From: Christoph Hellwig <hch@lst.de>
To: Kundan Kumar <kundan.kumar@samsung.com>
Cc: Christoph Hellwig <hch@lst.de>,
axboe@kernel.dk, willy@infradead.org,
linux-block@vger.kernel.org, joshi.k@samsung.com,
mcgrof@kernel.org, anuj20.g@samsung.com, nj.shetty@samsung.com,
c.gameti@samsung.com, gost.dev@samsung.com
Subject: Re: [PATCH] block : add larger order folio size instead of pages
Date: Sat, 27 Apr 2024 10:14:21 +0200 [thread overview]
Message-ID: <20240427081421.GA5666@lst.de> (raw)
In-Reply-To: <20240424132246.7ny74cec7cvphg5i@green245>
On Wed, Apr 24, 2024 at 06:52:46PM +0530, Kundan Kumar wrote:
> On 22/04/24 01:14PM, Christoph Hellwig wrote:
>>> + folio = page_folio(page);
>>> +
>>> + if (!folio_test_large(folio) ||
>>> + (bio_op(bio) == REQ_OP_ZONE_APPEND)) {
>>
>> I don't understand why you need this branch. All the arithmetics
>> below should also work just fine for non-large folios
>
> The branch helps to skip these calculations for zero order folio:
> A) folio_offset = (folio_page_idx(folio, page) << PAGE_SHIFT) + offset;
> B) folio_size(folio)
Well, we'll need to just handle folio and stop special casing
order 0 ones eventually.
> If we convert bio_iov_add_page() to bio_iov_add_folio()/bio_add_folio(),
> we see a decline of about 11% for 4K I/O. When mTHP is enabled we may get
> a large order folio even for a 4K I/O. The folio_offset may become larger
> than 4K and we endup using expensive mempool_alloc during nvme_map_data in
> NVMe driver[1].
>
> [1]
> static blk_status_t nvme_map_data(struct nvme_dev *dev, struct request *req,
> struct nvme_command *cmnd)
> {
> ...
> ...
> if (bv.bv_offset + bv.bv_len <= NVME_CTRL_PAGE_SIZE * 2)
We can replace this with:
if ((bv->bv_offset & (NVME_CTRL_PAGE_SIZE - 1)) + bv.bv_len <=
NVME_CTRL_PAGE_SIZE * 2)
as nvme_setup_prp_simple just masks away the high bits anyway.
next prev parent reply other threads:[~2024-04-27 8:14 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CGME20240419092428epcas5p4f63759b0efa1f12dfbcf13c67fa8d0f0@epcas5p4.samsung.com>
2024-04-19 9:17 ` [PATCH] block : add larger order folio size instead of pages Kundan Kumar
2024-04-19 14:16 ` Matthew Wilcox
2024-04-22 9:44 ` Kundan Kumar
2024-04-22 11:14 ` Christoph Hellwig
2024-04-24 13:22 ` Kundan Kumar
2024-04-24 17:04 ` Keith Busch
2024-04-27 8:14 ` Christoph Hellwig [this message]
2024-04-24 16:16 ` Keith Busch
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=20240427081421.GA5666@lst.de \
--to=hch@lst.de \
--cc=anuj20.g@samsung.com \
--cc=axboe@kernel.dk \
--cc=c.gameti@samsung.com \
--cc=gost.dev@samsung.com \
--cc=joshi.k@samsung.com \
--cc=kundan.kumar@samsung.com \
--cc=linux-block@vger.kernel.org \
--cc=mcgrof@kernel.org \
--cc=nj.shetty@samsung.com \
--cc=willy@infradead.org \
/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).