netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
To: davem@davemloft.net
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>,
	netdev@vger.kernel.org, nhorman@redhat.com, sassmann@redhat.com,
	jogreene@redhat.com, Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Subject: [net-next 09/11] i40e: Test memory before ethtool alloc succeeds
Date: Fri,  6 May 2016 00:03:46 -0700	[thread overview]
Message-ID: <1462518228-30527-10-git-send-email-jeffrey.t.kirsher@intel.com> (raw)
In-Reply-To: <1462518228-30527-1-git-send-email-jeffrey.t.kirsher@intel.com>

From: Jesse Brandeburg <jesse.brandeburg@intel.com>

When testing on systems with very limited amounts of RAM, a bug was
found where, while changing the number of descriptors using ethtool,
the driver didn't test the limits of system memory before permanently
assuming it would be able to get receive buffer memory.

Work around this issue by pre-allocation of the receive buffer
memory, in the "ghost" ring, which is then used during reinit
using the new ring length.

Change-Id: I92d7a5fb59a6c884b2efdd1ec652845f101c3359
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 34 +++++++++++++++++++++++---
 1 file changed, 31 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 6fd730ac..51a994d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -1274,6 +1274,13 @@ static int i40e_set_ringparam(struct net_device *netdev,
 		}
 
 		for (i = 0; i < vsi->num_queue_pairs; i++) {
+			/* this is to allow wr32 to have something to write to
+			 * during early allocation of Rx buffers
+			 */
+			u32 __iomem faketail = 0;
+			struct i40e_ring *ring;
+			u16 unused;
+
 			/* clone ring and setup updated count */
 			rx_rings[i] = *vsi->rx_rings[i];
 			rx_rings[i].count = new_rx_count;
@@ -1282,12 +1289,22 @@ static int i40e_set_ringparam(struct net_device *netdev,
 			 */
 			rx_rings[i].desc = NULL;
 			rx_rings[i].rx_bi = NULL;
+			rx_rings[i].tail = (u8 __iomem *)&faketail;
 			err = i40e_setup_rx_descriptors(&rx_rings[i]);
+			if (err)
+				goto rx_unwind;
+
+			/* now allocate the Rx buffers to make sure the OS
+			 * has enough memory, any failure here means abort
+			 */
+			ring = &rx_rings[i];
+			unused = I40E_DESC_UNUSED(ring);
+			err = i40e_alloc_rx_buffers(ring, unused);
+rx_unwind:
 			if (err) {
-				while (i) {
-					i--;
+				do {
 					i40e_free_rx_resources(&rx_rings[i]);
-				}
+				} while (i--);
 				kfree(rx_rings);
 				rx_rings = NULL;
 
@@ -1313,6 +1330,17 @@ static int i40e_set_ringparam(struct net_device *netdev,
 	if (rx_rings) {
 		for (i = 0; i < vsi->num_queue_pairs; i++) {
 			i40e_free_rx_resources(vsi->rx_rings[i]);
+			/* get the real tail offset */
+			rx_rings[i].tail = vsi->rx_rings[i]->tail;
+			/* this is to fake out the allocation routine
+			 * into thinking it has to realloc everything
+			 * but the recycling logic will let us re-use
+			 * the buffers allocated above
+			 */
+			rx_rings[i].next_to_use = 0;
+			rx_rings[i].next_to_clean = 0;
+			rx_rings[i].next_to_alloc = 0;
+			/* do a struct copy */
 			*vsi->rx_rings[i] = rx_rings[i];
 		}
 		kfree(rx_rings);
-- 
2.5.5

  parent reply	other threads:[~2016-05-06  7:06 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-05-06  7:03 [net-next 00/11][pull request] 40GbE Intel Wired LAN Driver Updates 2016-05-05 Jeff Kirsher
2016-05-06  7:03 ` [net-next 01/11] i40e/i40evf: Refactor tunnel interpretation Jeff Kirsher
2016-05-06  7:03 ` [net-next 02/11] i40e: Drop packet split receive routine Jeff Kirsher
2016-05-06  7:03 ` [net-next 03/11] i40e/i40evf: Remove reference to ring->dtype Jeff Kirsher
2016-05-06  7:03 ` [net-next 04/11] i40e: Refactor receive routine Jeff Kirsher
2016-05-06  7:03 ` [net-next 05/11] i40evf: Drop packet split " Jeff Kirsher
2016-05-06  7:03 ` [net-next 06/11] i40evf: refactor " Jeff Kirsher
2016-05-06  7:03 ` [net-next 07/11] i40e/i40evf: Remove unused hardware receive descriptor code Jeff Kirsher
2016-05-06  7:03 ` [net-next 08/11] i40evf: Allocate Rx buffers properly Jeff Kirsher
2016-05-06  7:03 ` Jeff Kirsher [this message]
2016-05-06  7:03 ` [net-next 10/11] i40e: fix misleading indentation Jeff Kirsher
2016-05-06  7:03 ` [net-next 11/11] i40e: constify i40e_client_ops structure Jeff Kirsher
2016-05-06 19:55 ` [net-next 00/11][pull request] 40GbE Intel Wired LAN Driver Updates 2016-05-05 David Miller

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=1462518228-30527-10-git-send-email-jeffrey.t.kirsher@intel.com \
    --to=jeffrey.t.kirsher@intel.com \
    --cc=davem@davemloft.net \
    --cc=jesse.brandeburg@intel.com \
    --cc=jogreene@redhat.com \
    --cc=netdev@vger.kernel.org \
    --cc=nhorman@redhat.com \
    --cc=sassmann@redhat.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).