DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Guo, Jia" <jia.guo@intel.com>
To: "Yu, DapengX" <dapengx.yu@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>,
	"Yu, DapengX" <dapengx.yu@intel.com>,
	"Zhang, Qi Z" <qi.z.zhang@intel.com>
Subject: Re: [dpdk-dev] [PATCH] net/ixgbe: fix fdirctrl register setting
Date: Tue, 15 Dec 2020 04:36:45 +0000	[thread overview]
Message-ID: <07e98f22a664460c9d3180119d034e28@intel.com> (raw)
In-Reply-To: <20201211013506.49885-1-dapengx.yu@intel.com>

Hi, dapeng

> -----Original Message-----
> From: dapengx.yu@intel.com <dapengx.yu@intel.com>
> Sent: Friday, December 11, 2020 9:35 AM
> To: Guo, Jia <jia.guo@intel.com>
> Cc: dev@dpdk.org; Yu, DapengX <dapengx.yu@intel.com>; Zhang, Qi Z
> <qi.z.zhang@intel.com>
> Subject: [PATCH] net/ixgbe: fix fdirctrl register setting
> 
> From: YU DAPENG <dapengx.yu@intel.com>
> 
> The function ixgbe_fdir_set_flexbytes_offset is used when create FDir rule
> for flexbytes. It set a register: FDIRCTRL.FLEX_OFFSET, which cause that even
> if the FDir flexbytes rule is destroyed, the rule still direct the packet and
> transfer it to the wrong place. It is because Setting FDIRCTRL shall only be
> permitted on Flow Director initialization flow or Clearing the Flow Director
> table, otherwise unexpected happens. In order to evade the limit, add code
> to make setting FDIRCTRL work without unexpected effects.
> 
> Fixes: f35fec63dde1 ("net/ixgbe: enable flex bytes for generic flow API")
> Cc: qi.z.zhang@intel.com
> 
> Signed-off-by: YU DAPENG <dapengx.yu@intel.com>
> ---
>  drivers/net/ixgbe/ixgbe_fdir.c | 23 +++++++++++++++++++++++
> drivers/net/ixgbe/ixgbe_flow.c |  7 +++----
>  2 files changed, 26 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/net/ixgbe/ixgbe_fdir.c b/drivers/net/ixgbe/ixgbe_fdir.c
> index a0fab5070..56dddd56b 100644
> --- a/drivers/net/ixgbe/ixgbe_fdir.c
> +++ b/drivers/net/ixgbe/ixgbe_fdir.c
> @@ -503,9 +503,32 @@ ixgbe_fdir_set_flexbytes_offset(struct rte_eth_dev
> *dev,
>  				uint16_t offset)
>  {
>  	struct ixgbe_hw *hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data-
> >dev_private);
> +	struct ixgbe_hw_fdir_info *fdir_info =
> +		IXGBE_DEV_PRIVATE_TO_FDIR_INFO(dev->data-
> >dev_private);
>  	uint32_t fdirctrl;
>  	int i;
> 
> +	if (fdir_info->flex_bytes_offset == offset)
> +		return 0;
> +
> +	fdir_info->flex_bytes_offset = offset;

Should this value assign at the end of the polling checking?
And could you kindly help to check is the timeout checking (i >= IXGBE_FDIR_INIT_DONE_POLL)
also need in the "ixgbe_fdir_set_flexbytes_offset"?

> +
> +	/*

/* -> /**

> +	 * 82599 adapters flow director init flow cannot be restarted,
> +	 * Workaround 82599 silicon errata by performing the following steps
> +	 * before re-writing the FDIRCTRL control register with the same
> value.
> +	 * - write 1 to bit 8 of FDIRCMD register &
> +	 * - write 0 to bit 8 of FDIRCMD register
> +	 */
> +	IXGBE_WRITE_REG(hw, IXGBE_FDIRCMD,
> +			(IXGBE_READ_REG(hw, IXGBE_FDIRCMD) |
> +			 IXGBE_FDIRCMD_CLEARHT));
> +	IXGBE_WRITE_FLUSH(hw);
> +	IXGBE_WRITE_REG(hw, IXGBE_FDIRCMD,
> +			(IXGBE_READ_REG(hw, IXGBE_FDIRCMD) &
> +			 ~IXGBE_FDIRCMD_CLEARHT));
> +	IXGBE_WRITE_FLUSH(hw);
> +

So you mean the above steps should be done before re-writing the FDIRCTRL, just like
the Flow Director initialization flow or Clearing the Flow Director, right?
If so, please refine your commit log and make it more clear, since Setting FDIRCTRL shall is not only be
permitted on Flow Director initialization flow or Clearing the Flow Director table.

>  	fdirctrl = IXGBE_READ_REG(hw, IXGBE_FDIRCTRL);
> 
>  	fdirctrl &= ~IXGBE_FDIRCTRL_FLEX_MASK; diff --git
> a/drivers/net/ixgbe/ixgbe_flow.c b/drivers/net/ixgbe/ixgbe_flow.c index
> 39f6ed73f..b37541d9b 100644
> --- a/drivers/net/ixgbe/ixgbe_flow.c
> +++ b/drivers/net/ixgbe/ixgbe_flow.c
> @@ -3137,8 +3137,6 @@ ixgbe_flow_create(struct rte_eth_dev *dev,
>  				rte_memcpy(&fdir_info->mask,
>  					&fdir_rule.mask,
>  					sizeof(struct ixgbe_hw_fdir_mask));
> -				fdir_info->flex_bytes_offset =
> -					fdir_rule.flex_bytes_offset;
> 
>  				if (fdir_rule.mask.flex_bytes_mask)
>  					ixgbe_fdir_set_flexbytes_offset(dev,
> @@ -3161,8 +3159,9 @@ ixgbe_flow_create(struct rte_eth_dev *dev,
>  				if (ret)
>  					goto out;
> 
> -				if (fdir_info->flex_bytes_offset !=
> -						fdir_rule.flex_bytes_offset)
> +				if (fdir_rule.mask.flex_bytes_mask &&
> +					(fdir_info->flex_bytes_offset !=

Suggest you check the line align. 

> +						fdir_rule.flex_bytes_offset))

This line is the same as above.

>  					goto out;
>  			}
>  		}
> --
> 2.26.2.windows.1


  parent reply	other threads:[~2020-12-15  4:36 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-11  1:35 dapengx.yu
2020-12-15  3:00 ` Zhou, JunX W
2020-12-15  4:36 ` Guo, Jia [this message]
2020-12-16  0:55   ` Yu, DapengX
2020-12-15 10:10 ` [dpdk-dev] [PATCH v2] " dapengx.yu
2020-12-22  6:51   ` Zhou, JunX W
2020-12-22  7:23     ` Guo, Jia
2021-01-05 12:09       ` [dpdk-dev] [dpdk-stable] " Ferruh Yigit

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=07e98f22a664460c9d3180119d034e28@intel.com \
    --to=jia.guo@intel.com \
    --cc=dapengx.yu@intel.com \
    --cc=dev@dpdk.org \
    --cc=qi.z.zhang@intel.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).