Received-SPF: pass (google.com: domain of linux-kernel+bounces-213178-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
From: Anand Khoje <anand.a.khoje@oracle.com>
To: linux-rdma@vger.kernel.org, linux-kernel@vger.kernel.org,
        netdev@vger.kernel.org
Cc: anand.a.khoje@oracle.com, rama.nichanamatlu@oracle.com,
        manjunath.b.patil@oracle.com
Subject: [PATCH v2] RDMA/mlx5 : Reclaim max 50K pages at once
Date: Thu, 13 Jun 2024 17:42:52 +0530
Message-ID: <20240613121252.93315-1-anand.a.khoje@oracle.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In non FLR context, at times CX-5 requests release of ~8 million FW pages.
This needs humongous number of cmd mailboxes, which to be released once
the pages are reclaimed. Release of humongous number of cmd mailboxes is
consuming cpu time running into many seconds. Which with non preemptible
kernels is leading to critical process starving on that cpu’s RQ.
To alleviate this, this change restricts the total number of pages
a worker will try to reclaim maximum 50K pages in one go.
The limit 50K is aligned with the current firmware capacity/limit of
releasing 50K pages at once per MLX5_CMD_OP_MANAGE_PAGES + MLX5_PAGES_TAKE
device command.

Our tests have shown significant benefit of this change in terms of
time consumed by dma_pool_free().
During a test where an event was raised by HCA
to release 1.3 Million pages, following observations were made:

- Without this change:
Number of mailbox messages allocated was around 20K, to accommodate
the DMA addresses of 1.3 million pages.
The average time spent by dma_pool_free() to free the DMA pool is between
16 usec to 32 usec.
           value  ------------- Distribution ------------- count
             256 |                                         0
             512 |@                                        287
            1024 |@@@                                      1332
            2048 |@                                        656
            4096 |@@@@@                                    2599
            8192 |@@@@@@@@@@                               4755
           16384 |@@@@@@@@@@@@@@@                          7545
           32768 |@@@@@                                    2501
           65536 |                                         0

- With this change:
Number of mailbox messages allocated was around 800; this was to
accommodate DMA addresses of only 50K pages.
The average time spent by dma_pool_free() to free the DMA pool in this case
lies between 1 usec to 2 usec.
           value  ------------- Distribution ------------- count
             256 |                                         0
             512 |@@@@@@@@@@@@@@@@@@                       346
            1024 |@@@@@@@@@@@@@@@@@@@@@@                   435
            2048 |                                         0
            4096 |                                         0
            8192 |                                         1
           16384 |                                         0

Signed-off-by: Anand Khoje <anand.a.khoje@oracle.com>
---
Changes in v2:
 - In v1, CPUs were yielded if more than 2 msec are spent in
   mlx5_free_cmd_msg(). The approach to limit the time spent is changed
   in this version.
---
 drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
index 1b38397..b1cf97d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c
@@ -482,12 +482,16 @@ static int reclaim_pages(struct mlx5_core_dev *dev, u32 func_id, int npages,
 	return err;
 }
 
+#define MAX_RECLAIM_NPAGES -50000
 static void pages_work_handler(struct work_struct *work)
 {
 	struct mlx5_pages_req *req = container_of(work, struct mlx5_pages_req, work);
 	struct mlx5_core_dev *dev = req->dev;
 	int err = 0;
 
+	if (req->npages < MAX_RECLAIM_NPAGES)
+		req->npages = MAX_RECLAIM_NPAGES;
+
 	if (req->release_all)
 		release_all_pages(dev, req->func_id, req->ec_function);
 	else if (req->npages < 0)
-- 
1.8.3.1