summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2024-11-15xfrm: Fix acquire state insertion.Steffen Klassert
A recent commit jumped over the dst hash computation and left the symbol uninitialized. Fix this by explicitly computing the dst hash before it is used. Fixes: 0045e3d80613 ("xfrm: Cache used outbound xfrm states at the policy.") Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
2024-11-14Merge branch 'net-make-rss-rxnfc-semantics-more-explicit'Jakub Kicinski
Edward Cree says: ==================== net: make RSS+RXNFC semantics more explicit The original semantics of ntuple filters with FLOW_RSS were not fully understood by all drivers, some ignoring the ring_cookie from the flow rule. Require this support to be explicitly declared by the driver for filters relying on it to be inserted, and add self- test coverage for this functionality. Also teach ethtool_check_max_channel() about this. ==================== Link: https://patch.msgid.link/cover.1731499021.git.ecree.xilinx@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14selftest: extend test_rss_context_queue_reconfigure for action additionEdward Cree
The combination of ntuple action (ring_cookie) and RSS context can cause an ntuple rule to target a higher queue than appears in any RSS indirection table or directly in the ntuple rule, since the two numbers are added together. Verify the logic that prevents reducing the queue count in this case. Signed-off-by: Edward Cree <ecree.xilinx@gmail.com> Link: https://patch.msgid.link/58276b800ab78c0a79c1918046ccae7fe45ba802.1731499022.git.ecree.xilinx@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14selftest: validate RSS+ntuple filters with nonzero ring_cookieEdward Cree
Test creates an ntuple filter with 'action 2' and an RSS context whose indirection table has entries 0 and 1. Resulting traffic should go to queues 2 and 3; verify that it never hits queues 0 and 1. Signed-off-by: Edward Cree <ecree.xilinx@gmail.com> Link: https://patch.msgid.link/114afdf4d2867f72ed27751e8e08fe8b128a8529.1731499022.git.ecree.xilinx@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14selftest: include dst-ip in ethtool ntuple rulesEdward Cree
sfc hardware does not support filters with only ipproto + dst-port; adding dst-ip to the flow spec allows the rss_ctx test to be run on these devices. Signed-off-by: Edward Cree <ecree.xilinx@gmail.com> Reviewed-by: Martin Habets <habetsm.xilinx@gmail.com> Link: https://patch.msgid.link/8e5d23c8f21310c23c080cc7bcd31b76f8fd3096.1731499022.git.ecree.xilinx@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14net: ethtool: account for RSS+RXNFC add semantics when checking channel countEdward Cree
In ethtool_check_max_channel(), the new RX count must not only cover the max queue indices in RSS indirection tables and RXNFC destinations separately, but must also, for RXNFC rules with FLOW_RSS, cover the sum of the destination queue and the maximum index in the associated RSS context's indirection table, since that is the highest queue that the rule can actually deliver traffic to. It could be argued that the max queue across all custom RSS contexts (ethtool_get_max_rss_ctx_channel()) need no longer be considered, since any context to which packets can actually be delivered will be targeted by some RXNFC rule and its max will thus be allowed for by ethtool_get_max_rxnfc_channel(). For simplicity we keep both checks, so even RSS contexts unused by any RXNFC rule must fit the channel count. Signed-off-by: Edward Cree <ecree.xilinx@gmail.com> Link: https://patch.msgid.link/43257d375434bef388e36181492aa4c458b88336.1731499022.git.ecree.xilinx@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14net: ethtool: only allow set_rxnfc with rss + ring_cookie if driver opts inEdward Cree
Ethtool ntuple filters with FLOW_RSS were originally defined as adding the base queue ID (ring_cookie) to the value from the indirection table, so that the same table could distribute over more than one set of queues when used by different filters. However, some drivers / hardware ignore the ring_cookie, and simply use the indirection table entries as queue IDs directly. Thus, for drivers which have not opted in by setting ethtool_ops.cap_rss_rxnfc_adds to declare that they support the original (addition) semantics, reject in ethtool_set_rxnfc any filter which combines FLOW_RSS and a nonzero ring. (For a ring_cookie of zero, both behaviours are equivalent.) Set the cap bit in sfc, as it is known to support this feature. Signed-off-by: Edward Cree <ecree.xilinx@gmail.com> Reviewed-by: Martin Habets <habetsm.xilinx@gmail.com> Link: https://patch.msgid.link/cc3da0844083b0e301a33092a6299e4042b65221.1731499022.git.ecree.xilinx@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14i40e: Fix handling changed priv flagsPeter Große
After assembling the new private flags on a PF, the operation to determine the changed flags uses the wrong bitmaps. Instead of xor-ing orig_flags with new_flags, it uses the still unchanged pf->flags, thus changed_flags is always 0. Fix it by using the correct bitmaps. The issue was discovered while debugging why disabling source pruning stopped working with release 6.7. Although the new flags will be copied to pf->flags later on in that function, disabling source pruning requires a reset of the PF, which was skipped due to this bug. Disabling source pruning: $ sudo ethtool --set-priv-flags eno1 disable-source-pruning on $ sudo ethtool --show-priv-flags eno1 Private flags for eno1: MFP : off total-port-shutdown : off LinkPolling : off flow-director-atr : on veb-stats : off hw-atr-eviction : off link-down-on-close : off legacy-rx : off disable-source-pruning: on disable-fw-lldp : off rs-fec : off base-r-fec : off vf-vlan-pruning : off Regarding reproducing: I observed the issue with a rather complicated lab setup, where * two VLAN interfaces are created on eno1 * each with a different MAC address assigned * each moved into a separate namespace * both VLANs are bridged externally, so they form a single layer 2 network The external bridge is done via a channel emulator adding packet loss and delay and the application in the namespaces tries to send/receive traffic and measure the performance. Sender and receiver are separated by namespaces, yet the network card "sees its own traffic" send back to it. To make that work, source pruning has to be disabled. Cc: stable@vger.kernel.org Fixes: 70756d0a4727 ("i40e: Use DECLARE_BITMAP for flags and hw_features fields in i40e_pf") Signed-off-by: Peter Große <pegro@friiks.de> Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@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> Link: https://patch.msgid.link/20241113210705.1296408-1-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14dt-bindings: net: sff,sfp: Fix "interrupts" property typoRob Herring (Arm)
The example has "interrupt" property which is not a defined property. It should be "interrupts" instead. "interrupts" also should not contain a phandle. Signed-off-by: Rob Herring (Arm) <robh@kernel.org> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Acked-by: Conor Dooley <conor.dooley@microchip.com> Link: https://patch.msgid.link/20241113225825.1785588-2-robh@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14dt-bindings: net: mdio-mux-gpio: Drop undocumented "marvell,reg-init"Rob Herring (Arm)
"marvell,reg-init" is not yet documented by schema. It's irrelevant to the example, so just drop it. Signed-off-by: Rob Herring (Arm) <robh@kernel.org> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Acked-by: Conor Dooley <conor.dooley@microchip.com> Link: https://patch.msgid.link/20241113225713.1784118-2-robh@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14net: sparx5: add missing lan969x Kconfig dependencyArnd Bergmann
The sparx5 switchdev driver can be built either with or without support for the Lan969x switch. However, it cannot be built-in when the lan969x driver is a loadable module because of a link-time dependency: arm-linux-gnueabi-ld: drivers/net/ethernet/microchip/sparx5/sparx5_main.o:(.rodata+0xd44): undefined reference to `lan969x_desc' Add a Kconfig dependency to reflect this in Kconfig, allowing all the valid configurations but forcing sparx5 to be a loadable module as well if lan969x is. Fixes: 98a01119608d ("net: sparx5: add compatible string for lan969x") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241113115513.4132548-1-arnd@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14selftests/bpf: Set test path for token/obj_priv_implicit_token_envvarIhor Solodrai
token/obj_priv_implicit_token_envvar test may fail in an environment where the process executing tests can not write to the root path. Example: https://github.com/libbpf/libbpf/actions/runs/11844507007/job/33007897936 Change default path used by the test to /tmp/bpf-token-fs, and make it runtime configurable via an environment variable. Signed-off-by: Ihor Solodrai <ihor.solodrai@pm.me> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241115003853.864397-1-ihor.solodrai@pm.me
2024-11-14net: enetc: clean up before returning in probe()Dan Carpenter
We recently added this error path. We need to call enetc_pci_remove() before returning. It cleans up the resources from enetc_pci_probe(). Fixes: 99100d0d9922 ("net: enetc: add preliminary support for i.MX95 ENETC PF") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Reviewed-by: Wei Fang <wei.fang@nxp.com> Link: https://patch.msgid.link/93888efa-c838-4682-a7e5-e6bf318e844e@stanley.mountain Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14net: phy: dp83869: fix status reporting for 1000base-x autonegotiationRomain Gantois
The DP83869 PHY transceiver supports converting from RGMII to 1000base-x. In this operation mode, autonegotiation can be performed, as described in IEEE802.3. The DP83869 has a set of fiber-specific registers located at offset 0xc00. When the transceiver is configured in RGMII-to-1000base-x mode, these registers are mapped onto offset 0, which should make reading the autonegotiation status transparent. However, the fiber registers at offset 0xc04 and 0xc05 follow the bit layout specified in Clause 37, and genphy_read_status() assumes a Clause 22 layout. Thus, genphy_read_status() doesn't properly read the capabilities advertised by the link partner, resulting in incorrect link parameters. Similarly, genphy_config_aneg() doesn't properly write advertised capabilities. Fix the 1000base-x autonegotiation procedure by replacing genphy_read_status() and genphy_config_aneg() with their Clause 37 equivalents. Fixes: a29de52ba2a1 ("net: dp83869: Add ability to advertise Fiber connection") Cc: stable@vger.kernel.org Signed-off-by: Romain Gantois <romain.gantois@bootlin.com> Link: https://patch.msgid.link/20241112-dp83869-1000base-x-v3-1-36005f4ab0d9@bootlin.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14mdio: Remove mdio45_ethtool_gset_npage()Alistair Francis
The mdio45_ethtool_gset_npage() function isn't called, so let's remove it. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Link: https://patch.msgid.link/20241112105430.438491-2-alistair@alistair23.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14include: mdio: Remove mdio45_ethtool_gset()Alistair Francis
mdio45_ethtool_gset() is never called, so let's remove it. Signed-off-by: Alistair Francis <alistair.francis@wdc.com> Link: https://patch.msgid.link/20241112105430.438491-1-alistair@alistair23.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14Merge tag 'for-netdev' of ↵Jakub Kicinski
https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next Martin KaFai Lau says: ==================== pull-request: bpf-next 2024-11-14 We've added 9 non-merge commits during the last 4 day(s) which contain a total of 3 files changed, 226 insertions(+), 84 deletions(-). The main changes are: 1) Fixes to bpf_msg_push/pop_data and test_sockmap. The changes has dependency on the other changes in the bpf-next/net branch, from Zijian Zhang. 2) Drop netns codes from mptcp test. Reuse the common helpers in test_progs, from Geliang Tang. * tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: bpf, sockmap: Fix sk_msg_reset_curr bpf, sockmap: Several fixes to bpf_msg_pop_data bpf, sockmap: Several fixes to bpf_msg_push_data selftests/bpf: Add more tests for test_txmsg_push_pop in test_sockmap selftests/bpf: Add push/pop checking for msg_verify_data in test_sockmap selftests/bpf: Fix total_bytes in msg_loop_rx in test_sockmap selftests/bpf: Fix SENDPAGE data logic in test_sockmap selftests/bpf: Add txmsg_pass to pull/push/pop in test_sockmap selftests/bpf: Drop netns helpers in mptcp ==================== Link: https://patch.msgid.link/20241114202832.3187927-1-martin.lau@linux.dev Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14Merge branch 'ipv4-prepare-bpf-helpers-to-flowi4_tos-conversion'Jakub Kicinski
Guillaume Nault says: ==================== ipv4: Prepare bpf helpers to .flowi4_tos conversion. Continue the process of making a dscp_t variable available when setting .flowi4_tos. This series focuses on the BPF helpers that initialise a struct flowi4 manually. The objective is to eventually convert .flowi4_tos to dscp_t, (to get type annotation and prevent ECN bits from interfering with DSCP). ==================== Link: https://patch.msgid.link/cover.1731064982.git.gnault@redhat.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14bpf: lwtunnel: Prepare bpf_lwt_xmit_reroute() to future .flowi4_tos conversion.Guillaume Nault
Use ip4h_dscp() to get the DSCP from the IPv4 header, then convert the dscp_t value to __u8 with inet_dscp_to_dsfield(). Then, when we'll convert .flowi4_tos to dscp_t, we'll just have to drop the inet_dscp_to_dsfield() call. Signed-off-by: Guillaume Nault <gnault@redhat.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> Link: https://patch.msgid.link/8338a12377c44f698a651d1ce357dd92bdf18120.1731064982.git.gnault@redhat.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14bpf: ipv4: Prepare __bpf_redirect_neigh_v4() to future .flowi4_tos conversion.Guillaume Nault
Use ip4h_dscp() to get the DSCP from the IPv4 header, then convert the dscp_t value to __u8 with inet_dscp_to_dsfield(). Then, when we'll convert .flowi4_tos to dscp_t, we'll just have to drop the inet_dscp_to_dsfield() call. Signed-off-by: Guillaume Nault <gnault@redhat.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> Link: https://patch.msgid.link/35eacc8955003e434afb1365d404193cc98a9579.1731064982.git.gnault@redhat.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14Merge branch 'tools-net-ynl-rework-async-notification-handling'Jakub Kicinski
Donald Hunter says: ==================== tools/net/ynl: rework async notification handling Revert patch 1bf70e6c3a53 which modified check_ntf() and instead add a new poll_ntf() with async notification semantics. See patch 2 for a detailed description. ==================== Link: https://patch.msgid.link/20241113090843.72917-1-donald.hunter@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14tools/net/ynl: add async notification handlingDonald Hunter
The notification handling in ynl is currently very simple, using sleep() to wait a period of time and then handling all the buffered messages in a single batch. This patch adds async notification handling so that messages can be processed as they are received. This makes it possible to use ynl as a library that supplies notifications in a timely manner. - Add poll_ntf() to be a generator that yields 1 notification at a time and blocks until a notification is available. - Add a --duration parameter to the CLI, with --sleep as an alias. ./tools/net/ynl/cli.py \ --spec <SPEC> --subscribe <TOPIC> [ --duration <SECS> ] The cli will report any notifications for duration seconds and then exit. If duration is not specified, then it will poll forever, until interrupted. Here is an example python snippet that shows how to use ynl as a library for receiving notifications: ynl = YnlFamily(f"{dir}/rt_route.yaml") ynl.ntf_subscribe('rtnlgrp-ipv4-route') for event in ynl.poll_ntf(): handle(event) Signed-off-by: Donald Hunter <donald.hunter@gmail.com> Link: https://patch.msgid.link/20241113090843.72917-3-donald.hunter@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14Revert "tools/net/ynl: improve async notification handling"Donald Hunter
This reverts commit 1bf70e6c3a5346966c25e0a1ff492945b25d3f80. This modification to check_ntf() is being reverted so that its behaviour remains equivalent to ynl_ntf_check() in the C YNL. Instead a new poll_ntf() will be added in a separate patch. Signed-off-by: Donald Hunter <donald.hunter@gmail.com> Link: https://patch.msgid.link/20241113090843.72917-2-donald.hunter@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14Input: ads7846 - increase xfer array size in 'struct ser_req'Nathan Chancellor
Clang warns (or errors with CONFIG_WERROR=y, trimmed for brevity): drivers/input/touchscreen/ads7846.c:412:2: error: array index 6 is past the end of the array (that has type 'struct spi_transfer[6]') [-Werror,-Warray-bounds] 412 | req->xfer[6].tx_buf = &req->scratch; | ^ ~ ... drivers/input/touchscreen/ads7846.c:416:2: error: array index 7 is past the end of the array (that has type 'struct spi_transfer[6]') [-Werror,-Warray-bounds] 416 | req->xfer[7].rx_buf = &req->scratch; | ^ ~ drivers/input/touchscreen/ads7846.c:334:2: note: array 'xfer' declared here 334 | struct spi_transfer xfer[6]; | ^ Increase the size of xfer in 'struct ser_req' to ensure that there is enough space for all necessary members and there are no out of bounds accesses, clearing up the warning. Fixes: 781a07da9bb9 ("Input: ads7846 - add dummy command register clearing cycle") Signed-off-by: Nathan Chancellor <nathan@kernel.org> Reviewed-by: Marek Vasut <marex@denx.de> Link: https://lore.kernel.org/r/20241111-input-ads7846-increase-xfer-array-size-v1-1-06cd92e9f20f@kernel.org Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
2024-11-14Input: fix the input_event struct documentationMaud Spierings
Fix the datatype of the value field of the input_event struct which is signed instead of unsigned. Signed-off-by: Maud Spierings <maud_spierings@hotmail.com> Link: https://lore.kernel.org/r/20241110-fix_input_doc-v1-1-745d5f908e61@hotmail.com Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
2024-11-14Merge branch ↵Jakub Kicinski
'net-phy-switch-eee_broken_modes-to-linkmode-bitmap-and-add-accessor' Heiner Kallweit says: ==================== net: phy: switch eee_broken_modes to linkmode bitmap and add accessor eee_broken_modes has a eee_cap1 register layout currently. This doesn't allow to flag e.g. 2.5Gbps or 5Gbps BaseT EEE as broken. To overcome this limitation switch eee_broken_modes to a linkmode bitmap. Add an accessor for the bitmap and use it in r8169. ==================== Link: https://patch.msgid.link/405734c5-0ed4-40e4-9ac9-91084b9536d6@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14r8169: copy vendor driver 2.5G/5G EEE advertisement constraintsHeiner Kallweit
Vendor driver r8125 doesn't advertise 2.5G EEE on RTL8125A, and r8126 doesn't advertise 5G EEE. Likely there are compatibility issues, therefore do the same in r8169. With this change we don't have to disable 2.5G EEE advertisement in rtl8125a_config_eee_phy() any longer. We use new phylib accessor phy_set_eee_broken() to mark the respective EEE modes as broken. Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Link: https://patch.msgid.link/ce185e10-8a2f-4cf8-a49b-fd8fb3c3c8a1@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14net: phy: add phy_set_eee_brokenHeiner Kallweit
Add an accessor for eee_broken_modes, so that drivers don't have to deal with phylib internals. Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Link: https://patch.msgid.link/0f8ee279-d40d-4489-a3b0-d993472d744a@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14net: phy: convert eee_broken_modes to a linkmode bitmapHeiner Kallweit
eee_broken_modes has a eee_cap1 register layout currently. This doen't allow to flag e.g. 2.5Gbps or 5Gbps BaseT EEE as broken. To overcome this limitation switch eee_broken_modes to a linkmode bitmap. Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Link: https://patch.msgid.link/dfe0c9ff-84b0-4328-86d7-e917ebc084a1@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-11-14sched_ext: Replace scx_next_task_picked() with switch_class() in commentZhao Mengmeng
scx_next_task_picked() has been replaced with siwtch_class(), but comment is still referencing old one, so replace it. Signed-off-by: Zhao Mengmeng <zhaomengmeng@kylinos.cn> Signed-off-by: Tejun Heo <tj@kernel.org>
2024-11-14iommu/arm-smmu-v3: Import IOMMUFD module namespaceNathan Chancellor
Commit 69d9b312f38a ("iommu/arm-smmu-v3: Support IOMMU_VIOMMU_ALLOC") started using _iommufd_object_alloc() without importing the IOMMUFD module namespace, resulting in a modpost warning: WARNING: modpost: module arm_smmu_v3 uses symbol _iommufd_object_alloc from namespace IOMMUFD, but does not import it. Commit d68beb276ba2 ("iommu/arm-smmu-v3: Support IOMMU_HWPT_INVALIDATE using a VIOMMU object") added another warning by using iommufd_viommu_find_dev(): WARNING: modpost: module arm_smmu_v3 uses symbol iommufd_viommu_find_dev from namespace IOMMUFD, but does not import it. Import the IOMMUFD module namespace to resolve the warnings. Fixes: 69d9b312f38a ("iommu/arm-smmu-v3: Support IOMMU_VIOMMU_ALLOC") Link: https://patch.msgid.link/r/20241114-arm-smmu-v3-import-iommufd-module-ns-v1-1-c551e7b972e9@kernel.org Signed-off-by: Nathan Chancellor <nathan@kernel.org> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2024-11-15powerpc/Makefile: Allow overriding CPPArnd Bergmann
Unlike all other arches, powerpc doesn't allow the user to override CPP, because it sets it unconditionally in the arch Makefile. This can lead to strange build failures. Instead add the required flags to KBUILD_CPPFLAGS, which are passed to CPP, CC and AS invocations by the generic Makefile logic. Reported-by: Arnd Bergmann <arnd@arndb.de> Tested-by: Nathan Chancellor <nathan@kernel.org> Reviewed-by: Nathan Chancellor <nathan@kernel.org> Closes: https://lore.kernel.org/all/20240607061629.530301-1-arnd@kernel.org Signed-off-by: Arnd Bergmann <arnd@arndb.de> [mpe: Rebase, write change log, add Arnd's SoB as communicated privately] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://patch.msgid.link/20241107112646.32401-1-mpe@ellerman.id.au
2024-11-14scftorture: Handle NULL argument passed to scf_add_to_free_list().Sebastian Andrzej Siewior
Dan reported that after the rework the newly introduced scf_add_to_free_list() may get a NULL pointer passed. This replaced kfree() which was fine with a NULL pointer but scf_add_to_free_list() isn't. Let scf_add_to_free_list() handle NULL pointer. Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Closes: https://lore.kernel.org/all/2375aa2c-3248-4ffa-b9b0-f0a24c50f237@stanley.mountain Fixes: 4788c861ad7e9 ("scftorture: Use a lock-less list to free memory.") Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2024-11-15powerpc-km82xx.c: replace of_node_put() with __freeDavid Hunter
The use of the __free macro allows the cleanup to be based on scope instead of on another function called later. This makes the cleanup automatic and less susceptible to errors later. Signed-off-by: David Hunter <david.hunter.linux@gmail.com> [mpe: Fix over-long line & change log wording] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://patch.msgid.link/20240709143553.117053-1-david.hunter.linux@gmail.com
2024-11-14mm, swap: fix allocation and scanning race with swapoffKairui Song
There are two flags used to synchronize allocation and scanning with swapoff: SWP_WRITEOK and SWP_SCANNING. SWP_WRITEOK: Swapoff will first unset this flag, at this point any further swap allocation or scanning on this device should just abort so no more new entries will be referencing this device. Swapoff will then unuse all existing swap entries. SWP_SCANNING: This flag is set when device is being scanned. Swapoff will wait for all scanner to stop before the final release of the swap device structures to avoid UAF. Note this flag is the highest used bit of si->flags so it could be added up arithmetically, if there are multiple scanner. commit 5f843a9a3a1e ("mm: swap: separate SSD allocation from scan_swap_map_slots()") ignored SWP_SCANNING and SWP_WRITEOK flags while separating cluster allocation path from the old allocation path. Add the flags back to fix swapoff race. The race is hard to trigger as si->lock prevents most parallel operations, but si->lock could be dropped for reclaim or discard. This issue is found during code review. This commit fixes this problem. For SWP_SCANNING, Just like before, set the flag before scan and remove it afterwards. For SWP_WRITEOK, there are several places where si->lock could be dropped, it will be error-prone and make the code hard to follow if we try to cover these places one by one. So just do one check before the real allocation, which is also very similar like before. With new cluster allocator it may waste a bit of time iterating the clusters but won't take long, and swapoff is not performance sensitive. Link: https://lkml.kernel.org/r/20241112083414.78174-1-ryncsn@gmail.com Fixes: 5f843a9a3a1e ("mm: swap: separate SSD allocation from scan_swap_map_slots()") Reported-by: "Huang, Ying" <ying.huang@intel.com> Closes: https://lore.kernel.org/linux-mm/87a5es3f1f.fsf@yhuang6-desk2.ccr.corp.intel.com/ Signed-off-by: Kairui Song <kasong@tencent.com> Cc: Barry Song <v-songbaohua@oppo.com> Cc: Chris Li <chrisl@kernel.org> Cc: Hugh Dickins <hughd@google.com> Cc: Kalesh Singh <kaleshsingh@google.com> Cc: Ryan Roberts <ryan.roberts@arm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-11-14clk: eyeq: add EyeQ6H west fixed factor clocksThéo Lebrun
Previous setup was: - pll-west clock registered from driver at of_clk_init(); - Both OCC and UART clocks registered from DT using fixed-factor-clock compatible. Now that drivers/clk/clk-eyeq.c supports registering fixed factors, use that capability to register west-per-occ and west-per-uart (giving them proper names at the same time). Also switch from hard-coded index 0 for pll-west to using the EQ6HC_WEST_PLL_PER constant by exposed dt-bindings headers. All get exposed at of_clk_init() because they get used by the AMBA PL011 serial ports. Those are instantiated before platform bus infrastructure. Signed-off-by: Théo Lebrun <theo.lebrun@bootlin.com> Link: https://lore.kernel.org/r/20241106-mbly-clk-v2-8-84cfefb3f485@bootlin.com Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14clk: eyeq: add EyeQ6H central fixed factor clocksThéo Lebrun
Previous setup was: - pll-cpu clock registered from driver at of_clk_init(); - occ-cpu clock registered from DT using fixed-factor-clock compatible. Now that drivers/clk/clk-eyeq.c supports registering fixed factors, use that capability to register occ-cpu. Also switch from hard-coded index 0 for pll-cpu to using the EQ6HC_CENTRAL_PLL_CPU constant by exposed dt-bindings headers. occ-cpu is exposed at of_clk_init() because it gets used by both the DT CPU nodes and the GIC timer. Signed-off-by: Théo Lebrun <theo.lebrun@bootlin.com> Link: https://lore.kernel.org/r/20241106-mbly-clk-v2-7-84cfefb3f485@bootlin.com Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14clk: eyeq: add EyeQ5 fixed factor clocksThéo Lebrun
Expose additional clocks on EyeQ5. Some indexes come from dt-bindings headers, others are private to the driver. Few clocks are early (of_clk_init() stage: core clocks and UART clock), others can wait until platform device probe. The source for this list is downstream Mobileye kernel and some internal documentation. Signed-off-by: Théo Lebrun <theo.lebrun@bootlin.com> Link: https://lore.kernel.org/r/20241106-mbly-clk-v2-6-84cfefb3f485@bootlin.com Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14clk: eyeq: add fixed factor clocks infrastructureThéo Lebrun
Driver can currently host two types of clocks: - PLLs derived directly from the main crystal (taken using a fwhandle). - Divider clocks derived from those PLLs. PLLs can be instantiated from of_clk_init() or platform device probe, using two separate clock providers. Divider clocks are all instantiated at platform device probe. Add a third type of clocks: fixed factors. Those can be instantiated at both stages. They can be parented to any clock from the driver. Early match data and match data store the list of fixed factor clocks. Signed-off-by: Théo Lebrun <theo.lebrun@bootlin.com> Link: https://lore.kernel.org/r/20241106-mbly-clk-v2-5-84cfefb3f485@bootlin.com Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14clk: eyeq: require clock index with phandle in all casesThéo Lebrun
We used to let compatibles with a single clock exposed to not have a cell. Switch away from that and enforce a cell in all cases. This is done at the same time as some compatibles (mobileye,eyeq6h-{central,west}-olb) go from one to more clocks exposed. Let's do the same switch and avoid future devicetree work if/when others follow. Signed-off-by: Théo Lebrun <theo.lebrun@bootlin.com> Link: https://lore.kernel.org/r/20241106-mbly-clk-v2-4-84cfefb3f485@bootlin.com Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14clk: fixed-factor: add clk_hw_register_fixed_factor_index() functionThéo Lebrun
Add non-devres version of clk_hw_register_fixed_factor(), with parent targeted using its index. Signed-off-by: Théo Lebrun <theo.lebrun@bootlin.com> Link: https://lore.kernel.org/r/20241106-mbly-clk-v2-3-84cfefb3f485@bootlin.com Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14dt-bindings: clock: eyeq: add more Mobileye EyeQ5/EyeQ6H clocksThéo Lebrun
Add #defines for Mobileye clock controller: - EyeQ5 core 0 thru 3 clocks. Internally: EQ5C_PLL_CPU: already exposed └── EQ5C_CPU_OCC: unexposed, no reason to do so ├── EQ5C_CPU_CORE0: new! ├── EQ5C_CPU_CORE1: new! ├── EQ5C_CPU_CORE2: new! └── EQ5C_CPU_CORE3: new! - EyeQ5 peripheral clocks. Internally: EQ5C_PLL_PER: already exposed ├── EQ5C_PER_OCC: new! │ ├── EQ5C_PER_SPI: new! │ ├── EQ5C_PER_I2C: new! │ ├── EQ5C_PER_GPIO: new! │ └── EQ5C_PER_UART: new! ├── EQ5C_PER_EMMC: new! └── EQ5C_PER_OCC_PCI: new! - EyeQ6H central OLB. Internally: EQ6HC_CENTRAL_PLL_CPU: new! └── EQ6HC_CENTRAL_CPU_OCC: new! - EyeQ6H west OLB. Internally: EQ6HC_WEST_PLL_PER: new! └── EQ6HC_WEST_PER_OCC: new! └── EQ6HC_WEST_PER_UART: new! Signed-off-by: Théo Lebrun <theo.lebrun@bootlin.com> Link: https://lore.kernel.org/r/20241106-mbly-clk-v2-2-84cfefb3f485@bootlin.com Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14dt-bindings: soc: mobileye: set `#clock-cells = <1>` for all compatiblesThéo Lebrun
Some compatibles expose a single clock. For those, we used to let them using `#clock-cells = <0>` (ie <&olb> reference rather than <&olb 0>). Switch away from that: enforce a cell for all compatibles. This is more straight forward, and avoids devicetree changes whenever a compatible goes from exposing a single clock to multiple ones. Also, dt-bindings get simpler. *This is an ABI break*. Change it while EyeQ5 platform support is at its infancy, without any user. More clocks might hide in each OLB as some registers are still unknown. Signed-off-by: Théo Lebrun <theo.lebrun@bootlin.com> Link: https://lore.kernel.org/r/20241106-mbly-clk-v2-1-84cfefb3f485@bootlin.com Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Reviewed-by: Rob Herring (Arm) <robh@kernel.org> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14clk: clk-axi-clkgen: make sure to enable the AXI bus clockNuno Sa
In order to access the registers of the HW, we need to make sure that the AXI bus clock is enabled. Hence let's increase the number of clocks by one. In order to keep backward compatibility and make sure old DTs still work we check if clock-names is available or not. If it is, then we can disambiguate between really having the AXI clock or a parent clock and so we can enable the bus clock. If not, we fallback to what was done before and don't explicitly enable the AXI bus clock. Note that if clock-names is given, the axi clock must be the last one in the phandle array (also enforced in the DT bindings) so that we can reuse as much code as possible. Fixes: 0e646c52cf0e ("clk: Add axi-clkgen driver") Signed-off-by: Nuno Sa <nuno.sa@analog.com> Link: https://lore.kernel.org/r/20241029-axi-clkgen-fix-axiclk-v2-2-bc5e0733ad76@analog.com Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14dt-bindings: clock: axi-clkgen: include AXI clkNuno Sa
In order to access the registers of the HW, we need to make sure that the AXI bus clock is enabled. Hence let's increase the number of clocks by one and add clock-names to differentiate between parent clocks and the bus clock. Fixes: 0e646c52cf0e ("clk: Add axi-clkgen driver") Signed-off-by: Nuno Sa <nuno.sa@analog.com> Link: https://lore.kernel.org/r/20241029-axi-clkgen-fix-axiclk-v2-1-bc5e0733ad76@analog.com Reviewed-by: Conor Dooley <conor.dooley@microchip.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14Merge tag 'v6.13-rockchip-clk1' of ↵Stephen Boyd
git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip into clk-bindings Pull a YAML conversion of the rk3328 clock controller binding from Heiko Stuebner. * tag 'v6.13-rockchip-clk1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip: dt-bindings: clock: convert rockchip,rk3328-cru.txt to YAML
2024-11-14clk: mmp: Add Marvell PXA1908 MPMU driverDuje Mihanović
Add driver for the MPMU controller block on Marvell's PXA1908 SoC. The driver is incomplete, currently only supporting the fixed PLL1; dynamic PLLs 2-4 and CPU/DDR/AXI clock support is missing. Signed-off-by: Duje Mihanović <duje.mihanovic@skole.hr> Link: https://lore.kernel.org/r/20241104-pxa1908-lkml-v13-8-e050609b8d6c@skole.hr Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14clk: mmp: Add Marvell PXA1908 APMU driverDuje Mihanović
Add driver for the APMU controller block found on Marvell's PXA1908 SoC. This driver is incomplete, lacking support for (at least) GPU, VPU, DSI and CCIC (camera related) clocks. Signed-off-by: Duje Mihanović <duje.mihanovic@skole.hr> Link: https://lore.kernel.org/r/20241104-pxa1908-lkml-v13-7-e050609b8d6c@skole.hr Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14clk: mmp: Add Marvell PXA1908 APBCP driverDuje Mihanović
Add driver for the APBCP controller block found on Marvell's PXA1908 SoC. Signed-off-by: Duje Mihanović <duje.mihanovic@skole.hr> Link: https://lore.kernel.org/r/20241104-pxa1908-lkml-v13-6-e050609b8d6c@skole.hr Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2024-11-14clk: mmp: Add Marvell PXA1908 APBC driverDuje Mihanović
Add driver for the APBC controller block found on Marvell's PXA1908 SoC. Signed-off-by: Duje Mihanović <duje.mihanovic@skole.hr> Link: https://lore.kernel.org/r/20241104-pxa1908-lkml-v13-5-e050609b8d6c@skole.hr Signed-off-by: Stephen Boyd <sboyd@kernel.org>
generated by cgit (git 2.34.1) at 2025-09-14 10:09:10 +0000