linux-kernel - [PATCH 1/1] RDMA/mlx5: align MR mem allocation size to power-of-two

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20230629213248.3184245-2-yzhong@purestorage.com>
Date:   Thu, 29 Jun 2023 15:32:48 -0600
From:   Yuanyuan Zhong <yzhong@...estorage.com>
To:     leon@...nel.org, jgg@...pe.ca
Cc:     cachen@...estorage.com, linux-rdma@...r.kernel.org,
        linux-kernel@...r.kernel.org,
        Yuanyuan Zhong <yzhong@...estorage.com>
Subject: [PATCH 1/1] RDMA/mlx5: align MR mem allocation size to power-of-two

The MR memory allocation requests extra bytes to guarantee that there
is enough space to find the memory aligned to MLX5_UMR_ALIGN.

For power-of-two sizes, the alignment can be guaranteed by kmalloc()
according to commit 59bb47985c1d ("mm, sl[aou]b: guarantee natural
alignment for kmalloc(power-of-two)").

So if target alignment is power-of-two and adding the extra bytes
crosses a power-of-two boundary, use the next power-of-two as the
allocation size.

Signed-off-by: Yuanyuan Zhong <yzhong@...estorage.com>
---
 drivers/infiniband/hw/mlx5/mr.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 2017ede100a6..92f35fafb2c0 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -1766,6 +1766,11 @@ mlx5_alloc_priv_descs(struct ib_device *device,
 	int ret;
 
 	add_size = max_t(int, MLX5_UMR_ALIGN - ARCH_KMALLOC_MINALIGN, 0);
+	if (is_power_of_2(MLX5_UMR_ALIGN) && add_size) {
+		int end = max_t(int, MLX5_UMR_ALIGN, roundup_pow_of_two(size));
+
+		add_size = min_t(int, end - size, add_size);
+	}
 
 	mr->descs_alloc = kzalloc(size + add_size, GFP_KERNEL);
 	if (!mr->descs_alloc)
-- 
2.34.1