Skip to content

Commit

Permalink
i40e: use priv flags to control packet split
Browse files Browse the repository at this point in the history
Ethtool priv flags implementation to enable or disable packet split, which
is a hardware feature that inspects headers and will put headers in a
separate DMA buffer from the payload data.  The driver was automatically
choosing to enable packet split in some cases and this gives the user the
ability to turn it off/on explicitly.

to query state:
ethtool --show-priv-flags ethx

to enable:
ethtool --set-priv-flags ethx packet-split on
to disable:
ethtool --set-priv-flags ethx packet-split off

Why would anyone want this?
	Because some environments benefit from header/data split in the receive
	buffer, and the driver defaults to one or the other depending on
	environment/kernel parameters.

Why didn't you implement a generic ethtool control for this feature?
	Because Intel hardware is the only hardware that supports header/data
	split.

Change-ID: I803121e1eecc9ccb2884031fd85dd1110b3af66d
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>
  • Loading branch information
Jesse Brandeburg authored and Jeff Kirsher committed Dec 3, 2015
1 parent d82acb3 commit 827de39
Show file tree
Hide file tree
Showing 2 changed files with 28 additions and 0 deletions.
1 change: 1 addition & 0 deletions drivers/net/ethernet/intel/i40e/i40e.h
Original file line number Diff line number Diff line change
Expand Up @@ -104,6 +104,7 @@
#define I40E_PRIV_FLAGS_LINKPOLL_FLAG BIT(1)
#define I40E_PRIV_FLAGS_FD_ATR BIT(2)
#define I40E_PRIV_FLAGS_VEB_STATS BIT(3)
#define I40E_PRIV_FLAGS_PS BIT(4)

#define I40E_NVM_VERSION_LO_SHIFT 0
#define I40E_NVM_VERSION_LO_MASK (0xff << I40E_NVM_VERSION_LO_SHIFT)
Expand Down
27 changes: 27 additions & 0 deletions drivers/net/ethernet/intel/i40e/i40e_ethtool.c
Original file line number Diff line number Diff line change
Expand Up @@ -231,6 +231,7 @@ static const char i40e_priv_flags_strings[][ETH_GSTRING_LEN] = {
"LinkPolling",
"flow-director-atr",
"veb-stats",
"packet-split",
};

#define I40E_PRIV_FLAGS_STR_LEN ARRAY_SIZE(i40e_priv_flags_strings)
Expand Down Expand Up @@ -2709,6 +2710,8 @@ static u32 i40e_get_priv_flags(struct net_device *dev)
I40E_PRIV_FLAGS_FD_ATR : 0;
ret_flags |= pf->flags & I40E_FLAG_VEB_STATS_ENABLED ?
I40E_PRIV_FLAGS_VEB_STATS : 0;
ret_flags |= pf->flags & I40E_FLAG_RX_PS_ENABLED ?
I40E_PRIV_FLAGS_PS : 0;

return ret_flags;
}
Expand All @@ -2723,6 +2726,26 @@ static int i40e_set_priv_flags(struct net_device *dev, u32 flags)
struct i40e_netdev_priv *np = netdev_priv(dev);
struct i40e_vsi *vsi = np->vsi;
struct i40e_pf *pf = vsi->back;
bool reset_required = false;

/* NOTE: MFP is not settable */

/* allow the user to control the method of receive
* buffer DMA, whether the packet is split at header
* boundaries into two separate buffers. In some cases
* one routine or the other will perform better.
*/
if ((flags & I40E_PRIV_FLAGS_PS) &&
!(pf->flags & I40E_FLAG_RX_PS_ENABLED)) {
pf->flags |= I40E_FLAG_RX_PS_ENABLED;
pf->flags &= ~I40E_FLAG_RX_1BUF_ENABLED;
reset_required = true;
} else if (!(flags & I40E_PRIV_FLAGS_PS) &&
(pf->flags & I40E_FLAG_RX_PS_ENABLED)) {
pf->flags &= ~I40E_FLAG_RX_PS_ENABLED;
pf->flags |= I40E_FLAG_RX_1BUF_ENABLED;
reset_required = true;
}

if (flags & I40E_PRIV_FLAGS_LINKPOLL_FLAG)
pf->flags |= I40E_FLAG_LINK_POLLING_ENABLED;
Expand All @@ -2745,6 +2768,10 @@ static int i40e_set_priv_flags(struct net_device *dev, u32 flags)
else
pf->flags &= ~I40E_FLAG_VEB_STATS_ENABLED;

/* if needed, issue reset to cause things to take effect */
if (reset_required)
i40e_do_reset(pf, BIT(__I40E_PF_RESET_REQUESTED));

return 0;
}

Expand Down

0 comments on commit 827de39

Please sign in to comment.