summaryrefslogtreecommitdiff
path: root/net
AgeCommit message (Collapse)Author
2023-10-23wifi: mac80211: don't drop all unprotected public action framesAvraham Stern
Not all public action frames have a protected variant. When MFP is enabled drop only public action frames that have a dual protected variant. Fixes: 76a3059cf124 ("wifi: mac80211: drop some unprotected action frames") Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20231016145213.2973e3c8d3bb.I6198b8d3b04cf4a97b06660d346caec3032f232a@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: cfg80211: fix assoc response warning on failed linksJohannes Berg
The warning here shouldn't be done before we even set the bss field (or should've used the input data). Move the assignment before the warning to fix it. We noticed this now because of Wen's bugfix, where the bug fixed there had previously hidden this other bug. Fixes: 53ad07e9823b ("wifi: cfg80211: support reporting failed links") Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: cfg80211: pass correct pointer to rdev_inform_bss()Ben Greear
Confusing struct member names here resulted in passing the wrong pointer, causing crashes. Pass the correct one. Fixes: eb142608e2c4 ("wifi: cfg80211: use a struct for inform_single_bss data") Signed-off-by: Ben Greear <greearb@candelatech.com> Link: https://lore.kernel.org/r/20231021154827.1142734-1-greearb@candelatech.com [rewrite commit message, add fixes] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: drop robust action frames before assocJohannes Berg
To be able to more easily understand the code, drop robust action frames before being associated, even if there's no MFP in the end, as they are Class 3 Frames and shouldn't be transmitted in the first place. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20231001125722.b2fd37083371.Ie9f4906e2f6c698989bce6681956ed2f9454f27c@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: cfg80211: Allow AP/P2PGO to indicate port authorization to peer ↵Vinayak Yadawad
STA/P2PClient In 4way handshake offload, cfg80211_port_authorized enables driver to indicate successful 4way handshake to cfg80211 layer. Currently this path of port authorization is restricted to interface type NL80211_IFTYPE_STATION and NL80211_IFTYPE_P2P_CLIENT. This patch extends the support for NL80211_IFTYPE_AP and NL80211_IFTYPE_P2P_GO interfaces to authorize peer STA/P2P_CLIENT, whenever authentication is offloaded on the AP/P2P_GO interface. Signed-off-by: Vinayak Yadawad <vinayak.yadawad@broadcom.com> Link: https://lore.kernel.org/r/dee3b0a2b4f617e932c90bff4504a89389273632.1695721435.git.vinayak.yadawad@broadcom.com Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: fix another key installation error pathJohannes Berg
Due to overlapping changes and merges, another error path ended up broken. Fix this one as well. Reported-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: rename struct cfg80211_rx_assoc_resp to ↵Kalle Valo
cfg80211_rx_assoc_resp_data make htmldocs warns: Documentation/driver-api/80211/cfg80211:48: ./include/net/cfg80211.h:7290: WARNING: Duplicate C declaration, also defined at cfg80211:7251. Declaration is '.. c:function:: void cfg80211_rx_assoc_resp (struct net_device *dev, struct cfg80211_rx_assoc_resp *data)'. This is because there's a function named cfg80211_rx_assoc_resp() and a struct named cfg80211_rx_assoc_resp, see previous patch for more info. To workaround this rename the struct to cfg80211_rx_assoc_resp_data. The parameter for the function is named 'data' anyway so the naming here is consistent. Compile tested only. Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231012114229.2931808-3-kvalo@kernel.org Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: rename ieee80211_tx_status() to ieee80211_tx_status_skb()Kalle Valo
make htmldocs warns: Documentation/driver-api/80211/mac80211:109: ./include/net/mac80211.h:5170: WARNING: Duplicate C declaration, also defined at mac80211:1117. Declaration is '.. c:function:: void ieee80211_tx_status (struct ieee80211_hw *hw, struct sk_buff *skb)'. This is because there's a function named ieee80211_tx_status() and a struct named ieee80211_tx_status. This has been discussed previously but no solution found: https://lore.kernel.org/all/20220521114629.6ee9fc06@coco.lan/ There's also a bug open for three years with no solution in sight: https://github.com/sphinx-doc/sphinx/pull/8313 So I guess we have no other solution than to a workaround this in the code, for example to rename the function to ieee80211_tx_status_skb() to avoid the name conflict. I got the idea for the name from ieee80211_tx_status_noskb() in which the skb is not provided as an argument, instead with ieee80211_tx_status_skb() the skb is provided. Compile tested only. Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com> Link: https://lore.kernel.org/r/20231012114229.2931808-2-kvalo@kernel.org Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: fix change_address deadlock during unregisterJohannes Berg
When using e.g. bonding, and doing a sequence such as # iw wlan0 set type __ap # ip link add name bond1 type bond # ip link set wlan0 master bond1 # iw wlan0 interface del we deadlock, since the wlan0 interface removal will cause bonding to reset the MAC address of wlan0. The locking would be somewhat difficult to fix, but since this only happens during removal, we can simply ignore the MAC address change at this time. Reported-by: syzbot+25b3a0b24216651bc2af@syzkaller.appspotmail.com Signed-off-by: Johannes Berg <johannes.berg@intel.com> Link: https://lore.kernel.org/r/20231012123447.9f9d7fd1f237.Ic3a5ef4391b670941a69cec5592aefc79d9c2890@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: Add __counted_by for struct ieee802_11_elems and use ↵Gustavo A. R. Silva
struct_size() 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 via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). While there, use struct_size() helper, instead of the open-coded version, to calculate the size for the allocation of the whole flexible structure including, of course, the flexible-array member. This code was found with the help of Coccinelle, and audited and fixed manually. Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org> Reviewed-by: Kees Cook <keescook@chromium.org> Link: https://lore.kernel.org/r/ZSQ/jcmTAf/PKHg/@work Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: remove unused argument of ieee80211_get_tdls_action()Dmitry Antipov
Remove unused 'hdr_size' argument of 'ieee80211_get_tdls_action()' and adjust 'ieee80211_report_used_skb()' accordingly. Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru> Link: https://lore.kernel.org/r/20231004153032.206134-1-dmantipov@yandex.ru Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: add link id to mgd_prepare_tx()Miri Korenblit
As we are moving to MLO and links terms, also the airtime protection will be done for a link rather than for a vif. Thus, some drivers will need to know for which link to protect airtime. Add link id as a parameter to the mgd_prepare_tx() callback. Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.c7fc59a6780b.Ic88a5037d31e184a2dce0b031ece1a0a93a3a9da@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: Check if we had first beacon with relevant linksMiri Korenblit
If there is a disassoc before the fisrt beacon we need to protect a session for the deauth frame. Currently we are checking if we had a beacon in the default link, which is wrong in a MLO connection and link id != 0. Fix this by checking all the active links, if none had a beacon then protect a session. If at least one link had a beacon there is no need for session protection. Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20231004120820.d290f0ab77b0.Ic1505cf3d60f74580d31efa7e52046947c490b85@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: flush STA queues on unauthorizationJohannes Berg
When the station is marked as no longer authorized, we shouldn't transmit to it any longer, but in particular we shouldn't be able to transmit to it after removing keys, which might lead to frames being sent out unencrypted depending on the exact hardware offload mechanism. Thus, instead of flushing only on station destruction, which covers only some cases, always flush on unauthorization. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.d47f528829e7.I96903652c7ee0c5c66891f8b2364383da8e45a1f@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: purge TX queues in flush_queues flowMiri Korenblit
When this flow is invoked with the "drop" parameter as true, we only drop the frames from the hw queues, but not from the sw queues. So when we call wake_queues() after hw queue purging, all the frames from the sw queues will be TX'ed, when what we actually want to do is to purge all queues in order to not TX anything... This can cause, for example, TXing data frames to the peer after the deauth frame was sent. Fix this by purging the sw queues in addition to the hw queues if the drop parameter is true. Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.8fc2ee23e56f.I8b3f6def9c28ea96261e2d31df8786986fb5385b@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: cfg80211: wext: convert return value to kernel-docJohannes Berg
Since I'm getting a warning here right now, fix the kernel-doc to be "Returns:" rather than just writing that out in the doc paragraph. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.ab3b9274bf07.If263f9f6726d6ad4661f8603e6a4485e0385d67f@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: fix a expired vs. cancel race in rocEmmanuel Grumbach
When the remain on channel is removed at the time it should have expired, we have a race: the driver could be handling the flow of the expiration while mac80211 is cancelling that very same remain on channel request. This wouldn't be problem in itself, but since mac80211 can send the next request to the driver in the cancellation flow, we can get to the following situation: CPU0 CPU1 expiration of roc in driver ieee80211_remain_on_channel_expired() Cancellation of the roc schedules a worker (hw_roc_done) Add next roc hw_roc_done_wk runs and ends the second roc prematurely. Since, by design, there is only one single request sent to the driver at a time, we can safely assume that after the cancel() request returns from the driver, we should not handle any worker that handles the expiration of the request. Cancel the hw_roc_done worker after the cancellation to make sure we start the next one with a clean slate. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.4e4469be20ac.Iab0525f5cc4698acf23eab98b8b1eec02099cde0@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: make mgd_protect_tdls_discover MLO-awareMiri Korenblit
Since userspace can choose now what link to establish the TDLS on, we should know on what channel to do session protection. Add a link id parameter to this callback. Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.ef12ce3eb835.If864f406cfd9e24f36a2b88fd13a37328633fcf9@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: Fix setting vif linksIlan Peer
When setting the interface links, ignore the change iff both the valid links and the dormant links did not change. This is needed to support cases where the valid links didn't change but the dormant links did. Fixes: 6d543b34dbcf ("wifi: mac80211: Support disabled links during association") Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.0357b6306587.I7dbfec347949b629fea680d246a650d6207ff217@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: cfg80211: Handle specific BSSID in 6GHz scanningIlan Peer
When the scan parameters for a 6GHz scan specify a unicast BSSID address, and the corresponding AP is found in the scan list, add a corresponding entry in the collocated AP list, so this AP would be directly probed even if it was not advertised as a collocated AP. This is needed for handling a scan request that is intended for a ML probe flow, where user space can requests a scan to retrieve information for other links in the AP MLD. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.54b954bc02ad.I1c072793d3d77a4c8fbbc64b4db5cce1bbb00382@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: mesh: fix some kdoc warningsBenjamin Berg
These were mostly missing or incorrectly tagged return values. Signed-off-by: Benjamin Berg <benjamin.berg@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.33fea2968c62.I41d197b570370ab7cad1405518512fdd36e08717@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: cfg80211: Include operating class 137 in 6GHz bandIlan Peer
Draft P802.11be_D3.1 added operating class to describe 320 MHz operation in the 6GHz band. Include this new operating class in ieee80211_operating_class_to_band(). Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.bed4a007d81b.I3eb4b8fe39c0c1a988c98a103b11a9f45a92b038@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: Rename and update IEEE80211_VIF_DISABLE_SMPS_OVERRIDEIlan Peer
EMLSR operation and SMPS operation cannot coexist. Thus, when EMLSR is enabled, all SMPS signaling towards the AP should be stopped (it is expected that the AP will consider SMPS to be off). Rename IEEE80211_VIF_DISABLE_SMPS_OVERRIDE to IEEE80211_VIF_EML_ACTIVE and use the flag as an indication from the driver that EMLSR is enabled. When EMLSR is enabled SMPS flows towards the AP MLD should be stopped. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.fb2c2f9a0645.If6df5357568abd623a081f0f33b07e63fb8bba99@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: handle debugfs when switching to/from MLOMiri Korenblit
In MLO, we have a per-link debugfs directory which contains the per-link files. In case of non-MLO we would like to put the per-link files in the netdev directory to keep it how it was before MLO. - Upon interface creation the netdev will be created with the per-link files in it. - Upon switching to MLO: delete the entire netdev directory and then recreate it without the per-link files. Then the per-link directories with the per-link files in it will be created in ieee80211_link_init() - Upon switching to non-MLO: delete the entire netdev directory (including the per-link directories) and recreate it with the per-link files in it. Note that this also aligns to always call the vif link debugfs method for the deflink as promised in the documentation, which wasn't done before. Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.082e698caca9.I5bef7b2026e0f58b4a958b3d1f459ac5baeccfc9@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: add a driver callback to add vif debugfsMiri Korenblit
Add a callback which the driver can use to add the vif debugfs. We used to have this back until commit d260ff12e776 ("mac80211: remove vif debugfs driver callbacks") where we thought that it will be easier to just add them during interface add/remove. However, now with multi-link, we want to have proper debugfs for drivers for multi-link where some files might be in the netdev for non-MLO connections, and in the links for MLO ones, so we need to do some reconstruction when switching the mode. Moving to this new call enables that and MLO drivers will have to use it for proper debugfs operation. Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.ac38913f6ab7.Iee731d746bb08fcc628fa776f337016a12dc62ac@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: don't recreate driver link debugfs in reconfigJohannes Berg
We can delete any that we want to remove, but we can't recreate the links as they already exist. Fixes: 170cd6a66d9a ("wifi: mac80211: add netdev per-link debugfs data and driver hook") Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.3d0214838421.I512a0ff86f631ff42bf25ea0cb2e8e8616794a94@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23wifi: mac80211: cleanup auth_data only if association continuesBenjamin Berg
If the association command fails then the authentication is still valid and it makes sense to keep it alive. Otherwise, we would currently get into an inconsistent state because mac80211 on the one hand is disconnected but on the other hand the state is not entirely cleared and a new authentication could not continue. Signed-off-by: Benjamin Berg <benjamin.berg@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230928172905.c9855f46ebc8.I7f3dcd4120a186484a91b87560e9b7201d40984f@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2023-10-23tcp: add TCPI_OPT_USEC_TSEric Dumazet
Add the ability to report in tcp_info.tcpi_options if a flow is using usec resolution in TCP TS val. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23tcp: add support for usec resolution in TCP TS valuesEric Dumazet
Back in 2015, Van Jacobson suggested to use usec resolution in TCP TS values. This has been implemented in our private kernels. Goals were : 1) better observability of delays in networking stacks. 2) better disambiguation of events based on TSval/ecr values. 3) building block for congestion control modules needing usec resolution. Back then we implemented a schem based on private SYN options to negotiate the feature. For upstream submission, we chose to use a route attribute, because this feature is probably going to be used in private networks [1] [2]. ip route add 10/8 ... features tcp_usec_ts Note that RFC 7323 recommends a "timestamp clock frequency in the range 1 ms to 1 sec per tick.", but also mentions "the maximum acceptable clock frequency is one tick every 59 ns." [1] Unfortunately RFC 7323 5.5 (Outdated Timestamps) suggests to invalidate TS.Recent values after a flow was idle for more than 24 days. This is the part making usec_ts a problem for peers following this recommendation for long living idle flows. [2] Attempts to standardize usec ts went nowhere: https://www.ietf.org/proceedings/97/slides/slides-97-tcpm-tcp-options-for-low-latency-00.pdf https://datatracker.ietf.org/doc/draft-wang-tcpm-low-latency-opt/ Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23tcp: add tcp_rtt_tsopt_us()Eric Dumazet
Before adding usec TS support, add tcp_rtt_tsopt_us() helper to factorize code. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23tcp: rename tcp_time_stamp() to tcp_time_stamp_ts()Eric Dumazet
This helper returns a TSval from a TCP socket. It currently calls tcp_time_stamp_ms() but will soon be able to return a usec based TSval, depending on an upcoming tp->tcp_usec_ts field. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23tcp: move tcp_ns_to_ts() to net/ipv4/syncookies.cEric Dumazet
tcp_ns_to_ts() is only used once from cookie_init_timestamp(). Also add the 'bool usec_ts' parameter to enable usec TS later. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23tcp: rename tcp_skb_timestamp()Eric Dumazet
This helper returns a 32bit TCP TSval from skb->tstamp. As we are going to support usec or ms units soon, rename it to tcp_skb_timestamp_ts() and add a boolean to select the unit. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23tcp: replace tcp_time_stamp_raw()Eric Dumazet
In preparation of usec TCP TS support, remove tcp_time_stamp_raw() in favor of tcp_clock_ts() helper. This helper will return a suitable 32bit result to feed TS values, depending on a socket field. Also add tcp_tw_tsval() and tcp_rsk_tsval() helpers to factorize the details. We do not yet support usec timestamps. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23tcp: introduce tcp_clock_ms()Eric Dumazet
It delivers current TCP time stamp in ms unit, and is used in place of confusing tcp_time_stamp_raw() It is the same family than tcp_clock_ns() and tcp_clock_ms(). tcp_time_stamp_raw() will be replaced later for TSval contexts with a more descriptive name. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23tcp: add tcp_time_stamp_ms() helperEric Dumazet
In preparation of adding usec TCP TS values, add tcp_time_stamp_ms() for contexts needing ms based values. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23tcp: fix cookie_init_timestamp() overflowsEric Dumazet
cookie_init_timestamp() is supposed to return a 64bit timestamp suitable for both TSval determination and setting of skb->tstamp. Unfortunately it uses 32bit fields and overflows after 2^32 * 10^6 nsec (~49 days) of uptime. Generated TSval are still correct, but skb->tstamp might be set far away in the past, potentially confusing other layers. tcp_ns_to_ts() is changed to return a full 64bit value, ts and ts_now variables are changed to u64 type, and TSMASK is removed in favor of shifts operations. While we are at it, change this sequence: ts >>= TSBITS; ts--; ts <<= TSBITS; ts |= options; to: ts -= (1UL << TSBITS); Fixes: 9a568de4818d ("tcp: switch TCP TS option (RFC 7323) to 1ms clock") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23ipv6: add new arguments to udp_tunnel6_dst_lookup()Beniamino Galvani
We want to make the function more generic so that it can be used by other UDP tunnel implementations such as geneve and vxlan. To do that, add the following arguments: - source and destination UDP port; - ifindex of the output interface, needed by vxlan; - the tos, because in some cases it is not taken from struct ip_tunnel_info (for example, when it's inherited from the inner packet); - the dst cache, because not all tunnel types (e.g. vxlan) want to use the one from struct ip_tunnel_info. With these parameters, the function no longer needs the full struct ip_tunnel_info as argument and we can pass only the relevant part of it (struct ip_tunnel_key). This is similar to what already done for IPv4 in commit 72fc68c6356b ("ipv4: add new arguments to udp_tunnel_dst_lookup()"). Suggested-by: Guillaume Nault <gnault@redhat.com> Signed-off-by: Beniamino Galvani <b.galvani@gmail.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23ipv6: remove "proto" argument from udp_tunnel6_dst_lookup()Beniamino Galvani
The function is now UDP-specific, the protocol is always IPPROTO_UDP. This is similar to what already done for IPv4 in commit 78f3655adcb5 ("ipv4: remove "proto" argument from udp_tunnel_dst_lookup()"). Suggested-by: Guillaume Nault <gnault@redhat.com> Signed-off-by: Beniamino Galvani <b.galvani@gmail.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23ipv6: rename and move ip6_dst_lookup_tunnel()Beniamino Galvani
At the moment ip6_dst_lookup_tunnel() is used only by bareudp. Ideally, other UDP tunnel implementations should use it, but to do so the function needs to accept new parameters that are specific for UDP tunnels, such as the ports. Prepare for these changes by renaming the function to udp_tunnel6_dst_lookup() and move it to file net/ipv6/ip6_udp_tunnel.c. This is similar to what already done for IPv4 in commit bf3fcbf7e7a0 ("ipv4: rename and move ip_route_output_tunnel()"). Suggested-by: Guillaume Nault <gnault@redhat.com> Signed-off-by: Beniamino Galvani <b.galvani@gmail.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23net: atm: Remove redundant check.Gavrilov Ilia
Checking the 'adev' variable is unnecessary, because 'cdev' has already been checked earlier. Found by InfoTeCS on behalf of Linux Verification Center (linuxtesting.org) with SVACE. Fixes: 656d98b09d57 ("[ATM]: basic sysfs support for ATM devices") Signed-off-by: Gavrilov Ilia <Ilia.Gavrilov@infotecs.ru> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-23xfrm Fix use after free in __xfrm6_udp_encap_rcv.Steffen Klassert
A recent patch changed xfrm6_udp_encap_rcv to not free the skb itself anymore but fogot the case where xfrm4_udp_encap_rcv is called subsequently. Fix this by moving the call to xfrm4_udp_encap_rcv from __xfrm6_udp_encap_rcv to xfrm6_udp_encap_rcv. Fixes: 221ddb723d90 ("xfrm: Support GRO for IPv6 ESP in UDP encapsulation") Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
2023-10-22tcp: fix wrong RTO timeout when received SACK renegingFred Chen
This commit fix wrong RTO timeout when received SACK reneging. When an ACK arrived pointing to a SACK reneging, tcp_check_sack_reneging() will rearm the RTO timer for min(1/2*srtt, 10ms) into to the future. But since the commit 62d9f1a6945b ("tcp: fix TLP timer not set when CA_STATE changes from DISORDER to OPEN") merged, the tcp_set_xmit_timer() is moved after tcp_fastretrans_alert()(which do the SACK reneging check), so the RTO timeout will be overwrited by tcp_set_xmit_timer() with icsk_rto instead of 1/2*srtt. Here is a packetdrill script to check this bug: 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 +0 bind(3, ..., ...) = 0 +0 listen(3, 1) = 0 // simulate srtt to 100ms +0 < S 0:0(0) win 32792 <mss 1000, sackOK,nop,nop,nop,wscale 7> +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 7> +.1 < . 1:1(0) ack 1 win 1024 +0 accept(3, ..., ...) = 4 +0 write(4, ..., 10000) = 10000 +0 > P. 1:10001(10000) ack 1 // inject sack +.1 < . 1:1(0) ack 1 win 257 <sack 1001:10001,nop,nop> +0 > . 1:1001(1000) ack 1 // inject sack reneging +.1 < . 1:1(0) ack 1001 win 257 <sack 9001:10001,nop,nop> // we expect rto fired in 1/2*srtt (50ms) +.05 > . 1001:2001(1000) ack 1 This fix remove the FLAG_SET_XMIT_TIMER from ack_flag when tcp_check_sack_reneging() set RTO timer with 1/2*srtt to avoid being overwrited later. Fixes: 62d9f1a6945b ("tcp: fix TLP timer not set when CA_STATE changes from DISORDER to OPEN") Signed-off-by: Fred Chen <fred.chenchen03@gmail.com> Reviewed-by: Neal Cardwell <ncardwell@google.com> Tested-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-20neighbour: fix various data-racesEric Dumazet
1) tbl->gc_thresh1, tbl->gc_thresh2, tbl->gc_thresh3 and tbl->gc_interval can be written from sysfs. 2) tbl->last_flush is read locklessly from neigh_alloc() 3) tbl->proxy_queue.qlen is read locklessly from neightbl_fill_info() 4) neightbl_fill_info() reads cpu stats that can be changed concurrently. Fixes: c7fb64db001f ("[NETLINK]: Neighbour table configuration and statistics via rtnetlink") Signed-off-by: Eric Dumazet <edumazet@google.com> Link: https://lore.kernel.org/r/20231019122104.1448310-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2023-10-20net: do not leave an empty skb in write queueEric Dumazet
Under memory stress conditions, tcp_sendmsg_locked() might call sk_stream_wait_memory(), thus releasing the socket lock. If a fresh skb has been allocated prior to this, we should not leave it in the write queue otherwise tcp_write_xmit() could panic. This apparently does not happen often, but a future change in __sk_mem_raise_allocated() that Shakeel and others are considering would increase chances of being hurt. Under discussion is to remove this controversial part: /* Fail only if socket is _under_ its sndbuf. * In this case we cannot block, so that we have to fail. */ if (sk->sk_wmem_queued + size >= sk->sk_sndbuf) { /* Force charge with __GFP_NOFAIL */ if (memcg_charge && !charged) { mem_cgroup_charge_skmem(sk->sk_memcg, amt, gfp_memcg_charge() | __GFP_NOFAIL); } return 1; } Fixes: fdfc5c8594c2 ("tcp: remove empty skb from write queue in error cases") Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Shakeel Butt <shakeelb@google.com> Link: https://lore.kernel.org/r/20231019112457.1190114-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2023-10-20ethtool: untangle the linkmode and ethtool headersJakub Kicinski
Commit 26c5334d344d ("ethtool: Add forced speed to supported link modes maps") added a dependency between ethtool.h and linkmode.h. The dependency in the opposite direction already exists so the new code was inserted in an awkward place. The reason for ethtool.h to include linkmode.h, is that ethtool_forced_speed_maps_init() is a static inline helper. That's not really necessary. Signed-off-by: Jakub Kicinski <kuba@kernel.org> Reviewed-by: Paul Greenwalt <paul.greenwalt@intel.com> Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-20net: fix IPSTATS_MIB_OUTPKGS increment in OutForwDatagrams.Heng Guo
Reproduce environment: network with 3 VM linuxs is connected as below: VM1<---->VM2(latest kernel 6.5.0-rc7)<---->VM3 VM1: eth0 ip: 192.168.122.207 MTU 1500 VM2: eth0 ip: 192.168.122.208, eth1 ip: 192.168.123.224 MTU 1500 VM3: eth0 ip: 192.168.123.240 MTU 1500 Reproduce: VM1 send 1400 bytes UDP data to VM3 using tools scapy with flags=0. scapy command: send(IP(dst="192.168.123.240",flags=0)/UDP()/str('0'*1400),count=1, inter=1.000000) Result: Before IP data is sent. ---------------------------------------------------------------------- root@qemux86-64:~# cat /proc/net/snmp Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests OutDiscards OutNoRoutes ReasmTimeout ReasmReqds ReasmOKs ReasmFails FragOKs FragFails FragCreates Ip: 1 64 11 0 3 4 0 0 4 7 0 0 0 0 0 0 0 0 0 ...... ---------------------------------------------------------------------- After IP data is sent. ---------------------------------------------------------------------- root@qemux86-64:~# cat /proc/net/snmp Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests OutDiscards OutNoRoutes ReasmTimeout ReasmReqds ReasmOKs ReasmFails FragOKs FragFails FragCreates Ip: 1 64 12 0 3 5 0 0 4 8 0 0 0 0 0 0 0 0 0 ...... ---------------------------------------------------------------------- "ForwDatagrams" increase from 4 to 5 and "OutRequests" also increase from 7 to 8. Issue description and patch: IPSTATS_MIB_OUTPKTS("OutRequests") is counted with IPSTATS_MIB_OUTOCTETS ("OutOctets") in ip_finish_output2(). According to RFC 4293, it is "OutOctets" counted with "OutTransmits" but not "OutRequests". "OutRequests" does not include any datagrams counted in "ForwDatagrams". ipSystemStatsOutOctets OBJECT-TYPE DESCRIPTION "The total number of octets in IP datagrams delivered to the lower layers for transmission. Octets from datagrams counted in ipIfStatsOutTransmits MUST be counted here. ipSystemStatsOutRequests OBJECT-TYPE DESCRIPTION "The total number of IP datagrams that local IP user- protocols (including ICMP) supplied to IP in requests for transmission. Note that this counter does not include any datagrams counted in ipSystemStatsOutForwDatagrams. So do patch to define IPSTATS_MIB_OUTPKTS to "OutTransmits" and add IPSTATS_MIB_OUTREQUESTS for "OutRequests". Add IPSTATS_MIB_OUTREQUESTS counter in __ip_local_out() for ipv4 and add IPSTATS_MIB_OUT counter in ip6_finish_output2() for ipv6. Test result with patch: Before IP data is sent. ---------------------------------------------------------------------- root@qemux86-64:~# cat /proc/net/snmp Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests OutDiscards OutNoRoutes ReasmTimeout ReasmReqds ReasmOKs ReasmFails FragOKs FragFails FragCreates OutTransmits Ip: 1 64 9 0 5 1 0 0 3 3 0 0 0 0 0 0 0 0 0 4 ...... root@qemux86-64:~# cat /proc/net/netstat ...... IpExt: InNoRoutes InTruncatedPkts InMcastPkts OutMcastPkts InBcastPkts OutBcastPkts InOctets OutOctets InMcastOctets OutMcastOctets InBcastOctets OutBcastOctets InCsumErrors InNoECTPkts InECT1Pkts InECT0Pkts InCEPkts ReasmOverlaps IpExt: 0 0 0 0 0 0 2976 1896 0 0 0 0 0 9 0 0 0 0 ---------------------------------------------------------------------- After IP data is sent. ---------------------------------------------------------------------- root@qemux86-64:~# cat /proc/net/snmp Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests OutDiscards OutNoRoutes ReasmTimeout ReasmReqds ReasmOKs ReasmFails FragOKs FragFails FragCreates OutTransmits Ip: 1 64 10 0 5 2 0 0 3 3 0 0 0 0 0 0 0 0 0 5 ...... root@qemux86-64:~# cat /proc/net/netstat ...... IpExt: InNoRoutes InTruncatedPkts InMcastPkts OutMcastPkts InBcastPkts OutBcastPkts InOctets OutOctets InMcastOctets OutMcastOctets InBcastOctets OutBcastOctets InCsumErrors InNoECTPkts InECT1Pkts InECT0Pkts InCEPkts ReasmOverlaps IpExt: 0 0 0 0 0 0 4404 3324 0 0 0 0 0 10 0 0 0 0 ---------------------------------------------------------------------- "ForwDatagrams" increase from 1 to 2 and "OutRequests" is keeping 3. "OutTransmits" increase from 4 to 5 and "OutOctets" increase 1428. Signed-off-by: Heng Guo <heng.guo@windriver.com> Reviewed-by: Kun Song <Kun.Song@windriver.com> Reviewed-by: Filip Pudak <filip.pudak@windriver.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-20netlink: add variable-length / auto integersJakub Kicinski
We currently push everyone to use padding to align 64b values in netlink. Un-padded nla_put_u64() doesn't even exist any more. The story behind this possibly start with this thread: https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/ where DaveM was concerned about the alignment of a structure containing 64b stats. If user space tries to access such struct directly: struct some_stats *stats = nla_data(attr); printf("A: %llu", stats->a); lack of alignment may become problematic for some architectures. These days we most often put every single member in a separate attribute, meaning that the code above would use a helper like nla_get_u64(), which can deal with alignment internally. Even for arches which don't have good unaligned access - access aligned to 4B should be pretty efficient. Kernel and well known libraries deal with unaligned input already. Padded 64b is quite space-inefficient (64b + pad means at worst 16B per attr vs 32b which takes 8B). It is also more typing: if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING, value, NETDEV_A_SOMETHING_PAD)) Create a new attribute type which will use 32 bits at netlink level if value is small enough (probably most of the time?), and (4B-aligned) 64 bits otherwise. Kernel API is just: if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value)) Calling this new type "just" sint / uint with no specific size will hopefully also make people more comfortable with using it. Currently telling people "don't use u8, you may need the bits, and netlink will round up to 4B, anyway" is the #1 comment we give to newcomers. In terms of netlink layout it looks like this: 0 4 8 12 16 32b: [nlattr][ u32 ] 64b: [ pad ][nlattr][ u64 ] uint(32) [nlattr][ u32 ] uint(64) [nlattr][ u64 ] Signed-off-by: Jakub Kicinski <kuba@kernel.org> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-20devlink: convert most of devlink_fmsg_*() to return voidPrzemek Kitszel
Since struct devlink_fmsg retains error by now (see 1st patch of this series), there is no longer need to keep returning it in each call. This is a separate commit to allow per-driver conversion to stop using those return values. Reviewed-by: Jiri Pirko <jiri@nvidia.com> Signed-off-by: Przemek Kitszel <przemyslaw.kitszel@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-20devlink: retain error in struct devlink_fmsgPrzemek Kitszel
Retain error value in struct devlink_fmsg, to relieve drivers from checking it after each call. Note that fmsg is an in-memory builder/buffer of formatted message, so it's not the case that half baked message was sent somewhere. We could find following scheme in multiple drivers: err = devlink_fmsg_obj_nest_start(fmsg); if (err) return err; err = devlink_fmsg_string_pair_put(fmsg, "src", src); if (err) return err; err = devlink_fmsg_something(fmsg, foo, bar); if (err) return err; // and so on... err = devlink_fmsg_obj_nest_end(fmsg); With retaining error API that translates to: devlink_fmsg_obj_nest_start(fmsg); devlink_fmsg_string_pair_put(fmsg, "src", src); devlink_fmsg_something(fmsg, foo, bar); // and so on... devlink_fmsg_obj_nest_end(fmsg); What means we check error just when is time to send. Possible error scenarios are developer error (API misuse) and memory exhaustion, both cases are good candidates to choose readability over fastest possible exit. Note that this patch keeps returning errors, to allow per-driver conversion to the new API, but those are not needed at this point already. This commit itself is an illustration of benefits for the dev-user, more of it will be in separate commits of the series. Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Reviewed-by: Jiri Pirko <jiri@nvidia.com> Signed-off-by: Przemek Kitszel <przemyslaw.kitszel@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
generated by cgit (git 2.34.1) at 2025-09-11 18:45:10 +0000