Hi Oliver,
Looking at the kernel code pci_mmap_resource()--->pci_mmap_fits(), it looks
like pci_mmap_fits() expects that the "number of pages" requested to map is
no larger than the actual size of the pci resource. But in
pci_uio_ioport_map() what is passed as the requested size is the end_addr
of
Hi,
We have a SRIOV+ DPDK 16.04 setup for a VM using KVM-qemu hypervisor. We
are unable to ping the VM. We are assigning the VM to a SRIOV vf and for
the DPDK enabled NIC on the VM to detect this VF, we are using KNI.
We used the KNI app available in the examples folder. Although the non
D
For some cases when reading from device are incorrect or image is
incorrect, this part of code causes crash due to division by zero.
Fixes: 8db9e2a1b232 ("i40e: base driver")
Signed-off-by: Jingjing Wu
CC: sta...@dpdk.org
---
drivers/net/i40e/base/i40e_common.c | 6 --
1 file changed, 4 ins
Big Endian platform will accidentally send the wrong
data to the firmware command. This patch fixes the issue.
Fixes: 788fc17b2dec ("i40e/base: support proxy config for X722")
Fixes: 3c89193a36fd ("i40e/base: support WOL config for X722")
Signed-off-by: Jingjing Wu
CC: sta...@dpdk.org
---
drive
remove X722_SUPPORT and I40E_NDIS_SUPPORT MACROs
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_adminq_cmd.h | 14 ---
drivers/net/i40e/base/i40e_common.c | 18 +-
drivers/net/i40e/base/i40e_devids.h | 2 --
drivers/net/i40e/base/i40e_nvm.c| 16 ---
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/Makefile | 2 +-
drivers/net/i40e/i40e_ethdev.c | 40
2 files changed, 1 insertion(+), 41 deletions(-)
diff --git a/drivers/net/i40e/Makefile b/drivers/net/i40e/Makefile
index 13085fb..66997b6 100644
-
We already define I40E_AQ_PHY_TYPE_EXT_25G* flags in the response adminq
structure above, and do not need to re-define these.
While we are here, replace 0X with 0x as normal style.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_adminq_cmd.h | 8 ++--
1 file changed, 2 insertions(
This patch adds I40E_NVMUPD_STATE_ERROR state for NVM update.
Without this patch driver has no possibility to return NVM image write
failure.This state is being set when ARQ rises error.
arq_last_status is also updated every time when ARQ event comes,
not only on error cases.
Signed-off-by: Jingji
Add a new adminq function that allows driver to configure per-VLAN
broadcast promiscuous mode, similar to how we handle unicast and
multicast promiscuous modes.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_common.c| 34 ++
drivers/net/i40e/base/i4
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_type.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/i40e/base/i40e_type.h
b/drivers/net/i40e/base/i40e_type.h
index 99e080e..3784c8f 100644
--- a/drivers/net/i40e/base/i40e_type.h
+++ b/drivers/net
The firmware expects the Port number to be in Little Endian format, and
the i40e_aq_add_udp_tunnel command clearly expects the udp_port variable
to be in Host order, as it uses CPU_TO_LE16(). It was recently
discovered in the Linux driver that we were passing a Big Endian port
number, which was the
By default the device clears all MAC filter information on PF Reset.
However, this will cause Wake-On-LAN to fail because the wake filters
are deleted on transition to D3 power state. To get around this,
firmware is adding functionality to preserve certain MAC filters during
PFR. These bits allow t
This patch implements a function to set a VSI to broadcast, multicast, and
unicast promiscuous mode all at once. This is specifically needed to set
the WoL/Proxy VSI created by FW to full promiscuous mode during power down
for WoL patterns and protocol offloads to function properly.
Signed-off-by:
Store the FEC status bits from the link up event into the
hw_link_info structure.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_common.c | 2 ++
drivers/net/i40e/base/i40e_type.h | 1 +
2 files changed, 3 insertions(+)
diff --git a/drivers/net/i40e/base/i40e_common.c
b/drivers/ne
Acquire NVM lock before reads on all devices. Previously, locks were
only used for X722 and later. Fixes an issue where simultaneous X710
NVM accesses were interfering with each other.
Fixes: 8db9e2a1b232 ("i40e: base driver")
Signed-off-by: Jingjing Wu
CC: sta...@dpdk.org
---
drivers/net/i40
This patch implements the clear Wake on LAN (WoL) filters admin queue
function which clears out ALL WoL patterns programmed into
the flex filters.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_adminq_cmd.h | 1 +
drivers/net/i40e/base/i40e_common.c | 20
dri
Define the values for the 25G PHY type bit-fields that match
reported values from firmware. There was a gap in the bit
fields but no corresponding gap i40e_aq_phy_type enum.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_type.h | 20
1 file changed, 16 insertions(
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_adminq_cmd.h | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/net/i40e/base/i40e_adminq_cmd.h
b/drivers/net/i40e/base/i40e_adminq_cmd.h
index 1884758..cef02b1 100644
--- a/drivers/net/i40e/base/i40e_a
Add FEC bits to the PHY capabilities AQ command struct. This is required
for 25GbE support. Change the name of the generic mod_type_ext field to
indicate that it is now used for handling FEC.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_adminq_cmd.h | 13 -
drivers/net/i
Microsoft recently removed the requirement for VFs to use the VMBus.
The Fort Park Windows VF has been changed to use only the hardware
mailbox, so the Hyper-V VF device ID can be removed.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_common.c | 1 -
drivers/net/i40e/base/i40e_devids
The PHY type value for unrecognized PHYs and cables was changed
based on firmware version number. Newer hardware use lower firmware
version numbers and this was causing some PHYs to be identified
as type 0x16 instead of 0xe (unknown).
Without this patch, newer card will incorrectly identify unknow
I40E_MAC_X710 was supposed to be for 10G and I40E_MAC_XL710
was supposed to be for 40G. But i40e_set_mac_type() sets
I40E_MAC_XL710 for all device IDS. I40E_MAC_X710 is not
used at all. Thus deprecating this extra macro.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_type.h | 1 -
1 f
Add logical_id to I40E_AQ_CAP_ID_MNG_MODE capability starting from major
version 2.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_common.c | 8
drivers/net/i40e/base/i40e_type.h | 4
2 files changed, 12 insertions(+)
diff --git a/drivers/net/i40e/base/i40e_common.c
To align with current memcpy use, replace existing legacy memcpy() calls
with i40e_memcpy() calls.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_common.c | 13 -
drivers/net/i40e/base/i40e_nvm.c| 7 ---
2 files changed, 12 insertions(+), 8 deletions(-)
diff --gi
Currently i40e_bus_info has PCI device and function info only. However
in log messages slot number (i.e hw->bus.device) is being printed
as bus number. Another field should be added to provide bus number
info and preserve existing information.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_common.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/net/i40e/base/i40e_common.c
b/drivers/net/i40e/base/i40e_common.c
index 2ad9448..7eea189 100644
--- a/drivers/net/i40e/base/i40e_common.c
+++ b/drivers/net/i40e/base
Group together the minimum set of offload capabilities that are always
supported by VF in base mode. This define would be used by PF to make
sure VF in base mode gets minimum of base capabilities.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_virtchnl.h | 4
1 file changed, 4 in
Some external PHYs require Clause22 and Clause45 method for
accessing registers. Mostly used for X722 support.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_common.c| 245 +++--
drivers/net/i40e/base/i40e_prototype.h | 16 ++-
drivers/net/i40e/base/i4
This patch fixes a problem where it could take a very
long time (>100 msec) to print the link down notification.
This problem is fixed by changing how often we update link
info from fw, when link is down. Without this patch, it can
take over 100msec to notify user link is down.
Fixes: e6691b428eb1
Add phy_type_ext copied from old setting to rpevents 25G PHY
types from being disabled when setting the flow control modes.
Fixes: 51131ae119 ("net/i40e/base: get PHY abilities for 25G")
Signed-off-by: Jingjing Wu
CC: sta...@dpdk.org
---
drivers/net/i40e/base/i40e_common.c | 1 +
1 file changed
Incorrect bit mask was used for testing "get link status" response.
Instead of I40E_AQ_LSE_ENABLE (which is actually 0x03) it should
be I40E_AQ_LSE_IS_ENABLED (which is defined as 0x01).
Fixes: 8db9e2a1b232 ("i40e: base driver")
Signed-off-by: Jingjing Wu
CC: sta...@dpdk.org
---
drivers/net/i40
Add ENCAP_CSUM offload negotiation flag. Currently VF assumes checksum
offload for encapsulated packets is supported by default. Going forward,
this feature needs to be negotiated with PF before advertising to the
stack. Hence, we need a flag to control it.
Signed-off-by: Jingjing Wu
---
drivers
This patch changes some assignments and removing the unnecessary
code to avoid error reported by static analysis tools.
Signed-off-by: Jingjing Wu
---
drivers/net/i40e/base/i40e_common.c | 4
drivers/net/i40e/base/i40e_lan_hmc.c | 5 -
2 files changed, 9 deletions(-)
diff --git a/driv
i40e base code upate. The main changes are:
- add clause22 and clause45 implementation for PHY registers accessing
- replace existing legacy memcpy() calls with i40e_memcpy() calls.
- use BIT() macro instead of bit fields
- add clear all WoL filters implementation
- add ERROR state for NVM upd
34 matches
Mail list logo
