[ Upstream commit 98f9fcdee35add80505b6c73f72de5f750d5c03c ]
When using hinic device as a bond slave device, and reading device stats
of master bond device, the kernel may hung.
The kernel panic calltrace as follows:
Kernel panic - not syncing: softlockup: hung tasks
Call trace:
native_queued_spin_lock_slowpath+0x1ec/0x31c
dev_get_stats+0x60/0xcc
dev_seq_printf_stats+0x40/0x120
dev_seq_show+0x1c/0x40
seq_read_iter+0x3c8/0x4dc
seq_read+0xe0/0x130
proc_reg_read+0xa8/0xe0
vfs_read+0xb0/0x1d4
ksys_read+0x70/0xfc
__arm64_sys_read+0x20/0x30
el0_svc_common+0x88/0x234
do_el0_svc+0x2c/0x90
el0_svc+0x1c/0x30
el0_sync_handler+0xa8/0xb0
el0_sync+0x148/0x180
And the calltrace of task that actually caused kernel hungs as follows:
__switch_to+124
__schedule+548
schedule+72
schedule_timeout+348
__down_common+188
__down+24
down+104
hinic_get_stats64+44 [hinic]
dev_get_stats+92
bond_get_stats+172 [bonding]
dev_get_stats+92
dev_seq_printf_stats+60
dev_seq_show+24
seq_read_iter+964
seq_read+220
proc_reg_read+164
vfs_read+172
ksys_read+108
__arm64_sys_read+28
el0_svc_common+132
do_el0_svc+40
el0_svc+24
el0_sync_handler+164
el0_sync+324
When getting device stats from bond, kernel will call bond_get_stats().
It first holds the spinlock bond->stats_lock, and then call
hinic_get_stats64() to collect hinic device's stats.
However, hinic_get_stats64() calls `down(&nic_dev->mgmt_lock)` to
protect its critical section, which may schedule current task out.
And if system is under high pressure, the task cannot be woken up
immediately, which eventually triggers kernel hung panic.
Since previous patch has replaced hinic_dev.tx_stats/rx_stats with local
variable in hinic_get_stats64(), there is nothing need to be protected
by lock, so just removing down()/up() is ok.
Fixes: edd384f682 ("net-next/hinic: Add ethtool and stats")
Signed-off-by: Qiao Ma <mqaio@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 67dffd3db98570af8ff54c934f7d14664c0d182a ]
Function hinic_get_stats64() will do two operations:
1. reads stats from every hinic_rxq/txq and accumulates them
2. calls hinic_rxq/txq_clean_stats() to clean every rxq/txq's stats
For hinic_get_stats64(), it could get right data, because it sums all
data to nic_dev->rx_stats/tx_stats.
But it is wrong for get_drv_queue_stats(), this function will read
hinic_rxq's stats, which have been cleared to zero by hinic_get_stats64().
I have observed hinic's cleanup operation by using such command:
> watch -n 1 "cat ethtool -S eth4 | tail -40"
Result before:
...
rxq7_pkts: 1
rxq7_bytes: 90
rxq7_errors: 0
rxq7_csum_errors: 0
rxq7_other_errors: 0
...
rxq9_pkts: 11
rxq9_bytes: 726
rxq9_errors: 0
rxq9_csum_errors: 0
rxq9_other_errors: 0
...
rxq11_pkts: 0
rxq11_bytes: 0
rxq11_errors: 0
rxq11_csum_errors: 0
rxq11_other_errors: 0
Result after a few seconds:
...
rxq7_pkts: 0
rxq7_bytes: 0
rxq7_errors: 0
rxq7_csum_errors: 0
rxq7_other_errors: 0
...
rxq9_pkts: 2
rxq9_bytes: 132
rxq9_errors: 0
rxq9_csum_errors: 0
rxq9_other_errors: 0
...
rxq11_pkts: 1
rxq11_bytes: 170
rxq11_errors: 0
rxq11_csum_errors: 0
rxq11_other_errors: 0
To solve this problem, we just keep every queue's total stats in their own
queue (aka hinic_{rxq|txq}), and simply sum all per-queue stats every time
calling hinic_get_stats64().
With that solution, there is no need to clean per-queue stats now,
and there is no need to maintain global hinic_dev.{tx|rx}_stats, too.
Fixes: edd384f682 ("net-next/hinic: Add ethtool and stats")
Signed-off-by: Qiao Ma <mqaio@linux.alibaba.com>
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 7c2c57263af41cfd8b5022274e6801542831bb69 ]
'vlan_bitmap' is a bitmap and is used as such. So allocate it with
devm_bitmap_zalloc() and its explicit bit size (i.e. VLAN_N_VID).
This avoids the need of the VLAN_BITMAP_SIZE macro which:
- needlessly has a 'nic_dev' parameter
- should be "long" (and not byte) aligned, so that the bitmap semantic
is respected
This is in fact not an issue because VLAN_N_VID is 4096 at the time
being, but devm_bitmap_zalloc() is less verbose and easier to understand.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Link: https://lore.kernel.org/r/6ff7b7d21414240794a77dc2456914412718a145.1656260842.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 67c3b611d92fc238c43734878bc3e232ab570c79 ]
Sending a PTP packet can imply to use the normal TX driver datapath but
invoked from the driver's ptp worker. The kernel generic TX code
disables softirqs and preemption before calling specific driver TX code,
but the ptp worker does not. Although current ptp driver functionality
does not require it, there are several reasons for doing so:
1) The invoked code is always executed with softirqs disabled for non
PTP packets.
2) Better if a ptp packet transmission is not interrupted by softirq
handling which could lead to high latencies.
3) netdev_xmit_more used by the TX code requires preemption to be
disabled.
Indeed a solution for dealing with kernel preemption state based on static
kernel configuration is not possible since the introduction of dynamic
preemption level configuration at boot time using the static calls
functionality.
Fixes: f79c957a0b ("drivers: net: sfc: use netdev_xmit_more helper")
Signed-off-by: Alejandro Lucero <alejandro.lucero-palau@amd.com>
Link: https://lore.kernel.org/r/20220726064504.49613-1-alejandro.lucero-palau@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit cc019545a238518fa9da1e2a889f6e1bb1005a63 upstream.
Currently loopback test is failiing due to the error returned from
ice_vsi_vlan_setup(). Skip calling it when preparing loopback VSI.
Fixes: 0e674aeb0b ("ice: Add handler for ethtool selftest")
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Tested-by: George Kuruvinakunnel <george.kuruvinakunnel@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 283d736ff7c7e96ac5b32c6c0de40372f8eb171e upstream.
Tx side sets EOP and RS bits on descriptors to indicate that a
particular descriptor is the last one and needs to generate an irq when
it was sent. These bits should not be checked on completion path
regardless whether it's the Tx or the Rx. DD bit serves this purpose and
it indicates that a particular descriptor is either for Rx or was
successfully Txed. EOF is also set as loopback test does not xmit
fragmented frames.
Look at (DD | EOF) bits setting in ice_lbtest_receive_frames() instead
of EOP and RS pair.
Fixes: 0e674aeb0b ("ice: Add handler for ethtool selftest")
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Tested-by: George Kuruvinakunnel <george.kuruvinakunnel@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 3666f666e99600518ab20982af04a078bbdad277 ]
While reading these knobs, they can be changed concurrently.
Thus, we need to add READ_ONCE() to their readers.
- tcp_sack
- tcp_window_scaling
- tcp_timestamps
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d7241f679a59cfe27f92cb5c6272cb429fb1f7ec ]
be_cmd_read_port_transceiver_data assumes that it is given a buffer that
is at least PAGE_DATA_LEN long, or twice that if the module supports SFF
8472. However, this is not always the case.
Fix this by passing the desired offset and length to
be_cmd_read_port_transceiver_data so that we only copy the bytes once.
Fixes: e36edd9d26 ("be2net: add ethtool "-m" option support")
Signed-off-by: Hristo Venev <hristo@venev.name>
Link: https://lore.kernel.org/r/20220716085134.6095-1-hristo@venev.name
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit f838a63369818faadec4ad1736cfbd20ab5da00e ]
Fix an issue when driver incorrectly detects state
of recovery process and erroneously reinitializes interrupts,
which results in a kernel error and call trace message.
The issue was caused by a combination of two factors:
1. Assuming the EMP reset issued after completing
firmware recovery means the whole recovery process is complete.
2. Erroneous reinitialization of interrupt vector after detecting
the above mentioned EMP reset.
Fixes (1) by changing how recovery state change is detected
and (2) by adjusting the conditional expression to ensure using proper
interrupt reinitialization method, depending on the situation.
Fixes: 4ff0ee1af0 ("i40e: Introduce recovery mode support")
Signed-off-by: Dawid Lukwinski <dawid.lukwinski@intel.com>
Signed-off-by: Jan Sokolowski <jan.sokolowski@intel.com>
Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Link: https://lore.kernel.org/r/20220715214542.2968762-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit a9f49e0060301a9bfebeca76739158d0cf91cdf6 ]
Fix memory leak caused by not handling dummy receive descriptor properly.
iavf_get_rx_buffer now sets the rx_buffer return value for dummy receive
descriptors. Without this patch, when the hardware writes a dummy
descriptor, iavf would not free the page allocated for the previous receive
buffer. This is an unlikely event but can still happen.
[Jesse: massaged commit message]
Fixes: efa14c3985 ("iavf: allow null RX descriptors")
Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit f4c7d8948e866918d61493264dbbd67e45ef2bda ]
Current stmmac driver will prepare/enable ptp_ref clock in
stmmac_init_tstamp_counter().
The stmmac_pltfr_noirq_suspend will disable it once in suspend flow.
But in resume flow,
stmmac_pltfr_noirq_resume --> stmmac_init_tstamp_counter
stmmac_resume --> stmmac_hw_setup --> stmmac_init_ptp --> stmmac_init_tstamp_counter
ptp_ref clock reference counter increases twice, which leads to unbalance
ptp clock when resume back.
Move ptp_ref clock prepare/enable out of stmmac_init_tstamp_counter to fix it.
Fixes: 0735e639f129d ("net: stmmac: skip only stmmac_ptp_register when resume from suspend")
Signed-off-by: Biao Huang <biao.huang@mediatek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 7bf9e18d9a5e99e3c83482973557e9f047b051e7 ]
While reading sysctl_ip_fwd_update_priority, it can be changed
concurrently. Thus, we need to add READ_ONCE() to its readers.
Fixes: 432e05d328 ("net: ipv4: Control SKB reprioritization after forwarding")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 7c1ddcee5311f3315096217881d2dbe47cc683f9 ]
The initially merged version of the igc driver code (via commit
146740f9ab, "igc: Add support for PF") contained the following
IGC_REMOVED checks in the igc_rd32/wr32() MMIO accessors:
u32 igc_rd32(struct igc_hw *hw, u32 reg)
{
u8 __iomem *hw_addr = READ_ONCE(hw->hw_addr);
u32 value = 0;
if (IGC_REMOVED(hw_addr))
return ~value;
value = readl(&hw_addr[reg]);
/* reads should not return all F's */
if (!(~value) && (!reg || !(~readl(hw_addr))))
hw->hw_addr = NULL;
return value;
}
And:
#define wr32(reg, val) \
do { \
u8 __iomem *hw_addr = READ_ONCE((hw)->hw_addr); \
if (!IGC_REMOVED(hw_addr)) \
writel((val), &hw_addr[(reg)]); \
} while (0)
E.g. igb has similar checks in its MMIO accessors, and has a similar
macro E1000_REMOVED, which is implemented as follows:
#define E1000_REMOVED(h) unlikely(!(h))
These checks serve to detect and take note of an 0xffffffff MMIO read
return from the device, which can be caused by a PCIe link flap or some
other kind of PCI bus error, and to avoid performing MMIO reads and
writes from that point onwards.
However, the IGC_REMOVED macro was not originally implemented:
#ifndef IGC_REMOVED
#define IGC_REMOVED(a) (0)
#endif /* IGC_REMOVED */
This led to the IGC_REMOVED logic to be removed entirely in a
subsequent commit (commit 3c215fb18e, "igc: remove IGC_REMOVED
function"), with the rationale that such checks matter only for
virtualization and that igc does not support virtualization -- but a
PCIe device can become detached even without virtualization being in
use, and without proper checks, a PCIe bus error affecting an igc
adapter will lead to various NULL pointer dereferences, as the first
access after the error will set hw->hw_addr to NULL, and subsequent
accesses will blindly dereference this now-NULL pointer.
This patch reinstates the IGC_REMOVED checks in igc_rd32/wr32(), and
implements IGC_REMOVED the way it is done for igb, by checking for the
unlikely() case of hw_addr being NULL. This change prevents the oopses
seen when a PCIe link flap occurs on an igc adapter.
Fixes: 146740f9ab ("igc: Add support for PF")
Signed-off-by: Lennert Buytenhek <buytenh@arista.com>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Acked-by: Sasha Neftin <sasha.neftin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit e5ec6a2513383fe2ecc2ee3b5f51d97acbbcd4d8 upstream.
mlxsw needs to distinguish nexthops with a gateway from connected
nexthops in order to write the former to the adjacency table of the
device. The check used to rely on the fact that nexthops with a gateway
have a 'link' scope whereas connected nexthops have a 'host' scope. This
is no longer correct after commit 747c14307214 ("ip: fix dflt addr
selection for connected nexthop").
Fix that by instead checking the address family of the gateway IP. This
is a more direct way and also consistent with the IPv6 counterpart in
mlxsw_sp_rt6_is_gateway().
Cc: stable@vger.kernel.org
Fixes: 747c14307214 ("ip: fix dflt addr selection for connected nexthop")
Fixes: 597cfe4fc3 ("nexthop: Add support for IPv4 nexthops")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 0f33250760384e05c36466b0a2f92f3c6007ba92 ]
Below commit claims that atlantic NIC requires to reset the device on pm
op, and had set the deep to true for all suspend/resume functions.
commit 1809c30b6e5a ("net: atlantic: always deep reset on pm op, fixing up my null deref regression")
So, we could remove deep parameter on suspend/resume functions without
any functional change.
Fixes: 1809c30b6e5a ("net: atlantic: always deep reset on pm op, fixing up my null deref regression")
Signed-off-by: Chia-Lin Kao (AceLan) <acelan.kao@canonical.com>
Link: https://lore.kernel.org/r/20220713111224.1535938-1-acelan.kao@canonical.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit ada74c5539eba06cf8b47d068f92e0b3963a9a6e ]
When creating VFs a kernel panic can happen when calling to
efx_ef10_try_update_nic_stats_vf.
When releasing a DMA coherent buffer, sometimes, I don't know in what
specific circumstances, it has to unmap memory with vunmap. It is
disallowed to do that in IRQ context or with BH disabled. Otherwise, we
hit this line in vunmap, causing the crash:
BUG_ON(in_interrupt());
This patch reenables BH to release the buffer.
Log messages when the bug is hit:
kernel BUG at mm/vmalloc.c:2727!
invalid opcode: 0000 [#1] PREEMPT SMP NOPTI
CPU: 6 PID: 1462 Comm: NetworkManager Kdump: loaded Tainted: G I --------- --- 5.14.0-119.el9.x86_64 #1
Hardware name: Dell Inc. PowerEdge R740/06WXJT, BIOS 2.8.2 08/27/2020
RIP: 0010:vunmap+0x2e/0x30
...skip...
Call Trace:
__iommu_dma_free+0x96/0x100
efx_nic_free_buffer+0x2b/0x40 [sfc]
efx_ef10_try_update_nic_stats_vf+0x14a/0x1c0 [sfc]
efx_ef10_update_stats_vf+0x18/0x40 [sfc]
efx_start_all+0x15e/0x1d0 [sfc]
efx_net_open+0x5a/0xe0 [sfc]
__dev_open+0xe7/0x1a0
__dev_change_flags+0x1d7/0x240
dev_change_flags+0x21/0x60
...skip...
Fixes: d778819609 ("sfc: DMA the VF stats only when requested")
Reported-by: Ma Yuying <yuma@redhat.com>
Signed-off-by: Íñigo Huguet <ihuguet@redhat.com>
Acked-by: Edward Cree <ecree.xilinx@gmail.com>
Link: https://lore.kernel.org/r/20220713092116.21238-1-ihuguet@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit ebe41da5d47ac0fff877e57bd14c54dccf168827 ]
Use after free is detected by kfence when disabling sriov. What was read
after being freed was vf->pci_dev: it was freed from pci_disable_sriov
and later read in efx_ef10_sriov_free_vf_vports, called from
efx_ef10_sriov_free_vf_vswitching.
Set the pointer to NULL at release time to not trying to read it later.
Reproducer and dmesg log (note that kfence doesn't detect it every time):
$ echo 1 > /sys/class/net/enp65s0f0np0/device/sriov_numvfs
$ echo 0 > /sys/class/net/enp65s0f0np0/device/sriov_numvfs
BUG: KFENCE: use-after-free read in efx_ef10_sriov_free_vf_vswitching+0x82/0x170 [sfc]
Use-after-free read at 0x00000000ff3c1ba5 (in kfence-#224):
efx_ef10_sriov_free_vf_vswitching+0x82/0x170 [sfc]
efx_ef10_pci_sriov_disable+0x38/0x70 [sfc]
efx_pci_sriov_configure+0x24/0x40 [sfc]
sriov_numvfs_store+0xfe/0x140
kernfs_fop_write_iter+0x11c/0x1b0
new_sync_write+0x11f/0x1b0
vfs_write+0x1eb/0x280
ksys_write+0x5f/0xe0
do_syscall_64+0x5c/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
kfence-#224: 0x00000000edb8ef95-0x00000000671f5ce1, size=2792, cache=kmalloc-4k
allocated by task 6771 on cpu 10 at 3137.860196s:
pci_alloc_dev+0x21/0x60
pci_iov_add_virtfn+0x2a2/0x320
sriov_enable+0x212/0x3e0
efx_ef10_sriov_configure+0x67/0x80 [sfc]
efx_pci_sriov_configure+0x24/0x40 [sfc]
sriov_numvfs_store+0xba/0x140
kernfs_fop_write_iter+0x11c/0x1b0
new_sync_write+0x11f/0x1b0
vfs_write+0x1eb/0x280
ksys_write+0x5f/0xe0
do_syscall_64+0x5c/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
freed by task 6771 on cpu 12 at 3170.991309s:
device_release+0x34/0x90
kobject_cleanup+0x3a/0x130
pci_iov_remove_virtfn+0xd9/0x120
sriov_disable+0x30/0xe0
efx_ef10_pci_sriov_disable+0x57/0x70 [sfc]
efx_pci_sriov_configure+0x24/0x40 [sfc]
sriov_numvfs_store+0xfe/0x140
kernfs_fop_write_iter+0x11c/0x1b0
new_sync_write+0x11f/0x1b0
vfs_write+0x1eb/0x280
ksys_write+0x5f/0xe0
do_syscall_64+0x5c/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
Fixes: 3c5eb87605 ("sfc: create vports for VFs and assign random MAC addresses")
Reported-by: Yanghang Liu <yanghliu@redhat.com>
Signed-off-by: Íñigo Huguet <ihuguet@redhat.com>
Acked-by: Martin Habets <habetsm.xilinx@gmail.com>
Link: https://lore.kernel.org/r/20220712062642.6915-1-ihuguet@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 49b9f431ff0d845a36be0b3ede35ec324f2e5fee ]
In ftgmac100_probe(), we should hold the refernece returned by
of_get_child_by_name() and use it to call of_node_put() for
reference balance.
Fixes: 39bfab8844 ("net: ftgmac100: Add support for DT phy-handle property")
Signed-off-by: Liang He <windhl@126.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 029c1c2059e9c4b38f97a06204cdecd10cfbeb8a ]
There is a long-standing issue with the Synopsys DWC Ethernet driver
for Tegra194 where random system crashes have been observed [0]. The
problem occurs when the split header feature is enabled in the stmmac
driver. In the bad case, a larger than expected buffer length is
received and causes the calculation of the total buffer length to
overflow. This results in a very large buffer length that causes the
kernel to crash. Why this larger buffer length is received is not clear,
however, the feedback from the NVIDIA design team is that the split
header feature is not supported for Tegra194. Therefore, disable split
header support for Tegra194 to prevent these random crashes from
occurring.
[0] https://lore.kernel.org/linux-tegra/b0b17697-f23e-8fa5-3757-604a86f3a095@nvidia.com/
Fixes: 67afd6d1cf ("net: stmmac: Add Split Header support and enable it in XGMAC cores")
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Link: https://lore.kernel.org/r/20220706083913.13750-1-jonathanh@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 452133dd580811f184e76b1402983182ee425298 ]
The existing capability check for vnic env counters only checks for
receive steering discards, although we need the counters update for the
exposed internal queue oob counter as well. This could result in the
latter counter not being updated correctly when the receive steering
discards counter is not supported.
Fix that by checking whether any counter is supported instead of only
the steering counter capability.
Fixes: 0cfafd4b4d ("net/mlx5e: Add device out of buffer counter")
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 6cc2714e85754a621219693ea8aa3077d6fca0cb ]
Use the correct constant (TLS_DRIVER_STATE_SIZE_TX) in the comparison
against the size of the private TX TLS driver context.
Fixes: df8d866770 ("net/mlx5e: kTLS, Use kernel API to extract private offload context")
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 1b18f09d31cfa7148df15a7d5c5e0e86f105f7d1 ]
During a reset, there may have been transmits in flight that are no
longer valid and cannot be fulfilled. Resetting and clearing the
queues is insufficient; each skb also needs to be explicitly freed
so that upper levels are not left waiting for confirmation of a
transmit that will never happen. If this happens frequently enough,
the apparent backlog will cause TCP to begin "congestion control"
unnecessarily, culminating in permanently decreased throughput.
Fixes: d7c0ef36bd ("ibmvnic: Free and re-allocate scrqs when tx/rx scrqs change")
Tested-by: Nick Child <nnac123@linux.ibm.com>
Reviewed-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Rick Lindsley <ricklind@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 1adb1563e7b7ec659379a18e607e8bc3522d8a78 ]
Dropped packets caused by too large frames were not included in
dropped RX packets statistics.
Issue was caused by not reading the GL_RXERR1 register. That register
stores count of packet which was have been dropped due to too large
size.
Fix it by reading GL_RXERR1 register for each interface.
Repro steps:
Send a packet larger than the set MTU to SUT
Observe rx statists: ethtool -S <interface> | grep rx | grep -v ": 0"
Fixes: 41a9e55c89 ("i40e: add missing VSI statistics")
Signed-off-by: Lukasz Cieplicki <lukaszx.cieplicki@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Tested-by: Gurucharan <gurucharanx.g@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Flooding of unregistered IP multicast has been broken (both to other
switch ports and to the CPU) since the ocelot driver introduction, and
up until commit 4cf35a2b627a ("net: mscc: ocelot: fix broken IP
multicast flooding"), a bug fix for commit 421741ea5672 ("net: mscc:
ocelot: offload bridge port flags to device") from v5.12.
The driver used to set PGID_MCIPV4 and PGID_MCIPV6 to the empty port
mask (0), which made unregistered IPv4/IPv6 multicast go nowhere, and
without ever modifying that port mask at runtime.
The expectation is that such packets are treated as broadcast, and
flooded according to the forwarding domain (to the CPU if the port is
standalone, or to the CPU and other bridged ports, if under a bridge).
Since the aforementioned commit, the limitation has been lifted by
responding to SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS events emitted by the
bridge. As for host flooding, DSA synthesizes another call to
ocelot_port_bridge_flags() on the NPI port which ensures that the CPU
gets the unregistered multicast traffic it might need, for example for
smcroute to work between standalone ports.
But between v4.18 and v5.12, IP multicast flooding has remained unfixed.
Delete the inexplicable premature optimization of clearing PGID_MCIPV4
and PGID_MCIPV6 as part of the init sequence, and allow unregistered IP
multicast to be flooded freely according to the forwarding domain
established by PGID_SRC, by explicitly programming PGID_MCIPV4 and
PGID_MCIPV6 towards all physical ports plus the CPU port module.
Fixes: a556c76adc ("net: mscc: Add initial Ocelot switch support")
Cc: stable@kernel.org
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 4e0effd9007ea0be31f7488611eb3824b4541554 ]
Intel I210 on some Intel Alder Lake platforms can only achieve ~750Mbps
Tx speed via iperf. The RR2DCDELAY shows around 0x2xxx DMA delay, which
will be significantly lower when 1) ASPM is disabled or 2) SoC package
c-state stays above PC3. When the RR2DCDELAY is around 0x1xxx the Tx
speed can reach to ~950Mbps.
According to the I210 datasheet "8.26.1 PCIe Misc. Register - PCIEMISC",
"DMA Idle Indication" doesn't seem to tie to DMA coalesce anymore, so
set it to 1b for "DMA is considered idle when there is no Rx or Tx AND
when there are no TLPs indicating that CPU is active detected on the
PCIe link (such as the host executes CSR or Configuration register read
or write operation)" and performing Tx should also fall under "active
CPU on PCIe link" case.
In addition to that, commit b6e0c419f0 ("igb: Move DMA Coalescing init
code to separate function.") seems to wrongly changed from enabling
E1000_PCIEMISC_LX_DECISION to disabling it, also fix that.
Fixes: b6e0c419f0 ("igb: Move DMA Coalescing init code to separate function.")
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Tested-by: Gurucharan <gurucharanx.g@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Link: https://lore.kernel.org/r/20220621221056.604304-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c3d184c83ff4b80167e34edfc3d21df424bf27ff ]
In current implementation ice_update_phy_type enables all link modes
for selected speed. This approach doesn't work for 1000M speeds,
because both copper (1000baseT) and optical (1000baseX) standards
cannot be enabled at once.
Fix this, by adding the function `ice_set_phy_type_from_speed()`
for 1000M speeds.
Fixes: 48cb27f2fd ("ice: Implement handlers for ethtool PHY/link operations")
Signed-off-by: Anatolii Gerasymenko <anatolii.gerasymenko@intel.com>
Tested-by: Gurucharan <gurucharanx.g@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 1b5d73fb862414106cf270a1a7300ce8ae77de83 upstream.
Enables PCIe PTM (Precision Time Measurement) support in the igc
driver. Notifies the PCI devices that PCIe PTM should be enabled.
PCIe PTM is similar protocol to PTP (Precision Time Protocol) running
in the PCIe fabric, it allows devices to report time measurements from
their internal clocks and the correlation with the PCIe root clock.
The i225 NIC exposes some registers that expose those time
measurements, those registers will be used, in later patches, to
implement the PTP_SYS_OFFSET_PRECISE ioctl().
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Dvora Fuxbrumer <dvorax.fuxbrumer@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Meng Tang <tangmeng@uniontech.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 4b7a632ac4e7101ceefee8484d5c2ca505d347b3 ]
Both RIF and ACL flow counters use a 24-bit SW-managed counter address to
communicate which counter they want to bind.
In a number of Spectrum FW releases, binding a RIF counter is broken and
slices the counter index to 16 bits. As a result, on Spectrum-2 and above,
no more than about 410 RIF counters can be effectively used. This
translates to 205 netdevices for which L3 HW stats can be enabled. (This
does not happen on Spectrum-1, because there are fewer counters available
overall and the counter index never exceeds 16 bits.)
Binding counters to ACLs does not have this issue. Therefore reorder the
counter allocation scheme so that RIF counters come first and therefore get
lower indices that are below the 16-bit barrier.
Fixes: 98e60dce4d ("Merge branch 'mlxsw-Introduce-initial-Spectrum-2-support'")
Reported-by: Maksym Yaremchuk <maksymy@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Link: https://lore.kernel.org/r/20220613125017.2018162-1-idosch@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit fd5855e6b1358e816710afee68a1d2bc685176ca ]
After PF reset and ethtool -t there was call trace in dmesg
sometimes leading to panic. When there was some time, around 5
seconds, between reset and test there were no errors.
Problem was that pf reset calls i40e_vsi_close in prep_for_reset
and ethtool -t calls i40e_vsi_close in diag_test. If there was not
enough time between those commands the second i40e_vsi_close starts
before previous i40e_vsi_close was done which leads to crash.
Add check to diag_test if pf is in reset and don't start offline
tests if it is true.
Add netif_info("testing failed") into unhappy path of i40e_diag_test()
Fixes: e17bc411ae ("i40e: Disable offline diagnostics if VFs are enabled")
Fixes: 510efb2682 ("i40e: Fix ethtool offline diagnostic with netqueues")
Signed-off-by: Michal Jaron <michalx.jaron@intel.com>
Signed-off-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Tested-by: Gurucharan <gurucharanx.g@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 0bb050670ac90a167ecfa3f9590f92966c9a3677 ]
If ADQ is enabled for a VF, then actual number of queue pair
is a number of currently available traffic classes for this VF.
Without this change the configuration of the Rx/Tx queues
fails with error.
Fixes: d29e0d233e ("i40e: missing input validation on VF message handling by the PF")
Signed-off-by: Grzegorz Szczurek <grzegorzx.szczurek@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Tested-by: Bharathi Sreenivas <bharathi.sreenivas@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c3238d36c3a2be0a29a9d848d6c51e1b14be6692 ]
Procedure of configure tc flower filters erroneously allows to create
filters on TC0 where unfiltered packets are also directed by default.
Issue was caused by insufficient checks of hw_tc parameter specifying
the hardware traffic class to pass matching packets to.
Fix checking hw_tc parameter which blocks creation of filters on TC0.
Fixes: 2f4b411a3d ("i40e: Enable cloud filters via tc-flower")
Signed-off-by: Grzegorz Szczurek <grzegorzx.szczurek@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Tested-by: Bharathi Sreenivas <bharathi.sreenivas@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 842c3b3ddc5f4d17275edbaa09e23d712bf8b915 ]
gcc-12 started warning about 'tracker' being used uninitialized:
drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c: In function ‘mlx5_do_bond’:
drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c:786:28: warning: ‘tracker’ is used uninitialized [-Wuninitialized]
786 | struct lag_tracker tracker;
| ^~~~~~~
which seems to be because it doesn't track how the use (and
initialization) is bound by the 'do_bond' flag.
But admittedly that 'do_bond' usage is fairly complicated, and involves
passing it around as an argument to helper functions, so it's somewhat
understandable that gcc doesn't see how that all works.
This function could be rewritten to make the use of that tracker
variable more obviously safe, but for now I'm just adding the forced
initialization of it.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 2f2c0d2919a14002760f89f4e02960c735a316d2 ]
When rx_flag == MTK_RX_FLAGS_HWLRO,
rx_data_len = MTK_MAX_LRO_RX_LENGTH(4096 * 3) > PAGE_SIZE.
netdev_alloc_frag is for alloction of page fragment only.
Reference to other drivers and Documentation/vm/page_frags.rst
Branch to use __get_free_pages when ring->frag_size > PAGE_SIZE.
Signed-off-by: Chen Lin <chen45464546@163.com>
Link: https://lore.kernel.org/r/1654692413-2598-1-git-send-email-chen45464546@163.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 7bb0fb7c63df95d6027dc50d6af3bc3bbbc25483 upstream.
When the promiscuous mode is enabled on a VF, the IXGBE_VMOLR_VPE
bit (VLAN Promiscuous Enable) is set. This means that the VF will
receive packets whose VLAN is not the same than the VLAN of the VF.
For instance, in this situation:
┌────────┐ ┌────────┐ ┌────────┐
│ │ │ │ │ │
│ │ │ │ │ │
│ VF0├────┤VF1 VF2├────┤VF3 │
│ │ │ │ │ │
└────────┘ └────────┘ └────────┘
VM1 VM2 VM3
vf 0: vlan 1000
vf 1: vlan 1000
vf 2: vlan 1001
vf 3: vlan 1001
If we tcpdump on VF3, we see all the packets, even those transmitted
on vlan 1000.
This behavior prevents to bridge VF1 and VF2 in VM2, because it will
create a loop: packets transmitted on VF1 will be received by VF2 and
vice-versa, and bridged again through the software bridge.
This patch remove the activation of VLAN Promiscuous when a VF enables
the promiscuous mode. However, the IXGBE_VMOLR_UPE bit (Unicast
Promiscuous) is kept, so that a VF receives all packets that has the
same VLAN, whatever the destination MAC address.
Fixes: 8443c1a4b1 ("ixgbe, ixgbevf: Add new mbox API xcast mode")
Cc: stable@vger.kernel.org
Cc: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 803e9895ea2b0fe80bc85980ae2d7a7e44037914 upstream.
After a VF requested to remove the promiscuous flag on an interface, the
broadcast packets are not received anymore. This breaks some protocols
like ARP.
In ixgbe_update_vf_xcast_mode(), we should keep the IXGBE_VMOLR_BAM
bit (Broadcast Accept) on promiscuous removal.
This flag is already set by default in ixgbe_set_vmolr() on VF reset.
Fixes: 8443c1a4b1 ("ixgbe, ixgbevf: Add new mbox API xcast mode")
Cc: stable@vger.kernel.org
Cc: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 11ec18b1d8d92b9df307d31950dcba0b3dd7283c ]
Every iteration of for_each_child_of_node() decrements
the reference count of the previous node.
When break from a for_each_child_of_node() loop,
we need to explicitly call of_node_put() on the child node when
not need anymore.
Add missing of_node_put() to avoid refcount leak.
Fixes: bbd2190ce9 ("Altera TSE: Add main and header file for Altera Ethernet Driver")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Link: https://lore.kernel.org/r/20220607041144.7553-1-linmq006@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8fa5e7b20e01042b14f8cd684d2da9b638460c74 ]
When combining two steering rules into one check
not only do they share the same actions but those
actions are also the same. This resolves an issue where
when creating two different rules with the same match
the actions are overwritten and one of the rules is deleted
a FW syndrome can be seen in dmesg.
mlx5_core 0000:03:00.0: mlx5_cmd_check:819:(pid 2105): DEALLOC_MODIFY_HEADER_CONTEXT(0x941) op_mod(0x0) failed, status bad resource state(0x9), syndrome (0x1ab444)
Fixes: 0d235c3fab ("net/mlx5: Add hash table to search FTEs in a flow-group")
Signed-off-by: Mark Bloch <mbloch@nvidia.com>
Reviewed-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8bf94e6414c9481bfa28269022688ab445d0081d ]
The current design does not arm the tracer if traces are available before
the tracer string database is fully loaded, leading to an unfunctional tracer.
This fix will rearm the tracer every time the FW triggers tracer event
regardless of the tracer strings database status.
Fixes: c71ad41ccb ("net/mlx5: FW tracer, events handling")
Signed-off-by: Feras Daoud <ferasda@nvidia.com>
Signed-off-by: Roy Novich <royno@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit f6279f113ad593971999c877eb69dc3d36a75894 ]
Some features (LRO, HW GRO) conflict with XDP. If there is an attempt to
enable such features while XDP is active, they will be set to `off
[requested on]`. In order to activate these features after XDP is turned
off, the driver needs to call netdev_update_features(). This commit adds
this missing call after XDP state changes.
Fixes: cf6e34c8c22f ("net/mlx5e: Properly block LRO when XDP is enabled")
Fixes: b0617e7b3500 ("net/mlx5e: Properly block HW GRO when XDP is enabled")
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 80b2bd737d0e833e6a2b77e482e5a714a79c86a4 ]
The call to mlx5dr_action_destroy() releases "action" memory. That
pointer is set to miss_action later and generates the following smatch
error:
drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c:53 set_miss_action()
warn: 'action' was already freed.
Make sure that the pointer is always valid by setting NULL after destroy.
Fixes: 6a48faeeca ("net/mlx5: Add direct rule fs_cmd implementation")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
