be shared across drivers as part of ethtool
kernel work.
Signed-off-by: Jacob Keller
---
Sorry aboutthe delay, our iwl queue maintainer Jeff is on vacation, and
discussion about how best to resolve this issue was/is ongoing in the
IWL list.
I opted to just move the contents of these
ter dereference, since both the for-loop and
memset in fq_codel_reset will be no-ops when flow_cnt is zero.
Signed-off-by: Jacob Keller
---
net/sched/sch_fq_codel.c | 25 ++---
1 file changed, 18 insertions(+), 7 deletions(-)
diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_f
ter dereference, since both the for-loop and
memset in fq_codel_reset will be no-ops when flow_cnt is zero.
Signed-off-by: Jacob Keller
---
net/sched/sch_fq_codel.c | 25 ++---
1 file changed, 18 insertions(+), 7 deletions(-)
diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_f
On Mon, Apr 2, 2018 at 7:05 AM, Bjorn Helgaas wrote:
> +/* PCIe speed to Mb/s reduced by encoding overhead */
> +#define PCIE_SPEED2MBS_ENC(speed) \
> + ((speed) == PCIE_SPEED_16_0GT ? (16000*(128/130)) : \
> +(speed) == PCIE_SPEED_8_0GT ? (8000*(128/130)) : \
mum number of xmit_more skbs to send in a sequence. This ensures
that all drivers benefit, and allows system administrators the option to
tune the value to their environment.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
Stray thoughts and further questions
Is this the rig
uot;i40e/i40evf: fix interrupt affinity bug", 2016-09-14)
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
Cc: sta...@vger.kernel.org # 4.10+
---
This updates the commit message for the original fix, and indicates that
it fixes a potential crash, as well as tagged the commit for stabl
ing.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
I found this as a result of debugging a race condition in the i40evf
driver, in which we assumed that netif_running() would not be true until
after dev->open() had been called and succeeded. Unfortunately we can't
hold t
.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
Cc: Stephen Rothwell <s...@canb.auug.org.au>
Cc: Bimmy Pujari <bimmy.puj...@intel.com>
---
Not sure if this was already in Jeff's queue, but since it's an obvious
fix for the issue found by Stephen, I thought I'd send it out now
to manually parse the value.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
rxclass.c | 16
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/rxclass.c b/rxclass.c
index c7bfebaf6e22..7f0e765d3b47 100644
--- a/rxclass.c
+++ b/rxclass.c
@@ -247,11 +
e to implement, and would be
a much more massive project.
Jacob Keller (2):
rxclass: use ethtool_get_flow_spec_ring to display queue and VF
ethtool: add support to specify ring_cookie as VF/queue
ethtool.8.in | 3 ++-
rxclass.c| 74 +++-
. This allows the user to more easily specify the matching
notation as it is understood by the kernel and some drivers. Since no
driver exists today with over 2,147,483,647 queues, this notation will
simply fail on drivers which don't recognize the new partitioning.
Signed-off-by: Jacob Keller <jacob.e.
-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
include/uapi/linux/ethtool.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index ed650eef9e54..a716112b2b01 100644
--- a/include/uapi/linux/ethtool.h
about, and then have ethtool core do the check for us to discard frames.
I haven't implemented this quite yet.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 4 ++--
drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool
about, and then have ethtool core do the check for us to discard frames.
I haven't implemented this quite yet.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 4 ++--
drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool
about, and then have ethtool core do the check for us to discard frames.
I haven't implemented this quite yet.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
I plan on using this helper when fixing the mask code for ntuple filters
in the Intel i40e driver. I wanted to see w
the expected behavior of userspace.
This change shouldn't break current or previous userspace as they have
consistently included the length correctly since as far back as we can
check.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
Reported-by: Alexander Duyck <alexander.du...@
the expected behavior of userspace.
This change shouldn't break current or previous userspace as they have
consistently included the length correctly since as far back as we can
check.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
Reported-by: Alexander Duyck <alexander.du...@
Add a sanity check to ensure that all requested channel sizes are within
bounds, which should reduce errors in driver implementation.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
net/core/ethtool.c | 15 +--
1 file changed, 13 insertions(+), 2 deletions(-)
diff
in this case should now no longer
occur, add a dev_err message to indicate the user that we did so.
Jacob Keller (4):
ethtool: correctly ensure {GS}CHANNELS doesn't conflict with GS{RXFH}
ethtool: ensure channel counts are within bounds during SCHANNELS
ethtool: can't set combined and tx/rx channel
Also print an error message incase we do have to reconfigure as this
should no longer happen anymore due to ethtool changes. If it somehow
does occur, user should be made aware of it.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
drivers/net/ethernet/intel/fm10k/fm10k_main.
a sanity check to prevent such requests. I am not 100%
sure if this is correct as I was not able to find documentation and
I think it currently depended on driver implementation.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
net/core/ethtool.c | 5 +
1 file changed, 5 inse
Rx flow table is
not going to assign flows to now disabled channels.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
include/linux/netdevice.h | 8 +++
net/core/ethtool.c| 59 +++
2 files changed, 67 insertions(+)
diff
Also print an error message incase we do have to reconfigure as this
should no longer happen anymore due to ethtool changes. If it somehow
does occur, user should be made aware of it.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
drivers/net/ethernet/intel/fm10k/fm10k_main.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
ethtool.8.in | 5 -
ethtool.c| 46 --
2 files changed, 36 insertions(+), 15 deletions(-)
diff --git a/ethtool.8.in b/ethtool.8.in
index eeffa70415b5..b7d56fbe4484 100644
--- a/e
separating combined vs asymmetric channels
* verified behavior using fm10k driver
Jacob Keller (3):
ethtool: correctly ensure {GS}CHANNELS doesn't conflict with GS{RXFH}
ethtool: ensure channel counts are within bounds during SCHANNELS
fm10k: don't reinitialize RSS flow table when RXFH
Add a sanity check to ensure that all requested channel sizes are within
bounds, which should reduce errors in driver implementation.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
net/core/ethtool.c | 13 +++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff
Rx flow table is
not going to assign flows to now disabled channels.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
include/linux/netdevice.h | 8 +++
net/core/ethtool.c| 55 +++
2 files changed, 63 insertions(+)
diff
if it would interfere
with the current redirection table.
Jacob Keller (2):
ethtool: support notifying drivers when user requests default rxfh
table
fm10k: correctly report error when changing number of channels
drivers/net/ethernet/intel/fm10k/fm10k.h | 2 ++
drivers/net/ethernet/intel
and go about
business.
If we don't have a way to properly indicate that we've reset to default
then we are not able to implement the proposed behavior, so this patch
adds a new method to properly indicate that we have reset to the default
indirection table.
Signed-off-by: Jacob Keller
on the reta table. This check is only enforced if the
user has manually configured the RSS indirection table, since the
driver should be free to repopulate its own default configuration.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
drivers/net/ethernet/intel/fm10k/f
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
ethtool-copy.h | 8 +++-
ethtool.c | 36 +++-
2 files changed, 42 insertions(+), 2 deletions(-)
diff --git a/ethtool-copy.h b/ethtool-copy.h
index d23ffc4c38b4..620dcea06d25 100644
--- a/e
may or may not apply the current settings and is free
to change the values as necessary. The current default is set to false,
as this is how most drivers appear to behave today.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
Cc: Lendacky, Thomas <thomas.lenda...@amd.com>
. If this is not
possible, we will indicate that the mode is now dynamic and reset to
driver defaults.
Signed-off-by: Jacob Keller <jacob.e.kel...@intel.com>
---
drivers/net/ethernet/intel/fm10k/fm10k.h | 1 +
drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c | 10 +-
drivers/net/ethernet/intel
driver was especially problematic, and
the current patch series attempts to resolve this.
It is possible to have most of the behavior handled purely in driver,
but then we lose any ability to communicate this to the user via
ethtool.
Jacob Keller (3):
ethtool: add dynamic flag to ETHTOOL_{GS}RXFH
34 matches
Mail list logo