[ Upstream commit 531d0d32de ]
gso_size is expected by the networking stack to be the size of the
payload (thus, not including ethernet/IP/TCP-headers). However, cqe_bcnt
is the full sized frame (including the headers). Dividing cqe_bcnt by
lro_num_seg will then give incorrect results.
For example, running a bpftrace higher up in the TCP-stack
(tcp_event_data_recv), we commonly have gso_size set to 1450 or 1451 even
though in reality the payload was only 1448 bytes.
This can have unintended consequences:
- In tcp_measure_rcv_mss() len will be for example 1450, but. rcv_mss
will be 1448 (because tp->advmss is 1448). Thus, we will always
recompute scaling_ratio each time an LRO-packet is received.
- In tcp_gro_receive(), it will interfere with the decision whether or
not to flush and thus potentially result in less gro'ed packets.
So, we need to discount the protocol headers from cqe_bcnt so we can
actually divide the payload by lro_num_seg to get the real gso_size.
v2:
- Use "(unsigned char *)tcp + tcp->doff * 4 - skb->data)" to compute header-len
(Tariq Toukan <tariqt@nvidia.com>)
- Improve commit-message (Gal Pressman <gal@nvidia.com>)
Fixes: e586b3b0ba ("net/mlx5: Ethernet Datapath files")
Signed-off-by: Christoph Paasch <cpaasch@openai.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Link: https://patch.msgid.link/20250715-cpaasch-pf-925-investigate-incorrect-gso_size-on-cx-7-nic-v2-1-e06c3475f3ac@openai.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 7727ec1523 ]
Add missing post-increment operators for byte pointers in the
loop that copies remaining bytes in xemaclite_aligned_read().
Without the increment, the same byte was written repeatedly
to the destination.
This update aligns with xemaclite_aligned_write()
Fixes: bb81b2ddfa ("net: add Xilinx emac lite device driver")
Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com>
Link: https://patch.msgid.link/20250710173849.2381003-1-alok.a.tiwari@oracle.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit ad4f6df4f3 upstream.
Add the upcoming ConnectX-10 device ID to the table of supported
PCI device IDs.
Cc: stable@vger.kernel.org
Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Reviewed-by: Eran Ben Elisha <eranbe@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/1752650969-148501-1-git-send-email-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit b74c2a2e9c ]
In bnxt_ets_validate(), the code incorrectly loops over all possible
traffic classes to check and add the ETS settings. Fix it to loop
over the configured traffic classes only.
The unconfigured traffic classes will default to TSA_ETS with 0
bandwidth. Looping over these unconfigured traffic classes may
cause the validation to fail and trigger this error message:
"rejecting ETS config starving a TC\n"
The .ieee_setets() will then fail.
Fixes: 7df4ae9fe8 ("bnxt_en: Implement DCBNL to support host-based DCBX.")
Reviewed-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Shravya KN <shravya.k-n@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Link: https://patch.msgid.link/20250710213938.1959625-2-michael.chan@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e81750b4e3 ]
The function ll_temac_ethtools_set_ringparam() incorrectly checked
rx_pending twice, once correctly for RX and once mistakenly in place
of tx_pending. This caused tx_pending to be left unchecked against
TX_BD_NUM_MAX.
As a result, invalid TX ring sizes may have been accepted or valid
ones wrongly rejected based on the RX limit, leading to potential
misconfiguration or unexpected results.
This patch corrects the condition to properly validate tx_pending.
Fixes: f7b261bfc3 ("net: ll_temac: Make RX/TX ring sizes configurable")
Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com>
Link: https://patch.msgid.link/20250710180621.2383000-1-alok.a.tiwari@oracle.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit dd4360c0e8 ]
Restrict the 100Mbit forced-mode workaround to link-down transitions
only, to prevent repeated link reset cycles in certain configurations.
The workaround was originally introduced to improve signal reliability
when switching cables between long and short distances. It temporarily
forces the PHY into 10 Mbps before returning to 100 Mbps.
However, when used with autonegotiating link partners (e.g., Intel i350),
executing this workaround on every link change can confuse the partner
and cause constant renegotiation loops. This results in repeated link
down/up transitions and the PHY never reaching a stable state.
Limit the workaround to only run during the PHY_NOLINK state. This ensures
it is triggered only once per link drop, avoiding disruptive toggling
while still preserving its intended effect.
Note: I am not able to reproduce the original issue that this workaround
addresses. I can only confirm that 100 Mbit mode works correctly in my
test setup. Based on code inspection, I assume the workaround aims to
reset some internal state machine or signal block by toggling speeds.
However, a PHY reset is already performed earlier in the function via
phy_init_hw(), which may achieve a similar effect. Without a reproducer,
I conservatively keep the workaround but restrict its conditions.
Fixes: e57cf3639c ("net: lan78xx: fix accessing the LAN7800's internal phy specific registers from the MAC driver")
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20250709130753.3994461-3-o.rempel@pengutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 01b8114b43 ]
The previous hardcoded definitions of NUM_RX_STATS and
NUM_TX_STATS were not updated when new fields were added
to the ibmvnic_{rx,tx}_queue_stats structures. Specifically,
commit 2ee73c54a6 ("ibmvnic: Add stat for tx direct vs tx
batched") added a fourth TX stat, but NUM_TX_STATS remained 3,
leading to a mismatch.
This patch replaces the static defines with dynamic sizeof-based
calculations to ensure the stat arrays are correctly sized.
This fixes incorrect indexing and prevents incomplete stat
reporting in tools like ethtool.
Fixes: 2ee73c54a6 ("ibmvnic: Add stat for tx direct vs tx batched")
Signed-off-by: Mingming Cao <mmc@linux.ibm.com>
Reviewed-by: Dave Marquardt <davemarq@linux.ibm.com>
Reviewed-by: Haren Myneni <haren@linux.ibm.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250709153332.73892-1-mmc@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 74b1ec9f5d ]
There is a potential NULL pointer dereference in zd_mac_tx_to_dev(). For
example, the following is possible:
T0 T1
zd_mac_tx_to_dev()
/* len == skb_queue_len(q) */
while (len > ZD_MAC_MAX_ACK_WAITERS) {
filter_ack()
spin_lock_irqsave(&q->lock, flags);
/* position == skb_queue_len(q) */
for (i=1; i<position; i++)
skb = __skb_dequeue(q)
if (mac->type == NL80211_IFTYPE_AP)
skb = __skb_dequeue(q);
spin_unlock_irqrestore(&q->lock, flags);
skb_dequeue() -> NULL
Since there is a small gap between checking skb queue length and skb being
unconditionally dequeued in zd_mac_tx_to_dev(), skb_dequeue() can return NULL.
Then the pointer is passed to zd_mac_tx_status() where it is dereferenced.
In order to avoid potential NULL pointer dereference due to situations like
above, check if skb is not NULL before passing it to zd_mac_tx_status().
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: 459c51ad6e ("zd1211rw: port to mac80211")
Signed-off-by: Daniil Dulov <d.dulov@aladdin.ru>
Link: https://patch.msgid.link/20250626114619.172631-1-d.dulov@aladdin.ru
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 9dfe110cc0 ]
Force a fixed MDI-X mode when auto-negotiation is disabled to prevent
link instability.
When forcing the link speed and duplex on a LAN9500 PHY (e.g., with
`ethtool -s eth0 autoneg off ...`) while leaving MDI-X control in auto
mode, the PHY fails to establish a stable link. This occurs because the
PHY's Auto-MDIX algorithm is not designed to operate when
auto-negotiation is disabled. In this state, the PHY continuously
toggles the TX/RX signal pairs, which prevents the link partner from
synchronizing.
This patch resolves the issue by detecting when auto-negotiation is
disabled. If the MDI-X control mode is set to 'auto', the driver now
forces a specific, stable mode (ETH_TP_MDI) to prevent the pair
toggling. This choice of a fixed MDI mode mirrors the behavior the
hardware would exhibit if the AUTOMDIX_EN strap were configured for a
fixed MDI connection.
Fixes: 05b35e7eb9 ("smsc95xx: add phylib support")
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Cc: Andre Edich <andre.edich@microchip.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20250703114941.3243890-4-o.rempel@pengutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit a141af8eb2 ]
Correct the Auto-MDIX configuration to ensure userspace settings are
respected when the feature is disabled by the AUTOMDIX_EN hardware strap.
The LAN9500 PHY allows its default MDI-X mode to be configured via a
hardware strap. If this strap sets the default to "MDI-X off", the
driver was previously unable to enable Auto-MDIX from userspace.
When handling the ETH_TP_MDI_AUTO case, the driver would set the
SPECIAL_CTRL_STS_AMDIX_ENABLE_ bit but neglected to set the required
SPECIAL_CTRL_STS_OVRRD_AMDIX_ bit. Without the override flag, the PHY
falls back to its hardware strap default, ignoring the software request.
This patch corrects the behavior by also setting the override bit when
enabling Auto-MDIX. This ensures that the userspace configuration takes
precedence over the hardware strap, allowing Auto-MDIX to be enabled
correctly in all scenarios.
Fixes: 05b35e7eb9 ("smsc95xx: add phylib support")
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Cc: Andre Edich <andre.edich@microchip.com>
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Link: https://patch.msgid.link/20250703114941.3243890-2-o.rempel@pengutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 2def09ead4 ]
The driver registered xdp_rxq_info structures via xdp_rxq_info_reg()
but failed to properly unregister them in error paths and during
removal.
Fixes: d678be1dc1 ("dpaa2-eth: add XDP_REDIRECT support")
Signed-off-by: Fushuai Wang <wangfushuai@baidu.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Link: https://patch.msgid.link/20250626133003.80136-1-wangfushuai@baidu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 315dbdd7cd ]
In xdp_linearize_page, when reading the following buffers from the ring,
we forget to check the received length with the true allocate size. This
can lead to an out-of-bound read. This commit adds that missing check.
Cc: <stable@vger.kernel.org>
Fixes: 4941d472bf ("virtio-net: do not reset during XDP set")
Signed-off-by: Bui Quang Minh <minhquangbui99@gmail.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Link: https://patch.msgid.link/20250630144212.48471-2-minhquangbui99@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 561aa0e22b ]
dma_map_XXX() functions return values DMA_MAPPING_ERROR as error values
which is often ~0. The error value should be tested with
dma_mapping_error().
This patch creates a new function in niu_ops to test if the mapping
failed. The test is fixed in niu_rbr_add_page(), added in
niu_start_xmit() and the successfully mapped pages are unmaped upon error.
Fixes: ec2deec1f3 ("niu: Fix to check for dma mapping errors.")
Signed-off-by: Thomas Fourier <fourier.thomas@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit aaf2b24803 ]
The comparison in enic_change_mtu() incorrectly used the current
netdev->mtu instead of the new new_mtu value when warning about
an MTU exceeding the port MTU. This could suppress valid warnings
or issue incorrect ones.
Fix the condition and log to properly reflect the new_mtu.
Fixes: ab123fe071 ("enic: handle mtu change for vf properly")
Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com>
Acked-by: John Daley <johndale@cisco.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250628145612.476096-1-alok.a.tiwari@oracle.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 42fd432fe6 ]
Update the Clause 37 Auto-Negotiation implementation to properly align
with the PCS hardware specifications:
- Fix incorrect bit settings in Link Status and Link Duplex fields
- Implement missing sequence steps 2 and 7
These changes ensure CL37 auto-negotiation protocol follows the exact
sequence patterns as specified in the hardware databook.
Fixes: 1bf40ada62 ("amd-xgbe: Add support for clause 37 auto-negotiation")
Signed-off-by: Raju Rangoju <Raju.Rangoju@amd.com>
Link: https://patch.msgid.link/20250630192636.3838291-1-Raju.Rangoju@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 0325143b59 ]
I226 devices advertise support for the PCI-E link L1.2 substate. However,
due to a hardware limitation, the exit latency from this low-power state
is longer than the packet buffer can tolerate under high traffic
conditions. This can lead to packet loss and degraded performance.
To mitigate this, disable the L1.2 substate. The increased power draw
between L1.1 and L1.2 is insignificant.
Fixes: 4354621173 ("igc: Add new device ID's")
Link: https://lore.kernel.org/intel-wired-lan/15248b4f-3271-42dd-8e35-02bfc92b25e1@intel.com
Signed-off-by: Vitaly Lifshits <vitaly.lifshits@intel.com>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Tested-by: Mor Bar-Gabay <morx.bar.gabay@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d72411d209 ]
The `dma_map_XXX()` functions can fail and must be checked using
`dma_mapping_error()`. This patch adds proper error handling for all
DMA mapping calls.
In `atl1_alloc_rx_buffers()`, if DMA mapping fails, the buffer is
deallocated and marked accordingly.
In `atl1_tx_map()`, previously mapped buffers are unmapped and the
packet is dropped on failure.
If `atl1_xmit_frame()` drops the packet, increment the tx_error counter.
Fixes: f3cc28c797 ("Add Attansic L1 ethernet driver.")
Signed-off-by: Thomas Fourier <fourier.thomas@gmail.com>
Link: https://patch.msgid.link/20250625141629.114984-2-fourier.thomas@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 3d5ecada04 upstream.
Fix the following warning:
drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c: In function ‘bnxt_hwrm_queue_cos2bw_cfg’:
cc1: error: writing 12 bytes into a region of size 1 [-Werror=stringop-overflow ]
In file included from drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c:19:
drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h:6045:17: note: destination object ‘unused_0’ of size 1
6045 | u8 unused_0;
Fix it by modifying struct hwrm_queue_cos2bw_cfg_input to use an array
of sub struct similar to the previous patch. This will eliminate the
pointer arithmetc to calculate the destination pointer passed to
memcpy().
Link: https://lore.kernel.org/netdev/CACKFLinikvXmKcxr4kjWO9TPYxTd2cb5agT1j=w9Qyj5-24s5A@mail.gmail.com/
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/r/20230807145720.159645-3-michael.chan@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit ac1b8c978a upstream.
Fix the following warning:
inlined from ‘bnxt_hwrm_queue_cos2bw_qcfg’ at drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c:165:3,
./include/linux/fortify-string.h:592:4: error: call to ‘__read_overflow2_field’ declared with attribute warning: detected read beyond size of field (2nd parameter); maybe use struct_group()? [-Werror]
__read_overflow2_field(q_size_field, size);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Modify the FW interface defintion of struct hwrm_queue_cos2bw_qcfg_output
to use an array of sub struct for the queue1 to queue7 fields. Note that
the layout of the queue0 fields are different and these are not part of
the array. This makes the code much cleaner by removing the pointer
arithmetic for memcpy().
Link: https://lore.kernel.org/netdev/20230727190726.1859515-2-kuba@kernel.org/
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/r/20230807145720.159645-2-michael.chan@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 833c4a8105 upstream.
Fix a W=1 warning with gcc 13.1:
In function ‘fortify_memcpy_chk’,
inlined from ‘bnxt_hwrm_queue_cos2bw_cfg’ at drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c:133:3:
include/linux/fortify-string.h:592:25: warning: call to ‘__read_overflow2_field’ declared with attribute warning: detected read beyond size of field (2nd parameter); maybe use struct_group()? [-Wattribute-warning]
592 | __read_overflow2_field(q_size_field, size);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The field group is already defined and starts at queue_id:
struct bnxt_cos2bw_cfg {
u8 pad[3];
struct_group_attr(cfg, __packed,
u8 queue_id;
__le32 min_bw;
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Link: https://lore.kernel.org/r/20230727190726.1859515-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 7b515f35a9 ]
enetc_hw.h provides two versions of _enetc_rd_reg64.
One which simply calls ioread64() when available.
And another that composes the 64-bit result from ioread32() calls.
In the second case the code appears to assume that each ioread32() call
returns a little-endian value. However both the shift and logical or
used to compose the return value would not work correctly on big endian
systems if this were the case. Moreover, this is inconsistent with the
first case where the return value of ioread64() is assumed to be in host
byte order.
It appears that the correct approach is for both versions to treat the
return value of ioread*() functions as being in host byte order. And
this patch corrects the ioread32()-based version to do so.
This is a bug but would only manifest on big endian systems
that make use of the ioread32-based implementation of _enetc_rd_reg64.
While all in-tree users of this driver are little endian and
make use of the ioread64-based implementation of _enetc_rd_reg64.
Thus, no in-tree user of this driver is affected by this bug.
Flagged by Sparse.
Compile tested only.
Fixes: 16eb4c85c9 ("enetc: Add ethtool statistics")
Closes: https://lore.kernel.org/all/AM9PR04MB850500D3FC24FE23DEFCEA158879A@AM9PR04MB8505.eurprd04.prod.outlook.com/
Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20250624-etnetc-le-v1-1-a73a95d96e4e@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e353b0854d ]
Before calling lan743x_ptp_io_event_clock_get(), the 'channel' value
is checked against the maximum value of PCI11X1X_PTP_IO_MAX_CHANNELS(8).
This seems correct and aligns with the PTP interrupt status register
(PTP_INT_STS) specifications.
However, lan743x_ptp_io_event_clock_get() writes to ptp->extts[] with
only LAN743X_PTP_N_EXTTS(4) elements, using channel as an index:
lan743x_ptp_io_event_clock_get(..., u8 channel,...)
{
...
/* Update Local timestamp */
extts = &ptp->extts[channel];
extts->ts.tv_sec = sec;
...
}
To avoid an out-of-bounds write and utilize all the supported GPIO
inputs, set LAN743X_PTP_N_EXTTS to 8.
Detected using the static analysis tool - Svace.
Fixes: 60942c397a ("net: lan743x: Add support for PTP-IO Event Input External Timestamp (extts)")
Signed-off-by: Alexey Kodanev <aleksei.kodanev@bell-sw.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Acked-by: Rengarajan S <rengarajan.s@microchip.com>
Link: https://patch.msgid.link/20250616113743.36284-1-aleksei.kodanev@bell-sw.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit b1de3c0df7 ]
The PTP_CMD_CTL is a self clearing register which controls the PTP clock
values. In the current implementation driver waits for a duration of 20
sec in case of HW failure to clear the PTP_CMD_CTL register bit. This
timeout of 20 sec is very long to recognize a HW failure, as it is
typically cleared in one clock(<16ns). Hence reducing the timeout to 1 sec
would be sufficient to conclude if there is any HW failure observed. The
usleep_range will sleep somewhere between 1 msec to 20 msec for each
iteration. By setting the PTP_CMD_CTL_TIMEOUT_CNT to 50 the max timeout
is extended to 1 sec.
Signed-off-by: Rengarajan S <rengarajan.s@microchip.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240502050300.38689-1-rengarajan.s@microchip.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Stable-dep-of: e353b0854d ("net: lan743x: fix potential out-of-bounds write in lan743x_ptp_io_event_clock_get()")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 5d3bc9e5e7 ]
This patch fixes an issue seen in a large-scale deployment under heavy
incoming pkts where the aRFS flow wrongly matches a flow and reprograms the
NIC with wrong settings. That mis-steering causes RX-path latency spikes
and noisy neighbor effects when many connections collide on the same
hash (some of our production servers have 20-30K connections).
set_rps_cpu() calls ndo_rx_flow_steer() with flow_id that is calculated by
hashing the skb sized by the per rx-queue table size. This results in
multiple connections (even across different rx-queues) getting the same
hash value. The driver steer function modifies the wrong flow to use this
rx-queue, e.g.: Flow#1 is first added:
Flow#1: <ip1, port1, ip2, port2>, Hash 'h', q#10
Later when a new flow needs to be added:
Flow#2: <ip3, port3, ip4, port4>, Hash 'h', q#20
The driver finds the hash 'h' from Flow#1 and updates it to use q#20. This
results in both flows getting un-optimized - packets for Flow#1 goes to
q#20, and then reprogrammed back to q#10 later and so on; and Flow #2
programming is never done as Flow#1 is matched first for all misses. Many
flows may wrongly share the same hash and reprogram rules of the original
flow each with their own q#.
Tested on two 144-core servers with 16K netperf sessions for 180s. Netperf
clients are pinned to cores 0-71 sequentially (so that wrong packets on q#s
72-143 can be measured). IRQs are set 1:1 for queues -> CPUs, enable XPS,
enable aRFS (global value is 144 * rps_flow_cnt).
Test notes about results from ice_rx_flow_steer():
---------------------------------------------------
1. "Skip:" counter increments here:
if (fltr_info->q_index == rxq_idx ||
arfs_entry->fltr_state != ICE_ARFS_ACTIVE)
goto out;
2. "Add:" counter increments here:
ret = arfs_entry->fltr_info.fltr_id;
INIT_HLIST_NODE(&arfs_entry->list_entry);
3. "Update:" counter increments here:
/* update the queue to forward to on an already existing flow */
Runtime comparison: original code vs with the patch for different
rps_flow_cnt values.
+-------------------------------+--------------+--------------+
| rps_flow_cnt | 512 | 2048 |
+-------------------------------+--------------+--------------+
| Ratio of Pkts on Good:Bad q's | 214 vs 822K | 1.1M vs 980K |
| Avoid wrong aRFS programming | 0 vs 310K | 0 vs 30K |
| CPU User | 216 vs 183 | 216 vs 206 |
| CPU System | 1441 vs 1171 | 1447 vs 1320 |
| CPU Softirq | 1245 vs 920 | 1238 vs 961 |
| CPU Total | 29 vs 22.7 | 29 vs 24.9 |
| aRFS Update | 533K vs 59 | 521K vs 32 |
| aRFS Skip | 82M vs 77M | 7.2M vs 4.5M |
+-------------------------------+--------------+--------------+
A separate TCP_STREAM and TCP_RR with 1,4,8,16,64,128,256,512 connections
showed no performance degradation.
Some points on the patch/aRFS behavior:
1. Enabling full tuple matching ensures flows are always correctly matched,
even with smaller hash sizes.
2. 5-6% drop in CPU utilization as the packets arrive at the correct CPUs
and fewer calls to driver for programming on misses.
3. Larger hash tables reduces mis-steering due to more unique flow hashes,
but still has clashes. However, with larger per-device rps_flow_cnt, old
flows take more time to expire and new aRFS flows cannot be added if h/w
limits are reached (rps_may_expire_flow() succeeds when 10*rps_flow_cnt
pkts have been processed by this cpu that are not part of the flow).
Fixes: 28bf26724f ("ice: Implement aRFS")
Signed-off-by: Krishna Kumar <krikku@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Rinitha S <sx.rinitha@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 5466491c9e ]
Some stress/negative firmware testing around devcmd(s) returning
EAGAIN found that the done bit could get out of sync in the
firmware when it wasn't cleared in a retry case.
While here, change the type of the local done variable to a bool
to match the return type from ionic_dev_cmd_done().
Fixes: ec8ee71473 ("ionic: stretch heartbeat detection")
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250609212827.53842-1-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 9c056ec6dd ]
The cn10k_free_matchall_ipolicer() calls the cn10k_map_unmap_rq_policer()
for each queue in a for loop without checking for any errors.
Check the return value of the cn10k_map_unmap_rq_policer() function during
each loop, and report a warning if the function fails.
Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250408032602.2909-1-vulab@iscas.ac.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 6a07e3af49 ]
It is desireable to push the hardware accelerator to also
process non-segmented TCP frames: we pass the skb->len
to the "TOE/TSO" offloader and it will handle them.
Without this quirk the driver becomes unstable and lock
up and and crash.
I do not know exactly why, but it is probably due to the
TOE (TCP offload engine) feature that is coupled with the
segmentation feature - it is not possible to turn one
part off and not the other, either both TOE and TSO are
active, or neither of them.
Not having the TOE part active seems detrimental, as if
that hardware feature is not really supposed to be turned
off.
The datasheet says:
"Based on packet parsing and TCP connection/NAT table
lookup results, the NetEngine puts the packets
belonging to the same TCP connection to the same queue
for the software to process. The NetEngine puts
incoming packets to the buffer or series of buffers
for a jumbo packet. With this hardware acceleration,
IP/TCP header parsing, checksum validation and
connection lookup are offloaded from the software
processing."
After numerous tests with the hardware locking up after
something between minutes and hours depending on load
using iperf3 I have concluded this is necessary to stabilize
the hardware.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://patch.msgid.link/20250408-gemini-ethernet-tso-always-v1-1-e669f932359c@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit a808691df3 ]
In case the rule already exists and another VSI wants to subscribe to it
new VSI list is being created and both VSIs are moved to it.
Currently, the check for already existing VSI with the same rule is done
based on fdw_id.hw_vsi_id, which applies only to LOOKUP_RX flag.
Change it to vsi_handle. This is software VSI ID, but it can be applied
here, because vsi_map itself is also based on it.
Additionally change return status in case the VSI already exists in the
VSI map to "Already exists". Such case should be handled by the caller.
Signed-off-by: Mateusz Pacuszka <mateuszx.pacuszka@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Signed-off-by: Larysa Zaremba <larysa.zaremba@intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Rafal Romanowski <rafal.romanowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 015bac5dac ]
When the device sends a specific input, an integer underflow can occur, leading
to MMIO write access to an invalid page.
Prevent the integer underflow by changing the type of related variables.
Signed-off-by: Kyungwook Boo <bookyungwook@gmail.com>
Link: https://lore.kernel.org/lkml/ffc91764-1142-4ba2-91b6-8c773f6f7095@gmail.com/T/
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Tested-by: Rinitha S <sx.rinitha@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 20c76dadc7 ]
FDB entries are allocated in an atomic context as they can be added from
the data path when learning is enabled.
After converting the FDB hash table to rhashtable, the insertion rate
will be much higher (*) which will entail a much higher rate of per-CPU
allocations via dst_cache_init().
When adding a large number of entries (e.g., 256k) in a batch, a small
percentage (< 0.02%) of these per-CPU allocations will fail [1]. This
does not happen with the current code since the insertion rate is low
enough to give the per-CPU allocator a chance to asynchronously create
new chunks of per-CPU memory.
Given that:
a. Only a small percentage of these per-CPU allocations fail.
b. The scenario where this happens might not be the most realistic one.
c. The driver can work correctly without dst caches. The dst_cache_*()
APIs first check that the dst cache was properly initialized.
d. The dst caches are not always used (e.g., 'tos inherit').
It seems reasonable to not treat these allocation failures as fatal.
Therefore, do not bail when dst_cache_init() fails and suppress warnings
by specifying '__GFP_NOWARN'.
[1] percpu: allocation failed, size=40 align=8 atomic=1, atomic alloc failed, no space left
(*) 97% reduction in average latency of vxlan_fdb_update() when adding
256k entries in a batch.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20250415121143.345227-14-idosch@nvidia.com
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4ecf56f4b6 ]
The MSE102x doesn't provide any interrupt register, so the only way
to handle the level interrupt is to fetch the whole packet from
the MSE102x internal buffer via SPI. So in cases the interrupt
handler fails to do this, it should return IRQ_NONE. This allows
the core to disable the interrupt in case the issue persists
and prevent an interrupt storm.
Signed-off-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://patch.msgid.link/20250509120435.43646-6-wahrenst@gmx.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit b86bcfee30 ]
As mlx4 has implemented skb_tx_timestamp() in mlx4_en_xmit(), the
SOFTWARE flag is surely needed when users are trying to get timestamp
information.
Signed-off-by: Jason Xing <kernelxing@tencent.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20250510093442.79711-1-kerneljasonxing@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 285ad74775 ]
Make sure the call of skb_tx_timestamp is as close as possible to the
doorbell.
Signed-off-by: Jason Xing <kernelxing@tencent.com>
Link: https://patch.msgid.link/20250510134812.48199-2-kerneljasonxing@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit cd2e7bae92 ]
Firmware requests 2 segments at first. The first segment is of 6799360
whose allocation fails due to dma remapping not available. The success
is returned to firmware. Then firmware asks for 22 smaller segments
instead of 2 big ones. Those get allocated successfully. At suspend/
hibernation time, these segments aren't freed as they will be reused
by firmware after resuming.
After resuming, the firmware asks for the 2 segments again with the
first segment of 6799360 size. Since chunk->vaddr is not NULL, the
type and size are compared with the previous type and size to know if
it can be reused or not. Unfortunately, it is detected that it cannot
be reused and this first smaller segment is freed. Then we continue to
allocate 6799360 size memory which fails and ath11k_qmi_free_target_mem_chunk()
is called which frees the second smaller segment as well. Later success
is returned to firmware which asks for 22 smaller segments again. But
as we had freed 2 segments already, we'll allocate the first 2 new
smaller segments again and reuse the remaining 20. Hence 20 small
segments are being reused instead of 22.
Add skip logic when vaddr is set, but size/type don't match. Use the
same skip and success logic as used when dma_alloc_coherent() fails.
By skipping, the possibility of resume failure due to kernel failing to
allocate memory for QMI can be avoided.
kernel: ath11k_pci 0000:03:00.0: failed to allocate dma memory for qmi (524288 B type 1)
ath11k_pci 0000:03:00.0: failed to allocate qmi target memory: -22
Tested-on: WCN6855 WLAN.HSP.1.1-03926.13-QCAHSPSWPL_V2_SILICONZ_CE-2.52297.6
Signed-off-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Reviewed-by: Baochen Qiang <quic_bqiang@quicinc.com>
Link: https://patch.msgid.link/20250428080242.466901-1-usama.anjum@collabora.com
Signed-off-by: Jeff Johnson <jeff.johnson@oss.qualcomm.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 12889ce926 ]
This patch synchronizes code that accesses from both user-space
and IRQ contexts. The `get_stats()` function can be called from both
context.
`dev->stats.tx_errors` and `dev->stats.collisions` are also updated
in the `tx_errors()` function. Therefore, these fields must also be
protected by synchronized.
There is no code that accessses `dev->stats.tx_errors` between the
previous and updated lines, so the updating point can be moved.
Signed-off-by: Moon Yeounsu <yyyynoom@gmail.com>
Link: https://patch.msgid.link/20250515075333.48290-1-yyyynoom@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 7011faebe5 ]
This allows mt7922 in hostapd mode to transmit up to 1.4 Gbps.
Signed-off-by: Samuel Williams <sam8641@gmail.com>
Link: https://patch.msgid.link/20250511005316.1118961-1-sam8641@gmail.com
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 3c0e4f606d ]
Adds support for:
- LiteOn WN4516R
- LiteOn WN4519R
Both use:
- A nonstandard USB connector
- Mediatek chipset MT7600U
- ASIC revision: 76320044
Disabled VHT support on ASIC revision 76320044:
This fixes the 5G connectibity issue on LiteOn WN4519R module
see https://github.com/openwrt/mt76/issues/971
And may also fix the 5G issues on the XBox One Wireless Adapter
see https://github.com/openwrt/mt76/issues/200
I have looked at the FCC info related to the MT7632U chip as mentioned in here:
https://github.com/openwrt/mt76/issues/459
These confirm the chipset does not support 'ac' mode and hence VHT should be turned of.
Signed-off-by: Henk Vergonet <henk.vergonet@gmail.com>
Acked-by: Lorenzo Bianconi <lorenzo@kernel.org>
Link: https://patch.msgid.link/20250418143914.31384-1-henk.vergonet@gmail.com
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit edb888d297 ]
Logic here always sets hdr->version to 2 if it is not a BE3 or Lancer chip,
even if it is BE2. Use 'else if' to prevent multiple assignments, setting
version 0 for BE2, version 1 for BE3 and Lancer, and version 2 for others.
Fixes potential incorrect version setting when BE2_chip and
BE3_chip/lancer_chip checks could both be true.
Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com>
Link: https://patch.msgid.link/20250519141731.691136-1-alok.a.tiwari@oracle.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 3b9935586a ]
Maximum OTP and EEPROM size for hearthstone PCI1xxxx devices are 8 Kb
and 64 Kb respectively. Adjust max size definitions and return correct
EEPROM length based on device. Also prevent out-of-bound read/write.
Signed-off-by: Rengarajan S <rengarajan.s@microchip.com>
Link: https://patch.msgid.link/20250523173326.18509-1-rengarajan.s@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 9ad0452c02 upstream.
In mii_nway_restart() the code attempts to call
mii->mdio_read which is ch9200_mdio_read(). ch9200_mdio_read()
utilises a local buffer called "buff", which is initialised
with control_read(). However "buff" is conditionally
initialised inside control_read():
if (err == size) {
memcpy(data, buf, size);
}
If the condition of "err == size" is not met, then
"buff" remains uninitialised. Once this happens the
uninitialised "buff" is accessed and returned during
ch9200_mdio_read():
return (buff[0] | buff[1] << 8);
The problem stems from the fact that ch9200_mdio_read()
ignores the return value of control_read(), leading to
uinit-access of "buff".
To fix this we should check the return value of
control_read() and return early on error.
Reported-by: syzbot <syzbot+3361c2d6f78a3e0892f9@syzkaller.appspotmail.com>
Closes: https://syzkaller.appspot.com/bug?extid=3361c2d6f78a3e0892f9
Tested-by: syzbot <syzbot+3361c2d6f78a3e0892f9@syzkaller.appspotmail.com>
Fixes: 4a476bd6d1 ("usbnet: New driver for QinHeng CH9200 devices")
Cc: stable@vger.kernel.org
Signed-off-by: Qasim Ijaz <qasdev00@gmail.com>
Link: https://patch.msgid.link/20250526183607.66527-1-qasdev00@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
