Merge branch 'linus' into sched/urgent, to resolve conflict

Conflicts:
kernel/sched/ext.c

There's a context conflict between this upstream commit:

3fdb9ebcec10 sched_ext: Start schedulers with

Merge branch 'linus' into sched/urgent, to resolve conflict

Conflicts:
kernel/sched/ext.c

There's a context conflict between this upstream commit:

3fdb9ebcec10 sched_ext: Start schedulers with consistent p->scx.slice values

... and this fix in sched/urgent:

98442f0ccd82 sched: Fix delayed_dequeue vs switched_from_fair()

Resolve it.

Signed-off-by: Ingo Molnar <mingo@kernel.org>

show more ...

Merge tag 'net-6.12-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Pull networking fixes from Jakub Kicinski:
"Including fixes from bluetooth and netfilter.

Current release - r

Merge tag 'net-6.12-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Pull networking fixes from Jakub Kicinski:
"Including fixes from bluetooth and netfilter.

Current release - regressions:

- dsa: sja1105: fix reception from VLAN-unaware bridges

- Revert "net: stmmac: set PP_FLAG_DMA_SYNC_DEV only if XDP is
enabled"

- eth: fec: don't save PTP state if PTP is unsupported

Current release - new code bugs:

- smc: fix lack of icsk_syn_mss with IPPROTO_SMC, prevent null-deref

- eth: airoha: update Tx CPU DMA ring idx at the end of xmit loop

- phy: aquantia: AQR115c fix up PMA capabilities

Previous releases - regressions:

- tcp: 3 fixes for retrans_stamp and undo logic

Previous releases - always broken:

- net: do not delay dst_entries_add() in dst_release()

- netfilter: restrict xtables extensions to families that are safe,
syzbot found a way to combine ebtables with extensions that are
never used by userspace tools

- sctp: ensure sk_state is set to CLOSED if hashing fails in
sctp_listen_start

- mptcp: handle consistently DSS corruption, and prevent corruption
due to large pmtu xmit"

* tag 'net-6.12-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (87 commits)
MAINTAINERS: Add headers and mailing list to UDP section
MAINTAINERS: consistently exclude wireless files from NETWORKING [GENERAL]
slip: make slhc_remember() more robust against malicious packets
net/smc: fix lacks of icsk_syn_mss with IPPROTO_SMC
ppp: fix ppp_async_encode() illegal access
docs: netdev: document guidance on cleanup patches
phonet: Handle error of rtnl_register_module().
mpls: Handle error of rtnl_register_module().
mctp: Handle error of rtnl_register_module().
bridge: Handle error of rtnl_register_module().
vxlan: Handle error of rtnl_register_module().
rtnetlink: Add bulk registration helpers for rtnetlink message handlers.
net: do not delay dst_entries_add() in dst_release()
mptcp: pm: do not remove closing subflows
mptcp: fallback when MPTCP opts are dropped after 1st data
tcp: fix mptcp DSS corruption due to large pmtu xmit
mptcp: handle consistently DSS corruption
net: netconsole: fix wrong warning
net: dsa: refuse cross-chip mirroring operations
net: fec: don't save PTP state if PTP is unsupported
...

show more ...

Merge branch 'ibmvnic-fix-for-send-scrq-direct'

Nick Child says:

====================
ibmvnic: Fix for send scrq direct

This is a v2 of a patchset (now just patch) which addresses a
bug in a new f

Merge branch 'ibmvnic-fix-for-send-scrq-direct'

Nick Child says:

====================
ibmvnic: Fix for send scrq direct

This is a v2 of a patchset (now just patch) which addresses a
bug in a new feature which is causing major link UP issues with
certain physical cards.

For a full summary of the issue:
1. During vnic initialization we get the following values from vnic
server regarding "Transmit / Receive Descriptor Requirement" (see
PAPR Table 584. CAPABILITIES Commands):
- LSO Tx frame = 0x0F , header offsets + L2, L3, L4 headers required
- CSO Tx frame = 0x0C , header offsets + L2 header required
- standard frame = 0x0C , header offsets + L2 header required
2. Assume we are dealing with only "standard frames" from now on (no
CSO, no LSO)
3. When using 100G backing device, we don't hand vnic server any header
information and TX is successful
4. When using 25G backing device, we don't hand vnic server any header
information and TX fails and we get "Adapter Error" transport events.
The obvious issue here is that vnic client should be respecting the 0X0C
header requirement for standard frames. But 100G cards will also give
0x0C despite the fact that we know TX works if we ignore it. That being
said, we still must respect values given from the managing server. Will
need to work with them going forward to hopefully get 100G cards to
return 0x00 for this bitstring so the performance gains of using
send_subcrq_direct can be continued.
====================

Link: https://patch.msgid.link/20241001163200.1802522-1-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

show more ...

ibmvnic: Inspect header requirements before using scrq direct

Previously, the TX header requirement for standard frames was ignored.
This requirement is a bitstring sent from the VIOS which maps to

ibmvnic: Inspect header requirements before using scrq direct

Previously, the TX header requirement for standard frames was ignored.
This requirement is a bitstring sent from the VIOS which maps to the
type of header information needed during TX. If no header information,
is needed then send subcrq direct can be used (which can be more
performant).

This bitstring was previously ignored for standard packets (AKA non LSO,
non CSO) due to the belief that the bitstring was over-cautionary. It
turns out that there are some configurations where the backing device
does need header information for transmission of standard packets. If
the information is not supplied then this causes continuous "Adapter
error" transport events. Therefore, this bitstring should be respected
and observed before considering the use of send subcrq direct.

Fixes: 74839f7a8268 ("ibmvnic: Introduce send sub-crq direct")
Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20241001163200.1802522-2-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

show more ...

Merge tag 'kvmarm-fixes-6.12-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD

KVM/arm64 fixes for 6.12, take #1

- Fix pKVM error path on init, making sure we do not chang

Merge tag 'kvmarm-fixes-6.12-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD

KVM/arm64 fixes for 6.12, take #1

- Fix pKVM error path on init, making sure we do not change critical
system registers as we're about to fail

- Make sure that the host's vector length is at capped by a value
common to all CPUs

- Fix kvm_has_feat*() handling of "negative" features, as the current
code is pretty broken

- Promote Joey to the status of official reviewer, while James steps
down -- hopefully only temporarly

show more ...

Merge tag 'asoc-fix-v6.12-rc1' of https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus

ASoC: Fixes for v6.12

A bunch of fixes here that came in during the merge window and t

Merge tag 'asoc-fix-v6.12-rc1' of https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus

ASoC: Fixes for v6.12

A bunch of fixes here that came in during the merge window and the first
week of release, plus some new quirks and device IDs. There's nothing
major here, it's a bit bigger than it might've been due to there being
no fixes sent during the merge window due to your vacation.

show more ...

Merge remote-tracking branch 'drm/drm-fixes' into drm-misc-fixes

Required for a panthor fix that broke when
FOP_UNSIGNED_OFFSET was added in place of FMODE_UNSIGNED_OFFSET.

Signed-off-by: Maarten L

Merge remote-tracking branch 'drm/drm-fixes' into drm-misc-fixes

Required for a panthor fix that broke when
FOP_UNSIGNED_OFFSET was added in place of FMODE_UNSIGNED_OFFSET.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>

show more ...

Merge branch 'next' into for-linus

Prepare input updates for 6.12 merge window.

Merge tag 'v6.10' into next

Sync up with mainline to bring in device_for_each_child_node_scoped()
and other newer APIs.

Merge tag 'v6.11-rc1' into clk-meson-next

Linux 6.11-rc1

Merge tag 'v6.11-rc3' into trace/ring-buffer/core

The "reserve_mem" kernel command line parameter has been pulled into
v6.11. Merge the latest -rc3 to allow the persistent ring buffer memory to
be a

Merge tag 'v6.11-rc3' into trace/ring-buffer/core

The "reserve_mem" kernel command line parameter has been pulled into
v6.11. Merge the latest -rc3 to allow the persistent ring buffer memory to
be able to be mapped at the address specified by the "reserve_mem" command
line parameter.

Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>

show more ...

Merge tag 'v6.11-rc1' into for-6.12

Linux 6.11-rc1

Merge drm/drm-next into drm-xe-next

Get drm-xe-next on v6.11-rc2 and synchronized with drm-intel-next for
the display side. This resolves the current conflict for the
enable_display module parameter

Merge drm/drm-next into drm-xe-next

Get drm-xe-next on v6.11-rc2 and synchronized with drm-intel-next for
the display side. This resolves the current conflict for the
enable_display module parameter and allows further pending refactors.

Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

show more ...

Merge drm/drm-next into drm-intel-next

Sync with v6.11-rc1 in general, and specifically get the new
BACKLIGHT_POWER_ constants for power states.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Merge drm/drm-next into drm-misc-next

Backmerging to get a late RC of v6.10 before moving into v6.11.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>

Merge tag 'net-next-6.12' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next

Pull networking updates from Jakub Kicinski:
"The zero-copy changes are relatively significant, but regres

Merge tag 'net-next-6.12' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next

Pull networking updates from Jakub Kicinski:
"The zero-copy changes are relatively significant, but regression risk
should be contained. The feature needs to be used to cause trouble.

Also it feels like we got an order of magnitude more semi-automated
"refactoring" chaff than usual, I wonder if it's just us.

Core & protocols:

- Support Device Memory TCP, ability to zero-copy receive TCP
payloads to a DMABUF region of memory while packet headers land
separately in normal kernel buffers, and TCP processes then as
usual.

- The ability to read the PTP PHC (Physical Hardware Clock) alongside
MONOTONIC_RAW timestamps with PTP_SYS_OFFSET_EXTENDED. Previously
only CLOCK_REALTIME was supported.

- Allow matching on all bits of IP DSCP for routing decisions.
Previously we only supported on matching TOS bits in IPv4 which is
a narrower interpretation of the same header field.

- Increase the range of weights used for multi-path routing from
8 bits to 16 bits.

- Add support for IPv6 PIO p flag in the Prefix Information Option
per draft-ietf-6man-pio-pflag.

- IPv6 IOAM6 support for new tunsrc encap mode for better
performance.

- Detect destinations which blackhole MPTCP traffic and avoid
initiating MPTCP connections to them for a certain period of time,
1h by default.

- Improve IPsec control path performance by removing the inexact
policies list.

- AF_VSOCK: add support for SIOCOUTQ ioctl.

- Add enum for reasons TCP reset was sent for easier tracing.

- Add SMC ringbufs usage statistics.

Drivers:

- Handle netconsole setup failures more gracefully, don't fail
loading, retain the specified target as disabled.

- Extend bonding's IPsec offload pass thru capabilities (ESN, stats).

Filtering:

- Add TCP_BPF_SOCK_OPS_CB_FLAGS to bpf_*sockopt() to address the case
when long-lived sockets miss a chance to set additional callbacks
if a sockops program was not attached early in their lifetime.

- Support using BPF skb helpers in tracepoints.

- Conntrack Netlink: support CTA_FILTER for flush.

- Improve SCTP support in nfnetlink_queue.

- Improve performance of large nftables flush transactions.

Things we sprinkled into general kernel code:

- selftests: support setting an "interpreter" for script files; make
it easy to run as separate cases tests where one "interpreter" is
fed various test descriptions (in our case packet sequences).

Driver API:

- Extend core and ethtool APIs to support many PHYs connected to a
single interface (PHY topologies).

- Extend cable diagnostics to specify whether Time Domain
Reflectometry (TDR) or Active Link Cable Diagnostic (ALCD) was
used.

- Add library for implementing MAC-PHY Ethernet drivers for SPI
devices compatible with Open Alliance 10BASE-T1x MAC-PHY Serial
Interface (TC6) standard.

- Add helpers to the PHY framework, for PHYs following the Open
Alliance standards:
- 1000BaseT1 link settings
- cable test and diagnostics

- Support listing / dumping all allocated RSS contexts.

- Add configuration for frequency Embedded SYNC in DPLL, which
magically embeds sync pulses into Ethernet signaling.

Device drivers:

- Ethernet high-speed NICs:
- Broadcom (bnxt):
- use better FW APIs for queue reset
- support QOS and TPID settings for the SR-IOV VLAN
- support dynamic MSI-X allocation
- Intel (100G, ice, idpf):
- ice: support PCIe subfunctions
- iavf: add support for TC U32 filters on VFs
- ice: support Embedded SYNC in DPLL
- nVidia/Mellanox (mlx5):
- support HW managed steering tables
- support PCIe PTM cross timestamping
- AMD/Pensando:
- ionic: use page_pool to increase Rx performance
- Cisco (enic):
- report per-queue statistics

- Ethernet virtual:
- Microsoft vNIC:
- mana: support configuring ring length
- netvsc: enable more channels on systems with many CPUs
- IBM veth:
- optimize polling to improve TCP_RR performance
- optimize performance of Tx handling
- VirtIO net:
- synchronize the operstate with the admin state to allow a
lower virtio-net to propagate the link status to an upper
device like macvlan

- Ethernet NICs consumer, and embedded:
- Add driver for Realtek automotive PCIe devices (RTL9054,
RTL9068, RTL9072, RTL9075, RTL9068, RTL9071)
- Add driver for Microchip LAN8650/1 10BASE-T1S MAC-PHY.
- Microchip:
- lan743x: use phylink - support WOL, EEE, pause, link settings
- add Wake-on-LAN support for KSZ87xx family
- add KSZ8895/KSZ8864 switch support
- factor out FDMA code and use it in sparx5 and lan966x
(including DCB support in both)
- Synopsys (stmmac):
- support frame preemption (configured using TC and ethtool)
- support Loongson DWMAC (GMAC v3.73)
- support RockChips RK3576 DWMAC
- TI:
- am65-cpsw: add multi queue RX support
- icssg-prueth: HSR offload support
- Cadence (macb):
- enable software (hrtimer based) IRQ coalescing by default
- Xilinx (axinet):
- expose HW statistics
- improve multicast filtering
- relax Rx checksum offload constraints
- MediaTek:
- mt7530: add EN7581 support
- Aspeed (ftgmac100):
- report link speed and duplex
- Intel:
- igc: add mqprio offload
- igc: report EEE configuration
- RealTek (r8169):
- add support for RTL8126A rev.b
- Vitesse (vsc73xx):
- implement FDB add/del/dump operations
- Freescale (fs_enet):
- use phylink

- Ethernet PHYs:
- vitesse: implement downshift and MDI-X in vsc73xx PHYs
- microchip: support LAN887x, supporting IEEE 802.3bw (100BASE-T1)
and IEEE 802.3bp (1000BASE-T1) specifications
- add Applied Micro QT2025 PHY driver (in Rust)
- add Motorcomm yt8821 2.5G Ethernet PHY driver

- CAN:
- add driver for Rockchip RK3568 CAN-FD controller
- flexcan: add wakeup support for imx95
- kvaser_usb: set hardware timestamp on transmitted packets

- WiFi:
- mac80211/cfg80211:
- EHT rate support in AQL airtime fairness
- handle DFS (radar detection) per link in Multi-Link Operation
- RealTek (rtw89):
- support RTL8852BT and 8852BE-VT (WiFi 6)
- support hardware rfkill
- support HW encryption in unicast management frames
- support Wake-on-WLAN with supported network detection
- RealTek (rtw89):
- improve Rx performance by using USB frame aggregation
- support USB 3 with RTL8822CU/RTL8822BU
- Intel (iwlwifi/mvm):
- offload RLC/SMPS functionality to firmware
- Marvell (mwifiex):
- add host based MLME to enable WPA3

- Bluetooth:
- add support for Amlogic HCI UART protocol
- add support for ISO data/packets to Intel and NXP drivers"

* tag 'net-next-6.12' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1303 commits)
net/mlx5: HWS, check the correct variable in hws_send_ring_alloc_sq()
netfilter: nft_socket: Fix a NULL vs IS_ERR() bug in nft_socket_cgroup_subtree_level()
ice: Fix a NULL vs IS_ERR() check in probe()
ice: Fix a couple NULL vs IS_ERR() bugs
net: ethernet: fs_enet: Make the per clock optional
net: ti: icssg-prueth: Add multicast filtering support in HSR mode
net: ti: icssg-prueth: Enable HSR Tx duplication, Tx Tag and Rx Tag offload
net: ti: icssg-prueth: Add support for HSR frame forward offload
net: ti: icssg-prueth: Stop hardcoding def_inc
net: ti: icss-iep: Move icss_iep structure
net: ibm: emac: get rid of wol_irq
net: ibm: emac: remove all waiting code
net: ibm: emac: replace of_get_property
net: ibm: emac: use netdev's phydev directly
net: ibm: emac: use devm for register_netdev
net: ibm: emac: remove mii_bus with devm
net: ibm: emac: use devm for of_iomap
net: ibm: emac: manage emac_irq with devm
net: ibm: emac: use devm for alloc_etherdev
octeontx2-af: debugfs: Add Channel info to RPM map
...

show more ...

Merge branch 'ibmvnic-ibmvnic-rr-patchset'

Nick Child says:

====================
ibmvnic: ibmvnic rr patchset

v1 - https://lore.kernel.org/netdev/20240801212340.132607-1-nnac123@linux.ibm.com/
v2

Merge branch 'ibmvnic-ibmvnic-rr-patchset'

Nick Child says:

====================
ibmvnic: ibmvnic rr patchset

v1 - https://lore.kernel.org/netdev/20240801212340.132607-1-nnac123@linux.ibm.com/
v2 - https://lore.kernel.org/netdev/20240806193706.998148-1-nnac123@linux.ibm.com/
====================

Link: https://patch.msgid.link/20240807211809.1259563-1-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

show more ...

ibmvnic: Perform tx CSO during send scrq direct

During initialization with the vnic server, a bitstring is communicated
to the client regarding header info needed during CSO (See "VNIC
Capabilities"

ibmvnic: Perform tx CSO during send scrq direct

During initialization with the vnic server, a bitstring is communicated
to the client regarding header info needed during CSO (See "VNIC
Capabilities" in PAPR). Most of the time, to be safe, vnic server
requests header info for CSO. When header info is needed, multiple TX
descriptors are required per skb; This limits the driver to use
send_subcrq_indirect instead of send_subcrq_direct.

Previously, the vnic server request for header info was ignored. This
allowed the use of send_sub_crq_direct. Transmissions were successful
because the bitstring returned by vnic server is broad and over
cautionary. It was observed that mlx backing devices could actually
transmit and handle CSO packets without the vnic server receiving
header info (despite the fact that the bitstring requested it).

There was a trust issue: The bitstring was overcautionary. This extra
precaution (requesting header info when the backing device may not use
it) comes at the cost of performance (using direct vs indirect hcalls
has a 30% delta in small packet RR transaction rate). So it has been
requested that the vnic server team tries to ensure that the bitstring
is more exact. In the meantime, disable CSO when it is possible to use
the skb in the send_subcrq_direct path. In other words, calculate the
checksum before handing the packet to FW when the packet is not
segmented and xmit_more is false.

Since the code path is only possible if the skb is non GSO and xmit_more
is false, the cost of doing checksum in the send_subcrq_direct path is
minimal. Any large segmented skb will have xmit_more set to true more
frequently and it is inexpensive to do checksumming on a small skb.
The worst-case workload would be a 9000 MTU TCP_RR test with close
to MTU sized packets (and TSO off). This allows xmit_more to be false
more frequently and open the code path up to use send_subcrq_direct.
Observing trace data (graph-time = 1) and packet rate with this workload
shows minimal performance degradation:

1. NIC does checksum w headers, safely use send_subcrq_indirect:
- Packet rate: 631k txs
- Trace data:
ibmvnic_xmit = 44344685.87 us / 6234576 hits = AVG 7.11 us
skb_checksum_help = 4.07 us / 2 hits = AVG 2.04 us
^ Notice hits, tracing this just for reassurance
ibmvnic_tx_scrq_flush = 33040649.69 us / 5638441 hits = AVG 5.86 us
send_subcrq_indirect = 37438922.24 us / 6030859 hits = AVG 6.21 us

2. NIC does checksum w/o headers, dangerously use send_subcrq_direct:
- Packet rate: 831k txs
- Trace data:
ibmvnic_xmit = 48940092.29 us / 8187630 hits = AVG 5.98 us
skb_checksum_help = 2.03 us / 1 hits = AVG 2.03
ibmvnic_tx_scrq_flush = 31141879.57 us / 7948960 hits = AVG 3.92 us
send_subcrq_indirect = 8412506.03 us / 728781 hits = AVG 11.54
^ notice hits is much lower b/c send_subcrq_direct was called
^ wasn't traceable

3. driver does checksum, safely use send_subcrq_direct (THIS PATCH):
- Packet rate: 829k txs
- Trace data:
ibmvnic_xmit = 56696077.63 us / 8066168 hits = AVG 7.03 us
skb_checksum_help = 8587456.16 us / 7526072 hits = AVG 1.14 us
ibmvnic_tx_scrq_flush = 30219545.55 us / 7782409 hits = AVG 3.88 us
send_subcrq_indirect = 8638326.44 us / 763693 hits = AVG 11.31 us

When the bitstring ever specifies that CSO does not require headers
(dependent on VIOS vnic server changes), then this patch should be
removed and replaced with one that investigates the bitstring before
using send_subcrq_direct.

Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Link: https://patch.msgid.link/20240807211809.1259563-8-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

show more ...

ibmvnic: Only record tx completed bytes once per handler

Byte Queue Limits depends on dql_completed being called once per tx
completion round in order to adjust its algorithm appropriately. The
dql-

ibmvnic: Only record tx completed bytes once per handler

Byte Queue Limits depends on dql_completed being called once per tx
completion round in order to adjust its algorithm appropriately. The
dql->limit value is an approximation of the amount of bytes that the NIC
can consume per irq interval. If this approximation is too high then the
NIC will become over-saturated. Too low and the NIC will starve.

The dql->limit depends on dql->prev-* stats to calculate an optimal
value. If dql_completed() is called more than once per irq handler then
those prev-* values become unreliable (because they are not an accurate
representation of the previous state of the NIC) resulting in a
sub-optimal limit value.

Therefore, move the call to netdev_tx_completed_queue() to the end of
ibmvnic_complete_tx().

When performing 150 sessions of TCP rr (request-response 1 byte packets)
workloads, one could observe:
PREVIOUSLY: - limit and inflight values hovering around 130
- transaction rate of around 750k pps.

NOW: - limit rises and falls in response to inflight (130-900)
- transaction rate of around 1M pps (33% improvement)

Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Link: https://patch.msgid.link/20240807211809.1259563-7-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

show more ...

ibmvnic: Introduce send sub-crq direct

Firmware supports two hcalls to send a sub-crq request:
H_SEND_SUB_CRQ_INDIRECT and H_SEND_SUB_CRQ. The indirect hcall allows
for submission of batched message

ibmvnic: Introduce send sub-crq direct

Firmware supports two hcalls to send a sub-crq request:
H_SEND_SUB_CRQ_INDIRECT and H_SEND_SUB_CRQ. The indirect hcall allows
for submission of batched messages while the other hcall is limited to
only one message. This protocol is defined in PAPR section 17.2.3.3.

Previously, the ibmvnic xmit function only used the indirect hcall. This
allowed the driver to batch it's skbs. A single skb can occupy a few
entries per hcall depending on if FW requires skb header information or
not. The FW only needs header information if the packet is segmented.

By this logic, if an skb is not GSO then it can fit in one sub-crq
message and therefore is a candidate for H_SEND_SUB_CRQ.
Batching skb transmission is only useful when there are more packets
coming down the line (ie netdev_xmit_more is true).

As it turns out, H_SEND_SUB_CRQ induces less latency than
H_SEND_SUB_CRQ_INDIRECT. Therefore, use H_SEND_SUB_CRQ where
appropriate.

Small latency gains seen when doing TCP_RR_150 (request/response
workload). Ftrace results (graph-time=1):
Previous:
ibmvnic_xmit = 29618270.83 us / 8860058.0 hits = AVG 3.34
ibmvnic_tx_scrq_flush = 21972231.02 us / 6553972.0 hits = AVG 3.35
Now:
ibmvnic_xmit = 22153350.96 us / 8438942.0 hits = AVG 2.63
ibmvnic_tx_scrq_flush = 15858922.4 us / 6244076.0 hits = AVG 2.54

Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Link: https://patch.msgid.link/20240807211809.1259563-6-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

show more ...

ibmvnic: Remove duplicate memory barriers in tx

send_subcrq_[in]direct() already has a dma memory barrier.
Remove the earlier one.

Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Link: https://pa

ibmvnic: Remove duplicate memory barriers in tx

send_subcrq_[in]direct() already has a dma memory barrier.
Remove the earlier one.

Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Link: https://patch.msgid.link/20240807211809.1259563-5-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

show more ...

ibmvnic: Reduce memcpys in tx descriptor generation

Previously when creating the header descriptors, the driver would:
1. allocate a temporary buffer on the stack (in build_hdr_descs_arr)
2. memcpy

ibmvnic: Reduce memcpys in tx descriptor generation

Previously when creating the header descriptors, the driver would:
1. allocate a temporary buffer on the stack (in build_hdr_descs_arr)
2. memcpy the header info into the temporary buffer (in build_hdr_data)
3. memcpy the temp buffer into a local variable (in create_hdr_descs)
4. copy the local variable into the return buffer (in create_hdr_descs)

Since, there is no opportunity for errors during this process, the temp
buffer is not needed and work can be done on the return buffer directly.

Repurpose build_hdr_data() to only calculate the header lengths. Rename
it to get_hdr_lens().
Edit create_hdr_descs() to read from the skb directly and copy directly
into the returned useful buffer.

The process now involves less memory and write operations while
also being more readable.

Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Link: https://patch.msgid.link/20240807211809.1259563-4-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

show more ...

ibmvnic: Use header len helper functions on tx

Use the header length helper functions rather than trying to calculate
it within the driver. There are defined functions for mac and network
headers (s

ibmvnic: Use header len helper functions on tx

Use the header length helper functions rather than trying to calculate
it within the driver. There are defined functions for mac and network
headers (skb_mac_header_len and skb_network_header_len) but no such
function exists for the transport header length.

Also, hdr_data was memset during allocation to all 0's so no need to
memset again.

Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Link: https://patch.msgid.link/20240807211809.1259563-3-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

show more ...

ibmvnic: Only replenish rx pool when resources are getting low

Previously, the driver would replenish the rx pool if the polling
function consumed less than the budget. The logic being that the driv

ibmvnic: Only replenish rx pool when resources are getting low

Previously, the driver would replenish the rx pool if the polling
function consumed less than the budget. The logic being that the driver
did not exhaust its budget so that must mean that the driver is not busy
and has cycles to spare for replenishing the pool.

So pool replenishment happens on every poll which did not consume
the budget. This can very costly during request-response tests.

In fact, an extra ~100pps can be seen in TCP_RR_150 tests when we remove
this conditional. Trace results (ftrace, graph-time=1) for the poll
function are below:
Previous results:
ibmvnic_poll = 64951846.0 us / 4167628.0 hits = AVG 15.58
replenish_rx_pool = 17602846.0 us / 4710437.0 hits = AVG 3.74
Now:
ibmvnic_poll = 57673941.0 us / 4791737.0 hits = AVG 12.04
replenish_rx_pool = 3938171.6 us / 4314.0 hits = AVG 912.88

While the replenish function takes longer, it is hit less frequently
meaning the ibmvnic_poll function, on average, is faster.

Furthermore, this change does not have a negative effect on
performance bandwidth/latency measurements.

Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Link: https://patch.msgid.link/20240807211809.1259563-2-nnac123@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

show more ...

Merge tag 'v6.10-rc6' into for-6.11/block-post

Pull in v6.10-rc6 to resolve a conflict for the integrity cleanups.

* tag 'v6.10-rc6': (778 commits)
Linux 6.10-rc6
ata: ahci: Clean up sysfs file

Merge tag 'v6.10-rc6' into for-6.11/block-post

Pull in v6.10-rc6 to resolve a conflict for the integrity cleanups.

* tag 'v6.10-rc6': (778 commits)
Linux 6.10-rc6
ata: ahci: Clean up sysfs file on error
ata: libata-core: Fix double free on error
ata,scsi: libata-core: Do not leak memory for ata_port struct members
ata: libata-core: Fix null pointer dereference on error
x86-32: fix cmpxchg8b_emu build error with clang
x86: stop playing stack games in profile_pc()
i2c: testunit: discard write requests while old command is running
i2c: testunit: don't erase registers after STOP
tty: mxser: Remove __counted_by from mxser_board.ports[]
randomize_kstack: Remove non-functional per-arch entropy filtering
string: kunit: add missing MODULE_DESCRIPTION() macros
ata: libata-core: Add ATA_HORKAGE_NOLPM for all Crucial BX SSD1 models
MAINTAINERS: Update IOMMU tree location
tools/power turbostat: Add local build_bug.h header for snapshot target
tools/power turbostat: Fix unc freq columns not showing with '-q' or '-l'
tools/power turbostat: option '-n' is ambiguous
drm/drm_file: Fix pid refcounting race
kallsyms: rework symbol lookup return codes
gpiolib: cdev: Ignore reconfiguration without direction
...

Signed-off-by: Jens Axboe <axboe@kernel.dk>

show more ...