spread for unbalanced nodes")
Reported-by: Xiaolong Ye <xiaolong.ye@intel.com>
Signed-off-by: Keith Busch <keith.busch@intel.com>
Link: http://lkml.kernel.org/r/1492104492-19943-1-git-send-email-keith.busch@intel.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The irq_create_affinity_masks routine is responsible for assigning a
number of interrupt vectors to CPUs. The optimal assignemnet will spread
requested vectors to all CPUs, with the fewest CPUs sharing a vector.
The algorithm may fail to assign some vectors to any CPUs if a node's
CPU count is lower than the average number of vectors per node. These
vectors are unusable and create an un-optimal spread.
Recalculate the number of vectors to assign at each node iteration by using
the remaining number of vectors and nodes to be assigned, not exceeding the
number of CPUs in that node. This will guarantee that every CPU is assigned
at least one vector.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: linux-nvme@lists.infradead.org
Link: http://lkml.kernel.org/r/1491247553-7603-1-git-send-email-keith.busch@intel.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Commit 34c3d9819f ("genirq/affinity: Provide smarter irq spreading
infrastructure") introduced a better IRQ spreading mechanism, taking
account of the available NUMA nodes in the machine.
Problem is that the algorithm of retrieving the nodemask iterates
"linearly" based on the number of online nodes - some architectures
present non-linear node distribution among the nodemask, like PowerPC.
If this is the case, the algorithm lead to a wrong node count number
and therefore to a bad/incomplete IRQ affinity distribution.
For example, this problem were found in a machine with 128 CPUs and two
nodes, namely nodes 0 and 8 (instead of 0 and 1, if it was linearly
distributed). This led to a wrong affinity distribution which then led to
a bad mq allocation for nvme driver.
Finally, we take the opportunity to fix a comment regarding the affinity
distribution when we have _more_ nodes than vectors.
Fixes: 34c3d9819f ("genirq/affinity: Provide smarter irq spreading infrastructure")
Reported-by: Gabriel Krisman Bertazi <gabriel@krisman.be>
Signed-off-by: Guilherme G. Piccoli <gpiccoli@linux.vnet.ibm.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Gabriel Krisman Bertazi <gabriel@krisman.be>
Reviewed-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Cc: linux-pci@vger.kernel.org
Cc: linuxppc-dev@lists.ozlabs.org
Cc: hch@lst.de
Link: http://lkml.kernel.org/r/1481738472-2671-1-git-send-email-gpiccoli@linux.vnet.ibm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The reserved vectors at the beginning and the end of the vector space get
cpu_possible_mask assigned as their affinity mask.
All other non-auto affine interrupts get the default irq affinity mask
assigned. Using cpu_possible_mask breaks that rule.
Treat them like any other interrupt and use irq_default_affinity as target
mask.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Christoph Hellwig <hch@lst.de>
The recent addition of reserved vectors at the beginning or the end of the
vector space did not take the reserved vectors at the beginning into
account for the various loop exit conditions. As a consequence the last
vectors of the spread area are not included into the spread algorithm and
are treated like the reserved vectors at the end of the vector space and
get the default affinity mask assigned.
Sum up the affinity vectors and the reserved vectors at the beginning and
use the sum as exit condition.
[ tglx: Fixed all conditions instead of only one and massaged changelog ]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: http://lkml.kernel.org/r/1479201178-29604-2-git-send-email-hch@lst.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Only calculate the affinity for the main I/O vectors, and skip the
pre or post vectors specified by struct irq_affinity.
Also remove the irq_affinity cpumask argument that has never been used.
If we ever need it in the future we can pass it through struct
irq_affinity.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Cc: linux-pci@vger.kernel.org
Link: http://lkml.kernel.org/r/1478654107-7384-4-git-send-email-hch@lst.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Only calculate the affinity for the main I/O vectors, and skip the pre or
post vectors specified by struct irq_affinity.
Also remove the irq_affinity cpumask argument that has never been used. If
we ever need it in the future we can pass it through struct irq_affinity.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Acked-by: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Cc: linux-pci@vger.kernel.org
Link: http://lkml.kernel.org/r/1478654107-7384-3-git-send-email-hch@lst.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The current irq spreading infrastructure is just looking at a cpumask and
tries to spread the interrupts over the mask. Thats suboptimal as it does
not take numa nodes into account.
Change the logic so the interrupts are spread across numa nodes and inside
the nodes. If there are more cpus than vectors per node, then we set the
affinity to several cpus. If HT siblings are available we take that into
account and try to set all siblings to a single vector.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: axboe@fb.com
Cc: keith.busch@intel.com
Cc: agordeev@redhat.com
Cc: linux-block@vger.kernel.org
Link: http://lkml.kernel.org/r/1473862739-15032-3-git-send-email-hch@lst.de
Without locking out CPU mask operations we might end up with an inconsistent
view of the cpumask in the function.
Fixes: 5e385a6ef3: "genirq: Add a helper to spread an affinity mask for MSI/MSI-X vectors"
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: http://lkml.kernel.org/r/1470924405-25728-1-git-send-email-hch@lst.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
