<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/drivers/nvme/host/nvme.h, branch v5.3-rc1</title>
<subtitle>Linux kernel for Apalis and Colibri modules</subtitle>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/'/>
<entry>
<title>nvme: set physical block size and optimal I/O size</title>
<updated>2019-07-09T21:15:37+00:00</updated>
<author>
<name>Bart Van Assche</name>
<email>bvanassche@acm.org</email>
</author>
<published>2019-06-28T16:53:31+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=81adb863349157c67ccec871e5ae5574600c50be'/>
<id>81adb863349157c67ccec871e5ae5574600c50be</id>
<content type='text'>
&gt;From the NVMe 1.4 spec:

NSFEAT bit 4 if set to 1: indicates that the fields NPWG, NPWA, NPDG, NPDA,
and NOWS are defined for this namespace and should be used by the host for
I/O optimization;
[ ... ]
Namespace Preferred Write Granularity (NPWG): This field indicates the
smallest recommended write granularity in logical blocks for this namespace.
This is a 0's based value. The size indicated should be less than or equal
to Maximum Data Transfer Size (MDTS) that is specified in units of minimum
memory page size. The value of this field may change if the namespace is
reformatted. The size should be a multiple of Namespace Preferred Write
Alignment (NPWA). Refer to section 8.25 for how this field is utilized to
improve performance and endurance.
[ ... ]
Each Write, Write Uncorrectable, or Write Zeroes commands should address a
multiple of Namespace Preferred Write Granularity (NPWG) (refer to Figure
245) and Stream Write Size (SWS) (refer to Figure 515) logical blocks (as
expressed in the NLB field), and the SLBA field of the command should be
aligned to Namespace Preferred Write Alignment (NPWA) (refer to Figure 245)
for best performance.

Signed-off-by: Bart Van Assche &lt;bvanassche@acm.org&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.com&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
&gt;From the NVMe 1.4 spec:

NSFEAT bit 4 if set to 1: indicates that the fields NPWG, NPWA, NPDG, NPDA,
and NOWS are defined for this namespace and should be used by the host for
I/O optimization;
[ ... ]
Namespace Preferred Write Granularity (NPWG): This field indicates the
smallest recommended write granularity in logical blocks for this namespace.
This is a 0's based value. The size indicated should be less than or equal
to Maximum Data Transfer Size (MDTS) that is specified in units of minimum
memory page size. The value of this field may change if the namespace is
reformatted. The size should be a multiple of Namespace Preferred Write
Alignment (NPWA). Refer to section 8.25 for how this field is utilized to
improve performance and endurance.
[ ... ]
Each Write, Write Uncorrectable, or Write Zeroes commands should address a
multiple of Namespace Preferred Write Granularity (NPWG) (refer to Figure
245) and Stream Write Size (SWS) (refer to Figure 515) logical blocks (as
expressed in the NLB field), and the SLBA field of the command should be
aligned to Namespace Preferred Write Alignment (NPWA) (refer to Figure 245)
for best performance.

Signed-off-by: Bart Van Assche &lt;bvanassche@acm.org&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.com&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme: enable to inject errors into admin commands</title>
<updated>2019-06-21T09:15:50+00:00</updated>
<author>
<name>Akinobu Mita</name>
<email>akinobu.mita@gmail.com</email>
</author>
<published>2019-06-09T14:17:01+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=f79d5fda4ea08c33a114087573d86f703149ee0e'/>
<id>f79d5fda4ea08c33a114087573d86f703149ee0e</id>
<content type='text'>
This enables to inject errors into the commands submitted to the admin
queue.

It is useful to test error handling in the controller initialization.

	# echo 100 &gt; /sys/kernel/debug/nvme0/fault_inject/probability
	# echo 1 &gt; /sys/kernel/debug/nvme0/fault_inject/times
	# echo 10 &gt; /sys/kernel/debug/nvme0/fault_inject/space
	# nvme reset /dev/nvme0
	# dmesg
	...
	nvme nvme0: Could not set queue count (16385)
	nvme nvme0: IO queues not created

Signed-off-by: Akinobu Mita &lt;akinobu.mita@gmail.com&gt;
Reviewed-by: Minwoo Im &lt;minwoo.im.dev@gmail.com&gt;
Reviewed-by: Sagi Grimberg &lt;sagi@grimberg.me&gt;
Reviewed-by: Chaitanya Kulkarni &lt;chaitanya.kulkarni@wdc.com&gt;
Reviewed-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This enables to inject errors into the commands submitted to the admin
queue.

It is useful to test error handling in the controller initialization.

	# echo 100 &gt; /sys/kernel/debug/nvme0/fault_inject/probability
	# echo 1 &gt; /sys/kernel/debug/nvme0/fault_inject/times
	# echo 10 &gt; /sys/kernel/debug/nvme0/fault_inject/space
	# nvme reset /dev/nvme0
	# dmesg
	...
	nvme nvme0: Could not set queue count (16385)
	nvme nvme0: IO queues not created

Signed-off-by: Akinobu Mita &lt;akinobu.mita@gmail.com&gt;
Reviewed-by: Minwoo Im &lt;minwoo.im.dev@gmail.com&gt;
Reviewed-by: Sagi Grimberg &lt;sagi@grimberg.me&gt;
Reviewed-by: Chaitanya Kulkarni &lt;chaitanya.kulkarni@wdc.com&gt;
Reviewed-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme: prepare for fault injection into admin commands</title>
<updated>2019-06-21T09:15:50+00:00</updated>
<author>
<name>Akinobu Mita</name>
<email>akinobu.mita@gmail.com</email>
</author>
<published>2019-06-20T06:49:02+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=a3646451edd52ba238cbe4f618aaf6eb9bf9d60c'/>
<id>a3646451edd52ba238cbe4f618aaf6eb9bf9d60c</id>
<content type='text'>
Currenlty fault injection support for nvme only enables to inject errors
into the commands submitted to I/O queues.

In preparation for fault injection into the admin commands, this makes
the helper functions independent of struct nvme_ns.

Signed-off-by: Akinobu Mita &lt;akinobu.mita@gmail.com&gt;
Reviewed-by: Minwoo Im &lt;minwoo.im.dev@gmail.com&gt;
Reviewed-by: Sagi Grimberg &lt;sagi@grimberg.me&gt;
Reviewed-by: Chaitanya Kulkarni &lt;chaitanya.kulkarni@wdc.com&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Currenlty fault injection support for nvme only enables to inject errors
into the commands submitted to I/O queues.

In preparation for fault injection into the admin commands, this makes
the helper functions independent of struct nvme_ns.

Signed-off-by: Akinobu Mita &lt;akinobu.mita@gmail.com&gt;
Reviewed-by: Minwoo Im &lt;minwoo.im.dev@gmail.com&gt;
Reviewed-by: Sagi Grimberg &lt;sagi@grimberg.me&gt;
Reviewed-by: Chaitanya Kulkarni &lt;chaitanya.kulkarni@wdc.com&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme: export get and set features</title>
<updated>2019-06-21T09:08:38+00:00</updated>
<author>
<name>Keith Busch</name>
<email>keith.busch@intel.com</email>
</author>
<published>2019-05-26T16:29:01+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=1a87ee657c530bb2f3e39e4ac184d48f5f959cda'/>
<id>1a87ee657c530bb2f3e39e4ac184d48f5f959cda</id>
<content type='text'>
Future use intends to make use of both, so export these functions. And
since their implementation is identical except for the opcode, provide a
new function that implement both.

[akinobu.mita@gmail.com&gt;: fix line over 80 characters]
Signed-off-by: Keith Busch &lt;keith.busch@intel.com&gt;
Signed-off-by: Akinobu Mita &lt;akinobu.mita@gmail.com&gt;
Reviewed-by: Chaitanya Kulkarni &lt;chaitanya.kulkarni@wdc.com&gt;
Reviewed-by: Minwoo Im &lt;minwoo.im.dev@gmail.com&gt;
Signed-off-by: Sagi Grimberg &lt;sagi@grimberg.me&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Future use intends to make use of both, so export these functions. And
since their implementation is identical except for the opcode, provide a
new function that implement both.

[akinobu.mita@gmail.com&gt;: fix line over 80 characters]
Signed-off-by: Keith Busch &lt;keith.busch@intel.com&gt;
Signed-off-by: Akinobu Mita &lt;akinobu.mita@gmail.com&gt;
Reviewed-by: Chaitanya Kulkarni &lt;chaitanya.kulkarni@wdc.com&gt;
Reviewed-by: Minwoo Im &lt;minwoo.im.dev@gmail.com&gt;
Signed-off-by: Sagi Grimberg &lt;sagi@grimberg.me&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme-pci: Sync queues on reset</title>
<updated>2019-05-17T17:04:34+00:00</updated>
<author>
<name>Keith Busch</name>
<email>keith.busch@intel.com</email>
</author>
<published>2019-05-14T20:46:09+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=d6135c3a1ec0cddda7b8b8e1b5b4abeeafd98289'/>
<id>d6135c3a1ec0cddda7b8b8e1b5b4abeeafd98289</id>
<content type='text'>
A controller with multiple namespaces may have multiple request_queues with
their own timeout work. If a controller fails with IO outstanding to
diffent namespaces, each request queue may attempt to handle it, so
ensure there is no previously scheduled timeout work executing prior to
starting controller initialization by synchronizing with each queue.

Reviewed-by: Minwoo Im &lt;minwoo.im.dev@gmail.com&gt;
Reviewed-by: Ming Lei &lt;ming.lei@redhat.com&gt;
Reviewed-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Keith Busch &lt;keith.busch@intel.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
A controller with multiple namespaces may have multiple request_queues with
their own timeout work. If a controller fails with IO outstanding to
diffent namespaces, each request queue may attempt to handle it, so
ensure there is no previously scheduled timeout work executing prior to
starting controller initialization by synchronizing with each queue.

Reviewed-by: Minwoo Im &lt;minwoo.im.dev@gmail.com&gt;
Reviewed-by: Ming Lei &lt;ming.lei@redhat.com&gt;
Reviewed-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Keith Busch &lt;keith.busch@intel.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme: mark nvme_core_init and nvme_core_exit static</title>
<updated>2019-05-01T13:18:47+00:00</updated>
<author>
<name>Christoph Hellwig</name>
<email>hch@lst.de</email>
</author>
<published>2019-04-30T15:37:43+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=893a74b7a76e6e9c5c7199e6aae946f090622fa2'/>
<id>893a74b7a76e6e9c5c7199e6aae946f090622fa2</id>
<content type='text'>
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Bart Van Assche &lt;bvanassche@acm.org&gt;
Reviewed-by: Keith Busch &lt;keith.busch@intel.com&gt;
Reviewed-by: Chaitanya Kulkarni &lt;chaitanya.kulkarni@wdc.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Bart Van Assche &lt;bvanassche@acm.org&gt;
Reviewed-by: Keith Busch &lt;keith.busch@intel.com&gt;
Reviewed-by: Chaitanya Kulkarni &lt;chaitanya.kulkarni@wdc.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme: disable Write Zeroes for qemu controllers</title>
<updated>2019-03-13T18:57:34+00:00</updated>
<author>
<name>Christoph Hellwig</name>
<email>hch@lst.de</email>
</author>
<published>2019-03-13T17:55:05+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=7b210e4ed5e281728243799c5e2b84d3f70d4dd1'/>
<id>7b210e4ed5e281728243799c5e2b84d3f70d4dd1</id>
<content type='text'>
Qemu started out with a broken implementation of Write Zeroes written
by yours truly.  Disable Write Zeroes on qemu for now, eventually
we need to go back and make all the qemu quirks version specific,
but that is left for another time.

Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Keith Busch &lt;keith.busch@intel.com&gt;
Tested-by: Ming Lei &lt;ming.lei@redhat.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@kernel.dk&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Qemu started out with a broken implementation of Write Zeroes written
by yours truly.  Disable Write Zeroes on qemu for now, eventually
we need to go back and make all the qemu quirks version specific,
but that is left for another time.

Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Keith Busch &lt;keith.busch@intel.com&gt;
Tested-by: Ming Lei &lt;ming.lei@redhat.com&gt;
Signed-off-by: Jens Axboe &lt;axboe@kernel.dk&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme: convert to SPDX identifiers</title>
<updated>2019-02-20T14:22:28+00:00</updated>
<author>
<name>Christoph Hellwig</name>
<email>hch@lst.de</email>
</author>
<published>2019-02-18T08:36:29+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=bc50ad7501dd3629af9aa423ed0d1eae0061bcf1'/>
<id>bc50ad7501dd3629af9aa423ed0d1eae0061bcf1</id>
<content type='text'>
Update license to use SPDX-License-Identifier instead of verbose license
text.

Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Sagi Grimberg &lt;sagi@grimberg.me&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Update license to use SPDX-License-Identifier instead of verbose license
text.

Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Sagi Grimberg &lt;sagi@grimberg.me&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme: unexport nvme_delete_ctrl_sync()</title>
<updated>2019-02-20T14:18:15+00:00</updated>
<author>
<name>Bart Van Assche</name>
<email>bvanassche@acm.org</email>
</author>
<published>2019-02-14T22:50:55+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=d84c4b024ac36c778fd6020fb8560447365a5478'/>
<id>d84c4b024ac36c778fd6020fb8560447365a5478</id>
<content type='text'>
Since nvme_delete_ctrl_sync() is not called from any other kernel module,
unexport it.

Signed-off-by: Bart Van Assche &lt;bvanassche@acm.org&gt;
Reviewed-by: Sagi Grimberg &lt;sagi@grimberg.me&gt;
Reviewed-by: Chaitanya Kulkarni &lt;chaitanya.kulkarni@wdc.com&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Since nvme_delete_ctrl_sync() is not called from any other kernel module,
unexport it.

Signed-off-by: Bart Van Assche &lt;bvanassche@acm.org&gt;
Reviewed-by: Sagi Grimberg &lt;sagi@grimberg.me&gt;
Reviewed-by: Chaitanya Kulkarni &lt;chaitanya.kulkarni@wdc.com&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme-multipath: round-robin I/O policy</title>
<updated>2019-02-20T14:17:49+00:00</updated>
<author>
<name>Hannes Reinecke</name>
<email>hare@suse.de</email>
</author>
<published>2019-02-18T10:43:26+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=75c10e73272484bc3a940a9c8e4ec39a7a1b8c21'/>
<id>75c10e73272484bc3a940a9c8e4ec39a7a1b8c21</id>
<content type='text'>
Implement a simple round-robin I/O policy for multipathing.  Path
selection is done in two rounds, first iterating across all optimized
paths, and if that doesn't return any valid paths, iterate over all
optimized and non-optimized paths.  If no paths are found, use the
existing algorithm.  Also add a sysfs attribute 'iopolicy' to switch
between the current NUMA-aware I/O policy and the 'round-robin' I/O
policy.

Signed-off-by: Hannes Reinecke &lt;hare@suse.com&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Implement a simple round-robin I/O policy for multipathing.  Path
selection is done in two rounds, first iterating across all optimized
paths, and if that doesn't return any valid paths, iterate over all
optimized and non-optimized paths.  If no paths are found, use the
existing algorithm.  Also add a sysfs attribute 'iopolicy' to switch
between the current NUMA-aware I/O policy and the 'round-robin' I/O
policy.

Signed-off-by: Hannes Reinecke &lt;hare@suse.com&gt;
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
</pre>
</div>
</content>
</entry>
</feed>
