- 04 Oct, 2023 28 commits
-
-
Jakub Kicinski authored
Hayes Wang says: ==================== r8152: modify rx_bottom v3: For patch #1, this patch is replaced. The new patch only break the loop, and keep that the driver would queue the rx packets. For patch #2, modify the code depends on patch #1. For work_down < budget, napi_get_frags() and napi_gro_frags() would be used. For the others, nothing is changed. v2: For patch #1, add comment, update commit message, and add Fixes tag. v1: These patches are used to improve rx_bottom(). ==================== Link: https://lore.kernel.org/r/20230926111714.9448-432-nic_swsd@realtek.comSigned-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Hayes Wang authored
Use napi_gro_frags() for the skb of fragments when the work_done is less than budget. Signed-off-by:
Hayes Wang <hayeswang@realtek.com> Link: https://lore.kernel.org/r/20230926111714.9448-434-nic_swsd@realtek.comSigned-off-by:
-
Hayes Wang authored
A bulk transfer of the USB may contain many packets. And, the total number of the packets in the bulk transfer may be more than budget. Originally, only budget packets would be handled by napi_gro_receive(), and the other packets would be queued in the driver for next schedule. This patch would break the loop about getting next bulk transfer, when the budget is exhausted. That is, only the current bulk transfer would be handled, and the other bulk transfers would be queued for next schedule. Besides, the packets which are more than the budget in the current bulk trasnfer would be still queued in the driver, as the original method. In addition, a bulk transfer wouldn't contain more than 400 packets, so the check of queue length is unnecessary. Therefore, I replace it with WARN_ON_ONCE(). Fixes: cf74eb5a ("eth: r8152: try to use a normal budget") Signed-off-by:
-
Jakub Kicinski authored
Kees Cook says: ==================== chelsio: Annotate structs with __counted_by This annotates several chelsio structures with the coming __counted_by attribute for bounds checking of flexible arrays at run-time. For more details, see commit dd06e72e ("Compiler Attributes: Add __counted_by macro"). ==================== Link: https://lore.kernel.org/r/20230929181042.work.990-kees@kernel.orgSigned-off-by:
-
Kees Cook authored
Prepare for the coming implementation by GCC and Clang of the __counted_by attribute. Flexible array members annotated with __counted_by can have their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). As found with Coccinelle[1], add __counted_by for struct smt_data. [1] https://github.com/kees/kernel-tools/blob/trunk/coccinelle/examples/counted_by.cocci Cc: Raju Rangoju <rajur@chelsio.com> Signed-off-by:
Kees Cook <keescook@chromium.org> Reviewed-by:
Gustavo A. R. Silva <gustavoars@kernel.org> Link: https://lore.kernel.org/r/20230929181149.3006432-5-keescook@chromium.orgSigned-off-by:
-
Kees Cook authored
Prepare for the coming implementation by GCC and Clang of the __counted_by attribute. Flexible array members annotated with __counted_by can have their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). As found with Coccinelle[1], add __counted_by for struct sched_table. [1] https://github.com/kees/kernel-tools/blob/trunk/coccinelle/examples/counted_by.cocci Cc: Raju Rangoju <rajur@chelsio.com> Signed-off-by:
-
Kees Cook authored
Prepare for the coming implementation by GCC and Clang of the __counted_by attribute. Flexible array members annotated with __counted_by can have their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). As found with Coccinelle[1], add __counted_by for struct cxgb4_tc_u32_table. [1] https://github.com/kees/kernel-tools/blob/trunk/coccinelle/examples/counted_by.cocci Cc: Raju Rangoju <rajur@chelsio.com> Signed-off-by:
-
Kees Cook authored
Prepare for the coming implementation by GCC and Clang of the __counted_by attribute. Flexible array members annotated with __counted_by can have their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). As found with Coccinelle[1], add __counted_by for struct clip_tbl. [1] https://github.com/kees/kernel-tools/blob/trunk/coccinelle/examples/counted_by.cocci Cc: Raju Rangoju <rajur@chelsio.com> Signed-off-by:
-
Kees Cook authored
Prepare for the coming implementation by GCC and Clang of the __counted_by attribute. Flexible array members annotated with __counted_by can have their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). As found with Coccinelle[1], add __counted_by for struct l2t_data. [1] https://github.com/kees/kernel-tools/blob/trunk/coccinelle/examples/counted_by.cocci Cc: Raju Rangoju <rajur@chelsio.com> Signed-off-by:
-
https://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-nextJakub Kicinski authored
Florian Westphal says: ==================== netfilter updates for net-next First patch, from myself, is a bug fix. The issue (connect timeout) is ancient, so I think its safe to give this more soak time given the esoteric conditions needed to trigger this. Also updates the existing selftest to cover this. Add netlink extacks when an update references a non-existent table/chain/set. This allows userspace to provide much better errors to the user, from Pablo Neira Ayuso. Last patch adds more policy checks to nf_tables as a better alternative to the existing runtime checks, from Phil Sutter. * tag 'nf-next-23-09-28' of https://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next: netfilter: nf_tables: Utilize NLA_POLICY_NESTED_ARRAY netfilter: nf_tables: missing extended netlink error in lookup functions selftests: netfilter: test nat source port clash resolution interaction with tcp early demux netfilter: nf_nat: undo erroneous tcp edemux lookup after port clash ==================== Link: https://lore.kernel.org/r/20230928144916.18339-1-fw@strlen.deSigned-off-by:
-
Geert Uytterhoeven authored
Fix a misspelling of "preceding". Signed-off-by:
Geert Uytterhoeven <geert+renesas@glider.be> Reviewed-by:
Simon Horman <horms@kernel.org> Link: https://lore.kernel.org/r/663b14d07d6d716ddc34482834d6b65a2f714cfb.1695903447.git.geert+renesas@glider.beSigned-off-by:
-
Dan Carpenter authored
There is a copy and paste error so this uses a valid pointer instead of an error pointer. Fixes: 09eeb3ae ("ptp_ocp: implement DPLL ops") Signed-off-by:
Dan Carpenter <dan.carpenter@linaro.org> Reviewed-by:
Vadim Fedorenko <vadim.fedorenko@linux.dev> Link: https://lore.kernel.org/r/5c581336-0641-48bd-88f7-51984c3b1f79@moroto.mountainSigned-off-by:
-
Russell King (Oracle) authored
Remove the mt753x_phylink_pcs_link_up() function for two reasons: 1) priv->pcs[i].pcs.neg_mode is set true, meaning it doesn't take a MLO_AN_FIXED anymore, but one of PHYLINK_PCS_NEG_*. However, this is inconsequential due to... 2) priv->pcs[port].pcs.ops is always initialised to point at mt7530_pcs_ops, which does not have a pcs_link_up() member. So, let's remove mt753x_phylink_pcs_link_up() entirely. Signed-off-by:
Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by:
Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by:
Florian Fainelli <florian.fainelli@broadcom.com> Link: https://lore.kernel.org/r/E1qlTQS-008BWe-Va@rmk-PC.armlinux.org.ukSigned-off-by:
-
Greg Kroah-Hartman authored
The COPS Appletalk support is very old, never said to actually work properly, and the firmware code for the devices are under a very suspect license. Remove it all to clear up the license issue, if it is still needed and actually used by anyone, we can add it back later once the license is cleared up. Reported-by:
Prarit Bhargava <prarit@redhat.com> Cc: jschlst@samba.org Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Acked-by:
Christoph Hellwig <hch@lst.de> Acked-by:
Vitaly Kuznetsov <vkuznets@redhat.com> Link: https://lore.kernel.org/r/20230927090029.44704-2-gregkh@linuxfoundation.orgSigned-off-by:
-
Jakub Kicinski authored
This reverts commit e76d44fe. We no longer accept drivers extending their use of the legacy SR-IOV configuration APIs. Users should move to bridge offload. Link: https://lore.kernel.org/r/20231004112243.41cb6351@kernel.org/Signed-off-by:
-
Fabio Estevam authored
The imx8dl FEC has the same programming model as the one on the imx8qxp. Add the imx8dl compatible string. Signed-off-by:
Fabio Estevam <festevam@denx.de> Acked-by:
Conor Dooley <conor.dooley@microchip.com> Link: https://lore.kernel.org/r/20230926111017.320409-1-festevam@gmail.comSigned-off-by:
-
Jacob Keller authored
The recent support for DPLL introduced by commit 8a3a565f ("ice: add admin commands to access cgu configuration") and commit d7999f5e ("ice: implement dpll interface to control cgu") broke linking the ice driver if CONFIG_PTP_1588_CLOCK=n: ld: vmlinux.o: in function `ice_init_feature_support': (.text+0x8702b8): undefined reference to `ice_is_phy_rclk_present' ld: (.text+0x8702cd): undefined reference to `ice_is_cgu_present' ld: (.text+0x8702d9): undefined reference to `ice_is_clock_mux_present_e810t' ld: vmlinux.o: in function `ice_dpll_init_info_direct_pins': ice_dpll.c:(.text+0x894167): undefined reference to `ice_cgu_get_pin_freq_supp' ld: ice_dpll.c:(.text+0x894197): undefined reference to `ice_cgu_get_pin_name' ld: ice_dpll.c:(.text+0x8941a8): undefined reference to `ice_cgu_get_pin_type' ld: vmlinux.o: in function `ice_dpll_update_state': ice_dpll.c:(.text+0x894494): undefined reference to `ice_get_cgu_state' ld: vmlinux.o: in function `ice_dpll_init': (.text+0x8953d5): undefined reference to `ice_get_cgu_rclk_pin_info' The first commit broke things by calling functions in ice_init_feature_support that are compiled as part of ice_ptp_hw.o, including: * ice_is_phy_rclk_present * ice_is_clock_mux_present_e810t * ice_is_cgU_present The second commit continued the break by calling several CGU functions defined in ice_ptp_hw.c in the DPLL code. Because the ice_dpll.c file is compiled unconditionally, it will not link when CONFIG_PTP_1588_CLOCK=n. It might be possible to break this dependency and expose those functions without CONFIG_PTP_1588_CLOCK, but that is not clear to me. For the DPLL case, simply compile ice_dpll.o only when we have CONFIG_PTP_1588_CLOCK. Add stub no-op implementation of ice_dpll_init() and ice_dpll_uninit() when CONFIG_PTP_1588_CLOCK=n into ice_dpll.h The other functions are part of checking the netlist to see if hardware features are enabled. These checks don't really belong in ice_ptp_hw.c, and make more sense as part of the ice_common.c file. We already have ice_is_gps_in_netlist() in ice_common.c which is doing a similar check. Move the functions into ice_common.c and rename them to have the similar postfix of "in_netlist()" to be more expressive of what they are actually checking. This also makes the ice_find_netlist_node only called from within ice_common.c, so its safe to mark it static and stop declaring it in the ice_common.h header as well. Fixes: 8a3a565f ("ice: add admin commands to access cgu configuration") Fixes: d7999f5e ("ice: implement dpll interface to control cgu") Reported-by:
kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202309191214.TaYEct4H-lkp@intel.comSigned-off-by:
Jacob Keller <jacob.e.keller@intel.com> Reviewed-by:
Przemek Kitszel <przemyslaw.kitszel@intel.com> Signed-off-by:
Tony Nguyen <anthony.l.nguyen@intel.com> Reviewed-by:
-
David S. Miller authored
Michael Chan says: ==================== bnxt_en: hwmon and SRIOV updates The first 7 patches are v2 of the hwmon patches posted about 6 weeks ago on Aug 14. The last 2 patches are SRIOV related updates. Link to v1 hwmon patches: https://lore.kernel.org/netdev/20230815045658.80494-11-michael.chan@broadcom.com/ ==================== Reviewed-by:
David S. Miller <davem@davemloft.net>
-
Vikas Gupta authored
Newer versions of firmware will pre-reserve 1 VNIC for every possible PF and VF function. Update the driver logic to take this into account when assigning VNICs to the VFs. These pre-reserved VNICs for the inactive VFs should be subtracted from the global pool before assigning them to the active VFs. Not doing so may cause discrepancies that ultimately may cause some VFs to have insufficient VNICs to support features such as aRFS. Signed-off-by:
Vikas Gupta <vikas.gupta@broadcom.com> Signed-off-by:
Michael Chan <michael.chan@broadcom.com> Signed-off-by:
-
Sreekanth Reddy authored
Add these missing settings in the .ndo_set_vf_vlan() method. Older firmware does not support the TPID setting so check for proper support. Remove the unused BNXT_VF_QOS flag. Signed-off-by:
Sreekanth Reddy <sreekanth.reddy@broadcom.com> Signed-off-by:
-
Kalesh AP authored
Newer FW will send a new async event when it detects that the chip's temperature has crossed the configured threshold value. The driver will now notify hwmon and will log a warning message. Link: https://lore.kernel.org/netdev/20230815045658.80494-13-michael.chan@broadcom.com/ Cc: Jean Delvare <jdelvare@suse.com> Cc: Guenter Roeck <linux@roeck-us.net> Cc: linux-hwmon@vger.kernel.org Signed-off-by:
Kalesh AP <kalesh-anakkur.purayil@broadcom.com> Signed-off-by:
Guenter Roeck <linux@roeck-us.net> Signed-off-by:
-
Kalesh AP authored
Implement the sysfs attributes directly in the driver for shutdown threshold temperature and pass an extra attribute group to the hwmon core when registering the hwmon device. Link: https://lore.kernel.org/netdev/20230815045658.80494-12-michael.chan@broadcom.com/ Cc: Jean Delvare <jdelvare@suse.com> Cc: Guenter Roeck <linux@roeck-us.net> Cc: linux-hwmon@vger.kernel.org Reviewed-by:
Andy Gospodarek <andrew.gospodarek@broadcom.com> Signed-off-by:
-
Kalesh AP authored
HWRM_TEMP_MONITOR_QUERY response now indicates various threshold temperatures. Expose these threshold temperatures through the hwmon sysfs using this mapping: hwmon_temp_max : bp->warn_thresh_temp hwmon_temp_crit : bp->crit_thresh_temp hwmon_temp_emergency : bp->fatal_thresh_temp hwmon_temp_max_alarm : temp >= bp->warn_thresh_temp hwmon_temp_crit_alarm : temp >= bp->crit_thresh_temp hwmon_temp_emergency_alarm : temp >= bp->fatal_thresh_temp Link: https://lore.kernel.org/netdev/20230815045658.80494-12-michael.chan@broadcom.com/ Cc: Jean Delvare <jdelvare@suse.com> Cc: Guenter Roeck <linux@roeck-us.net> Cc: linux-hwmon@vger.kernel.org Signed-off-by:
-
Kalesh AP authored
The use of hwmon_device_register_with_groups() is deprecated. Modified the driver to use hwmon_device_register_with_info(). Driver currently exports only temp1_input through hwmon sysfs interface. But FW has been modified to report more threshold temperatures and driver want to report them through the hwmon interface. Cc: Jean Delvare <jdelvare@suse.com> Cc: Guenter Roeck <linux@roeck-us.net> Cc: linux-hwmon@vger.kernel.org Signed-off-by:
-
Kalesh AP authored
This is in preparation for upcoming patches in the series. Driver has to expose more threshold temperatures through the hwmon sysfs interface. More code will be added and do not want to overload bnxt.c. Reviewed-by:
-
Kalesh AP authored
Driver currently does hwmon device register and unregister in open and close() respectively. As a result, user will not be able to query hwmon temperature when interface is in ifdown state. Enhance it by moving the hwmon register/unregister to the probe/remove functions. Reviewed-by:
Somnath Kotur <somnath.kotur@broadcom.com> Signed-off-by:
-
Michael Chan authored
The main changes are the additional thermal thresholds in hwrm_temp_monitor_query_output and the new async event to report thermal errors. Signed-off-by:
-
Lukas Bulwahn authored
Commit 9431063a ("dpll: core: Add DPLL framework base functions") adds the section DPLL SUBSYSTEM in MAINTAINERS and includes a file entry to the non-existing file 'include/net/dpll.h'. Hence, ./scripts/get_maintainer.pl --self-test=patterns complains about a broken reference. Looking at the file stat of the commit above, this entry clearly intended to refer to 'include/linux/dpll.h'. Adjust this header file entry in DPLL SUBSYSTEM. Signed-off-by:
Lukas Bulwahn <lukas.bulwahn@gmail.com> Reviewed-by:
-
- 03 Oct, 2023 12 commits
-
-
Patrick Rohr authored
This change adds a sysctl to opt-out of RFC4862 section 5.5.3e's valid lifetime derivation mechanism. RFC4862 section 5.5.3e prescribes that the valid lifetime in a Router Advertisement PIO shall be ignored if it less than 2 hours and to reset the lifetime of the corresponding address to 2 hours. An in-progress 6man draft (see draft-ietf-6man-slaac-renum-07 section 4.2) is currently looking to remove this mechanism. While this draft has not been moving particularly quickly for other reasons, there is widespread consensus on section 4.2 which updates RFC4862 section 5.5.3e. Cc: Maciej Żenczykowski <maze@google.com> Cc: Lorenzo Colitti <lorenzo@google.com> Cc: Jen Linkova <furry@google.com> Signed-off-by:
Patrick Rohr <prohr@google.com> Reviewed-by:
Jiri Pirko <jiri@nvidia.com> Reviewed-by:
David Ahern <dsahern@kernel.org> Link: https://lore.kernel.org/r/20230925214711.959704-1-prohr@google.comSigned-off-by:
-
Jakub Kicinski authored
Bagas Sanjaya says: ==================== Documentation fixes for dpll subsystem Here is a mini docs fixes for dpll subsystem. The fixes are all code block-related. This series is triggered because I was emailed by kernel test robot, alerting htmldocs warnings (see patch [1/2]). [1]: https://lore.kernel.org/all/20230918093240.29824-1-bagasdotme@gmail.com/ ==================== Link: https://lore.kernel.org/r/20230928052708.44820-1-bagasdotme@gmail.comSigned-off-by:
-
Bagas Sanjaya authored
DPLL_CMD_PIN_GET netlink command output for mux-type pins looks ugly with normal paragraph formatting. Format it as a code block instead. Signed-off-by:
Bagas Sanjaya <bagasdotme@gmail.com> Acked-by:
Randy Dunlap <rdunlap@infradead.org> Tested-by:
-
Bagas Sanjaya authored
kernel test robot and Stephen Rothwell report htmldocs warnings: Documentation/driver-api/dpll.rst:427: WARNING: Error in "code-block" directive: maximum 1 argument(s) allowed, 18 supplied. .. code-block:: c <snipped>... Documentation/driver-api/dpll.rst:444: WARNING: Error in "code-block" directive: maximum 1 argument(s) allowed, 21 supplied. .. code-block:: c <snipped>... Documentation/driver-api/dpll.rst:474: WARNING: Error in "code-block" directive: maximum 1 argument(s) allowed, 12 supplied. .. code-block:: c <snipped>... Fix these above by adding missing blank line separator after code-block directive. Reported-by:
Stephen Rothwell <sfr@canb.auug.org.au> Closes: https://lore.kernel.org/linux-next/20230918131521.155e9e63@canb.auug.org.au/ Fixes: dbb291f1 ("dpll: documentation on DPLL subsystem interface") Signed-off-by:
-
Jakub Kicinski authored
Przemek Kitszel says: ==================== introduce DEFINE_FLEX() macro Add DEFINE_FLEX() macro, that helps on-stack allocation of structures with trailing flex array member. Expose __struct_size() macro which reads size of data allocated by DEFINE_FLEX(). Accompany new macros introduction with actual usage, in the ice driver - hence targeting for netdev tree. Obvious benefits include simpler resulting code, less heap usage, less error checking. Less obvious is the fact that compiler has more room to optimize, and as a whole, even with more stuff on the stack, we end up with overall better (smaller) report from bloat-o-meter: add/remove: 8/6 grow/shrink: 7/18 up/down: 2211/-2270 (-59) (individual results in each patch). ==================== Link: https://lore.kernel.org/r/20230912115937.1645707-1-przemyslaw.kitszel@intel.comSigned-off-by:
-
Przemek Kitszel authored
Use DEFINE_FLEX() macro for 1-elem flex array members of ice_switch.c Signed-off-by:
-
Przemek Kitszel authored
Use DEFINE_FLEX() macro for 1-elem flex array use case of struct ice_aqc_dis_txq_item. Signed-off-by:
-
Przemek Kitszel authored
Use DEFINE_FLEX() macro for 1-elem flex array use case of struct ice_aqc_add_tx_qgrp. Signed-off-by:
-
Przemek Kitszel authored
Use DEFINE_FLEX() macro for constant-num-of-elems (4) flex array members of ice_ddp.c Signed-off-by:
-
Przemek Kitszel authored
Remove two arguments of ice_aq_move_sched_elems(). Last of them was always NULL, and @grps_req was always 1. Assuming @grps_req to be one, allows us to use DEFINE_FLEX() macro, what removes some need for heap allocations. Signed-off-by:
-
Przemek Kitszel authored
Replace array+size params of ice_sched_remove_elems:() by just single u32, as all callers are using it with "1". This enables moving from heap-based, to stack-based allocation, what is also more elegant thanks to DEFINE_FLEX() macro. Signed-off-by:
-
Przemek Kitszel authored
Add DEFINE_FLEX() macro for on-stack allocations of structs with flexible array member. Expose __struct_size() macro outside of fortify-string.h, as it could be used to read size of structs allocated by DEFINE_FLEX(). Move __member_size() alongside it. -Kees Using underlying array for on-stack storage lets us to declare known-at-compile-time structures without kzalloc(). Actual usage for ice driver is in following patches of the series. Missing __has_builtin() workaround is moved up to serve also assembly compilation with m68k-linux-gcc, see [1]. Error was (note the .S file extension): In file included from ../include/linux/linkage.h:5, from ../arch/m68k/fpsp040/skeleton.S:40: ../include/linux/compiler_types.h:331:5: warning: "__has_builtin" is not defined, evaluates to 0 [-Wundef] 331 | #if __has_builtin(__builtin_dynamic_object_size) | ^~~~~~~~~~~~~ ../include/linux/compiler_types.h:331:18: error: missing binary operator before token "(" 331 | #if __has_builtin(__builtin_dynamic_object_size) | ^ [1] https://lore.kernel.org/netdev/202308112122.OuF0YZqL-lkp@intel.com/Co-developed-by:
-
