From: John Hubbard <jhubbard@nvidia.com>
To: Matthew Wilcox <willy@infradead.org>
Cc: linux-xfs@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-f2fs-devel@lists.sourceforge.net, cluster-devel@redhat.com,
linux-mm@kvack.org, ocfs2-devel@oss.oracle.com,
linux-fsdevel@vger.kernel.org, linux-ext4@vger.kernel.org,
linux-erofs@lists.ozlabs.org, linux-btrfs@vger.kernel.org
Subject: Re: [PATCH v7 09/24] mm: Put readahead pages in cache earlier
Date: Thu, 20 Feb 2020 20:19:39 -0800 [thread overview]
Message-ID: <7abd9e60-bcc0-7474-4535-51ec9fe3be5b@nvidia.com> (raw)
In-Reply-To: <20200221034304.GC24185@bombadil.infradead.org>
On 2/20/20 7:43 PM, Matthew Wilcox wrote:
> On Thu, Feb 20, 2020 at 07:19:58PM -0800, John Hubbard wrote:
>>> +static inline struct page *readahead_page(struct readahead_control *rac)
>>> +{
>>> + struct page *page;
>>> +
>>> + BUG_ON(rac->_batch_count > rac->_nr_pages);
>>> + rac->_nr_pages -= rac->_batch_count;
>>> + rac->_index += rac->_batch_count;
>>> + rac->_batch_count = 0;
>>
>>
>> Is it intentional, to set rac->_batch_count twice (here, and below)? The
>> only reason I can see is if a caller needs to use ->_batch_count in the
>> "return NULL" case, which doesn't seem to come up...
>
> Ah, but it does. Not in this patch, but the next one ...
>
> + if (aops->readahead) {
> + aops->readahead(rac);
> + /* Clean up the remaining pages */
> + while ((page = readahead_page(rac))) {
> + unlock_page(page);
> + put_page(page);
> + }
>
> In the normal case, the ->readahead method will consume all the pages,
> and we need readahead_page() to do nothing if it is called again.
>
>>> + if (!rac->_nr_pages)
>>> + return NULL;
>
> ... admittedly I could do:
>
> if (!rac->_nr_pages) {
> rac->_batch_count = 0;
> return NULL;
> }
>
> which might be less confusing.
Yes, that would be a nice bit of polish if you end up doing another revision for other
reasons.
>
>>> @@ -130,23 +129,23 @@ static void read_pages(struct readahead_control *rac, struct list_head *pages,
>>> readahead_count(rac));
>>> /* Clean up the remaining pages */
>>> put_pages_list(pages);
>>> - goto out;
>>> - }
>>> -
>>> - for (page_idx = 0; page_idx < readahead_count(rac); page_idx++) {
>>> - struct page *page = lru_to_page(pages);
>>> - list_del(&page->lru);
>>> - if (!add_to_page_cache_lru(page, rac->mapping, page->index,
>>> - gfp))
>>> + rac->_index += rac->_nr_pages;
>>> + rac->_nr_pages = 0;
>>> + } else {
>>> + while ((page = readahead_page(rac))) {
>>> aops->readpage(rac->file, page);
>>> - put_page(page);
>>> + put_page(page);
>>> + }
>>> }
>>>
>>> -out:
>>> blk_finish_plug(&plug);
>>>
>>> BUG_ON(!list_empty(pages));
>>> - rac->_nr_pages = 0;
>>> + BUG_ON(readahead_count(rac));
>>> +
>>> +out:
>>> + /* If we were called due to a conflicting page, skip over it */
>>
>> Tiny documentation nit: What if we were *not* called due to a conflicting page?
>> (And what is a "conflicting page", in this context, btw?) The next line unconditionally
>> moves the index ahead, so the "if" part of the comment really confuses me.
>
> By the end of the series, read_pages() is called in three places:
>
> 1. if (page && !xa_is_value(page)) {
> read_pages(&rac, &page_pool);
>
> 2. } else if (add_to_page_cache_lru(page, mapping, index + i,
> gfp_mask) < 0) {
> put_page(page);
> read_pages(&rac, &page_pool);
>
> 3. read_pages(&rac, &page_pool);
>
> In the first two cases, there's an existing page in the page cache
> (which conflicts with this readahead operation), and so we need to
> advance index. In the third case, we're exiting the function, so it
> does no harm to advance index one further.
OK, I see. As you know, I tend toward maybe over-documenting, but what about
adding just a *few* hints to help new readers, like this approximately (maybe
it should be pared down):
diff --git a/mm/readahead.c b/mm/readahead.c
index 9fb5f77dcf69..0dd5b09c376e 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -114,6 +114,10 @@ int read_cache_pages(struct address_space *mapping, struct list_head *pages,
EXPORT_SYMBOL(read_cache_pages);
+/*
+ * Read pages into the page cache, OR skip over a page if it is already in the
+ * page cache.
+ */
static void read_pages(struct readahead_control *rac, struct list_head *pages)
{
const struct address_space_operations *aops = rac->mapping->a_ops;
@@ -152,7 +156,11 @@ static void read_pages(struct readahead_control *rac, struct list_head *pages)
BUG_ON(readahead_count(rac));
out:
- /* If we were called due to a conflicting page, skip over it */
+ /*
+ * This routine might have been called in order to skip over a page
+ * that is already in the page cache. And for other cases, the index is
+ * ignored by the caller. So just increment unconditionally:
+ */
rac->_index++;
}
?
>
>>> + } else if (add_to_page_cache_lru(page, mapping, index + i,
>>> + gfp_mask) < 0) {
>>
>> I still think you'll want to compare against !=0, rather than < 0, here.
>
> I tend to prefer < 0 when checking for an error value in case the function
> decides to start using positive numbers to mean something. I don't think
> it's a particularly important preference though (after all, returning 1
> might mean "failed, but for this weird reason rather than an errno").
>
>>> + put_page(page);
>>> + read_pages(&rac, &page_pool);
>>
>> Doing a read_pages() in the error case is because...actually, I'm not sure yet.
>> Why do we do this? Effectively it's a retry?
>
> Same as the reason we call read_pages() if we found a page in the page
> cache earlier -- we're sending down a set of pages which are consecutive
> in the file's address space, and now we have to skip one. At least one ;-)
>
Got it. Finally. :)
thanks,
--
John Hubbard
NVIDIA
next prev parent reply other threads:[~2020-02-21 4:20 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-19 21:00 [PATCH v7 00/23] Change readahead API Matthew Wilcox
2020-02-19 21:00 ` [PATCH v7 01/24] mm: Move readahead prototypes from mm.h Matthew Wilcox
2020-02-21 2:43 ` John Hubbard
2020-02-21 21:48 ` Matthew Wilcox
2020-02-22 0:15 ` John Hubbard
2020-02-24 21:32 ` Christoph Hellwig
2020-02-19 21:00 ` [PATCH v7 02/24] mm: Return void from various readahead functions Matthew Wilcox
2020-02-24 21:33 ` Christoph Hellwig
2020-02-19 21:00 ` [PATCH v7 03/24] mm: Ignore return value of ->readpages Matthew Wilcox
2020-02-19 21:00 ` [PATCH v7 04/24] mm: Move readahead nr_pages check into read_pages Matthew Wilcox
2020-02-20 14:36 ` Zi Yan
2020-02-21 4:24 ` John Hubbard
2020-02-24 21:34 ` Christoph Hellwig
2020-02-19 21:00 ` [PATCH v7 05/24] mm: Use readahead_control to pass arguments Matthew Wilcox
2020-02-24 21:36 ` Christoph Hellwig
2020-02-19 21:00 ` [PATCH v7 06/24] mm: Rename various 'offset' parameters to 'index' Matthew Wilcox
2020-02-21 2:21 ` John Hubbard
2020-02-21 3:27 ` John Hubbard
2020-02-19 21:00 ` [PATCH v7 07/24] mm: rename readahead loop variable to 'i' Matthew Wilcox
2020-02-19 21:00 ` [PATCH v7 08/24] mm: Remove 'page_offset' from readahead loop Matthew Wilcox
2020-02-21 2:48 ` John Hubbard
2020-02-24 21:37 ` Christoph Hellwig
2020-02-19 21:00 ` [PATCH v7 09/24] mm: Put readahead pages in cache earlier Matthew Wilcox
2020-02-21 3:19 ` John Hubbard
2020-02-21 3:43 ` Matthew Wilcox
2020-02-21 4:19 ` John Hubbard [this message]
2020-02-24 21:40 ` Christoph Hellwig
2020-02-19 21:00 ` [PATCH v7 10/24] mm: Add readahead address space operation Matthew Wilcox
2020-02-20 15:00 ` Zi Yan
2020-02-20 15:10 ` Matthew Wilcox
2020-02-21 4:30 ` John Hubbard
2020-02-24 21:41 ` Christoph Hellwig
2020-02-19 21:00 ` [PATCH v7 11/24] mm: Move end_index check out of readahead loop Matthew Wilcox
2020-02-21 3:50 ` John Hubbard
2020-02-21 15:35 ` Matthew Wilcox
2020-02-21 19:41 ` John Hubbard
2020-02-19 21:00 ` [PATCH v7 12/24] mm: Add page_cache_readahead_unbounded Matthew Wilcox
2020-02-24 21:53 ` [Cluster-devel] " Christoph Hellwig
2020-02-19 21:00 ` [PATCH v7 13/24] fs: Convert mpage_readpages to mpage_readahead Matthew Wilcox
2020-02-24 21:54 ` [Cluster-devel] " Christoph Hellwig
2020-02-19 21:00 ` [PATCH v7 14/24] btrfs: Convert from readpages to readahead Matthew Wilcox
2020-02-20 9:42 ` Johannes Thumshirn
2020-02-20 13:48 ` Matthew Wilcox
2020-02-20 15:46 ` Christoph Hellwig
2020-02-20 15:54 ` Matthew Wilcox
2020-02-20 15:57 ` Christoph Hellwig
2020-02-24 21:43 ` Christoph Hellwig
2020-02-24 21:54 ` Matthew Wilcox
2020-02-24 21:57 ` Christoph Hellwig
2020-02-19 21:00 ` [PATCH v7 15/24] erofs: Convert uncompressed files " Matthew Wilcox
2020-02-19 21:00 ` [PATCH v7 16/24] erofs: Convert compressed " Matthew Wilcox
2020-02-19 21:00 ` [PATCH v7 17/24] ext4: Convert " Matthew Wilcox
2020-02-19 21:00 ` [PATCH v7 18/24] ext4: Pass the inode to ext4_mpage_readpages Matthew Wilcox
2020-02-19 21:00 ` [PATCH v7 19/24] f2fs: Convert from readpages to readahead Matthew Wilcox
2020-02-19 21:00 ` [PATCH v7 20/24] fuse: " Matthew Wilcox
2020-02-19 21:01 ` [PATCH v7 21/24] iomap: Restructure iomap_readpages_actor Matthew Wilcox
2020-02-20 15:47 ` Christoph Hellwig
2020-02-20 16:24 ` Matthew Wilcox
2020-02-24 22:17 ` Christoph Hellwig
2020-02-25 1:49 ` Matthew Wilcox
2020-02-22 0:44 ` Darrick J. Wong
2020-02-22 1:54 ` Matthew Wilcox
2020-02-23 17:55 ` Darrick J. Wong
2020-02-19 21:01 ` [PATCH v7 22/24] iomap: Convert from readpages to readahead Matthew Wilcox
2020-02-20 15:49 ` Christoph Hellwig
2020-02-20 16:57 ` Matthew Wilcox
2020-02-22 1:00 ` Darrick J. Wong
2020-02-24 4:33 ` Matthew Wilcox
2020-02-24 16:52 ` Darrick J. Wong
2020-02-22 1:03 ` Darrick J. Wong
2020-02-22 1:09 ` Matthew Wilcox
2020-02-19 21:01 ` [PATCH v7 23/24] mm: Document why we don't set PageReadahead Matthew Wilcox
2020-02-19 21:01 ` [PATCH v7 24/24] mm: Use memalloc_nofs_save in readahead path Matthew Wilcox
2020-02-20 17:54 ` [PATCH v7 00/23] Change readahead API David Sterba
2020-02-20 22:39 ` Matthew Wilcox
2020-02-21 11:59 ` David Sterba
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=7abd9e60-bcc0-7474-4535-51ec9fe3be5b@nvidia.com \
--to=jhubbard@nvidia.com \
--cc=cluster-devel@redhat.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-erofs@lists.ozlabs.org \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-f2fs-devel@lists.sourceforge.net \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-xfs@vger.kernel.org \
--cc=ocfs2-devel@oss.oracle.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).