From: Dan Williams <dan.j.williams@intel.com>
To: Pankaj Gupta <pagupta@redhat.com>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
KVM list <kvm@vger.kernel.org>,
Qemu Developers <qemu-devel@nongnu.org>,
linux-nvdimm <linux-nvdimm@ml01.01.org>, Jan Kara <jack@suse.cz>,
Stefan Hajnoczi <stefanha@redhat.com>,
Rik van Riel <riel@surriel.com>,
Nitesh Narayan Lal <nilal@redhat.com>,
Kevin Wolf <kwolf@redhat.com>,
Paolo Bonzini <pbonzini@redhat.com>,
"Zwisler, Ross" <ross.zwisler@intel.com>,
David Hildenbrand <david@redhat.com>,
Xiao Guangrong <xiaoguangrong.eric@gmail.com>,
Christoph Hellwig <hch@infradead.org>,
"Michael S. Tsirkin" <mst@redhat.com>,
niteshnarayanlal@hotmail.com, lcapitulino@redhat.com,
Igor Mammedov <imammedo@redhat.com>,
Eric Blake <eblake@redhat.com>
Subject: Re: [PATCH 2/3] libnvdimm: nd_region flush callback support
Date: Fri, 21 Sep 2018 17:43:58 -0700 [thread overview]
Message-ID: <CAPcyv4js9hWfXPK++a0DhDwtk1pBazMvb9H-X2OQy18uemUhGw@mail.gmail.com> (raw)
In-Reply-To: <20180831133019.27579-3-pagupta@redhat.com>
On Fri, Aug 31, 2018 at 6:32 AM Pankaj Gupta <pagupta@redhat.com> wrote:
>
> This patch adds functionality to perform flush from guest
> to host over VIRTIO. We are registering a callback based
> on 'nd_region' type. virtio_pmem driver requires this special
> flush function. For rest of the region types we are registering
> existing flush function. Report error returned by host fsync
> failure to userspace.
>
> Signed-off-by: Pankaj Gupta <pagupta@redhat.com>
This looks ok to me, just some nits below.
> ---
> drivers/acpi/nfit/core.c | 7 +++++--
> drivers/nvdimm/claim.c | 3 ++-
> drivers/nvdimm/pmem.c | 12 ++++++++----
> drivers/nvdimm/region_devs.c | 12 ++++++++++--
> include/linux/libnvdimm.h | 4 +++-
> 5 files changed, 28 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c
> index b072cfc..cd63b69 100644
> --- a/drivers/acpi/nfit/core.c
> +++ b/drivers/acpi/nfit/core.c
> @@ -2216,6 +2216,7 @@ static void write_blk_ctl(struct nfit_blk *nfit_blk, unsigned int bw,
> {
> u64 cmd, offset;
> struct nfit_blk_mmio *mmio = &nfit_blk->mmio[DCR];
> + struct nd_region *nd_region = nfit_blk->nd_region;
>
> enum {
> BCW_OFFSET_MASK = (1ULL << 48)-1,
> @@ -2234,7 +2235,7 @@ static void write_blk_ctl(struct nfit_blk *nfit_blk, unsigned int bw,
> offset = to_interleave_offset(offset, mmio);
>
> writeq(cmd, mmio->addr.base + offset);
> - nvdimm_flush(nfit_blk->nd_region);
> + nd_region->flush(nd_region);
I would keep the indirect function call override inside of
nvdimm_flush. Then this hunk can go away...
>
> if (nfit_blk->dimm_flags & NFIT_BLK_DCR_LATCH)
> readq(mmio->addr.base + offset);
> @@ -2245,6 +2246,7 @@ static int acpi_nfit_blk_single_io(struct nfit_blk *nfit_blk,
> unsigned int lane)
> {
> struct nfit_blk_mmio *mmio = &nfit_blk->mmio[BDW];
> + struct nd_region *nd_region = nfit_blk->nd_region;
> unsigned int copied = 0;
> u64 base_offset;
> int rc;
> @@ -2283,7 +2285,8 @@ static int acpi_nfit_blk_single_io(struct nfit_blk *nfit_blk,
> }
>
> if (rw)
> - nvdimm_flush(nfit_blk->nd_region);
> + nd_region->flush(nd_region);
> +
>
...ditto, no need to touch this code.
> rc = read_blk_stat(nfit_blk, lane) ? -EIO : 0;
> return rc;
> diff --git a/drivers/nvdimm/claim.c b/drivers/nvdimm/claim.c
> index fb667bf..49dce9c 100644
> --- a/drivers/nvdimm/claim.c
> +++ b/drivers/nvdimm/claim.c
> @@ -262,6 +262,7 @@ static int nsio_rw_bytes(struct nd_namespace_common *ndns,
> {
> struct nd_namespace_io *nsio = to_nd_namespace_io(&ndns->dev);
> unsigned int sz_align = ALIGN(size + (offset & (512 - 1)), 512);
> + struct nd_region *nd_region = to_nd_region(ndns->dev.parent);
> sector_t sector = offset >> 9;
> int rc = 0;
>
> @@ -301,7 +302,7 @@ static int nsio_rw_bytes(struct nd_namespace_common *ndns,
> }
>
> memcpy_flushcache(nsio->addr + offset, buf, size);
> - nvdimm_flush(to_nd_region(ndns->dev.parent));
> + nd_region->flush(nd_region);
For this you would need to teach nsio_rw_bytes() that the flush can fail.
>
> return rc;
> }
> diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c
> index 6071e29..ba57cfa 100644
> --- a/drivers/nvdimm/pmem.c
> +++ b/drivers/nvdimm/pmem.c
> @@ -201,7 +201,8 @@ static blk_qc_t pmem_make_request(struct request_queue *q, struct bio *bio)
> struct nd_region *nd_region = to_region(pmem);
>
> if (bio->bi_opf & REQ_PREFLUSH)
> - nvdimm_flush(nd_region);
> + bio->bi_status = nd_region->flush(nd_region);
> +
Let's have nvdimm_flush() return 0 or -EIO if it fails since thats
what nsio_rw_bytes() expects, and you'll need to translate that to:
BLK_STS_IOERR
>
> do_acct = nd_iostat_start(bio, &start);
> bio_for_each_segment(bvec, bio, iter) {
> @@ -216,7 +217,7 @@ static blk_qc_t pmem_make_request(struct request_queue *q, struct bio *bio)
> nd_iostat_end(bio, start);
>
> if (bio->bi_opf & REQ_FUA)
> - nvdimm_flush(nd_region);
> + bio->bi_status = nd_region->flush(nd_region);
Same comment.
>
> bio_endio(bio);
> return BLK_QC_T_NONE;
> @@ -517,6 +518,7 @@ static int nd_pmem_probe(struct device *dev)
> static int nd_pmem_remove(struct device *dev)
> {
> struct pmem_device *pmem = dev_get_drvdata(dev);
> + struct nd_region *nd_region = to_region(pmem);
>
> if (is_nd_btt(dev))
> nvdimm_namespace_detach_btt(to_nd_btt(dev));
> @@ -528,14 +530,16 @@ static int nd_pmem_remove(struct device *dev)
> sysfs_put(pmem->bb_state);
> pmem->bb_state = NULL;
> }
> - nvdimm_flush(to_nd_region(dev->parent));
> + nd_region->flush(nd_region);
Not needed if the indirect function call moves inside nvdimm_flush().
>
> return 0;
> }
>
> static void nd_pmem_shutdown(struct device *dev)
> {
> - nvdimm_flush(to_nd_region(dev->parent));
> + struct nd_region *nd_region = to_nd_region(dev->parent);
> +
> + nd_region->flush(nd_region);
> }
>
> static void nd_pmem_notify(struct device *dev, enum nvdimm_event event)
> diff --git a/drivers/nvdimm/region_devs.c b/drivers/nvdimm/region_devs.c
> index fa37afc..a170a6b 100644
> --- a/drivers/nvdimm/region_devs.c
> +++ b/drivers/nvdimm/region_devs.c
> @@ -290,7 +290,7 @@ static ssize_t deep_flush_store(struct device *dev, struct device_attribute *att
> return rc;
> if (!flush)
> return -EINVAL;
> - nvdimm_flush(nd_region);
> + nd_region->flush(nd_region);
Let's pass the error code through if the flush fails.
>
> return len;
> }
> @@ -1065,6 +1065,11 @@ static struct nd_region *nd_region_create(struct nvdimm_bus *nvdimm_bus,
> dev->of_node = ndr_desc->of_node;
> nd_region->ndr_size = resource_size(ndr_desc->res);
> nd_region->ndr_start = ndr_desc->res->start;
> + if (ndr_desc->flush)
> + nd_region->flush = ndr_desc->flush;
> + else
> + nd_region->flush = nvdimm_flush;
> +
We'll need to rename the existing nvdimm_flush() to generic_nvdimm_flush().
> nd_device_register(dev);
>
> return nd_region;
> @@ -1109,7 +1114,7 @@ EXPORT_SYMBOL_GPL(nvdimm_volatile_region_create);
> * nvdimm_flush - flush any posted write queues between the cpu and pmem media
> * @nd_region: blk or interleaved pmem region
> */
> -void nvdimm_flush(struct nd_region *nd_region)
> +int nvdimm_flush(struct nd_region *nd_region)
> {
> struct nd_region_data *ndrd = dev_get_drvdata(&nd_region->dev);
> int i, idx;
> @@ -1133,7 +1138,10 @@ void nvdimm_flush(struct nd_region *nd_region)
> if (ndrd_get_flush_wpq(ndrd, i, 0))
> writeq(1, ndrd_get_flush_wpq(ndrd, i, idx));
> wmb();
> +
> + return 0;
> }
> +
Needless newline.
> EXPORT_SYMBOL_GPL(nvdimm_flush);
>
> /**
> diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
> index 097072c..3af7177 100644
> --- a/include/linux/libnvdimm.h
> +++ b/include/linux/libnvdimm.h
> @@ -115,6 +115,7 @@ struct nd_mapping_desc {
> int position;
> };
>
> +struct nd_region;
> struct nd_region_desc {
> struct resource *res;
> struct nd_mapping_desc *mapping;
> @@ -126,6 +127,7 @@ struct nd_region_desc {
> int numa_node;
> unsigned long flags;
> struct device_node *of_node;
> + int (*flush)(struct nd_region *nd_region);
> };
>
> struct device;
> @@ -201,7 +203,7 @@ unsigned long nd_blk_memremap_flags(struct nd_blk_region *ndbr);
> unsigned int nd_region_acquire_lane(struct nd_region *nd_region);
> void nd_region_release_lane(struct nd_region *nd_region, unsigned int lane);
> u64 nd_fletcher64(void *addr, size_t len, bool le);
> -void nvdimm_flush(struct nd_region *nd_region);
> +int nvdimm_flush(struct nd_region *nd_region);
> int nvdimm_has_flush(struct nd_region *nd_region);
> int nvdimm_has_cache(struct nd_region *nd_region);
>
> --
> 2.9.3
>
next prev parent reply other threads:[~2018-09-22 0:43 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-08-31 13:30 [PATCH 0/3] kvm "fake DAX" device Pankaj Gupta
[not found] ` <20180831133019.27579-1-pagupta-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2018-08-31 13:30 ` [PATCH 1/3] nd: move nd_region to common header Pankaj Gupta
[not found] ` <20180831133019.27579-2-pagupta-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2018-09-22 0:47 ` Dan Williams
[not found] ` <CAPcyv4jFimkVnVuzza5TCG=KvY88KZnXzH4GNEgUBbTouprzJA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2018-09-24 11:40 ` Pankaj Gupta
2018-08-31 13:30 ` [PATCH 2/3] libnvdimm: nd_region flush callback support Pankaj Gupta
2018-09-04 15:29 ` kbuild test robot
[not found] ` <20180904152917.GE17047-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2018-09-05 8:40 ` Pankaj Gupta
2018-09-22 0:43 ` Dan Williams [this message]
2018-09-24 11:07 ` Pankaj Gupta
2018-08-31 13:30 ` [PATCH 3/3] virtio-pmem: Add virtio pmem driver Pankaj Gupta
[not found] ` <20180831133019.27579-4-pagupta-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2018-09-04 15:17 ` kbuild test robot
2018-09-05 8:34 ` Pankaj Gupta
2018-09-05 12:02 ` kbuild test robot
2018-09-12 16:54 ` Luiz Capitulino
2018-09-13 6:58 ` [Qemu-devel] " Pankaj Gupta
[not found] ` <831225077.12817716.1536821901550.JavaMail.zimbra-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2018-09-13 12:19 ` Luiz Capitulino
2018-09-14 12:13 ` Pankaj Gupta
2018-09-22 1:08 ` Dan Williams
2018-09-24 9:41 ` Pankaj Gupta
2018-09-27 13:06 ` Pankaj Gupta
2018-09-27 15:55 ` Dan Williams
2018-08-31 13:30 ` [PATCH] qemu: Add virtio pmem device Pankaj Gupta
[not found] ` <20180831133019.27579-5-pagupta-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2018-09-12 16:57 ` Luiz Capitulino
2018-09-13 7:06 ` Pankaj Gupta
[not found] ` <563893075.12819183.1536822387535.JavaMail.zimbra-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2018-09-13 12:22 ` Luiz Capitulino
2018-09-20 11:21 ` David Hildenbrand
[not found] ` <2721c3ee-88d1-a8e9-1f1e-ffc3eef1d1ca-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2018-09-20 12:03 ` [Qemu-devel] " Pankaj Gupta
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=CAPcyv4js9hWfXPK++a0DhDwtk1pBazMvb9H-X2OQy18uemUhGw@mail.gmail.com \
--to=dan.j.williams@intel.com \
--cc=david@redhat.com \
--cc=eblake@redhat.com \
--cc=hch@infradead.org \
--cc=imammedo@redhat.com \
--cc=jack@suse.cz \
--cc=kvm@vger.kernel.org \
--cc=kwolf@redhat.com \
--cc=lcapitulino@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvdimm@ml01.01.org \
--cc=mst@redhat.com \
--cc=nilal@redhat.com \
--cc=niteshnarayanlal@hotmail.com \
--cc=pagupta@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=riel@surriel.com \
--cc=ross.zwisler@intel.com \
--cc=stefanha@redhat.com \
--cc=xiaoguangrong.eric@gmail.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).