Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:7 as permitted sender) client-ip=2620:137:e000::3:7;
Date:   Wed, 13 Dec 2023 13:09:23 -0800
From:   Andrew Morton <akpm@linux-foundation.org>
To:     Sidhartha Kumar <sidhartha.kumar@oracle.com>
Cc:     linux-kernel@vger.kernel.org, linux-mm@kvack.org,
        maple-tree@lists.infradead.org, willy@infradead.org,
        liam.howlett@oracle.com, zhangpeng.00@bytedance.com,
        stable@vger.kernel.org
Subject: Re: [PATCH V2] maple_tree: do not preallocate nodes for slot stores
Message-Id: <20231213130923.cc00317b4ebab1b0864d8b42@linux-foundation.org>
In-Reply-To: <31d24c29-5b7c-328c-b830-276acab43203@oracle.com>
References: <20231213205058.386589-1-sidhartha.kumar@oracle.com>
        <31d24c29-5b7c-328c-b830-276acab43203@oracle.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Wed, 13 Dec 2023 13:03:29 -0800 Sidhartha Kumar <sidhartha.kumar@oracle.com> wrote:

> On 12/13/23 12:50 PM, Sidhartha Kumar wrote:
> > mas_preallocate() defaults to requesting 1 node for preallocation and then
> > ,depending on the type of store, will update the request variable. There
> > isn't a check for a slot store type, so slot stores are preallocating the
> > default 1 node. Slot stores do not require any additional nodes, so add a
> > check for the slot store case that will bypass node_count_gfp(). Update
> > the tests to reflect that slot stores do not require allocations.
> > 
> > User visible effects of this bug include increased memory usage from the
> > unneeded node that was allocated.
> > 
> > Fixes: 0b8bb544b1a7 ("maple_tree: update mas_preallocate() testing")
> > Cc: <stable@vger.kernel.org> # 6.6+
> > Signed-off-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>
> > ---
> > v1->v2:
> > 	fix coding style per Matthew and Andrew
> > 	use wr_mas->node_end to fix build error
> > 
> 
> When this is merged to mm-unstable could the following fixlet be applied to be 
> compatible with Liam's series[1]:
> 

Yup, already on it.  Liam's series is now in mm-stable so for now I'll
leave a bisection hole.  Maybe I'll fold it later if an mm-stable
rebase/rebuild is needed,

From: Andrew Morton <akpm@linux-foundation.org>
Subject: lib/maple_tree.c: fix build error due to hotfix alteration
Date: Wed Dec 13 12:59:49 PM PST 2023

Commit 0de56e38b307 ("maple_tree: use maple state end for write
operations") was broken by a later patch "maple_tree: do not preallocate
nodes for slot stores".  But the later patch was scheduled ahead of
0de56e38b307, for 6.7-rc.

This fixlet undoes the damage.

Fixes: 0de56e38b307 ("maple_tree: use maple state end for write operations")
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Sidhartha Kumar <sidhartha.kumar@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 lib/maple_tree.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/lib/maple_tree.c~lib-maple_treec-fix-build-error-due-to-hotfix-alteration
+++ a/lib/maple_tree.c
@@ -5477,7 +5477,7 @@ int mas_preallocate(struct ma_state *mas
 	node_size = mas_wr_new_end(&wr_mas);
 
 	/* Slot store, does not require additional nodes */
-	if (node_size == wr_mas.node_end) {
+	if (node_size == mas->end) {
 		/* reuse node */
 		if (!mt_in_rcu(mas->tree))
 			return 0;
_