From: Saeed Mahameed <saeed@kernel.org>
To: "David S. Miller" <davem@davemloft.net>,
Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
Eric Dumazet <edumazet@google.com>
Cc: Saeed Mahameed <saeedm@nvidia.com>,
netdev@vger.kernel.org, Tariq Toukan <tariqt@nvidia.com>,
Gal Pressman <gal@nvidia.com>,
Leon Romanovsky <leonro@nvidia.com>,
Mohamed Khalfella <mkhalfella@purestorage.com>,
Yuanyuan Zhong <yzhong@purestorage.com>,
Moshe Shemesh <moshe@nvidia.com>
Subject: [net 2/8] net/mlx5: Added cond_resched() to crdump collection
Date: Wed, 25 Sep 2024 13:20:07 -0700 [thread overview]
Message-ID: <20240925202013.45374-3-saeed@kernel.org> (raw)
In-Reply-To: <20240925202013.45374-1-saeed@kernel.org>
From: Mohamed Khalfella <mkhalfella@purestorage.com>
Collecting crdump involves reading vsc registers from pci config space
of mlx device, which can take long time to complete. This might result
in starving other threads waiting to run on the cpu.
Numbers I got from testing ConnectX-5 Ex MCX516A-CDAT in the lab:
- mlx5_vsc_gw_read_block_fast() was called with length = 1310716.
- mlx5_vsc_gw_read_fast() reads 4 bytes at a time. It was not used to
read the entire 1310716 bytes. It was called 53813 times because
there are jumps in read_addr.
- On average mlx5_vsc_gw_read_fast() took 35284.4ns.
- In total mlx5_vsc_wait_on_flag() called vsc_read() 54707 times.
The average time for each call was 17548.3ns. In some instances
vsc_read() was called more than one time when the flag was not set.
As expected the thread released the cpu after 16 iterations in
mlx5_vsc_wait_on_flag().
- Total time to read crdump was 35284.4ns * 53813 ~= 1.898s.
It was seen in the field that crdump can take more than 5 seconds to
complete. During that time mlx5_vsc_wait_on_flag() did not release the
cpu because it did not complete 16 iterations. It is believed that pci
config reads were slow. Adding cond_resched() every 128 register read
improves the situation. In the common case the, crdump takes ~1.8989s,
the thread yields the cpu every ~4.51ms. If crdump takes ~5s, the thread
yields the cpu every ~18.0ms.
Fixes: 8b9d8baae1de ("net/mlx5: Add Crdump support")
Reviewed-by: Yuanyuan Zhong <yzhong@purestorage.com>
Signed-off-by: Mohamed Khalfella <mkhalfella@purestorage.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.c | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.c
index d0b595ba6110..432c98f2626d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/pci_vsc.c
@@ -24,6 +24,11 @@
pci_write_config_dword((dev)->pdev, (dev)->vsc_addr + (offset), (val))
#define VSC_MAX_RETRIES 2048
+/* Reading VSC registers can take relatively long time.
+ * Yield the cpu every 128 registers read.
+ */
+#define VSC_GW_READ_BLOCK_COUNT 128
+
enum {
VSC_CTRL_OFFSET = 0x4,
VSC_COUNTER_OFFSET = 0x8,
@@ -273,6 +278,7 @@ int mlx5_vsc_gw_read_block_fast(struct mlx5_core_dev *dev, u32 *data,
{
unsigned int next_read_addr = 0;
unsigned int read_addr = 0;
+ unsigned int count = 0;
while (read_addr < length) {
if (mlx5_vsc_gw_read_fast(dev, read_addr, &next_read_addr,
@@ -280,6 +286,10 @@ int mlx5_vsc_gw_read_block_fast(struct mlx5_core_dev *dev, u32 *data,
return read_addr;
read_addr = next_read_addr;
+ if (++count == VSC_GW_READ_BLOCK_COUNT) {
+ cond_resched();
+ count = 0;
+ }
}
return length;
}
--
2.46.1
next prev parent reply other threads:[~2024-09-25 20:20 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-25 20:20 [pull request][net 0/8] mlx5 fixes 2024-09-25 Saeed Mahameed
2024-09-25 20:20 ` [net 1/8] net/mlx5: Fix error path in multi-packet WQE transmit Saeed Mahameed
2024-10-03 0:30 ` patchwork-bot+netdevbpf
2024-09-25 20:20 ` Saeed Mahameed [this message]
2024-09-25 20:20 ` [net 3/8] net/mlx5e: Fix NULL deref in mlx5e_tir_builder_alloc() Saeed Mahameed
2024-09-25 20:20 ` [net 4/8] net/mlx5: Fix wrong reserved field in hca_cap_2 in mlx5_ifc Saeed Mahameed
2024-09-25 20:20 ` [net 5/8] net/mlx5: HWS, fixed double-free in error flow of creating SQ Saeed Mahameed
2024-09-25 20:20 ` [net 6/8] net/mlx5: HWS, changed E2BIG error to a negative return code Saeed Mahameed
2024-09-25 20:20 ` [net 7/8] net/mlx5e: SHAMPO, Fix overflow of hd_per_wq Saeed Mahameed
2024-09-25 20:20 ` [net 8/8] net/mlx5e: Fix crash caused by calling __xfrm_state_delete() twice Saeed Mahameed
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=20240925202013.45374-3-saeed@kernel.org \
--to=saeed@kernel.org \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=gal@nvidia.com \
--cc=kuba@kernel.org \
--cc=leonro@nvidia.com \
--cc=mkhalfella@purestorage.com \
--cc=moshe@nvidia.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=saeedm@nvidia.com \
--cc=tariqt@nvidia.com \
--cc=yzhong@purestorage.com \
/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).