Issue:
1) dev_queue_xmit acquired the per queue tx lock and called driver
transmit routine.
2) During the transmit - common interrupt asserted which tries to
acquire the same tx lock which resulted in lock recursion.
Fix: Move the lock acquire to tasklet context under common isr.
Bug 3773016
Change-Id: I7cfd49beb1238286d3bccd9e4b9ccc054c4f6d30
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2770227
Reviewed-by: Narayan Reddy <narayanr@nvidia.com>
Reviewed-by: Bitan Biswas <bbiswas@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Issue: In the current scenrio, when the remote
interface is brought down and up. There is no way
of getting the link up again with the remote device
which in turn fails the data transfers even though
interface from other end is brought up.
Fix: When interface is gone down on the other side,
MAC receives local faults and in this scenario
no data should be sent to MAC, so disable network
queues and initiate the lane bring up process for
monitoring the link status. Once the link is up,
re enable the network queues for data transfers.
Also added support for enabling driver logs using
msglvl of ethtool
Bug 3744088
Bug 3654543
Bug 3665378
Change-Id: I16cdee74e4e3ff6cd176924575f5005f385d4c5d
Signed-off-by: Narayan Reddy <narayanr@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2730876
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com>
Reviewed-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
GVS: Gerrit_Virtual_Submit
Issue: If WOL is enabled after running "./ethtool -s eth0 wol g"
command there will not be any eapol frames transaction, hence macsec
will be disabled. As macsec is disabled before suspend we are not
enabling it back once resumed. Because of this eapol frames are also
sent with sec-tag as byp_lut is not programmed.
Fix: Instead of depending on the macsec state, check if there is any
supplicant running, if there is any supplicant then enable the macsec
post resume.
Bug 3577131
Change-Id: I6018b94411034eebba34f5f0a9520907305c573e
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2694446
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
GVS: Gerrit_Virtual_Submit
Issue: Observed Tx timestamp timeout from application since
workqueue is not getting scheduled at the driver layer to consume
the Tx timestamp from the OSI layer which resulted in tx_timeout
from APP layer.
Fix: Invoke the timestamp consume function in tx done instead
of scheduling through workqueue. Schedule the work queue only
if the timestamp is not found.
Bug 3658701
Change-Id: If97aebc1980f685e3cc34f135ace91298ee78657
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2732444
Reviewed-by: Troy Kong <troyk@nvidia.com>
Reviewed-by: Narayan Reddy <narayanr@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com>
Reviewed-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-by: Bitan Biswas <bbiswas@nvidia.com>
Tested-by: Troy Kong <troyk@nvidia.com>
GVS: Gerrit_Virtual_Submit
Issue: Name of the nl_family is nv_macsec for all the interfaces
hence even if the macsec is enabled for one interface and disabled for
other interfaces, supplicant can be launched for all the interfaces.
Fix: Use the interface name for the nl_family name hence supplicant can
check if macsec is enabled in DT before initializing supplicant for that
particular interface
Bug 3673434
Change-Id: I33d62a0ab2b74f7f53abcd8e6629c1b013eede86
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2726483
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Issue: While ethernet going into suspend all registers
of controllers saved through save_registers IOCTL and
same will be restored during resume. Register restoring
without following sequence will lead to multiple issues.
Fix: For every dynamic configuration save the input
parameters and use the same parameters through API's
to restore the controller configuration. API approach
will follow the specific sequence for programming the
controller registers.
Bug 3665476
Change-Id: Ib321937faa07eeb086e00ebfce8d0d0b8aeb1f55
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2663393
Reviewed-by: Narayan Reddy <narayanr@nvidia.com>
Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com>
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
GVS: Gerrit_Virtual_Submit
Issue:
OSD Driver maintains a local MC address list to avoid the programming
old addresses again into HW when new MC address list passed from the
network stack.
Its observed MC address index for a specific address is different once a
MC address deleted from the list. Because of this new MC address which
is passed in new list is overwriting the old MC address.
Fix:
Program the entire list every time when network stack passes it instead
of maintaining local list at driver.
Bug 3609583
Change-Id: I97cb3ea4d35641ad156d16c54c89e8593ad64459
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2702851
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Narayan Reddy <narayanr@nvidia.com>
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com>
GVS: Gerrit_Virtual_Submit
1) Add threaded interrupt handler function to call EPL API
to report error.
2) add hsi_enable sysfs interface to enable HSI feature at runtime
to enable:
echo enable > /sys/devices/platform/./nvethernet/hsi_enable
to disable:
echo disable > /sys/devices/platform/./nvethernet/hsi_enable
3) For mac RX CRC and checksum error, error reporting frequency is
every one second for HSI_ERROR_THRESHOLD error count
4) Added DT property “nvidia,hsi_err_time_threshold” to specify time
threshold for error reporting
5) Added DT property "nvidia,hsi_err_count_threshold" to specify
error count threshold for error reporting
Bug 3543410
Change-Id: I73ef1a674a846295e02deff7e07954e8dada9792
Signed-off-by: Om Prakash Singh <omp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2675040
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com>
Reviewed-by: Narayan Reddy <narayanr@nvidia.com>
Reviewed-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
GVS: Gerrit_Virtual_Submit
Issue: If MTU is increased after Supplicant is initialized
we are not updating the MACSEC MTU so the frames will get
dropped as the MACSEC MTU is lesser than the frames received
Fix: Changes to update the MACSEC MTU along with MAC MTU
Bug 3577143
Change-Id: I8cdd8ce331cbf40139e6f486439113d5b29a0687
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2686289
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Issue: for MGBE with 10 channel enabled TCP tx
is failing with 16k MTU because payload checksum
errors getting incremented due to lesser MTL buffer
available for each dma channel
Fix: Allow MTU more than 9000 for single channel
configuration only
Bug 3536247
Change-Id: I675342cb401723aaa58171f00721786b4d69f57d
Signed-off-by: Om Prakash Singh <omp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2672802
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Issue: Adjust link from the phy subsystem can be called
multiple times either with link down or up. If its called
second time with same link down then code tries to disable
the same clks again which results in clock warinigs.
Fix: Maintain the state of the clk and then disable/enable
the clock based on the state.
Bug 3585098
Change-Id: I7d259e6066c7b67d0a970122eb0841e4b5871b43
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2688290
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Issue: MAC loopback is not getting enabled with below sequence
step1> enable MAC loopback via ioctl call
step2> bring down then bring up interface
step3> re-enable MAC loopback via ioctl call
Fix:
Reset MAC loopback variable if it is enabled, while
interface is going down. So that IOCTL path can
reconfigure the MAC loopback settings in MAC HW.
Bug 3445596
Change-Id: I01ce6551aaded409af02183e166396a7597c3b51
Signed-off-by: Narayan Reddy <narayanr@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2640912
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Issue:
- Workqueue for Tx timestamp is running without
delay, use many CPU cycle at hard isr thread
priority.
- Incorrect return value from ether_handle_tso()
Fix:
- Update code to use delayed workqueue
- Return correct value
Bug 200780891
Bug 3400623
Change-Id: I2095a0634e079bf870ef87cd2de1d35fe24bafd4
Signed-off-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2610986
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Issue:
1) usuage of SUPPORTED_* in PHY driver for setting
the link mode parameters leading data corruption. So
replaced SUPPORTED_* with linkmode_*_bit in PHY driver.
2) Redundant code of setting Pause of PHY
Fix:
1) Change SUPPORTED_* to ETHTOOL_LINK_MODE_*_BIT for
consistent check between PHY and Ethernet driver for K5.10.
2) removed pause setting of PHY, since this is already
taken care in corresponding PHY drivers.
Bug 200781153
Change-Id: Ic6a212dd4560fb3a06a4f60be3bdd33620a38197
Signed-off-by: Narayan Reddy <narayanr@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2607712
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Issue: mdio bus is registered in probe and before exiting
probe ethernet clocks are disabled to save power. When SC7
is initiated then PHY framework invokes a PHY suspend which
triggers the PHY register write. Since clocks are not enabled
PHY register write is ignored and returned failure which
inturn prevented the entire system to enter into SC7
Fix: Move mdio bus registration to open so that PHY reads/writes
will not be invoked before bringing up the ethernet interface.
Bug 3368603
Change-Id: Idc74be76f47ca1cb607502a4572cb2001d42903a
Signed-off-by: Narayan Reddy <narayanr@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2607300
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com>
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Issue:
To save more power PHY RXC will be disabled when PHY in LPI state by
setting 10th bit in below register. This is standard PHY register.
- PCS Control 1 register (MMD device 3, Address 0x00)
When EEE enabled above register being programmed (10th bit got set) and
it created issue while accessing the MAC registers through indirect
addressing since RXC stopped from PHY.
Fix:
Don't stop RXC in LPI for Orin EQOS
Bug 200776300
Bug 200765092
Change-Id: I4150eebe8e2a5ab79ad5b7180232adad2331e315
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2599137
Reviewed-by: svc_kernel_abi <svc_kernel_abi@nvidia.com>
Reviewed-by: Narayan Reddy <narayanr@nvidia.com>
Reviewed-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-by: Bitan Biswas <bbiswas@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
