From: Alexander Duyck <alexander.h.duyck@linux.intel.com>
To: Jeff Kirsher <jeffrey.t.kirsher@intel.com>, davem@davemloft.net
Cc: netdev@vger.kernel.org, nhorman@redhat.com, sassmann@redhat.com,
Gregg Leventhal <gleventhal@janestreet.com>,
Andrew Bowers <andrewx.bowers@intel.com>
Subject: Re: [net-next 11/14] ixgbe: Prevent u8 wrapping of ITR value to something less than 10us
Date: Tue, 10 Sep 2019 15:19:13 -0700 [thread overview]
Message-ID: <91551885b0731a9e6121a3afe2c9a6cc912f61d5.camel@linux.intel.com> (raw)
In-Reply-To: <20190910163434.2449-12-jeffrey.t.kirsher@intel.com>
On Tue, 2019-09-10 at 09:34 -0700, Jeff Kirsher wrote:
> From: Alexander Duyck <alexander.h.duyck@linux.intel.com>
>
> There were a couple cases where the ITR value generated via the adaptive
> ITR scheme could exceed 126. This resulted in the value becoming either 0
> or something less than 10. Switching back and forth between a value less
> than 10 and a value greater than 10 can cause issues as certain hardware
> features such as RSC to not function well when the ITR value has dropped
> that low.
One quick thing we can add on here is:
Fixes: b4ded8327fea ("ixgbe: Update adaptive ITR algorithm")
This is likely something that we may want to backport to stable.
> Reported-by: Gregg Leventhal <gleventhal@janestreet.com>
> Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
> Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> ---
> drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
> index dc034f4e8cf6..a5398b691aa8 100644
> --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
> +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
> @@ -2623,7 +2623,7 @@ static void ixgbe_update_itr(struct ixgbe_q_vector *q_vector,
> /* 16K ints/sec to 9.2K ints/sec */
> avg_wire_size *= 15;
> avg_wire_size += 11452;
> - } else if (avg_wire_size <= 1980) {
> + } else if (avg_wire_size < 1968) {
> /* 9.2K ints/sec to 8K ints/sec */
> avg_wire_size *= 5;
> avg_wire_size += 22420;
> @@ -2656,6 +2656,8 @@ static void ixgbe_update_itr(struct ixgbe_q_vector *q_vector,
> case IXGBE_LINK_SPEED_2_5GB_FULL:
> case IXGBE_LINK_SPEED_1GB_FULL:
> case IXGBE_LINK_SPEED_10_FULL:
> + if (avg_wire_size > 8064)
> + avg_wire_size = 8064;
> itr += DIV_ROUND_UP(avg_wire_size,
> IXGBE_ITR_ADAPTIVE_MIN_INC * 64) *
> IXGBE_ITR_ADAPTIVE_MIN_INC;
next prev parent reply other threads:[~2019-09-10 22:19 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-10 16:34 [net-next 00/14][pull request] Intel Wired LAN Driver Updates 2019-09-10 Jeff Kirsher
2019-09-10 16:34 ` [net-next 01/14] ixgbe: fix memory leaks Jeff Kirsher
2019-09-10 16:34 ` [net-next 02/14] i40e: check __I40E_VF_DISABLE bit in i40e_sync_filters_subtask Jeff Kirsher
2019-09-10 16:34 ` [net-next 03/14] ixgbe: use skb_get_queue_mapping in tx path Jeff Kirsher
2019-09-10 16:34 ` [net-next 04/14] i40e: use ktime_get_real_ts64 instead of ktime_to_timespec64 Jeff Kirsher
2019-09-10 16:34 ` [net-next 05/14] i40e: remove I40E_AQC_ADD_CLOUD_FILTER_OIP Jeff Kirsher
2019-09-10 16:34 ` [net-next 06/14] i40e: mark additional missing bits as reserved Jeff Kirsher
2019-09-10 16:34 ` [net-next 07/14] i40e: fix missed "Negotiated" string in i40e_print_link_message() Jeff Kirsher
2019-09-10 16:34 ` [net-next 08/14] i40e: Fix message for other card without FEC Jeff Kirsher
2019-09-10 16:34 ` [net-next 09/14] i40e: use BIT macro to specify the cloud filter field flags Jeff Kirsher
2019-09-10 16:34 ` [net-next 10/14] i40e: clear __I40E_VIRTCHNL_OP_PENDING on invalid min Tx rate Jeff Kirsher
2019-09-10 16:34 ` [net-next 11/14] ixgbe: Prevent u8 wrapping of ITR value to something less than 10us Jeff Kirsher
2019-09-10 22:19 ` Alexander Duyck [this message]
2019-09-11 8:24 ` David Miller
2019-09-10 16:34 ` [net-next 12/14] iavf: fix MAC address setting for VFs when filter is rejected Jeff Kirsher
2019-09-10 16:34 ` [net-next 13/14] net/ixgbevf: make array api static const, makes object smaller Jeff Kirsher
2019-09-10 16:34 ` [net-next 14/14] i40e: fix potential RX buffer starvation for AF_XDP Jeff Kirsher
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=91551885b0731a9e6121a3afe2c9a6cc912f61d5.camel@linux.intel.com \
--to=alexander.h.duyck@linux.intel.com \
--cc=andrewx.bowers@intel.com \
--cc=davem@davemloft.net \
--cc=gleventhal@janestreet.com \
--cc=jeffrey.t.kirsher@intel.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