From: Jan Kara <jack@suse.cz>
To: Dan Williams <dan.j.williams@intel.com>
Cc: axboe@fb.com, jack@suse.cz, linux-nvdimm@lists.01.org,
david@fromorbit.com, linux-kernel@vger.kernel.org,
ross.zwisler@linux.intel.com, willy@linux.intel.com,
akpm@linux-foundation.org, hch@lst.de
Subject: Re: [PATCH 2/5] dax: increase granularity of dax_clear_blocks() operations
Date: Thu, 22 Oct 2015 11:26:14 +0200 [thread overview]
Message-ID: <20151022092614.GD14445@quack.suse.cz> (raw)
In-Reply-To: <20151022064154.12700.90545.stgit@dwillia2-desk3.amr.corp.intel.com>
On Thu 22-10-15 02:41:54, Dan Williams wrote:
> dax_clear_blocks is currently performing a cond_resched() after every
> PAGE_SIZE memset. We need not check so frequently, for example md-raid
> only calls cond_resched() at stripe granularity. Also, in preparation
> for introducing a dax_map_atomic() operation that temporarily pins a dax
> mapping move the call to cond_resched() to the outer loop.
>
> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The patch looks good to me. You can add:
Reviewed-by: Jan Kara <jack@suse.com>
Honza
> ---
> fs/dax.c | 27 ++++++++++++---------------
> 1 file changed, 12 insertions(+), 15 deletions(-)
>
> diff --git a/fs/dax.c b/fs/dax.c
> index 5dc33d788d50..f8e543839e5c 100644
> --- a/fs/dax.c
> +++ b/fs/dax.c
> @@ -28,6 +28,7 @@
> #include <linux/sched.h>
> #include <linux/uio.h>
> #include <linux/vmstat.h>
> +#include <linux/sizes.h>
>
> int dax_clear_blocks(struct inode *inode, sector_t block, long size)
> {
> @@ -38,24 +39,20 @@ int dax_clear_blocks(struct inode *inode, sector_t block, long size)
> do {
> void __pmem *addr;
> unsigned long pfn;
> - long count;
> + long count, sz;
>
> - count = bdev_direct_access(bdev, sector, &addr, &pfn, size);
> + sz = min_t(long, size, SZ_1M);
> + count = bdev_direct_access(bdev, sector, &addr, &pfn, sz);
> if (count < 0)
> return count;
> - BUG_ON(size < count);
> - while (count > 0) {
> - unsigned pgsz = PAGE_SIZE - offset_in_page(addr);
> - if (pgsz > count)
> - pgsz = count;
> - clear_pmem(addr, pgsz);
> - addr += pgsz;
> - size -= pgsz;
> - count -= pgsz;
> - BUG_ON(pgsz & 511);
> - sector += pgsz / 512;
> - cond_resched();
> - }
> + if (count < sz)
> + sz = count;
> + clear_pmem(addr, sz);
> + addr += sz;
> + size -= sz;
> + BUG_ON(sz & 511);
> + sector += sz / 512;
> + cond_resched();
> } while (size);
>
> wmb_pmem();
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
WARNING: multiple messages have this Message-ID (diff)
From: Jan Kara <jack@suse.cz>
To: Dan Williams <dan.j.williams@intel.com>
Cc: axboe@fb.com, jack@suse.cz, linux-nvdimm@ml01.01.org,
david@fromorbit.com, linux-kernel@vger.kernel.org,
ross.zwisler@linux.intel.com, willy@linux.intel.com,
akpm@linux-foundation.org, hch@lst.de
Subject: Re: [PATCH 2/5] dax: increase granularity of dax_clear_blocks() operations
Date: Thu, 22 Oct 2015 11:26:14 +0200 [thread overview]
Message-ID: <20151022092614.GD14445@quack.suse.cz> (raw)
In-Reply-To: <20151022064154.12700.90545.stgit@dwillia2-desk3.amr.corp.intel.com>
On Thu 22-10-15 02:41:54, Dan Williams wrote:
> dax_clear_blocks is currently performing a cond_resched() after every
> PAGE_SIZE memset. We need not check so frequently, for example md-raid
> only calls cond_resched() at stripe granularity. Also, in preparation
> for introducing a dax_map_atomic() operation that temporarily pins a dax
> mapping move the call to cond_resched() to the outer loop.
>
> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The patch looks good to me. You can add:
Reviewed-by: Jan Kara <jack@suse.com>
Honza
> ---
> fs/dax.c | 27 ++++++++++++---------------
> 1 file changed, 12 insertions(+), 15 deletions(-)
>
> diff --git a/fs/dax.c b/fs/dax.c
> index 5dc33d788d50..f8e543839e5c 100644
> --- a/fs/dax.c
> +++ b/fs/dax.c
> @@ -28,6 +28,7 @@
> #include <linux/sched.h>
> #include <linux/uio.h>
> #include <linux/vmstat.h>
> +#include <linux/sizes.h>
>
> int dax_clear_blocks(struct inode *inode, sector_t block, long size)
> {
> @@ -38,24 +39,20 @@ int dax_clear_blocks(struct inode *inode, sector_t block, long size)
> do {
> void __pmem *addr;
> unsigned long pfn;
> - long count;
> + long count, sz;
>
> - count = bdev_direct_access(bdev, sector, &addr, &pfn, size);
> + sz = min_t(long, size, SZ_1M);
> + count = bdev_direct_access(bdev, sector, &addr, &pfn, sz);
> if (count < 0)
> return count;
> - BUG_ON(size < count);
> - while (count > 0) {
> - unsigned pgsz = PAGE_SIZE - offset_in_page(addr);
> - if (pgsz > count)
> - pgsz = count;
> - clear_pmem(addr, pgsz);
> - addr += pgsz;
> - size -= pgsz;
> - count -= pgsz;
> - BUG_ON(pgsz & 511);
> - sector += pgsz / 512;
> - cond_resched();
> - }
> + if (count < sz)
> + sz = count;
> + clear_pmem(addr, sz);
> + addr += sz;
> + size -= sz;
> + BUG_ON(sz & 511);
> + sector += sz / 512;
> + cond_resched();
> } while (size);
>
> wmb_pmem();
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
next prev parent reply other threads:[~2015-10-22 9:26 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-22 6:41 [PATCH 0/5] block, dax: updates for 4.4 Dan Williams
2015-10-22 6:41 ` Dan Williams
2015-10-22 6:41 ` [PATCH 1/5] pmem, dax: clean up clear_pmem() Dan Williams
2015-10-22 6:41 ` Dan Williams
2015-10-22 6:41 ` [PATCH 2/5] dax: increase granularity of dax_clear_blocks() operations Dan Williams
2015-10-22 6:41 ` Dan Williams
2015-10-22 9:26 ` Jan Kara [this message]
2015-10-22 9:26 ` Jan Kara
2015-10-22 6:41 ` [PATCH 3/5] block, dax: fix lifetime of in-kernel dax mappings with dax_map_atomic() Dan Williams
2015-10-22 6:41 ` Dan Williams
2015-10-22 6:42 ` [PATCH 4/5] block: introduce file_bd_inode() Dan Williams
2015-10-22 6:42 ` Dan Williams
2015-10-22 9:45 ` Jan Kara
2015-10-22 9:45 ` Jan Kara
2015-10-22 15:41 ` Dan Williams
2015-10-22 15:41 ` Dan Williams
2015-10-22 6:42 ` [PATCH 5/5] block: enable dax for raw block devices Dan Williams
2015-10-22 6:42 ` Dan Williams
2015-10-22 9:35 ` Jan Kara
2015-10-22 9:35 ` Jan Kara
2015-10-22 16:05 ` Williams, Dan J
2015-10-22 16:05 ` Williams, Dan J
2015-10-22 21:08 ` Jan Kara
2015-10-22 21:08 ` Jan Kara
2015-10-22 23:41 ` Williams, Dan J
2015-10-22 23:41 ` Williams, Dan J
2015-10-24 12:21 ` Jan Kara
2015-10-24 12:21 ` Jan Kara
2015-10-23 23:32 ` Dan Williams
2015-10-23 23:32 ` Dan Williams
2015-10-24 14:49 ` Jan Kara
2015-10-24 14:49 ` Jan Kara
2015-10-25 21:22 ` Dave Chinner
2015-10-25 21:22 ` Dave Chinner
2015-10-26 2:48 ` Dan Williams
2015-10-26 2:48 ` Dan Williams
2015-10-26 6:23 ` Dave Chinner
2015-10-26 6:23 ` Dave Chinner
2015-10-26 7:20 ` Jan Kara
2015-10-26 7:20 ` Jan Kara
2015-10-26 8:56 ` Dan Williams
2015-10-26 8:56 ` Dan Williams
2015-10-26 22:19 ` Dave Chinner
2015-10-26 22:19 ` Dave Chinner
2015-10-27 22:55 ` Ross Zwisler
2015-10-27 22:55 ` Ross Zwisler
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=20151022092614.GD14445@quack.suse.cz \
--to=jack@suse.cz \
--cc=akpm@linux-foundation.org \
--cc=axboe@fb.com \
--cc=dan.j.williams@intel.com \
--cc=david@fromorbit.com \
--cc=hch@lst.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvdimm@lists.01.org \
--cc=ross.zwisler@linux.intel.com \
--cc=willy@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.