From: Sergey Senozhatsky <senozhatsky@chromium.org>
To: Andrew Morton <akpm@linux-foundation.org>,
Yosry Ahmed <yosry.ahmed@linux.dev>
Cc: Minchan Kim <minchan@kernel.org>, Nhat Pham <nphamcs@gmail.com>,
Johannes Weiner <hannes@cmpxchg.org>,
Brian Geffon <bgeffon@google.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
Sergey Senozhatsky <senozhatsky@chromium.org>,
Herbert Xu <herbert@gondor.apana.org.au>
Subject: [PATCH] zsmalloc: introduce SG-list based object read API
Date: Tue, 13 Jan 2026 12:46:45 +0900 [thread overview]
Message-ID: <20260113034645.2729998-1-senozhatsky@chromium.org> (raw)
Currently, zsmalloc performs address linearization on read
(which sometimes requires memcpy() to a local buffer).
Not all zsmalloc users need a linear address. For example,
Crypto API supports SG-list, performing linearization under
the hood, if needed. In addition, some compressors can have
native SG-list support, completely avoiding the linearization
step.
Provide an SG-list based zsmalloc read API:
- zs_obj_read_sg_begin()
- zs_obj_read_sg_end()
This API allows callers to obtain an SG representation
of the object (one entry for objects that are contained
in a single page and two entries for spanning objects),
avoiding the need for a bounce buffer and memcpy.
Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
---
include/linux/zsmalloc.h | 4 +++
mm/zsmalloc.c | 65 ++++++++++++++++++++++++++++++++++++++++
2 files changed, 69 insertions(+)
diff --git a/include/linux/zsmalloc.h b/include/linux/zsmalloc.h
index 5565c3171007..11e614663dd3 100644
--- a/include/linux/zsmalloc.h
+++ b/include/linux/zsmalloc.h
@@ -22,6 +22,7 @@ struct zs_pool_stats {
};
struct zs_pool;
+struct scatterlist;
struct zs_pool *zs_create_pool(const char *name);
void zs_destroy_pool(struct zs_pool *pool);
@@ -43,6 +44,9 @@ void *zs_obj_read_begin(struct zs_pool *pool, unsigned long handle,
size_t mem_len, void *local_copy);
void zs_obj_read_end(struct zs_pool *pool, unsigned long handle,
size_t mem_len, void *handle_mem);
+int zs_obj_read_sg_begin(struct zs_pool *pool, unsigned long handle,
+ struct scatterlist *sg, size_t mem_len);
+void zs_obj_read_sg_end(struct zs_pool *pool, unsigned long handle);
void zs_obj_write(struct zs_pool *pool, unsigned long handle,
void *handle_mem, size_t mem_len);
diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
index 16d5587a052a..5abb8bc0956a 100644
--- a/mm/zsmalloc.c
+++ b/mm/zsmalloc.c
@@ -30,6 +30,7 @@
#include <linux/highmem.h>
#include <linux/string.h>
#include <linux/slab.h>
+#include <linux/scatterlist.h>
#include <linux/spinlock.h>
#include <linux/sprintf.h>
#include <linux/shrinker.h>
@@ -1146,6 +1147,70 @@ void zs_obj_read_end(struct zs_pool *pool, unsigned long handle,
}
EXPORT_SYMBOL_GPL(zs_obj_read_end);
+int zs_obj_read_sg_begin(struct zs_pool *pool, unsigned long handle,
+ struct scatterlist *sg, size_t mem_len)
+{
+ struct zspage *zspage;
+ struct zpdesc *zpdesc;
+ unsigned long obj, off;
+ unsigned int obj_idx;
+ struct size_class *class;
+
+ /* Guarantee we can get zspage from handle safely */
+ read_lock(&pool->lock);
+ obj = handle_to_obj(handle);
+ obj_to_location(obj, &zpdesc, &obj_idx);
+ zspage = get_zspage(zpdesc);
+
+ /* Make sure migration doesn't move any pages in this zspage */
+ zspage_read_lock(zspage);
+ read_unlock(&pool->lock);
+
+ class = zspage_class(pool, zspage);
+ off = offset_in_page(class->size * obj_idx);
+
+ if (!ZsHugePage(zspage))
+ off += ZS_HANDLE_SIZE;
+
+ if (off + mem_len <= PAGE_SIZE) {
+ /* this object is contained entirely within a page */
+ sg_init_table(sg, 1);
+ sg_set_page(sg, zpdesc_page(zpdesc), mem_len, off);
+ } else {
+ size_t sizes[2];
+
+ /* this object spans two pages */
+ sizes[0] = PAGE_SIZE - off;
+ sizes[1] = mem_len - sizes[0];
+
+ sg_init_table(sg, 2);
+ sg_set_page(sg, zpdesc_page(zpdesc), sizes[0], off);
+
+ zpdesc = get_next_zpdesc(zpdesc);
+ sg = sg_next(sg);
+
+ sg_set_page(sg, zpdesc_page(zpdesc), sizes[1], 0);
+ }
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(zs_obj_read_sg_begin);
+
+void zs_obj_read_sg_end(struct zs_pool *pool, unsigned long handle)
+{
+ struct zspage *zspage;
+ struct zpdesc *zpdesc;
+ unsigned long obj;
+ unsigned int obj_idx;
+
+ obj = handle_to_obj(handle);
+ obj_to_location(obj, &zpdesc, &obj_idx);
+ zspage = get_zspage(zpdesc);
+
+ zspage_read_unlock(zspage);
+}
+EXPORT_SYMBOL_GPL(zs_obj_read_sg_end);
+
void zs_obj_write(struct zs_pool *pool, unsigned long handle,
void *handle_mem, size_t mem_len)
{
--
2.52.0.457.g6b5491de43-goog
next reply other threads:[~2026-01-13 3:47 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-13 3:46 Sergey Senozhatsky [this message]
2026-01-13 4:30 ` Herbert Xu
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=20260113034645.2729998-1-senozhatsky@chromium.org \
--to=senozhatsky@chromium.org \
--cc=akpm@linux-foundation.org \
--cc=bgeffon@google.com \
--cc=hannes@cmpxchg.org \
--cc=herbert@gondor.apana.org.au \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=minchan@kernel.org \
--cc=nphamcs@gmail.com \
--cc=yosry.ahmed@linux.dev \
/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