From: Dan Williams <dan.j.williams@intel.com>
To: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: "linux-nvdimm@lists.01.org" <linux-nvdimm@ml01.01.org>,
Linux ACPI <linux-acpi@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2 2/2] libnvdimm, region: sysfs trigger for nvdimm_flush()
Date: Tue, 25 Apr 2017 09:38:34 -0700 [thread overview]
Message-ID: <CAPcyv4iRPeA=BEx4X+B651amfKt4m4rAXiq3KhAaZ2=rD0bjQA@mail.gmail.com> (raw)
In-Reply-To: <20170425163707.GA11773@linux.intel.com>
On Tue, Apr 25, 2017 at 9:37 AM, Ross Zwisler
<ross.zwisler@linux.intel.com> wrote:
> On Mon, Apr 24, 2017 at 04:50:01PM -0700, Dan Williams wrote:
>> The nvdimm_flush() mechanism helps to reduce the impact of an ADR
>> (asynchronous-dimm-refresh) failure. The ADR mechanism handles flushing
>> platform WPQ (write-pending-queue) buffers when power is removed. The
>> nvdimm_flush() mechanism performs that same function on-demand.
>>
>> When a pmem namespace is associated with a block device, an
>> nvdimm_flush() is triggered with every block-layer REQ_FUA, or REQ_FLUSH
>> request. These requests are typically associated with filesystem
>> metadata updates. However, when a namespace is in device-dax mode,
>> userspace (think database metadata) needs another path to perform the
>> same flushing. In other words this is not required to make data
>> persistent, but in the case of metadata it allows for a smaller failure
>> domain in the unlikely event of an ADR failure.
>>
>> The new 'flush' attribute is visible when the individual DIMMs backing a
>> given interleave-set are described by platform firmware. In ACPI terms
>> this is "NVDIMM Region Mapping Structures" and associated "Flush Hint
>> Address Structures". Reads return "1" if the region supports triggering
>> WPQ flushes on all DIMMs. Reads return "0" the flush operation is a
>> platform nop, and in that case the attribute is read-only.
>>
>> Cc: Jeff Moyer <jmoyer@redhat.com>
>> Cc: Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>
>> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
>> ---
>> drivers/nvdimm/region_devs.c | 41 +++++++++++++++++++++++++++++++++++++++++
>> 1 file changed, 41 insertions(+)
>>
>> diff --git a/drivers/nvdimm/region_devs.c b/drivers/nvdimm/region_devs.c
>> index 24abceda986a..c48f3eddce2d 100644
>> --- a/drivers/nvdimm/region_devs.c
>> +++ b/drivers/nvdimm/region_devs.c
>> @@ -255,6 +255,35 @@ static ssize_t size_show(struct device *dev,
>> }
>> static DEVICE_ATTR_RO(size);
>>
>> +static ssize_t flush_show(struct device *dev,
>> + struct device_attribute *attr, char *buf)
>> +{
>> + struct nd_region *nd_region = to_nd_region(dev);
>> +
>> + /*
>> + * NOTE: in the nvdimm_has_flush() error case this attribute is
>> + * not visible.
>> + */
>> + return sprintf(buf, "%d\n", nvdimm_has_flush(nd_region));
>> +}
>> +
>> +static ssize_t flush_store(struct device *dev, struct device_attribute *attr,
>> + const char *buf, size_t len)
>> +{
>> + bool flush;
>> + int rc = strtobool(buf, &flush);
>> + struct nd_region *nd_region = to_nd_region(dev);
>> +
>> + if (rc)
>> + return rc;
>> + if (!flush)
>> + return -EINVAL;
>
> Is there a benefit to verifying whether the user actually pushed a "1" into
> our flush sysfs entry? Why have an -EINVAL error case at all?
>
> Flushing is non-destructive and we don't actually need the user to give us any
> data, so it seems simpler to just have this code flush, regardless of what
> input we received.
I want to be specific so that in the future if we decide that we want
to have "0" or some other value have a different meaning of "1" we
won't need to contend with userspace that may be expecting any random
value to work.
next prev parent reply other threads:[~2017-04-25 16:38 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-04-24 23:49 [PATCH v2 0/2] libnvdimm: export wpq flush interface Dan Williams
2017-04-24 23:49 ` [PATCH v2 1/2] libnvdimm, region: fix flush hint detection crash Dan Williams
2017-04-26 19:43 ` Jeff Moyer
2017-04-26 20:04 ` Dan Williams
2017-04-24 23:50 ` [PATCH v2 2/2] libnvdimm, region: sysfs trigger for nvdimm_flush() Dan Williams
2017-04-25 16:37 ` Ross Zwisler
2017-04-25 16:38 ` Dan Williams [this message]
2017-04-25 20:17 ` [PATCH v3] " Dan Williams
2017-04-26 20:38 ` Jeff Moyer
2017-04-26 23:00 ` Dan Williams
2017-04-27 13:45 ` Jeff Moyer
2017-04-27 16:56 ` Dan Williams
2017-04-27 18:41 ` Jeff Moyer
2017-04-27 19:17 ` Dan Williams
2017-04-27 19:21 ` Dan Williams
2017-04-27 19:43 ` Jeff Moyer
2017-04-27 19:40 ` Jeff Moyer
2017-04-27 20:02 ` Dan Williams
2017-04-27 21:36 ` Dan Williams
2017-04-26 23:36 ` [PATCH v4] " Dan Williams
2017-04-27 22:17 ` [PATCH v5] " Dan Williams
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAPcyv4iRPeA=BEx4X+B651amfKt4m4rAXiq3KhAaZ2=rD0bjQA@mail.gmail.com' \
--to=dan.j.williams@intel.com \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvdimm@ml01.01.org \
--cc=ross.zwisler@linux.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).