mirror of
https://github.com/nxp-imx/linux-imx.git
synced 2025-07-18 23:29:57 +02:00
ef901b38d3
298 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Linus Torvalds
|
daa121128a |
dma-mapping updates for Linux 6.10
- optimize DMA sync calls when they are no-ops (Alexander Lobakin) - fix swiotlb padding for untrusted devices (Michael Kelley) - add documentation for swiotb (Michael Kelley) -----BEGIN PGP SIGNATURE----- iQI/BAABCgApFiEEgdbnc3r/njty3Iq9D55TZVIEUYMFAmZLV+gLHGhjaEBsc3Qu ZGUACgkQD55TZVIEUYPO7hAAlKuXigzwcrVEUnfRGRdaZ28xbmffyC1dPfw8HRZe xJqvD51aJ/VOoOCcUyt3hNLEQHwtjEk4eM0xGcAASMdwceU58doJCcDJBpbbgbDK CPKJgBLQBC1JfAJUpRiJkV4RsudRhAyndIzUPVgkz0WObpEgDpfO0ClHRF/0Pavy 1sBFVFMbB1ewb/D8ffpp+DWfwrwu0oMC3A2LkYu2F5SQFWuVOpbNemrnZ6K2ckPt 2mcLpJ308+sti8Ka/LrI2akU8JCLYMYDQnue/44v3X3Gm63cMcEx/fj5M5x6m71n P+cxAkjsGDHybnfjbUvR842to8msRsH4CI4Zbb69+5HDlWSadM8JhQd74oeii6o6 RiGPrrFEk7vCxFOkUsqGFYMykEX+71wXfQ1Mpp/b4QgdqBLkxW4ozQ3Ya7ASUs2z TLLmQvIXtYKGnyU+RdOkvS6piHjd4wVHOhuGVdXqVT7WrbaPeovY4TNSTV2ZA1gE 9Y5RCdrX9xeGGNjsYXKwsWGvXVsm6UTQmQVUsatQb3ic+K3S6tQR9pwzk0HmhMuM BscWHSAEL7T8ZZ5Ydph45Cw/6xdH7LggD+nRtLcdAuzCika12eabZHsO0DrF533n qXYOjZOgsMEZWICynxq6+EGQKGWY+F+GyKDMU2w2Es5OgMa9Bqb40aSF+Q887s96 xwI= =Pa8W -----END PGP SIGNATURE----- Merge tag 'dma-mapping-6.10-2024-05-20' of git://git.infradead.org/users/hch/dma-mapping Pull dma-mapping updates from Christoph Hellwig: - optimize DMA sync calls when they are no-ops (Alexander Lobakin) - fix swiotlb padding for untrusted devices (Michael Kelley) - add documentation for swiotb (Michael Kelley) * tag 'dma-mapping-6.10-2024-05-20' of git://git.infradead.org/users/hch/dma-mapping: dma: fix DMA sync for drivers not calling dma_set_mask*() xsk: use generic DMA sync shortcut instead of a custom one page_pool: check for DMA sync shortcut earlier page_pool: don't use driver-set flags field directly page_pool: make sure frag API fields don't span between cachelines iommu/dma: avoid expensive indirect calls for sync operations dma: avoid redundant calls for sync operations dma: compile-out DMA sync op calls when not used iommu/dma: fix zeroing of bounce buffer padding used by untrusted devices swiotlb: remove alloc_size argument to swiotlb_tbl_map_single() Documentation/core-api: add swiotlb documentation |
||
|
Corinna Vinschen
|
86167183a1 |
igc: fix a log entry using uninitialized netdev
During successful probe, igc logs this:
[ 5.133667] igc 0000:01:00.0 (unnamed net_device) (uninitialized): PHC added
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
The reason is that igc_ptp_init() is called very early, even before
register_netdev() has been called. So the netdev_info() call works
on a partially uninitialized netdev.
Fix this by calling igc_ptp_init() after register_netdev(), right
after the media autosense check, just as in igb. Add a comment,
just as in igb.
Now the log message is fine:
[ 5.200987] igc 0000:01:00.0 eth0: PHC added
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Reviewed-by: Hariprasad Kelam <hkelam@marvell.com>
Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
||
|
Alexander Lobakin
|
163943ac00 |
xsk: use generic DMA sync shortcut instead of a custom one
XSk infra's been using its own DMA sync shortcut to try avoiding redundant function calls. Now that there is a generic one, remove the custom implementation and rely on the generic helpers. xsk_buff_dma_sync_for_cpu() doesn't need the second argument anymore, remove it. Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com> Signed-off-by: Christoph Hellwig <hch@lst.de> |
||
|
Eric Dumazet
|
1eb2cded45 |
net: annotate writes on dev->mtu from ndo_change_mtu()
Simon reported that ndo_change_mtu() methods were never
updated to use WRITE_ONCE(dev->mtu, new_mtu) as hinted
in commit
|
||
|
Song Yoong Siang
|
15fd021bc4 |
igc: Add Tx hardware timestamp request for AF_XDP zero-copy packet
This patch adds support to per-packet Tx hardware timestamp request to
AF_XDP zero-copy packet via XDP Tx metadata framework. Please note that
user needs to enable Tx HW timestamp capability via igc_ioctl() with
SIOCSHWTSTAMP cmd before sending xsk Tx hardware timestamp request.
Same as implementation in RX timestamp XDP hints kfunc metadata, Timer 0
(adjustable clock) is used in xsk Tx hardware timestamp. i225/i226 have
four sets of timestamping registers. *skb and *xsk_tx_buffer pointers
are used to indicate whether the timestamping register is already occupied.
Furthermore, a boolean variable named xsk_pending_ts is used to hold the
transmit completion until the tx hardware timestamp is ready. This is
because, for i225/i226, the timestamp notification event comes some time
after the transmit completion event. The driver will retrigger hardware irq
to clean the packet after retrieve the tx hardware timestamp.
Besides, xsk_meta is added into struct igc_tx_timestamp_request as a hook
to the metadata location of the transmit packet. When the Tx timestamp
interrupt is fired, the interrupt handler will copy the value of Tx hwts
into metadata location via xsk_tx_metadata_complete().
This patch is tested with tools/testing/selftests/bpf/xdp_hw_metadata
on Intel ADL-S platform. Below are the test steps and results.
Test Step 1: Run xdp_hw_metadata app
./xdp_hw_metadata <iface> > /dev/shm/result.log
Test Step 2: Enable Tx hardware timestamp
hwstamp_ctl -i <iface> -t 1 -r 1
Test Step 3: Run ptp4l and phc2sys for time synchronization
Test Step 4: Generate UDP packets with 1ms interval for 10s
trafgen --dev <iface> '{eth(da=<addr>), udp(dp=9091)}' -t 1ms -n 10000
Test Step 5: Rerun Step 1-3 with 10s iperf3 as background traffic
Test Step 6: Rerun Step 1-4 with 10s iperf3 as background traffic
Based on iperf3 results below, the impact of holding tx completion to
throughput is not observable.
Result of last UDP packet (no. 10000) in Step 4:
poll: 1 (0) skip=99 fail=0 redir=10000
xsk_ring_cons__peek: 1
0x5640a37972d0: rx_desc[9999]->addr=f2110 addr=f2110 comp_addr=f2110 EoP
rx_hash: 0x2049BE1D with RSS type:0x1
HW RX-time: 1679819246792971268 (sec:1679819246.7930) delta to User RX-time sec:0.0000 (14.990 usec)
XDP RX-time: 1679819246792981987 (sec:1679819246.7930) delta to User RX-time sec:0.0000 (4.271 usec)
No rx_vlan_tci or rx_vlan_proto, err=-95
0x5640a37972d0: ping-pong with csum=ab19 (want 315b) csum_start=34 csum_offset=6
0x5640a37972d0: complete tx idx=9999 addr=f010
HW TX-complete-time: 1679819246793036971 (sec:1679819246.7930) delta to User TX-complete-time sec:0.0001 (77.656 usec)
XDP RX-time: 1679819246792981987 (sec:1679819246.7930) delta to User TX-complete-time sec:0.0001 (132.640 usec)
HW RX-time: 1679819246792971268 (sec:1679819246.7930) delta to HW TX-complete-time sec:0.0001 (65.703 usec)
0x5640a37972d0: complete rx idx=10127 addr=f2110
Result of iperf3 without tx hwts request in step 5:
[ ID] Interval Transfer Bitrate Retr
[ 5] 0.00-10.00 sec 2.74 GBytes 2.36 Gbits/sec 0 sender
[ 5] 0.00-10.05 sec 2.74 GBytes 2.34 Gbits/sec receiver
Result of iperf3 running parallel with trafgen command in step 6:
[ ID] Interval Transfer Bitrate Retr
[ 5] 0.00-10.00 sec 2.74 GBytes 2.36 Gbits/sec 0 sender
[ 5] 0.00-10.04 sec 2.74 GBytes 2.34 Gbits/sec receiver
Co-developed-by: Lai Peter Jun Ann <jun.ann.lai@intel.com>
Signed-off-by: Lai Peter Jun Ann <jun.ann.lai@intel.com>
Signed-off-by: Song Yoong Siang <yoong.siang.song@intel.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Link: https://lore.kernel.org/r/20240424210256.3440903-1-anthony.l.nguyen@intel.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
||
|
Jakub Kicinski
|
2bd87951de |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Cross-merge networking fixes after downstream PR. Conflicts: drivers/net/ethernet/ti/icssg/icssg_prueth.c net/mac80211/chan.c |
||
|
Lukas Wunner
|
c04d1b9ecc |
igc: Fix LED-related deadlock on driver unbind
Roman reports a deadlock on unplug of a Thunderbolt docking station
containing an Intel I225 Ethernet adapter.
The root cause is that led_classdev's for LEDs on the adapter are
registered such that they're device-managed by the netdev. That
results in recursive acquisition of the rtnl_lock() mutex on unplug:
When the driver calls unregister_netdev(), it acquires rtnl_lock(),
then frees the device-managed resources. Upon unregistering the LEDs,
netdev_trig_deactivate() invokes unregister_netdevice_notifier(),
which tries to acquire rtnl_lock() again.
Avoid by using non-device-managed LED registration.
Stack trace for posterity:
schedule+0x6e/0xf0
schedule_preempt_disabled+0x15/0x20
__mutex_lock+0x2a0/0x750
unregister_netdevice_notifier+0x40/0x150
netdev_trig_deactivate+0x1f/0x60 [ledtrig_netdev]
led_trigger_set+0x102/0x330
led_classdev_unregister+0x4b/0x110
release_nodes+0x3d/0xb0
devres_release_all+0x8b/0xc0
device_del+0x34f/0x3c0
unregister_netdevice_many_notify+0x80b/0xaf0
unregister_netdev+0x7c/0xd0
igc_remove+0xd8/0x1e0 [igc]
pci_device_remove+0x3f/0xb0
Fixes:
|
||
|
Sasha Neftin
|
6f31d6b643 |
igc: Refactor runtime power management flow
Following the corresponding discussion [1] and [2] refactor the 'igc_open' method and avoid taking the rtnl_lock() during the 'igc_resume' method. The rtnl_lock is held by the upper layer and could lead to the deadlock during resuming from a runtime power management flow. Notify the stack of the actual queue counts 'netif_set_real_num_*_queues' outside the '_igc_open' wrapper. This notification doesn't have to be called on each resume. Test: 1. Disconnect the ethernet cable 2. Enable the runtime power management via file system: echo auto > /sys/devices/pci0000\.../power/control 3. Check the device state (lspci -s <device> -vvv | grep -i Status) Link: https://lore.kernel.org/netdev/20231206113934.8d7819857574.I2deb5804 ef1739a2af307283d320ef7d82456494@changeid/#r [1] Link: https://lore.kernel.org/netdev/20211125074949.5f897431@kicinski-fedo ra-pc1c0hjn.dhcp.thefacebook.com/t/ [2] Signed-off-by: Sasha Neftin <sasha.neftin@intel.com> Tested-by: Naama Meir <naamax.meir@linux.intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> |
||
|
Jesse Brandeburg
|
75a3f93b53 |
net: intel: implement modern PM ops declarations
Switch the Intel networking drivers to use the new power management ops declaration formats and macros, which allows us to drop __maybe_unused, as well as a bunch of ifdef checking CONFIG_PM. This is safe to do because the compiler drops the unused functions, verified by checking for any of the power management function symbols being present in System.map for a build without CONFIG_PM. If a driver has runtime PM, define the ops with pm_ptr(), and if the driver has Simple PM, use pm_sleep_ptr(), as well as the new versions of the macros for declaring the members of the pm_ops structs. Checked with network-enabled allnoconfig, allyesconfig, allmodconfig on x64_64. Reviewed-by: Alan Brady <alan.brady@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> |
||
|
Jakub Kicinski
|
6e9b01909a |
net: remove gfp_mask from napi_alloc_skb()
__napi_alloc_skb() is napi_alloc_skb() with the added flexibility of choosing gfp_mask. This is a NAPI function, so GFP_ATOMIC is implied. The only practical choice the caller has is whether to set __GFP_NOWARN. But that's a false choice, too, allocation failures in atomic context will happen, and printing warnings in logs, effectively for a packet drop, is both too much and very likely non-actionable. This leads me to a conclusion that most uses of napi_alloc_skb() are simply misguided, and should use __GFP_NOWARN in the first place. We also have a "standard" way of reporting allocation failures via the queue stat API (qstats::rx-alloc-fail). The direct motivation for this patch is that one of the drivers used at Meta calls napi_alloc_skb() (so prior to this patch without __GFP_NOWARN), and the resulting OOM warning is the top networking warning in our fleet. Reviewed-by: Alexander Lobakin <aleksander.lobakin@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://lore.kernel.org/r/20240327040213.3153864-1-kuba@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Kurt Kanzenbach
|
47ce2956c7 |
igc: Remove stale comment about Tx timestamping
The initial igc Tx timestamping implementation used only one register for retrieving Tx timestamps. Commit |
||
|
Jakub Kicinski
|
ed1f164038 |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Merge in late fixes to prepare for the 6.9 net-next PR. Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Jakub Kicinski
|
e3afe5dd3a |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Cross-merge networking fixes after downstream PR. No conflicts. Adjacent changes: net/core/page_pool_user.c |
||
|
Vinicius Costa Gomes
|
244ae992e3 |
igc: Fix missing time sync events
Fix "double" clearing of interrupts, which can cause external events
or timestamps to be missed.
The IGC_TSIRC Time Sync Interrupt Cause register can be cleared in two
ways, by either reading it or by writing '1' into the specific cause
bit. This is documented in section 8.16.1.
The following flow was used:
1. read IGC_TSIRC into 'tsicr';
2. handle the interrupts present in 'tsirc' and mark them in 'ack';
3. write 'ack' into IGC_TSICR;
As both (1) and (3) will clear the interrupt cause, if the same
interrupt happens again between (1) and (3) it will be ignored,
causing events to be missed.
Remove the extra clear in (3).
Fixes:
|
||
|
Florian Kauer
|
ef27f655b4 |
igc: avoid returning frame twice in XDP_REDIRECT
When a frame can not be transmitted in XDP_REDIRECT
(e.g. due to a full queue), it is necessary to free
it by calling xdp_return_frame_rx_napi.
However, this is the responsibility of the caller of
the ndo_xdp_xmit (see for example bq_xmit_all in
kernel/bpf/devmap.c) and thus calling it inside
igc_xdp_xmit (which is the ndo_xdp_xmit of the igc
driver) as well will lead to memory corruption.
In fact, bq_xmit_all expects that it can return all
frames after the last successfully transmitted one.
Therefore, break for the first not transmitted frame,
but do not call xdp_return_frame_rx_napi in igc_xdp_xmit.
This is equally implemented in other Intel drivers
such as the igb.
There are two alternatives to this that were rejected:
1. Return num_frames as all the frames would have been
transmitted and release them inside igc_xdp_xmit.
While it might work technically, it is not what
the return value is meant to represent (i.e. the
number of SUCCESSFULLY transmitted packets).
2. Rework kernel/bpf/devmap.c and all drivers to
support non-consecutively dropped packets.
Besides being complex, it likely has a negative
performance impact without a significant gain
since it is anyway unlikely that the next frame
can be transmitted if the previous one was dropped.
The memory corruption can be reproduced with
the following script which leads to a kernel panic
after a few seconds. It basically generates more
traffic than a i225 NIC can transmit and pushes it
via XDP_REDIRECT from a virtual interface to the
physical interface where frames get dropped.
#!/bin/bash
INTERFACE=enp4s0
INTERFACE_IDX=`cat /sys/class/net/$INTERFACE/ifindex`
sudo ip link add dev veth1 type veth peer name veth2
sudo ip link set up $INTERFACE
sudo ip link set up veth1
sudo ip link set up veth2
cat << EOF > redirect.bpf.c
SEC("prog")
int redirect(struct xdp_md *ctx)
{
return bpf_redirect($INTERFACE_IDX, 0);
}
char _license[] SEC("license") = "GPL";
EOF
clang -O2 -g -Wall -target bpf -c redirect.bpf.c -o redirect.bpf.o
sudo ip link set veth2 xdp obj redirect.bpf.o
cat << EOF > pass.bpf.c
SEC("prog")
int pass(struct xdp_md *ctx)
{
return XDP_PASS;
}
char _license[] SEC("license") = "GPL";
EOF
clang -O2 -g -Wall -target bpf -c pass.bpf.c -o pass.bpf.o
sudo ip link set $INTERFACE xdp obj pass.bpf.o
cat << EOF > trafgen.cfg
{
/* Ethernet Header */
0xe8, 0x6a, 0x64, 0x41, 0xbf, 0x46,
0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF,
const16(ETH_P_IP),
/* IPv4 Header */
0b01000101, 0, # IPv4 version, IHL, TOS
const16(1028), # IPv4 total length (UDP length + 20 bytes (IP header))
const16(2), # IPv4 ident
0b01000000, 0, # IPv4 flags, fragmentation off
64, # IPv4 TTL
17, # Protocol UDP
csumip(14, 33), # IPv4 checksum
/* UDP Header */
10, 0, 1, 1, # IP Src - adapt as needed
10, 0, 1, 2, # IP Dest - adapt as needed
const16(6666), # UDP Src Port
const16(6666), # UDP Dest Port
const16(1008), # UDP length (UDP header 8 bytes + payload length)
csumudp(14, 34), # UDP checksum
/* Payload */
fill('W', 1000),
}
EOF
sudo trafgen -i trafgen.cfg -b3000MB -o veth1 --cpp
Fixes:
|
||
|
Eric Dumazet
|
80bfab79b8 |
net: adopt skb_network_offset() and similar helpers
This is a cleanup patch, making code a bit more concise.
1) Use skb_network_offset(skb) in place of
(skb_network_header(skb) - skb->data)
2) Use -skb_network_offset(skb) in place of
(skb->data - skb_network_header(skb))
3) Use skb_transport_offset(skb) in place of
(skb_transport_header(skb) - skb->data)
4) Use skb_inner_transport_offset(skb) in place of
(skb_inner_transport_header(skb) - skb->data)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Edward Cree <ecree.xilinx@gmail.com> # for sfc
Signed-off-by: David S. Miller <davem@davemloft.net>
|
||
|
Kurt Kanzenbach
|
ea578703b0 |
igc: Add support for LEDs on i225/i226
Add support for LEDs on i225/i226. The LEDs can be controlled via sysfs from user space using the netdev trigger. The LEDs are named as igc-<bus><device>-<led> to be easily identified. Offloading link speed and activity are supported. Other modes are simulated in software by using on/off. Tested on Intel i225. Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Tested-by: Naama Meir <naamax.meir@linux.intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Link: https://lore.kernel.org/r/20240213184138.1483968-1-anthony.l.nguyen@intel.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
Kurt Kanzenbach
|
b747102594 |
igc: Unify filtering rule fields
All filtering parameters such as EtherType and VLAN TCI are stored in host byte order except for the VLAN EtherType. Unify it. Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de> Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Tested-by: Naama Meir <naamax.meir@linux.intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> |
||
|
Kurt Kanzenbach
|
5edcf51d0b |
igc: Use netdev printing functions for flex filters
All igc filter implementations use netdev_*() printing functions except for the flex filters. Unify it. Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de> Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Tested-by: Naama Meir <naamax.meir@linux.intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> |
||
|
Kurt Kanzenbach
|
50534a5577 |
igc: Use reverse xmas tree
Use reverse xmas tree coding style convention in igc_add_flex_filter(). Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de> Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Tested-by: Naama Meir <naamax.meir@linux.intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> |
||
|
Jesse Brandeburg
|
a8e0c7a680 |
igc: field get conversion
Refactor the igc driver to use FIELD_GET() for mask and shift reads, which reduces lines of code and adds clarity of intent. This code was generated by the following coccinelle/spatch script and then manually repaired in a later patch. @get@ constant shift,mask; type T; expression a; @@ -((T)((a) & mask) >> shift) +FIELD_GET(mask, a) and applied via: spatch --sp-file field_prep.cocci --in-place --dir \ drivers/net/ethernet/intel/ Cc: Julia Lawall <Julia.Lawall@inria.fr> Reviewed-by: Marcin Szycik <marcin.szycik@linux.intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> |
||
|
Jesse Brandeburg
|
c82e64868a |
igc: field prep conversion
Refactor igc driver to use FIELD_PREP(), which reduces lines of code and adds clarity of intent. This code was generated by the following coccinelle/spatch script and then manually repaired in a later patch. @prep2@ constant shift,mask; type T; expression a; @@ -(((T)(a) << shift) & mask) +FIELD_PREP(mask, a) @prep@ constant shift,mask; type T; expression a; @@ -((T)((a) << shift) & mask) +FIELD_PREP(mask, a) Cc: Julia Lawall <Julia.Lawall@inria.fr> Cc: Sasha Neftin <sasha.neftin@intel.com> Reviewed-by: Marcin Szycik <marcin.szycik@linux.intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> |
||
|
Vinicius Costa Gomes
|
069b142f58 |
igc: Add support for PTP .getcyclesx64()
Add support for using Timer 1 (i225/i226 have 4 timer registers) as a free-running clock (the "cycles" clock) in addition to Timer 0 (the default, "adjustable clock"). The objective is to allow taprio/etf offloading to coexist with PTP vclocks. Besides the implementation of .getcyclesx64() for i225/i226, to keep timestamping working when vclocks are in use, we also need to add support for TX and RX timestamping using the free running timer, when the requesting socket is bound to a vclock. On the RX side, i225/i226 can be configured to store the values of two timers in the received packet metadata area, so it's a matter of configuring the right registers and retrieving the right timestamp. The TX is a bit more involved because the hardware stores a single timestamp (with the selected timer in the TX descriptor) into one of the timestamp registers. Note some changes at how the timestamps are done for RX, the conversion and adjustment of timestamps are now done closer to the consumption of the timestamp instead of near the reception. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Naama Meir <naamax.meir@linux.intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> |
||
|
Vinicius Costa Gomes
|
fbe5677859 |
igc: Simplify setting flags in the TX data descriptor
We can re-use the IGC_SET_FLAG() macro to simplify setting some values in the TX data descriptor. With the macro it's easier to get the meaning of the operations. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Tested-by: Naama Meir <naamax.meir@linux.intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> |
||
|
Justin Stitt
|
d10d64ad01 |
igc: replace deprecated strncpy with strscpy
`strncpy` is deprecated for use on NUL-terminated destination strings [1] and as such we should prefer more robust and less ambiguous string interfaces. We expect netdev->name to be NUL-terminated based on its use with format strings: | if (q_vector->rx.ring && q_vector->tx.ring) | sprintf(q_vector->name, "%s-TxRx-%u", netdev->name, Furthermore, we do not need NUL-padding as netdev is already zero-allocated: | netdev = alloc_etherdev_mq(sizeof(struct igc_adapter), | IGC_MAX_TX_QUEUES); ... alloc_etherdev() -> alloc_etherdev_mq() -> alloc_etherdev_mqs() -> alloc_netdev_mqs() ... | p = kvzalloc(alloc_size, GFP_KERNEL_ACCOUNT | __GFP_RETRY_MAYFAIL); Considering the above, a suitable replacement is `strscpy` [2] due to the fact that it guarantees NUL-termination on the destination buffer without unnecessarily NUL-padding. Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1] Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html [2] Link: https://github.com/KSPP/linux/issues/90 Signed-off-by: Justin Stitt <justinstitt@google.com> Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel) Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Link: https://lore.kernel.org/r/20231017190411.2199743-10-jacob.e.keller@intel.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Vinicius Costa Gomes
|
cb47b1f679 |
igc: Fix infinite initialization loop with early XDP redirect
When an XDP redirect happens before the link is ready, that
transmission will not finish and will timeout, causing an adapter
reset. If the redirects do not stop, the adapter will not stop
resetting.
Wait for the driver to signal that there's a carrier before allowing
transmissions to proceed.
Previous code was relying that when __IGC_DOWN is cleared, the NIC is
ready to transmit as all the queues are ready, what happens is that
the carrier presence will only be signaled later, after the watchdog
workqueue has a chance to run. And during this interval (between
clearing __IGC_DOWN and the watchdog running) if any transmission
happens the timeout is emitted (detected by igc_tx_timeout()) which
causes the reset, with the potential for the infinite loop.
Fixes:
|
||
|
Vinicius Costa Gomes
|
3ed247e789 |
igc: Add support for multiple in-flight TX timestamps
Add support for using the four sets of timestamping registers that
i225/i226 have available for TX.
In some workloads, where multiple applications request hardware
transmission timestamps, it was possible that some of those requests
were denied because the only in use register was already occupied.
This is also in preparation to future support for hardware
timestamping with multiple PTP domains. With multiple domains chances
of multiple TX timestamps being requested at the same time increase.
Before:
$ sudo ./ntpperf -i enp3s0 -m 10:22:22:22:22:21 -d 192.168.1.3 -s 172.18.0.0/16 -I -H -o 37
| responses | TX timestamp offset (ns)
rate clients | lost invalid basic xleave | min mean max stddev
1000 100 0.00% 0.00% 0.00% 100.00% +1 +41 +73 13
1500 150 0.00% 0.00% 0.00% 100.00% +9 +49 +87 15
2250 225 0.00% 0.00% 0.00% 100.00% +9 +42 +79 13
3375 337 0.00% 0.00% 0.00% 100.00% +11 +46 +81 13
5062 506 0.00% 0.00% 0.00% 100.00% +7 +44 +80 13
7593 759 0.00% 0.00% 0.00% 100.00% +9 +44 +79 12
11389 1138 0.00% 0.00% 0.00% 100.00% +14 +51 +87 13
17083 1708 0.00% 0.00% 0.00% 100.00% +1 +41 +80 14
25624 2562 0.00% 0.00% 0.00% 100.00% +11 +50 +5107 51
38436 3843 0.00% 0.00% 0.00% 100.00% -2 +36 +7843 38
57654 5765 0.00% 0.00% 0.00% 100.00% +4 +42 +10503 69
86481 8648 0.00% 0.00% 0.00% 100.00% +11 +54 +5492 65
129721 12972 0.00% 0.00% 0.00% 100.00% +31 +2680 +6942 2606
194581 16384 16.79% 0.00% 0.87% 82.34% +73 +4444 +15879 3116
291871 16384 35.05% 0.00% 1.53% 63.42% +188 +5381 +17019 3035
437806 16384 54.95% 0.00% 2.55% 42.50% +233 +6302 +13885 2846
After:
$ sudo ./ntpperf -i enp3s0 -m 10:22:22:22:22:21 -d 192.168.1.3 -s 172.18.0.0/16 -I -H -o 37
| responses | TX timestamp offset (ns)
rate clients | lost invalid basic xleave | min mean max stddev
1000 100 0.00% 0.00% 0.00% 100.00% -20 +12 +43 13
1500 150 0.00% 0.00% 0.00% 100.00% -23 +18 +57 14
2250 225 0.00% 0.00% 0.00% 100.00% -2 +33 +67 13
3375 337 0.00% 0.00% 0.00% 100.00% +1 +38 +76 13
5062 506 0.00% 0.00% 0.00% 100.00% +9 +52 +93 14
7593 759 0.00% 0.00% 0.00% 100.00% +11 +47 +82 13
11389 1138 0.00% 0.00% 0.00% 100.00% -9 +27 +74 13
17083 1708 0.00% 0.00% 0.00% 100.00% -13 +25 +66 14
25624 2562 0.00% 0.00% 0.00% 100.00% -8 +28 +65 13
38436 3843 0.00% 0.00% 0.00% 100.00% -13 +28 +69 13
57654 5765 0.00% 0.00% 0.00% 100.00% -11 +32 +71 14
86481 8648 0.00% 0.00% 0.00% 100.00% +2 +44 +83 14
129721 12972 15.36% 0.00% 0.35% 84.29% -2 +2248 +22907 4252
194581 16384 42.98% 0.00% 1.98% 55.04% -4 +5278 +65039 5856
291871 16384 54.33% 0.00% 2.21% 43.46% -3 +6306 +22608 5665
We can see that with 4 registers, as expected, we are able to handle a
increasing number of requests more consistently, but as soon as all
registers are in use, the decrease in quality of service happens in a
sharp step.
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Muhammad Husaini Zulkifli <muhammad.husaini.zulkifli@intel.com>
Reviewed-by: Kurt Kanzenbach <kurt@linutronix.de>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
||
|
Jakub Kicinski
|
4d016ae42e |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Cross-merge networking fixes after downstream PR. No conflicts. Adjacent changes: drivers/net/ethernet/intel/igc/igc_main.c |
||
|
Muhammad Husaini Zulkifli
|
06b412589e |
igc: Add lock to safeguard global Qbv variables
Access to shared variables through hrtimer requires locking in order
to protect the variables because actions to write into these variables
(oper_gate_closed, admin_gate_closed, and qbv_transition) might potentially
occur simultaneously. This patch provides a locking mechanisms to avoid
such scenarios.
Fixes:
|
||
|
Jakub Kicinski
|
014acf2668 |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Cross-merge networking fixes after downstream PR. No conflicts or adjacent changes. Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Muhammad Husaini Zulkifli
|
d4a7ce6421 |
igc: Fix Kernel Panic during ndo_tx_timeout callback
The Xeon validation group has been carrying out some loaded tests with various HW configurations, and they have seen some transmit queue time out happening during the test. This will cause the reset adapter function to be called by igc_tx_timeout(). Similar race conditions may arise when the interface is being brought down and up in igc_reinit_locked(), an interrupt being generated, and igc_clean_tx_irq() being called to complete the TX. When the igc_tx_timeout() function is invoked, this patch will turn off all TX ring HW queues during igc_down() process. TX ring HW queues will be activated again during the igc_configure_tx_ring() process when performing the igc_up() procedure later. This patch also moved existing igc_disable_tx_ring_hw() to avoid using forward declaration. Kernel trace: [ 7678.747813] ------------[ cut here ]------------ [ 7678.757914] NETDEV WATCHDOG: enp1s0 (igc): transmit queue 2 timed out [ 7678.770117] WARNING: CPU: 0 PID: 13 at net/sched/sch_generic.c:525 dev_watchdog+0x1ae/0x1f0 [ 7678.784459] Modules linked in: xt_conntrack nft_chain_nat xt_MASQUERADE xt_addrtype nft_compat nf_tables nfnetlink br_netfilter bridge stp llc overlay dm_mod emrcha(PO) emriio(PO) rktpm(PO) cegbuf_mod(PO) patch_update(PO) se(PO) sgx_tgts(PO) mktme(PO) keylocker(PO) svtdx(PO) svfs_pci_hotplug(PO) vtd_mod(PO) davemem(PO) svmabort(PO) svindexio(PO) usbx2(PO) ehci_sched(PO) svheartbeat(PO) ioapic(PO) sv8259(PO) svintr(PO) lt(PO) pcierootport(PO) enginefw_mod(PO) ata(PO) smbus(PO) spiflash_cdf(PO) arden(PO) dsa_iax(PO) oobmsm_punit(PO) cpm(PO) svkdb(PO) ebg_pch(PO) pch(PO) sviotargets(PO) svbdf(PO) svmem(PO) svbios(PO) dram(PO) svtsc(PO) targets(PO) superio(PO) svkernel(PO) cswitch(PO) mcf(PO) pentiumIII_mod(PO) fs_svfs(PO) mdevdefdb(PO) svfs_os_services(O) ixgbe mdio mdio_devres libphy emeraldrapids_svdefs(PO) regsupport(O) libnvdimm nls_cp437 snd_hda_codec_realtek snd_hda_codec_generic ledtrig_audio snd_hda_intel snd_intel_dspcfg snd_hda_codec snd_hwdep x86_pkg_temp_thermal snd_hda_core snd_pcm snd_timer isst_if_mbox_pci [ 7678.784496] input_leds isst_if_mmio sg snd isst_if_common soundcore wmi button sad9(O) drm fuse backlight configfs efivarfs ip_tables x_tables vmd sdhci led_class rtl8150 r8152 hid_generic pegasus mmc_block usbhid mmc_core hid megaraid_sas ixgb igb i2c_algo_bit ice i40e hpsa scsi_transport_sas e1000e e1000 e100 ax88179_178a usbnet xhci_pci sd_mod xhci_hcd t10_pi crc32c_intel crc64_rocksoft igc crc64 crc_t10dif usbcore crct10dif_generic ptp crct10dif_common usb_common pps_core [ 7679.200403] RIP: 0010:dev_watchdog+0x1ae/0x1f0 [ 7679.210201] Code: 28 e9 53 ff ff ff 4c 89 e7 c6 05 06 42 b9 00 01 e8 17 d1 fb ff 44 89 e9 4c 89 e6 48 c7 c7 40 ad fb 81 48 89 c2 e8 52 62 82 ff <0f> 0b e9 72 ff ff ff 65 8b 05 80 7d 7c 7e 89 c0 48 0f a3 05 0a c1 [ 7679.245438] RSP: 0018:ffa00000001f7d90 EFLAGS: 00010282 [ 7679.256021] RAX: 0000000000000000 RBX: ff11000109938440 RCX: 0000000000000000 [ 7679.268710] RDX: ff11000361e26cd8 RSI: ff11000361e1b880 RDI: ff11000361e1b880 [ 7679.281314] RBP: ffa00000001f7da8 R08: ff1100035f8fffe8 R09: 0000000000027ffb [ 7679.293840] R10: 0000000000001f0a R11: ff1100035f840000 R12: ff11000109938000 [ 7679.306276] R13: 0000000000000002 R14: dead000000000122 R15: ffa00000001f7e18 [ 7679.318648] FS: 0000000000000000(0000) GS:ff11000361e00000(0000) knlGS:0000000000000000 [ 7679.332064] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 7679.342757] CR2: 00007ffff7fca168 CR3: 000000013b08a006 CR4: 0000000000471ef8 [ 7679.354984] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 7679.367207] DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400 [ 7679.379370] PKRU: 55555554 [ 7679.386446] Call Trace: [ 7679.393152] <TASK> [ 7679.399363] ? __pfx_dev_watchdog+0x10/0x10 [ 7679.407870] call_timer_fn+0x31/0x110 [ 7679.415698] expire_timers+0xb2/0x120 [ 7679.423403] run_timer_softirq+0x179/0x1e0 [ 7679.431532] ? __schedule+0x2b1/0x820 [ 7679.439078] __do_softirq+0xd1/0x295 [ 7679.446426] ? __pfx_smpboot_thread_fn+0x10/0x10 [ 7679.454867] run_ksoftirqd+0x22/0x30 [ 7679.462058] smpboot_thread_fn+0xb7/0x160 [ 7679.469670] kthread+0xcd/0xf0 [ 7679.476097] ? __pfx_kthread+0x10/0x10 [ 7679.483211] ret_from_fork+0x29/0x50 [ 7679.490047] </TASK> [ 7679.495204] ---[ end trace 0000000000000000 ]--- [ 7679.503179] igc 0000:01:00.0 enp1s0: Register Dump [ 7679.511230] igc 0000:01:00.0 enp1s0: Register Name Value [ 7679.519892] igc 0000:01:00.0 enp1s0: CTRL 181c0641 [ 7679.528782] igc 0000:01:00.0 enp1s0: STATUS 40280683 [ 7679.537551] igc 0000:01:00.0 enp1s0: CTRL_EXT 10000040 [ 7679.546284] igc 0000:01:00.0 enp1s0: MDIC 180a3800 [ 7679.554942] igc 0000:01:00.0 enp1s0: ICR 00000081 [ 7679.563503] igc 0000:01:00.0 enp1s0: RCTL 04408022 [ 7679.571963] igc 0000:01:00.0 enp1s0: RDLEN[0-3] 00001000 00001000 00001000 00001000 [ 7679.583075] igc 0000:01:00.0 enp1s0: RDH[0-3] 00000068 000000b6 0000000f 00000031 [ 7679.594162] igc 0000:01:00.0 enp1s0: RDT[0-3] 00000066 000000b2 |
||
|
Jakub Kicinski
|
59be3baa8d |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Cross-merge networking fixes after downstream PR. No conflicts or adjacent changes. Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Florian Kauer
|
78adb4bcf9 |
igc: Prevent garbled TX queue with XDP ZEROCOPY
In normal operation, each populated queue item has
next_to_watch pointing to the last TX desc of the packet,
while each cleaned item has it set to 0. In particular,
next_to_use that points to the next (necessarily clean)
item to use has next_to_watch set to 0.
When the TX queue is used both by an application using
AF_XDP with ZEROCOPY as well as a second non-XDP application
generating high traffic, the queue pointers can get in
an invalid state where next_to_use points to an item
where next_to_watch is NOT set to 0.
However, the implementation assumes at several places
that this is never the case, so if it does hold,
bad things happen. In particular, within the loop inside
of igc_clean_tx_irq(), next_to_clean can overtake next_to_use.
Finally, this prevents any further transmission via
this queue and it never gets unblocked or signaled.
Secondly, if the queue is in this garbled state,
the inner loop of igc_clean_tx_ring() will never terminate,
completely hogging a CPU core.
The reason is that igc_xdp_xmit_zc() reads next_to_use
before acquiring the lock, and writing it back
(potentially unmodified) later. If it got modified
before locking, the outdated next_to_use is written
pointing to an item that was already used elsewhere
(and thus next_to_watch got written).
Fixes:
|
||
|
Muhammad Husaini Zulkifli
|
d3750076d4 |
igc: Add TransmissionOverrun counter
Add TransmissionOverrun as per defined by IEEE 802.1Q Bridges. TransmissionOverrun counter shall be incremented if the implementation detects that a frame from a given queue is still being transmitted by the MAC when that gate-close event for that queue occurs. This counter is utilised by the Certification conformance test to inform the user application whether any packets are currently being transmitted on a particular queue during a gate-close event. Intel Discrete I225/I226 have a mechanism to not transmit a packets if the gate open time is insufficient for the packet transmission by setting the Strict_End bit. Thus, it is expected for this counter to be always zero at this moment. Inspired from enetc_taprio_stats() and enetc_taprio_queue_stats(), now driver also report the tx_overruns counter per traffic class. User can get this counter by using below command: 1) tc -s qdisc show dev <interface> root 2) tc -s class show dev <interface> Test Result (Before): class mq :1 root Sent 1289 bytes 20 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 class mq :2 root Sent 124 bytes 2 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 class mq :3 root Sent 46028 bytes 86 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 class mq :4 root Sent 2596 bytes 14 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 Test Result (After): class taprio 100:1 root Sent 8491 bytes 38 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 Transmit overruns: 0 class taprio 100:2 root Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 Transmit overruns: 0 class taprio 100:3 root Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 Transmit overruns: 0 class taprio 100:4 root Sent 994 bytes 11 pkt (dropped 0, overlimits 0 requeues 1) backlog 0b 0p requeues 1 Transmit overruns: 0 Signed-off-by: Muhammad Husaini Zulkifli <muhammad.husaini.zulkifli@intel.com> Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com> Tested-by: Naama Meir <naamax.meir@linux.intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Link: https://lore.kernel.org/r/20230714201428.1718097-1-anthony.l.nguyen@intel.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
Florian Kauer
|
0bcc62858d |
igc: Fix inserting of empty frame for launchtime
The insertion of an empty frame was introduced with commit |
||
|
Florian Kauer
|
c1bca9ac0b |
igc: Fix launchtime before start of cycle
It is possible (verified on a running system) that frames are processed
by igc_tx_launchtime with a txtime before the start of the cycle
(baset_est).
However, the result of txtime - baset_est is written into a u32,
leading to a wrap around to a positive number. The following
launchtime > 0 check will only branch to executing launchtime = 0
if launchtime is already 0.
Fix it by using a s32 before checking launchtime > 0.
Fixes:
|
||
|
Florian Kauer
|
e5d88c53d0 |
igc: Handle already enabled taprio offload for basetime 0
Since commit |
||
|
Florian Kauer
|
82ff5f29b7 |
igc: Do not enable taprio offload for invalid arguments
Only set adapter->taprio_offload_enable after validating the arguments.
Otherwise, it stays set even if the offload was not enabled.
Since the subsequent code does not get executed in case of invalid
arguments, it will not be read at first.
However, by activating and then deactivating another offload
(e.g. ETF/TX launchtime offload), taprio_offload_enable is read
and erroneously keeps the offload feature of the NIC enabled.
This can be reproduced as follows:
# TAPRIO offload (flags == 0x2) and negative base-time leading to expected -ERANGE
sudo tc qdisc replace dev enp1s0 parent root handle 100 stab overhead 24 taprio \
num_tc 1 \
map 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 \
queues 1@0 \
base-time -1000 \
sched-entry S 01 300000 \
flags 0x2
# IGC_TQAVCTRL is 0x0 as expected (iomem=relaxed for reading register)
sudo pcimem /sys/bus/pci/devices/0000:01:00.0/resource0 0x3570 w*1
# Activate ETF offload
sudo tc qdisc replace dev enp1s0 parent root handle 6666 mqprio \
num_tc 3 \
map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 \
queues 1@0 1@1 2@2 \
hw 0
sudo tc qdisc add dev enp1s0 parent 6666:1 etf \
clockid CLOCK_TAI \
delta 500000 \
offload
# IGC_TQAVCTRL is 0x9 as expected
sudo pcimem /sys/bus/pci/devices/0000:01:00.0/resource0 0x3570 w*1
# Deactivate ETF offload again
sudo tc qdisc delete dev enp1s0 parent 6666:1
# IGC_TQAVCTRL should now be 0x0 again, but is observed as 0x9
sudo pcimem /sys/bus/pci/devices/0000:01:00.0/resource0 0x3570 w*1
Fixes:
|
||
|
Florian Kauer
|
8046063df8 |
igc: Rename qbv_enable to taprio_offload_enable
In the current implementation the flags adapter->qbv_enable
and IGC_FLAG_TSN_QBV_ENABLED have a similar name, but do not
have the same meaning. The first one is used only to indicate
taprio offload (i.e. when igc_save_qbv_schedule was called),
while the second one corresponds to the Qbv mode of the hardware.
However, the second one is also used to support the TX launchtime
feature, i.e. ETF qdisc offload. This leads to situations where
adapter->qbv_enable is false, but the flag IGC_FLAG_TSN_QBV_ENABLED
is set. This is prone to confusion.
The rename should reduce this confusion. Since it is a pure
rename, it has no impact on functionality.
Fixes:
|
||
|
Tan Tee Min
|
25102893e4 |
igc: Include the length/type field and VLAN tag in queueMaxSDU
IEEE 802.1Q does not have clear definitions of what constitutes an
SDU (Service Data Unit), but IEEE Std 802.3 clause 3.1.2 does define
the MAC service primitives and clause 3.2.7 does define the MAC Client
Data for Q-tagged frames.
It shows that the mac_service_data_unit (MSDU) does NOT contain the
preamble, destination and source address, or FCS. The MSDU does contain
the length/type field, MAC client data, VLAN tag and any padding
data (prior to the FCS).
Thus, the maximum 802.3 frame size that is allowed to be transmitted
should be QueueMaxSDU (MSDU) + 16 (6 byte SA + 6 byte DA + 4 byte FCS).
Fixes:
|
||
|
Muhammad Husaini Zulkifli
|
175c241288 |
igc: Fix TX Hang issue when QBV Gate is closed
If a user schedules a Gate Control List (GCL) to close one of
the QBV gates while also transmitting a packet to that closed gate,
TX Hang will be happen. HW would not drop any packet when the gate
is closed and keep queuing up in HW TX FIFO until the gate is re-opened.
This patch implements the solution to drop the packet for the closed
gate.
This patch will also reset the adapter to perform SW initialization
for each 1st Gate Control List (GCL) to avoid hang.
This is due to the HW design, where changing to TSN transmit mode
requires SW initialization. Intel Discrete I225/6 transmit mode
cannot be changed when in dynamic mode according to Software User
Manual Section 7.5.2.1. Subsequent Gate Control List (GCL) operations
will proceed without a reset, as they already are in TSN Mode.
Step to reproduce:
DUT:
1) Configure GCL List with certain gate close.
BASE=$(date +%s%N)
tc qdisc replace dev $IFACE parent root handle 100 taprio \
num_tc 4 \
map 0 1 2 3 3 3 3 3 3 3 3 3 3 3 3 3 \
queues 1@0 1@1 1@2 1@3 \
base-time $BASE \
sched-entry S 0x8 500000 \
sched-entry S 0x4 500000 \
flags 0x2
2) Transmit the packet to closed gate. You may use udp_tai
application to transmit UDP packet to any of the closed gate.
./udp_tai -i <interface> -P 100000 -p 90 -c 1 -t <0/1> -u 30004
Fixes:
|
||
|
Muhammad Husaini Zulkifli
|
cca28ceac7 |
igc: Remove delay during TX ring configuration
Remove unnecessary delay during the TX ring configuration.
This will cause delay, especially during link down and
link up activity.
Furthermore, old SKUs like as I225 will call the reset_adapter
to reset the controller during TSN mode Gate Control List (GCL)
setting. This will add more time to the configuration of the
real-time use case.
It doesn't mentioned about this delay in the Software User Manual.
It might have been ported from legacy code I210 in the past.
Fixes:
|
||
|
Muhammad Husaini Zulkifli
|
ed89b74d2d |
igc: Add condition for qbv_config_change_errors counter
Add condition to increase the qbv counter during taprio qbv
configuration only.
There might be a case when TC already been setup then user configure
the ETF/CBS qdisc and this counter will increase if no condition above.
Fixes:
|
||
|
Jakub Kicinski
|
3674fbf045 |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Merge in late fixes to prepare for the 6.5 net-next PR. Signed-off-by: Jakub Kicinski <kuba@kernel.org> |
||
|
Vinicius Costa Gomes
|
afa141583d |
igc: Retrieve TX timestamp during interrupt handling
When the interrupt is handled, the TXTT_0 bit in the TSYNCTXCTL
register should already be set and the timestamp value already loaded
in the appropriate register.
This simplifies the handling, and reduces the latency for retrieving
the TX timestamp, which increase the amount of TX timestamps that can
be handled in a given time period.
As the "work" function doesn't run in a workqueue anymore, rename it
to something more sensible, a event handler.
Using ntpperf[1] we can see the following performance improvements:
Before:
$ sudo ./ntpperf -i enp3s0 -m 10:22:22:22:22:21 -d 192.168.1.3 -s 172.18.0.0/16 -I -H -o -37
| responses | TX timestamp offset (ns)
rate clients | lost invalid basic xleave | min mean max stddev
1000 100 0.00% 0.00% 0.00% 100.00% -56 +9 +52 19
1500 150 0.00% 0.00% 0.00% 100.00% -40 +30 +75 22
2250 225 0.00% 0.00% 0.00% 100.00% -11 +29 +72 15
3375 337 0.00% 0.00% 0.00% 100.00% -18 +40 +88 22
5062 506 0.00% 0.00% 0.00% 100.00% -19 +23 +77 15
7593 759 0.00% 0.00% 0.00% 100.00% +7 +47 +5168 43
11389 1138 0.00% 0.00% 0.00% 100.00% -11 +41 +5240 39
17083 1708 0.00% 0.00% 0.00% 100.00% +19 +60 +5288 50
25624 2562 0.00% 0.00% 0.00% 100.00% +1 +56 +5368 58
38436 3843 0.00% 0.00% 0.00% 100.00% -84 +12 +8847 66
57654 5765 0.00% 0.00% 100.00% 0.00%
86481 8648 0.00% 0.00% 100.00% 0.00%
129721 12972 0.00% 0.00% 100.00% 0.00%
194581 16384 0.00% 0.00% 100.00% 0.00%
291871 16384 27.35% 0.00% 72.65% 0.00%
437806 16384 50.05% 0.00% 49.95% 0.00%
After:
$ sudo ./ntpperf -i enp3s0 -m 10:22:22:22:22:21 -d 192.168.1.3 -s 172.18.0.0/16 -I -H -o -37
| responses | TX timestamp offset (ns)
rate clients | lost invalid basic xleave | min mean max stddev
1000 100 0.00% 0.00% 0.00% 100.00% -44 +0 +61 19
1500 150 0.00% 0.00% 0.00% 100.00% -6 +39 +81 16
2250 225 0.00% 0.00% 0.00% 100.00% -22 +25 +69 15
3375 337 0.00% 0.00% 0.00% 100.00% -28 +15 +56 14
5062 506 0.00% 0.00% 0.00% 100.00% +7 +78 +143 27
7593 759 0.00% 0.00% 0.00% 100.00% -54 +24 +144 47
11389 1138 0.00% 0.00% 0.00% 100.00% -90 -33 +28 21
17083 1708 0.00% 0.00% 0.00% 100.00% -50 -2 +35 14
25624 2562 0.00% 0.00% 0.00% 100.00% -62 +7 +66 23
38436 3843 0.00% 0.00% 0.00% 100.00% -33 +30 +5395 36
57654 5765 0.00% 0.00% 100.00% 0.00%
86481 8648 0.00% 0.00% 100.00% 0.00%
129721 12972 0.00% 0.00% 100.00% 0.00%
194581 16384 19.50% 0.00% 80.50% 0.00%
291871 16384 35.81% 0.00% 64.19% 0.00%
437806 16384 55.40% 0.00% 44.60% 0.00%
[1] https://github.com/mlichvar/ntpperf
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Reviewed-by: Kurt Kanzenbach <kurt@linutronix.de>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
||
|
Vinicius Costa Gomes
|
ce58c7cc8b |
igc: Check if hardware TX timestamping is enabled earlier
Before requesting a packet transmission to be hardware timestamped,
check if the user has TX timestamping enabled. Fixes an issue that if
a packet was internally forwarded to the NIC, and it had the
SKBTX_HW_TSTAMP flag set, the driver would mark that timestamp as
skipped.
In reality, that timestamp was "not for us", as TX timestamp could
never be enabled in the NIC.
Checking if the TX timestamping is enabled earlier has a secondary
effect that when TX timestamping is disabled, there's no need to check
for timestamp timeouts.
We should only take care to free any pending timestamp when TX
timestamping is disabled, as that skb would never be released
otherwise.
Fixes:
|
||
|
Vinicius Costa Gomes
|
9c50e2b150 |
igc: Fix race condition in PTP tx code
Currently, the igc driver supports timestamping only one tx packet at a
time. During the transmission flow, the skb that requires hardware
timestamping is saved in adapter->ptp_tx_skb. Once hardware has the
timestamp, an interrupt is delivered, and adapter->ptp_tx_work is
scheduled. In igc_ptp_tx_work(), we read the timestamp register, update
adapter->ptp_tx_skb, and notify the network stack.
While the thread executing the transmission flow (the user process
running in kernel mode) and the thread executing ptp_tx_work don't
access adapter->ptp_tx_skb concurrently, there are two other places
where adapter->ptp_tx_skb is accessed: igc_ptp_tx_hang() and
igc_ptp_suspend().
igc_ptp_tx_hang() is executed by the adapter->watchdog_task worker
thread which runs periodically so it is possible we have two threads
accessing ptp_tx_skb at the same time. Consider the following scenario:
right after __IGC_PTP_TX_IN_PROGRESS is set in igc_xmit_frame_ring(),
igc_ptp_tx_hang() is executed. Since adapter->ptp_tx_start hasn't been
written yet, this is considered a timeout and adapter->ptp_tx_skb is
cleaned up.
This patch fixes the issue described above by adding the ptp_tx_lock to
protect access to ptp_tx_skb and ptp_tx_start fields from igc_adapter.
Since igc_xmit_frame_ring() called in atomic context by the networking
stack, ptp_tx_lock is defined as a spinlock, and the irq safe variants
of lock/unlock are used.
With the introduction of the ptp_tx_lock, the __IGC_PTP_TX_IN_PROGRESS
flag doesn't provide much of a use anymore so this patch gets rid of it.
Fixes:
|
||
|
Jakub Kicinski
|
173780ff18 |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Cross-merge networking fixes after downstream PR. Conflicts: include/linux/mlx5/driver.h |
||
|
Vinicius Costa Gomes
|
c080fe262f |
igc: Fix possible system crash when loading module
Guarantee that when probe() is run again, PTM and PCI busmaster will be
in the same state as it was if the driver was never loaded.
Avoid an i225/i226 hardware issue that PTM requests can be made even
though PCI bus mastering is not enabled. These unexpected PTM requests
can crash some systems.
So, "force" disable PTM and busmastering before removing the driver,
so they can be re-enabled in the right order during probe(). This is
more like a workaround and should be applicable for i225 and i226, in
any platform.
Fixes:
|
||
|
Muhammad Husaini Zulkifli
|
e43516f597 |
igc: Clean the TX buffer and TX descriptor ring
There could be a race condition during link down where interrupt
being generated and igc_clean_tx_irq() been called to perform the
TX completion. Properly clear the TX buffer/descriptor ring and
disable the TX Queue ring in igc_free_tx_resources() to avoid that.
Kernel trace:
[ 108.237177] Hardware name: Intel Corporation Tiger Lake Client Platform/TigerLake U DDR4 SODIMM RVP, BIOS TGLIFUI1.R00.4204.A00.2105270302 05/27/2021
[ 108.237178] RIP: 0010:refcount_warn_saturate+0x55/0x110
[ 108.242143] RSP: 0018:ffff9e7980003db0 EFLAGS: 00010286
[ 108.245555] Code: 84 bc 00 00 00 c3 cc cc cc cc 85 f6 74 46 80 3d 20 8c 4d 01 00 75 ee 48 c7 c7 88 f4 03 ab c6 05 10 8c 4d 01 01 e8 0b 10 96 ff <0f> 0b c3 cc cc cc cc 80 3d fc 8b 4d 01 00 75 cb 48 c7 c7 b0 f4 03
[ 108.250434]
[ 108.250434] RSP: 0018:ffff9e798125f910 EFLAGS: 00010286
[ 108.254358] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
[ 108.259325]
[ 108.259325] RAX: 0000000000000000 RBX: ffff8ddb935b8000 RCX: 0000000000000027
[ 108.261868] RDX: ffff8de250a28800 RSI: ffff8de250a1c580 RDI: ffff8de250a1c580
[ 108.265538] RDX: 0000000000000027 RSI: 0000000000000002 RDI: ffff8de250a9c588
[ 108.265539] RBP: ffff8ddb935b8000 R08: ffffffffab2655a0 R09: ffff9e798125f898
[ 108.267914] RBP: ffff8ddb8a5b8d80 R08: 0000005648eba354 R09: 0000000000000000
[ 108.270196] R10: 0000000000000001 R11: 000000002d2d2d2d R12: ffff9e798125f948
[ 108.270197] R13: ffff9e798125fa1c R14: ffff8ddb8a5b8d80 R15: 7fffffffffffffff
[ 108.273001] R10: 000000002d2d2d2d R11: 000000002d2d2d2d R12: ffff8ddb8a5b8ed4
[ 108.276410] FS: 00007f605851b740(0000) GS:ffff8de250a80000(0000) knlGS:0000000000000000
[ 108.280597] R13: 00000000000002ac R14: 00000000ffffff99 R15: ffff8ddb92561b80
[ 108.282966] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 108.282967] CR2: 00007f053c039248 CR3: 0000000185850003 CR4: 0000000000f70ee0
[ 108.286206] FS: 0000000000000000(0000) GS:ffff8de250a00000(0000) knlGS:0000000000000000
[ 108.289701] PKRU: 55555554
[ 108.289702] Call Trace:
[ 108.289704] <TASK>
[ 108.293977] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 108.297562] sock_alloc_send_pskb+0x20c/0x240
[ 108.301494] CR2: 00007f053c03a168 CR3: 0000000184394002 CR4: 0000000000f70ef0
[ 108.301495] PKRU: 55555554
[ 108.306464] __ip_append_data.isra.0+0x96f/0x1040
[ 108.309441] Call Trace:
[ 108.309443] ? __pfx_ip_generic_getfrag+0x10/0x10
[ 108.314927] <IRQ>
[ 108.314928] sock_wfree+0x1c7/0x1d0
[ 108.318078] ? __pfx_ip_generic_getfrag+0x10/0x10
[ 108.320276] skb_release_head_state+0x32/0x90
[ 108.324812] ip_make_skb+0xf6/0x130
[ 108.327188] skb_release_all+0x16/0x40
[ 108.330775] ? udp_sendmsg+0x9f3/0xcb0
[ 108.332626] napi_consume_skb+0x48/0xf0
[ 108.334134] ? xfrm_lookup_route+0x23/0xb0
[ 108.344285] igc_poll+0x787/0x1620 [igc]
[ 108.346659] udp_sendmsg+0x9f3/0xcb0
[ 108.360010] ? ttwu_do_activate+0x40/0x220
[ 108.365237] ? __pfx_ip_generic_getfrag+0x10/0x10
[ 108.366744] ? try_to_wake_up+0x289/0x5e0
[ 108.376987] ? sock_sendmsg+0x81/0x90
[ 108.395698] ? __pfx_process_timeout+0x10/0x10
[ 108.395701] sock_sendmsg+0x81/0x90
[ 108.409052] __napi_poll+0x29/0x1c0
[ 108.414279] ____sys_sendmsg+0x284/0x310
[ 108.419507] net_rx_action+0x257/0x2d0
[ 108.438216] ___sys_sendmsg+0x7c/0xc0
[ 108.439723] __do_softirq+0xc1/0x2a8
[ 108.444950] ? finish_task_switch+0xb4/0x2f0
[ 108.452077] irq_exit_rcu+0xa9/0xd0
[ 108.453584] ? __schedule+0x372/0xd00
[ 108.460713] common_interrupt+0x84/0xa0
[ 108.467840] ? clockevents_program_event+0x95/0x100
[ 108.474968] </IRQ>
[ 108.482096] ? do_nanosleep+0x88/0x130
[ 108.489224] <TASK>
[ 108.489225] asm_common_interrupt+0x26/0x40
[ 108.496353] ? __rseq_handle_notify_resume+0xa9/0x4f0
[ 108.503478] RIP: 0010:cpu_idle_poll+0x2c/0x100
[ 108.510607] __sys_sendmsg+0x5d/0xb0
[ 108.518687] Code: 05 e1 d9 c8 00 65 8b 15 de 64 85 55 85 c0 7f 57 e8 b9 ef ff ff fb 65 48 8b 1c 25 00 cc 02 00 48 8b 03 a8 08 74 0b eb 1c f3 90 <48> 8b 03 a8 08 75 13 8b 05 77 63 cd 00 85 c0 75 ed e8 ce ec ff ff
[ 108.525817] do_syscall_64+0x44/0xa0
[ 108.531563] RSP: 0018:ffffffffab203e70 EFLAGS: 00000202
[ 108.538693] entry_SYSCALL_64_after_hwframe+0x72/0xdc
[ 108.546775]
[ 108.546777] RIP: 0033:0x7f605862b7f7
[ 108.549495] RAX: 0000000000000001 RBX: ffffffffab20c940 RCX: 000000000000003b
[ 108.551955] Code: 0e 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b9 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 89 54 24 1c 48 89 74 24 10
[ 108.554068] RDX: 4000000000000000 RSI: 000000002da97f6a RDI: 00000000002b8ff4
[ 108.559816] RSP: 002b:00007ffc99264058 EFLAGS: 00000246
[ 108.564178] RBP: 0000000000000000 R08: 00000000002b8ff4 R09: ffff8ddb01554c80
[ 108.571302] ORIG_RAX: 000000000000002e
[ 108.571303] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f605862b7f7
[ 108.574023] R10: 000000000000015b R11: 000000000000000f R12: ffffffffab20c940
[ 108.574024] R13: 0000000000000000 R14: ffff8de26fbeef40 R15: ffffffffab20c940
[ 108.578727] RDX: 0000000000000000 RSI: 00007ffc992640a0 RDI: 0000000000000003
[ 108.578728] RBP: 00007ffc99264110 R08: 0000000000000000 R09: 175f48ad1c3a9c00
[ 108.581187] do_idle+0x62/0x230
[ 108.585890] R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffc992642d8
[ 108.585891] R13: 00005577814ab2ba R14: 00005577814addf0 R15: 00007f605876d000
[ 108.587920] cpu_startup_entry+0x1d/0x20
[ 108.591422] </TASK>
[ 108.596127] rest_init+0xc5/0xd0
[ 108.600490] ---[ end trace 0000000000000000 ]---
Test Setup:
DUT:
- Change mac address on DUT Side. Ensure NIC not having same MAC Address
- Running udp_tai on DUT side. Let udp_tai running throughout the test
Example:
./udp_tai -i enp170s0 -P 100000 -p 90 -c 1 -t 0 -u 30004
Host:
- Perform link up/down every 5 second.
Result:
Kernel panic will happen on DUT Side.
Fixes:
|