netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Bhupesh Sharma <bhsharma@redhat.com>,
	kexec@lists.infradead.org, Ariel Elior <aelior@marvell.com>,
	GR-everest-linux-l2@marvell.com,
	Manish Chopra <manishc@marvell.com>,
	"David S . Miller" <davem@davemloft.net>,
	Sasha Levin <sashal@kernel.org>,
	netdev@vger.kernel.org
Subject: [PATCH AUTOSEL 4.19 070/106] net: qed*: Reduce RX and TX default ring count when running inside kdump kernel
Date: Mon,  8 Jun 2020 19:22:02 -0400	[thread overview]
Message-ID: <20200608232238.3368589-70-sashal@kernel.org> (raw)
In-Reply-To: <20200608232238.3368589-1-sashal@kernel.org>

From: Bhupesh Sharma <bhsharma@redhat.com>

[ Upstream commit 73e030977f7884dbe1be0018bab517e8d02760f8 ]

Normally kdump kernel(s) run under severe memory constraint with the
basic idea being to save the crashdump vmcore reliably when the primary
kernel panics/hangs.

Currently the qed* ethernet driver ends up consuming a lot of memory in
the kdump kernel, leading to kdump kernel panic when one tries to save
the vmcore via ssh/nfs (thus utilizing the services of the underlying
qed* network interfaces).

An example OOM message log seen in the kdump kernel can be seen here
[1], with crashkernel size reservation of 512M.

Using tools like memstrack (see [2]), we can track the modules taking up
the bulk of memory in the kdump kernel and organize the memory usage
output as per 'highest allocator first'. An example log for the OOM case
indicates that the qed* modules end up allocating approximately 216M
memory, which is a large part of the total crashkernel size:

 dracut-pre-pivot[676]: ======== Report format module_summary: ========
 dracut-pre-pivot[676]: Module qed using 149.6MB (2394 pages), peak allocation 149.6MB (2394 pages)
 dracut-pre-pivot[676]: Module qede using 65.3MB (1045 pages), peak allocation 65.3MB (1045 pages)

This patch reduces the default RX and TX ring count from 1024 to 64
when running inside kdump kernel, which leads to a significant memory
saving.

An example log with the patch applied shows the reduced memory
allocation in the kdump kernel:
 dracut-pre-pivot[674]: ======== Report format module_summary: ========
 dracut-pre-pivot[674]: Module qed using 141.8MB (2268 pages), peak allocation 141.8MB (2268 pages)
 <..snip..>
[dracut-pre-pivot[674]: Module qede using 4.8MB (76 pages), peak allocation 4.9MB (78 pages)

Tested crashdump vmcore save via ssh/nfs protocol using underlying qed*
network interface after applying this patch.

[1] OOM log:
------------

 kworker/0:6: page allocation failure: order:6,
 mode:0x60c0c0(GFP_KERNEL|__GFP_COMP|__GFP_ZERO), nodemask=(null)
 kworker/0:6 cpuset=/ mems_allowed=0
 CPU: 0 PID: 145 Comm: kworker/0:6 Not tainted 4.18.0-109.el8.aarch64 #1
 Hardware name: To be filled by O.E.M. Saber/Saber, BIOS 0ACKL025
 01/18/2019
 Workqueue: events work_for_cpu_fn
 Call trace:
  dump_backtrace+0x0/0x188
  show_stack+0x24/0x30
  dump_stack+0x90/0xb4
  warn_alloc+0xf4/0x178
  __alloc_pages_nodemask+0xcac/0xd58
  alloc_pages_current+0x8c/0xf8
  kmalloc_order_trace+0x38/0x108
  qed_iov_alloc+0x40/0x248 [qed]
  qed_resc_alloc+0x224/0x518 [qed]
  qed_slowpath_start+0x254/0x928 [qed]
   __qede_probe+0xf8/0x5e0 [qede]
  qede_probe+0x68/0xd8 [qede]
  local_pci_probe+0x44/0xa8
  work_for_cpu_fn+0x20/0x30
  process_one_work+0x1ac/0x3e8
  worker_thread+0x44/0x448
  kthread+0x130/0x138
  ret_from_fork+0x10/0x18
  Cannot start slowpath
  qede: probe of 0000:05:00.1 failed with error -12

[2]. Memstrack tool: https://github.com/ryncsn/memstrack

Cc: kexec@lists.infradead.org
Cc: linux-kernel@vger.kernel.org
Cc: Ariel Elior <aelior@marvell.com>
Cc: GR-everest-linux-l2@marvell.com
Cc: Manish Chopra <manishc@marvell.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Bhupesh Sharma <bhsharma@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/qlogic/qede/qede.h      |  2 ++
 drivers/net/ethernet/qlogic/qede/qede_main.c | 11 +++++++++--
 2 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h
index dc3be8a4acf4..2bdc410d1144 100644
--- a/drivers/net/ethernet/qlogic/qede/qede.h
+++ b/drivers/net/ethernet/qlogic/qede/qede.h
@@ -550,12 +550,14 @@ int qede_add_tc_flower_fltr(struct qede_dev *edev, __be16 proto,
 #define RX_RING_SIZE		((u16)BIT(RX_RING_SIZE_POW))
 #define NUM_RX_BDS_MAX		(RX_RING_SIZE - 1)
 #define NUM_RX_BDS_MIN		128
+#define NUM_RX_BDS_KDUMP_MIN	63
 #define NUM_RX_BDS_DEF		((u16)BIT(10) - 1)
 
 #define TX_RING_SIZE_POW	13
 #define TX_RING_SIZE		((u16)BIT(TX_RING_SIZE_POW))
 #define NUM_TX_BDS_MAX		(TX_RING_SIZE - 1)
 #define NUM_TX_BDS_MIN		128
+#define NUM_TX_BDS_KDUMP_MIN	63
 #define NUM_TX_BDS_DEF		NUM_TX_BDS_MAX
 
 #define QEDE_MIN_PKT_LEN		64
diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index 0d8e39ffbcd1..1aabb2e7a38b 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -29,6 +29,7 @@
  * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
+#include <linux/crash_dump.h>
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/version.h>
@@ -730,8 +731,14 @@ static struct qede_dev *qede_alloc_etherdev(struct qed_dev *cdev,
 	edev->dp_module = dp_module;
 	edev->dp_level = dp_level;
 	edev->ops = qed_ops;
-	edev->q_num_rx_buffers = NUM_RX_BDS_DEF;
-	edev->q_num_tx_buffers = NUM_TX_BDS_DEF;
+
+	if (is_kdump_kernel()) {
+		edev->q_num_rx_buffers = NUM_RX_BDS_KDUMP_MIN;
+		edev->q_num_tx_buffers = NUM_TX_BDS_KDUMP_MIN;
+	} else {
+		edev->q_num_rx_buffers = NUM_RX_BDS_DEF;
+		edev->q_num_tx_buffers = NUM_TX_BDS_DEF;
+	}
 
 	DP_INFO(edev, "Allocated netdev with %d tx queues and %d rx queues\n",
 		info->num_queues, info->num_queues);
-- 
2.25.1


  parent reply	other threads:[~2020-06-08 23:55 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-08 23:20 [PATCH AUTOSEL 4.19 001/106] ath9x: Fix stack-out-of-bounds Write in ath9k_hif_usb_rx_cb Sasha Levin
2020-06-08 23:20 ` [PATCH AUTOSEL 4.19 002/106] ath9k: Fix use-after-free Write in ath9k_htc_rx_msg Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 008/106] net: atlantic: make hw_get_regs optional Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 009/106] net: ena: fix error returning in ena_com_get_hash_function() Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 015/106] ixgbe: Fix XDP redirect on archs with PAGE_SIZE above 4K Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 017/106] Bluetooth: Add SCO fallback for invalid LMP parameters error Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 025/106] batman-adv: Revert "disable ethtool link speed detection when auto negotiation off" Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 029/106] net: vmxnet3: fix possible buffer overflow caused by bad DMA value in vmxnet3_get_rss() Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 031/106] ath9k: Fix use-after-free Read in ath9k_wmi_ctrl_rx Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 032/106] ath9k: Fix general protection fault in ath9k_hif_usb_rx_cb Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 033/106] brcmfmac: fix wrong location to get firmware feature Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 035/106] e1000: Distribute switch variables for initialization Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 043/106] netfilter: nft_nat: return EOPNOTSUPP if type or flags are not supported Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 044/106] selftests/bpf: Fix memory leak in extract_build_id() Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 045/106] net: bcmgenet: set Rx mode before starting netif Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 051/106] net: lpc-enet: fix error return code in lpc_mii_init() Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 053/106] net: allwinner: Fix use correct return type for ndo_start_xmit() Sasha Levin
2020-06-08 23:21 ` [PATCH AUTOSEL 4.19 055/106] ath9k_htc: Silence undersized packet warnings Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 068/106] ath10k: Remove msdu from idr when management pkt send fails Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 069/106] wcn36xx: Fix error handling path in 'wcn36xx_probe()' Sasha Levin
2020-06-08 23:22 ` Sasha Levin [this message]
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 071/106] mt76: avoid rx reorder buffer overflow Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 073/106] net: dsa: mt7530: set CPU port to fallback mode Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 074/106] veth: Adjust hard_start offset on redirect XDP frames Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 075/106] net/mlx5e: IPoIB, Drop multicast packets that this interface sent Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 076/106] rtlwifi: Fix a double free in _rtl_usb_tx_urb_setup() Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 077/106] mwifiex: Fix memory corruption in dump_station Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 087/106] net: ethernet: fec: move GPR register offset and bit into DT Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 088/106] macvlan: Skip loopback packets in RX handler Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 097/106] ixgbe: fix signed-integer-overflow warning Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 104/106] vxlan: Avoid infinite loop when suppressing NS messages with invalid options Sasha Levin
2020-06-08 23:22 ` [PATCH AUTOSEL 4.19 105/106] net_failover: fixed rollback in net_failover_open() Sasha Levin

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=20200608232238.3368589-70-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=GR-everest-linux-l2@marvell.com \
    --cc=aelior@marvell.com \
    --cc=bhsharma@redhat.com \
    --cc=davem@davemloft.net \
    --cc=kexec@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=manishc@marvell.com \
    --cc=netdev@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    /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).