Issue: IOCTLs from supplicanta pplication are not reaching Nvidia
Ethernet driver in kernel 6.1
Fix: Filled the correct policy while registering with netlink. Added
Cipher Suite attribute to genl macsec policy
Ported from -
https://git-master.nvidia.com/r/c/nvethernet-docs/+/3126784
Bug 4316709
Change-Id: I7d0b86d14b1d5bb9aff34b79d635bfccd7443ebf
The correct module license string for drivers is "GPL" and not "GPL
V2". Note that "GPL" in the module license indicates GPLv2 according to
the kernel documentation. Using "GPL V2" causes the following kernel
warns to be printed when the module is loaded ...
r8168: module license 'GPL V2' taints kernel.
Disabling lock debugging due to kernel taint
r8168: module license taints kernel.
Fix this by correcting the license string for dummy drivers.
Bug 4606591
Bug 4471899
Change-Id: Ib785b2d3c90234e279cb1d804bf9a9842e6e5084
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3141628
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
r8168 driver build is disabled on K6.9 and later. This is
causing the packaging failure where module is not available.
Add dummy driver when real driver is not available to make
packaging success.
bug 4628651
Change-Id: I8ade511488fab856f63e84c1ee51986286b2fa2d
Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3138401
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Revanth Kumar Uppala <ruppala@nvidia.com>
Issue:
Ethernet mplan stability tests at -20C for 10Mbps fails with rate 15%
Basically the GPHY 10M power saving includes the below parameters,
clock speed down, pll off and reference voltage off.
During debugging it was found that enable/disable pll circuit
frequently when in 10M low data traffic (such as idle mode) may have
a corner case and plays a part in this issue repro.
Fix:
So plloff saving function should be disable (do not have to open it) for nvidia -20C mplan test case.
Bug 3946623
Change-Id: Ifabe9e26e840537520d66acca106b37d3c285722
Signed-off-by: Revanth Kumar Uppala <ruppala@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3123359
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
For Linux v6.9, the structure 'ethtool_eee' was replaced by
'ethtool_keee' and this new structure uses bitmaps for the structure
fields 'supported', 'advertised' and 'lp_advertised' as opposed to
'u32' types.
A test has been added to the conftest script to detect the presence of
the ethtool_keee structure and so use the definitions generated by
conftest to compile the r8168 accordingly. Note that because the
'supported', 'advertised' and 'lp_advertised' fields are no longer 'u32'
types use the functions mii_eee_cap1_mod_linkmode_t() and
linkmode_to_mii_eee_cap1_t() for working with the bitmap types.
Bug 4471899
Change-Id: I44053d36499261c2aec855ea396bf5654b207ff6
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3110495
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Laxman Dewangan <ldewangan@nvidia.com>
Tested-by: Laxman Dewangan <ldewangan@nvidia.com>
The 'adjfine' method was first introduced in Linux v4.10 by commit
d8d263541913 ("ptp: Introduce a high resolution frequency adjustment
method.") and the 'adjfreq' method was finally removed in Linux v6.2.
Given that 'adjfine' has been supported since Linux v4.10, drop the
deprecated 'adjfreq' method completely to simplify the nvethernet
driver. This fixes build issues for 3rd party kernels that removed the
'adjfreq' method and are older than Linux v6.2.
Bug 3936429
Change-Id: I897bbd3ba92b11d8132c5762d2618b4e82ad44aa
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3128119
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Revanth Kumar Uppala <ruppala@nvidia.com>
For Linux v6.9, the structure 'ethtool_eee' was replaced by
'ethtool_keee' and this new structure uses bitmaps for the structure
fields 'supported', 'advertised' and 'lp_advertised' as opposed to
'u32' types.
Add a test to conftest to detect the presence of the new 'ethtool_keee'
structure and update the nvethernet driver to use the various
linkmode_xxx helper functions for accessing the bitmap structure fields.
Bug 4471899
Change-Id: I275becfa665a36ed90e8bc227eaa4122825faa6a
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3110494
Reviewed-by: Paritosh Dixit <paritoshd@nvidia.com>
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
- Deploy latest supported MLNX MST drivers version-4.26.1-3
https://www.mellanox.com/downloads/MFT/mft-4.26.1-3-arm64-deb.tgz
- Fixes to enable build: fix path used for gcc -I for nnt_driver header
searching but use default path *.c files.
- Use new supported syntax for all objs required to be linked to build
kernel module for each of mst_pci, mst_pciconf & bf3_livefish which
uses <module>-y to set dependent .o objs and ccflags-y to declare the
local cflags.
- Remove makefile-specific targets so that the default kernel-src's
make modules apply to these modules.
- Add symlinks to nnt_driver to each of mst_pci, mst_pciconf this is
required to individually build the nnt_driver objs and use them in
parallel builds because nnt_driver build is disabled.
- This base support CL also includes the fixes from CLs at topic
https://git-master.nvidia.com/r/q/topic:mst_fix.
bug 4192483
bug 4312056
bug 4538284
bug 4538284
bug 4539757
Change-Id: I6bd72f11cc7a8bf1dd8abba6385ea1d471befa02
Signed-off-by: Bharath H S <bhs@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3069978
(cherry picked from commit 73bffe884a1127164b3c0e74b0e4a5b8a6cf4d71)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3069973
Reviewed-by: Laxman Dewangan <ldewangan@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
In Linux v6.9-rc1, the typedef of skb_frag_t is updated to a new struct
skb_frag. Struct skb_frag does not include bv_offset and bv_page
members. Hence, use functions skb_frag_off and skb_frag_page to get
offset and page respectively. Use of functions skb_frag_off and
skb_frag_page is backward compatible with older kernel versions that
we use.
Bug 4471899
Signed-off-by: Paritosh Dixit <paritoshd@nvidia.com>
Change-Id: I5d14c750c5b4cd28a2abeda95d7c11bf975ef1b8
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3104114
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Revanth Kumar Uppala <ruppala@nvidia.com>
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Existing implementation uses a single page_pool
for all Rx DMA channels. As by default all irqs
are on CPU 0, this does not cause any issue.
Routing an irq to some other CPU causes race
conditon for page allocation from page_pool which
leads to random memory corruption.
[ 158.416637] Call trace:
[ 158.416644] arm_lpae_map_pages+0xb4/0x1e0
[ 158.416649] arm_smmu_map_pages+0x8c/0x160
[ 158.416661] __iommu_map+0xf8/0x2b0
[ 158.416677] iommu_map_atomic+0x58/0xb0
[ 158.416683] __iommu_dma_map+0xac/0x150
[ 158.416687] iommu_dma_map_page+0xf4/0x220
[ 158.416690] dma_map_page_attrs+0x1e8/0x260
[ 158.416727] page_pool_dma_map+0x48/0xd0
[ 158.416750] __page_pool_alloc_pages_slow+0xc4/0x390
[ 158.416757] page_pool_alloc_pages+0x64/0x90
[ 158.416762] ether_padctrl_mii_rx_pins+0x164/0x9d0 [nvethernet]
[ 158.416807] ether_padctrl_mii_rx_pins+0x478/0x9d0 [nvethernet]
[ 158.416822] osi_process_rx_completions+0x284/0x4d0 [nvethernet]
[ 158.416832] 0xffffa26218b8f71c
[ 158.416855] __napi_poll+0x48/0x230
[ 158.416871] net_rx_action+0xf4/0x290
[ 158.416875] __do_softirq+0x130/0x3e8
[ 158.416889] __irq_exit_rcu+0xe8/0x110
This change creates a page_pool per Rx DMA channel.
This ensures there is no race conditon for page
alloc/dealloc from each DMA napi context.
Bug 4541158
Change-Id: I12668ee7d824fd30d54a874bbbdf190d02943478
Signed-off-by: Aniruddha Paul <anpaul@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3091494
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
If the userspace service 'irqbalance' is installed then the nvethernet
driver crashes when there is network activity. To avoid this crash set
the IRQF_NOBALANCING flash for the VM interrupts. No performance
degradation is observed when running iperf3 with a 1Gbps link.
Long-term the nvethernet driver still needs to be fixed to allow IRQ
balancing.
Bug 4293378
Bug 4541158
Change-Id: I0aa4ee28e36c7d273f14ff043544e72d3e988bd3
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3087525
(cherry picked from commit f9ed570c2b)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3088936
Reviewed-by: Revanth Kumar Uppala <ruppala@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
