- 12 Dec, 2023 17 commits
-
-
Ilpo Järvinen authored
Use pcie_capability_read_word() for reading LNKSTA and remove the custom define that matches to PCI_EXP_LNKSTA. As only single user for cap_offset remains, replace it with a call to pci_pcie_cap(). Instead of e1000_adapter, make local variable out of pci_dev because both users are interested in it. Signed-off-by:
Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Reviewed-by:
Jonathan Cameron <Jonathan.Cameron@huawei.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>
-
Ilpo Järvinen authored
e1000e has own copy of PCI Negotiated Link Width field defines. Use the ones from include/uapi/linux/pci_regs.h instead of the custom ones and remove the custom ones and convert to FIELD_GET(). Suggested-by:
-
Ilpo Järvinen authored
Use FIELD_GET() to extract PCIe Negotiated Link Width field instead of custom masking and shifting. Signed-off-by:
-
Paolo Abeni authored
Linus Walleij says: ==================== net: dsa: realtek: Two RTL8366RB fixes These minor fixes were found while digging into other issues: a weirdly named variable and bogus MTU handling. Fix it up. Signed-off-by:
Linus Walleij <linus.walleij@linaro.org> ==================== Link: https://lore.kernel.org/r/20231209-rtl8366rb-mtu-fix-v1-0-df863e2b2b2a@linaro.orgSigned-off-by:
Paolo Abeni <pabeni@redhat.com>
-
Linus Walleij authored
The MTU callbacks are in layer 1 size, so for example 1500 bytes is a normal setting. Cache this size, and only add the layer 2 framing right before choosing the setting. On the CPU port this will however include the DSA tag since this is transmitted from the parent ethernet interface! Add the layer 2 overhead such as ethernet and VLAN framing and FCS before selecting the size in the register. This will make the code easier to understand. The rtl8366rb_max_mtu() callback returns a bogus MTU just subtracting the CPU tag, which is the only thing we should NOT subtract. Return the correct layer 1 max MTU after removing headers and checksum. Signed-off-by:
Alvin Šipraga <alsi@bang-olufsen.dk> Reviewed-by:
Florian Fainelli <florian.fainelli@broadcom.com> Signed-off-by:
-
Linus Walleij authored
Rename the register name to RTL8366RB instead of the bogus RTL8368S (internal product name?) Signed-off-by:
Luiz Angelo Daros de Luca <luizluca@gmail.com> Reviewed-by:
-
Ahelenia Ziemiańska authored
$ modinfo dnsresolver dns_resolver | grep name modinfo: ERROR: Module dnsresolver not found. filename: /lib/modules/6.1.0-9-amd64/kernel/net/dns_resolver/dns_resolver.ko name: dns_resolver Signed-off-by:
Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz> Link: https://lore.kernel.org/r/gh4sxphjxbo56n2spgmc66vtazyxgiehpmv5f2gkvgicy6f4rs@tarta.nabijaczleweli.xyzSigned-off-by:
-
Andy Shevchenko authored
The driver is using iowriteXX()/ioreadXX() APIs which are LE IO accessors simplified as 1. Convert given value _from_ CPU _to_ LE 2. Write it to the device as is The dev_addr is a byte stream, but because the driver uses 16-bit IO accessors, it wants to perform double conversion on BE CPUs, but it took it wrong, as it effectivelly does two times _from_ CPU _to_ LE. What it has to do is to consider dev_addr as an array of LE16 and hence do _from_ LE _to_ CPU conversion, followed by implied _from_ CPU _to_ LE in the iowrite16(). To achieve that, use get_unaligned_le16(). This will make it correct and allows to avoid sparse warning as reported by LKP. Reported-by:
kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202312030058.hfZPTXd7-lkp@intel.com/Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Link: https://lore.kernel.org/r/20231208153327.3306798-1-andriy.shevchenko@linux.intel.comSigned-off-by:
-
Swarup Laxman Kotiaklapudi authored
Add some missing(not all) attributes in devlink.yaml. Signed-off-by:
Swarup Laxman Kotiaklapudi <swarupkotikalapudi@gmail.com> Suggested-by:
Jiri Pirko <jiri@resnulli.us> Reviewed-by:
Jiri Pirko <jiri@nvidia.com> Link: https://lore.kernel.org/r/20231208182515.1206616-1-swarupkotikalapudi@gmail.comSigned-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
Pedro Tammela says: ==================== net/sched: conditional notification of events for cls and act This is an optimization we have been leveraging on P4TC but we believe it will benefit rtnl users in general. It's common to allocate an skb, build a notification message and then broadcast an event. In the absence of any user space listeners, these resources (cpu and memory operations) are wasted. In cases where the subsystem is lockless (such as in tc-flower) this waste is more prominent. For the scenarios where the rtnl_lock is held it is not as prominent. The idea is simple. Build and send the notification iif: - The user requests via NLM_F_ECHO or - Someone is listening to the rtnl group (tc mon) On a simple test with tc-flower adding 1M entries, using just a single core, there's already a noticeable difference in the cycles spent in tc_new_tfilter with this patchset. before: - 43.68% tc_new_tfilter + 31.73% fl_change + 6.35% tfilter_notify + 1.62% nlmsg_notify 0.66% __tcf_qdisc_find.part.0 0.64% __tcf_chain_get 0.54% fl_get + 0.53% tcf_proto_lookup_ops after: - 39.20% tc_new_tfilter + 34.58% fl_change 0.69% __tcf_qdisc_find.part.0 0.67% __tcf_chain_get + 0.61% tcf_proto_lookup_ops Note, the above test is using iproute2:tc which execs a shell. We expect people using netlink directly to observe even greater reductions. The qdisc side needs some refactoring of the notification routines to fit in this new model, so they will be sent in a later patchset. ==================== Link: https://lore.kernel.org/r/20231208192847.714940-1-pctammela@mojatatu.comSigned-off-by: -
Pedro Tammela authored
As of today tc-filter/chain events are unconditionally built and sent to RTNLGRP_TC. As with the introduction of rtnl_notify_needed we can check before-hand if they are really needed. This will help to alleviate system pressure when filters are concurrently added without the rtnl lock as in tc-flower. Reviewed-by:
Pedro Tammela <pctammela@mojatatu.com> Link: https://lore.kernel.org/r/20231208192847.714940-8-pctammela@mojatatu.comSigned-off-by:
-
Pedro Tammela authored
This argument is never called while set to true, so remove it as there's no need for it. Signed-off-by:
-
Pedro Tammela authored
As of today tc-action events are unconditionally built and sent to RTNLGRP_TC. As with the introduction of rtnl_notify_needed we can check before-hand if they are really needed. Signed-off-by:
-
Pedro Tammela authored
Use max() in a couple of places that are open coding it with the ternary operator. Signed-off-by:
-
Pedro Tammela authored
This is a convenience helper for routines handling conditional rtnl events, that is code that might send a notification depending on rtnl_has_listeners/rtnl_notify_needed. Instead of: if (skb) rtnetlink_send(...) Use: rtnetlink_maybe_send(...) Reviewed-by: -
Victor Nogueira authored
Building on the rtnl_has_listeners helper, add the rtnl_notify_needed helper to check if we can bail out early in the notification routines. Reviewed-by:
Victor Nogueira <victor@mojatatu.com> Signed-off-by:
-
Jamal Hadi Salim authored
As of today, rtnl code creates a new skb and unconditionally fills and broadcasts it to the relevant group. For most operations this is okay and doesn't waste resources in general. When operations are done without the rtnl_lock, as in tc-flower, such skb allocation, message fill and no-op broadcasting can happen in all cores of the system, which contributes to system pressure and wastes precious cpu cycles when no one will receive the built message. Introduce this helper so rtnetlink operations can simply check if someone is listening and then proceed if necessary. Reviewed-by:
Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by:
-
- 11 Dec, 2023 5 commits
-
-
David S. Miller authored
Eric Dumazet says: ==================== ipv6: more data-race annotations Small follow up series, taking care of races around np->mcast_oif and np->ucast_oif. ==================== Signed-off-by:David S. Miller <davem@davemloft.net>
-
Eric Dumazet authored
np->ucast_oif is read locklessly in some contexts. Make all accesses to this field lockless, adding appropriate annotations. This also makes setsockopt( IPV6_UNICAST_IF ) lockless. Signed-off-by:
Eric Dumazet <edumazet@google.com> Reviewed-by:
David Ahern <dsahern@kernel.org> Signed-off-by:
-
Eric Dumazet authored
np->mcast_oif is read locklessly in some contexts. Make all accesses to this field lockless, adding appropriate annotations. This also makes setsockopt( IPV6_MULTICAST_IF ) lockless. Signed-off-by:
-
Johannes Berg authored
This reverts commit b8dbbbc5 ("net: rtnetlink: remove local list in __linkwatch_run_queue()"). It's evidently broken when there's a non-urgent work that gets added back, and then the loop can never finish. While reverting, add a note about that. Reported-by:
Marek Szyprowski <m.szyprowski@samsung.com> Fixes: b8dbbbc5 ("net: rtnetlink: remove local list in __linkwatch_run_queue()") Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
-
Fei Qin authored
The device supports UDP hardware segmentation offload, which helps improving the performance. Thus, this patch adds support for UDP segmentation offload from the driver side. Signed-off-by:
Fei Qin <fei.qin@corigine.com> Signed-off-by:
Louis Peens <louis.peens@corigine.com> Signed-off-by:
-
- 10 Dec, 2023 13 commits
-
-
David S. Miller authored
Yoshihiro Shimoda says: ==================== net: rswitch: Add jumbo frames support This patch series is based on the latest net-next.git / main branch. Changes from v3: https://lore.kernel.org/all/20231204012058.3876078-1-yoshihiro.shimoda.uh@renesas.com/ - Based on the latest net-next.git / main branch. - Modify for code consistancy in the patch 3/9. - Add a condition in the patch 3/9. - Fix usage of dma_addr in the patch 8/9. Changes from v2: https://lore.kernel.org/all/20231201054655.3731772-1-yoshihiro.shimoda.uh@renesas.com/ - Based on the latest net-next.git / main branch. - Fix using a variable in the patch 8/9. - Add Reviewed-by tag in the patch 1/9. Changes from v1: https://lore.kernel.org/all/20231127115334.3670790-1-yoshihiro.shimoda.uh@renesas.com/ - Based on the latest net-next.git / main branch. - Fix commit descriptions (s/near the future/the near future/). ==================== Signed-off-by:
-
Yoshihiro Shimoda authored
Allow jumbo frames by changing maximum MTU size and number of RX queues. Signed-off-by:
Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com> Signed-off-by:
-
Yoshihiro Shimoda authored
If the driver would like to transmit a jumbo frame like 2KiB or more, it should be split into multiple queues. In the near future, to support this, add handling specific descriptor types F{START,MID,END}. However, such jumbo frames will not happen yet because the maximum MTU size is still default for now. Signed-off-by: -
Yoshihiro Shimoda authored
If this hardware receives a jumbo frame like 2KiB or more, it will be split into multiple queues. In the near future, to support this, add handling specific descriptor types F{START,MID,END}. However, such jumbo frames will not happen yet because the maximum MTU size is still default for now. Signed-off-by: -
Yoshihiro Shimoda authored
To support jumbo frames, set GWMDNC register with acceptable maximum values for TX and RX. Signed-off-by:
-
Yoshihiro Shimoda authored
If the driver would like to transmit a jumbo frame like 2KiB or more, it should be split into multiple queues. In the near future, to support this, add a setting ext descriptor function to improve code readability. Signed-off-by:
-
Yoshihiro Shimoda authored
If the driver would like to transmit a jumbo frame like 2KiB or more, it should be split into multiple queues. In the near future, to support this, add unmap_addrs array to unmap dma mapping address instead of dma address in each TX descriptor because the descriptors may not have the top dma address. Signed-off-by:
-
Yoshihiro Shimoda authored
If this hardware receives a jumbo frame like 2KiB or more, it will be split into multiple queues. In the near future, to support this, use build_skb() instead of netdev_alloc_skb_ip_align(). Signed-off-by:
-
Yoshihiro Shimoda authored
Array index should not be negative, so use unsigned int for descriptors related array index. Signed-off-by:
-
Yoshihiro Shimoda authored
Drop unused argument and return value of rswitch_tx_free() to simplify the code. Signed-off-by:
Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by:
-
Justin Stitt authored
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 mdiodev->modalias to be NUL-terminated based on its usage with strcmp(): | return strcmp(mdiodev->modalias, drv->name) == 0; Moreover, mdiodev->modalias is already zero-allocated: | mdiodev = kzalloc(sizeof(*mdiodev), GFP_KERNEL); ... which means the NUL-padding strncpy provides is not necessary. 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 Cc: linux-hardening@vger.kernel.org Signed-off-by:
Justin Stitt <justinstitt@google.com> Reviewed-by:
Kees Cook <keescook@chromium.org> Signed-off-by:
-
Justin Stitt authored
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 fw_info->fw_file_name to be NUL-terminated based on its use within _request_firmware_prepare() wherein `name` refers to it: | if (firmware_request_builtin_buf(firmware, name, dbuf, size)) { | dev_dbg(device, "using built-in %s\n", name); | return 0; /* assigned */ | } ... and with firmware_request_builtin() also via `name`: | if (strcmp(name, b_fw->name) == 0) { There is no evidence that NUL-padding is required. Additionally replace size macro (QLC_FW_FILE_NAME_LEN) with sizeof(fw_info->fw_file_name) to more directly tie the maximum buffer size to the destination buffer: 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 Cc: linux-hardening@vger.kernel.org Signed-off-by: -
justinstitt@google.com authored
`strncpy` is deprecated for use on NUL-terminated destination strings [1] and as such we should prefer more robust and less ambiguous string interfaces. A suitable replacement is `strscpy` [2] due to the fact that it guarantees NUL-termination on the destination buffer without unnecessarily NUL-padding. host_info allocation is done in ena_com_allocate_host_info() via dma_alloc_coherent() and is not zero initialized by alloc_etherdev_mq(). However zero initialization of the destination doesn't matter in this case, because strscpy() guarantees a NULL termination. 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 Cc: linux-hardening@vger.kernel.org Signed-off-by:
Arthur Kiyanovski <akiyano@amazon.com> Reviewed-by:
-
- 09 Dec, 2023 2 commits
-
-
Eric Dumazet authored
My prior patch went a bit too far, because apparently fib6_has_expires() could be true while f6i->gc_link is not hashed yet. fib6_set_expires_locked() can indeed set RTF_EXPIRES while f6i->fib6_table is NULL. Original syzbot reports were about corruptions caused by dangling f6i->gc_link. Fixes: 5a08d006 ("ipv6: add debug checks in fib6_info_release()") Reported-by: syzbot+c15aa445274af8674f41@syzkaller.appspotmail.com Signed-off-by:
-
Johannes Berg authored
My previous patch added a call to linkwatch_sync_dev(), but that of course needs to be called under RTNL, which I missed earlier, but now saw RCU warnings from. Fix that by acquiring the RTNL in a similar fashion to how other files do it here. Fixes: facd15df ("net: core: synchronize link-watch when carrier is queried") Signed-off-by:
-
- 08 Dec, 2023 3 commits
-
-
Kees Cook authored
The return value from nla_len() is never expected to be negative, and can never be more than struct nlattr::nla_len (a u16). Adjust the prototype on the function. This will let GCC's value range optimization passes know that the return can never be negative, and can never be larger than u16. As recently discussed[1], this silences the following warning in GCC 12+: net/wireless/nl80211.c: In function 'nl80211_set_cqm_rssi.isra': net/wireless/nl80211.c:12892:17: warning: 'memcpy' specified bound 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=] 12892 | memcpy(cqm_config->rssi_thresholds, thresholds, | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 12893 | flex_array_size(cqm_config, rssi_thresholds, | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 12894 | n_thresholds)); | ~~~~~~~~~~~~~~ A future change would be to clamp the subtraction to make sure it never wraps around if nla_len is somehow less than NLA_HDRLEN, which would have the additional benefit of being defensive in the face of nlattr corruption or logic errors. Reported-by: -
Sean Nyekjaer authored
Correct the use of define DSA_TAG_PROTO_LAN937X_VALUE to DSA_TAG_PROTO_LAN937X to improve readability. Signed-off-by:
Sean Nyekjaer <sean@geanix.com> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Acked-by:
Arun Ramadoss <arun.ramadoss@microchip.com> Link: https://lore.kernel.org/r/20231206160124.1935451-1-sean@geanix.comSigned-off-by:
-
David Laight authored
Commit 227b60f5 added a seqlock to ensure that the low and high port numbers were always updated together. This is overkill because the two 16bit port numbers can be held in a u32 and read/written in a single instruction. More recently 91d0b78c added support for finer per-socket limits. The user-supplied value is 'high << 16 | low' but they are held separately and the socket options protected by the socket lock. Use a u32 containing 'high << 16 | low' for both the 'net' and 'sk' fields and use READ_ONCE()/WRITE_ONCE() to ensure both values are always updated together. Change (the now trival) inet_get_local_port_range() to a static inline to optimise the calling code. (In particular avoiding returning integers by reference.) Signed-off-by:
David Laight <david.laight@aculab.com> Reviewed-by:
Mat Martineau <martineau@kernel.org> Reviewed-by:
Kuniyuki Iwashima <kuniyu@amazon.com> Link: https://lore.kernel.org/r/4e505d4198e946a8be03fb1b4c3072b0@AcuMS.aculab.comSigned-off-by:
-
