From: Minchan Kim <minchan@kernel.org>
To: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Nitin Gupta <ngupta@vflare.org>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCHv4 4/9] zsmalloc: make huge class watermark zs_pool member
Date: Thu, 10 Nov 2022 14:25:34 -0800 [thread overview]
Message-ID: <Y216Xq8V8InFQyms@google.com> (raw)
In-Reply-To: <20221031054108.541190-5-senozhatsky@chromium.org>
On Mon, Oct 31, 2022 at 02:41:03PM +0900, Sergey Senozhatsky wrote:
> We will permit per-pool configuration of pages per-zspage value,
> which changes characteristics of the classes and moves around
> huge class size watermark. Thus huge class size needs to be
> a per-pool variable.
I think part of code in previous patch should move here since
you are creating the feature in this patch:
BTW, I am wondering we really need to jump the per-pool config
option over global general golden ratio and/or smarter approach
to optimize transparently depending on how much memory we have
wasted.
>
> Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org>
> ---
> mm/zsmalloc.c | 10 +++++-----
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
> index 5f79223e7bfe..d329bd673baa 100644
> --- a/mm/zsmalloc.c
> +++ b/mm/zsmalloc.c
> @@ -178,7 +178,6 @@ static struct dentry *zs_stat_root;
> * (see: fix_fullness_group())
> */
> static const int fullness_threshold_frac = 4;
> -static size_t huge_class_size;
>
> struct size_class {
> spinlock_t lock;
> @@ -227,6 +226,7 @@ struct zs_pool {
>
> u32 num_size_classes;
> u32 min_alloc_size;
> + size_t huge_class_size;
>
> struct zs_pool_stats stats;
>
> @@ -1350,7 +1350,7 @@ EXPORT_SYMBOL_GPL(zs_unmap_object);
> */
> size_t zs_huge_class_size(struct zs_pool *pool)
> {
> - return huge_class_size;
> + return pool->huge_class_size;
> }
> EXPORT_SYMBOL_GPL(zs_huge_class_size);
>
> @@ -2262,8 +2262,8 @@ struct zs_pool *zs_create_pool(const char *name)
> * endup in the huge class.
> */
> if (pages_per_zspage != 1 && objs_per_zspage != 1 &&
> - !huge_class_size) {
> - huge_class_size = size;
> + !pool->huge_class_size) {
> + pool->huge_class_size = size;
> /*
> * The object uses ZS_HANDLE_SIZE bytes to store the
> * handle. We need to subtract it, because zs_malloc()
> @@ -2273,7 +2273,7 @@ struct zs_pool *zs_create_pool(const char *name)
> * class because it grows by ZS_HANDLE_SIZE extra bytes
> * right before class lookup.
> */
> - huge_class_size -= (ZS_HANDLE_SIZE - 1);
> + pool->huge_class_size -= (ZS_HANDLE_SIZE - 1);
> }
>
> /*
> --
> 2.38.1.273.g43a17bfeac-goog
>
next prev parent reply other threads:[~2022-11-10 22:25 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-31 5:40 [PATCHv4 0/9] zsmalloc/zram: configurable zspage size Sergey Senozhatsky
2022-10-31 5:41 ` [PATCHv4 1/9] zram: add size class equals check into recompression Sergey Senozhatsky
2022-10-31 5:41 ` [PATCHv4 2/9] zsmalloc: turn zspage order into runtime variable Sergey Senozhatsky
2022-11-10 21:59 ` Minchan Kim
2022-11-11 10:38 ` Sergey Senozhatsky
2022-11-11 17:09 ` Minchan Kim
2022-11-14 3:55 ` Sergey Senozhatsky
2022-10-31 5:41 ` [PATCHv4 3/9] zsmalloc: move away from page order defines Sergey Senozhatsky
2022-11-10 22:02 ` Minchan Kim
2022-10-31 5:41 ` [PATCHv4 4/9] zsmalloc: make huge class watermark zs_pool member Sergey Senozhatsky
2022-11-10 22:25 ` Minchan Kim [this message]
2022-11-11 1:07 ` Sergey Senozhatsky
2022-10-31 5:41 ` [PATCHv4 5/9] zram: huge size watermark cannot be global Sergey Senozhatsky
2022-10-31 5:41 ` [PATCHv4 6/9] zsmalloc: pass limit on pages per-zspage to zs_create_pool() Sergey Senozhatsky
2022-11-09 6:24 ` Sergey Senozhatsky
2022-11-11 17:14 ` Minchan Kim
2022-11-11 2:10 ` Minchan Kim
2022-11-11 10:32 ` Sergey Senozhatsky
2022-10-31 5:41 ` [PATCHv4 7/9] zram: add pages_per_pool_page device attribute Sergey Senozhatsky
2022-11-09 4:34 ` Sergey Senozhatsky
2022-10-31 5:41 ` [PATCHv4 8/9] Documentation: document zram pages_per_pool_page attribute Sergey Senozhatsky
2022-11-11 2:20 ` Minchan Kim
2022-11-11 10:34 ` Sergey Senozhatsky
2022-10-31 5:41 ` [PATCHv4 9/9] zsmalloc: break out of loop when found perfect zspage order Sergey Senozhatsky
2022-11-10 22:44 ` [PATCHv4 0/9] zsmalloc/zram: configurable zspage size Minchan Kim
2022-11-11 0:56 ` Sergey Senozhatsky
2022-11-11 17:03 ` Minchan Kim
2022-11-14 3:53 ` Sergey Senozhatsky
2022-11-14 7:55 ` Sergey Senozhatsky
2022-11-14 8:37 ` Sergey Senozhatsky
2022-11-15 6:01 ` Sergey Senozhatsky
2022-11-15 7:59 ` Sergey Senozhatsky
2022-11-15 23:23 ` Minchan Kim
2022-11-16 0:52 ` Sergey Senozhatsky
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=Y216Xq8V8InFQyms@google.com \
--to=minchan@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=ngupta@vflare.org \
--cc=senozhatsky@chromium.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).