- Dec 14, 2021
-
-
Johannes Berg authored
When we reconfigure, the driver might do some things to complete the reconfiguration. It's strange and could be broken in some cases because we restart other works (e.g. remain-on-channel and TX) before this happens, yet only start queues later. Change this to do the reconfig complete when reconfiguration is actually complete, not when we've already started doing other things again. For iwlwifi, this should fix a race where the reconfig can race with TX, for ath10k and ath11k that also use this it won't make a difference because they just start queues there, and mac80211 also stopped the queues and will restart them later as before. Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/iwlwifi.20211129152938.cab99f22fe19.Iefe494687f15fd85f77c1b989d1149c8efdfdc36@changeid Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Johannes Berg authored
Mark TXQs as having seen transmit while they were stopped if we bail out of drv_wake_tx_queue() due to reconfig, so that the queue wake after this will make them catch up. This is particularly necessary for when TXQs are used for management packets since those TXQs won't see a lot of traffic that'd make them catch up later. Cc: stable@vger.kernel.org Fixes: 4856bfd2 ("mac80211: do not call driver wake_tx_queue op during reconfig") Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/iwlwifi.20211129152938.4573a221c0e1.I0d1d5daea3089be3fc0dccc92991b0f8c5677f0c@changeid Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Mordechay Goodstein authored
Currently channel context is updated only after station got an update about new assoc state, this results in station using the old channel context. Fix this by moving the update channel context before updating station, enabling the driver to immediately use the updated channel context in the new assoc state. Signed-off-by:
Mordechay Goodstein <mordechay.goodstein@intel.com> Signed-off-by:
Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/iwlwifi.20211129152938.1c80c17ffd8a.I94ae31378b363c1182cfdca46c4b7e7165cff984@changeid Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Ilan Peer authored
Instead of using the hard-coded value of '100' use the correct scan IEs length as calculated during HW registration to mac80211. Signed-off-by:
Ilan Peer <ilan.peer@intel.com> Signed-off-by:
Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/iwlwifi.20211129152938.0a82d6891719.I8ded1f2e0bccb9e71222c945666bcd86537f2e35@changeid Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Johannes Berg authored
We should be doing the HE capabilities lookup based on the full interface type so if P2P doesn't have HE but client has it doesn't get confused. Fix that. Fixes: 2ab45876 ("mac80211: add support for the ADDBA extension element") Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/iwlwifi.20211129152938.010fc1d61137.If3a468145f29d670cb00a693bed559d8290ba693@changeid Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Johannes Berg authored
Before attempting to parse an extended element, verify that the extended element ID is present. Fixes: 41cbb0f5 ("mac80211: add support for HE") Reported-by:
<syzbot+59bdff68edce82e393b6@syzkaller.appspotmail.com> Link: https://lore.kernel.org/r/20211211201023.f30a1b128c07.I5cacc176da94ba316877c6e10fe3ceec8b4dbd7d@changeid Cc: stable@vger.kernel.org Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Ilan Peer authored
The function cfg80211_reg_can_beacon_relax() expects wiphy mutex to be held when it is being called. However, when reg_leave_invalid_chans() is called the mutex is not held. Fix it by acquiring the lock before calling the function. Fixes: a05829a7 ("cfg80211: avoid holding the RTNL when calling the driver") Signed-off-by:
Ilan Peer <ilan.peer@intel.com> Signed-off-by:
Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/iwlwifi.20211202152831.527686cda037.I40ad9372a47cbad53b4aae7b5a6ccc0dc3fddf8b@changeid Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Johannes Berg authored
When we call ieee80211_agg_start_txq(), that will in turn call schedule_and_wake_txq(). Called from ieee80211_stop_tx_ba_cb() this is done under sta->lock, which leads to certain circular lock dependencies, as reported by Chris Murphy: https://lore.kernel.org/r/CAJCQCtSXJ5qA4bqSPY=oLRMbv-irihVvP7A2uGutEbXQVkoNaw@mail.gmail.com In general, ieee80211_agg_start_txq() is usually not called with sta->lock held, only in this one place. But it's always called with sta->ampdu_mlme.mtx held, and that's therefore clearly sufficient. Change ieee80211_stop_tx_ba_cb() to also call it without the sta->lock held, by factoring it out of ieee80211_remove_tid_tx() (which is only called in this one place). This breaks the locking chain and makes it less likely that we'll have similar locking chain problems in the future. Fixes: ba8c3d6f ("mac80211: add an intermediate software queue implementation") Reported-by:
Chris Murphy <lists@colorremedies.com> Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
Luca Coelho <luciano.coelho@intel.com> Link: https://lore.kernel.org/r/iwlwifi.20211202152554.f519884c8784.I555fef8e67d93fff3d9a304886c4a9f8b322e591@changeid Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Finn Behrens authored
This removes the previously unused reload flag, which was introduced in 1eda9191 . The request is handled as NL80211_REGDOM_SET_BY_CORE, which is parsed unconditionally. Reported-by:
kernel test robot <lkp@intel.com> Reported-by:
Nathan Chancellor <nathan@kernel.org> Fixes: 1eda9191 ("nl80211: reset regdom when reloading regdb") Link: https://lore.kernel.org/all/YaZuKYM5bfWe2Urn@archlinux-ax161/ Signed-off-by:
Finn Behrens <me@kloenk.de> Reviewed-by:
Nathan Chancellor <nathan@kernel.org> Link: https://lore.kernel.org/r/YadvTolO8rQcNCd/@gimli.kloenk.dev Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Felix Fietkau authored
Sending them out on a different queue can cause a race condition where a number of packets in the queue may be discarded by the receiver, because the ADDBA request is sent too early. This affects any driver with software A-MPDU setup which does not allocate packet seqno in hardware on tx, regardless of whether iTXQ is used or not. The only driver I've seen that explicitly deals with this issue internally is mwl8k. Cc: stable@vger.kernel.org Signed-off-by:
Felix Fietkau <nbd@nbd.name> Link: https://lore.kernel.org/r/20211202124533.80388-1-nbd@nbd.name Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
- Nov 29, 2021
-
-
Finn Behrens authored
Reload the regdom when the regulatory db is reloaded. Otherwise, the user had to change the regulatoy domain to a different one and then reset it to the correct one to have a new regulatory db take effect after a reload. Signed-off-by:
Finn Behrens <fin@nyantec.com> Link: https://lore.kernel.org/r/YaIIZfxHgqc/UTA7@gimli.kloenk.dev [edit commit message] Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Johannes Berg authored
As pointed out by Stephen, add the missing docs. Reported-by:
Stephen Rothwell <sfr@canb.auug.org.au> Link: https://lore.kernel.org/r/20211129091948.1327ec82beab.Iecc5975406a3028d35c65ff8d2dec31a693888d3@changeid Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
- Nov 26, 2021
-
-
Ahmed Zaki authored
The following is from a system that went OOM due to a memory leak: wlan0: Allocated STA 74:83:c2:64:0b:87 wlan0: Allocated STA 74:83:c2:64:0b:87 wlan0: IBSS finish 74:83:c2:64:0b:87 (---from ieee80211_ibss_add_sta) wlan0: Adding new IBSS station 74:83:c2:64:0b:87 wlan0: moving STA 74:83:c2:64:0b:87 to state 2 wlan0: moving STA 74:83:c2:64:0b:87 to state 3 wlan0: Inserted STA 74:83:c2:64:0b:87 wlan0: IBSS finish 74:83:c2:64:0b:87 (---from ieee80211_ibss_work) wlan0: Adding new IBSS station 74:83:c2:64:0b:87 wlan0: moving STA 74:83:c2:64:0b:87 to state 2 wlan0: moving STA 74:83:c2:64:0b:87 to state 3 . . wlan0: expiring inactive not authorized STA 74:83:c2:64:0b:87 wlan0: moving STA 74:83:c2:64:0b:87 to state 2 wlan0: moving STA 74:83:c2:64:0b:87 to state 1 wlan0: Removed STA 74:83:c2:64:0b:87 wlan0: Destroyed STA 74:83:c2:64:0b:87 The ieee80211_ibss_finish_sta() is called twice on the same STA from 2 different locations. On the second attempt, the allocated STA is not destroyed creating a kernel memory leak. This is happening because sta_info_insert_finish() does not call sta_info_free() the second time when the STA already exists (returns -EEXIST). Note that the caller sta_info_insert_rcu() assumes STA is destroyed upon errors. Same fix is applied to -ENOMEM. Signed-off-by:
Ahmed Zaki <anzaki@gmail.com> Link: https://lore.kernel.org/r/20211002145329.3125293-1-anzaki@gmail.com [change the error path label to use the existing code] Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Xing Song authored
Mesh forwarding requires that the fwd_skb->dev is set up for TX handling, otherwise the following warning will be generated, so set it up for the pending frames. [ 72.835674 ] WARNING: CPU: 0 PID: 1193 at __skb_flow_dissect+0x284/0x1298 [ 72.842379 ] Modules linked in: ksmbd pppoe ppp_async l2tp_ppp ... [ 72.962020 ] CPU: 0 PID: 1193 Comm: kworker/u5:1 Tainted: P S 5.4.137 #0 [ 72.969938 ] Hardware name: MT7622_MT7531 RFB (DT) [ 72.974659 ] Workqueue: napi_workq napi_workfn [ 72.979025 ] pstate: 60000005 (nZCv daif -PAN -UAO) [ 72.983822 ] pc : __skb_flow_dissect+0x284/0x1298 [ 72.988444 ] lr : __skb_flow_dissect+0x54/0x1298 [ 72.992977 ] sp : ffffffc010c738c0 [ 72.996293 ] x29: ffffffc010c738c0 x28: 0000000000000000 [ 73.001615 ] x27: 000000000000ffc2 x26: ffffff800c2eb818 [ 73.006937 ] x25: ffffffc010a987c8 x24: 00000000000000ce [ 73.012259 ] x23: ffffffc010c73a28 x22: ffffffc010a99c60 [ 73.017581 ] x21: 000000000000ffc2 x20: ffffff80094da800 [ 73.022903 ] x19: 0000000000000000 x18: 0000000000000014 [ 73.028226 ] x17: 00000000084d16af x16: 00000000d1fc0bab [ 73.033548 ] x15: 00000000715f6034 x14: 000000009dbdd301 [ 73.038870 ] x13: 00000000ea4dcbc3 x12: 0000000000000040 [ 73.044192 ] x11: 000000000eb00ff0 x10: 0000000000000000 [ 73.049513 ] x9 : 000000000eb00073 x8 : 0000000000000088 [ 73.054834 ] x7 : 0000000000000000 x6 : 0000000000000001 [ 73.060155 ] x5 : 0000000000000000 x4 : 0000000000000000 [ 73.065476 ] x3 : ffffffc010a98000 x2 : 0000000000000000 [ 73.070797 ] x1 : 0000000000000000 x0 : 0000000000000000 [ 73.076120 ] Call trace: [ 73.078572 ] __skb_flow_dissect+0x284/0x1298 [ 73.082846 ] __skb_get_hash+0x7c/0x228 [ 73.086629 ] ieee80211_txq_may_transmit+0x7fc/0x17b8 [mac80211] [ 73.092564 ] ieee80211_tx_prepare_skb+0x20c/0x268 [mac80211] [ 73.098238 ] ieee80211_tx_pending+0x144/0x330 [mac80211] [ 73.103560 ] tasklet_action_common.isra.16+0xb4/0x158 [ 73.108618 ] tasklet_action+0x2c/0x38 [ 73.112286 ] __do_softirq+0x168/0x3b0 [ 73.115954 ] do_softirq.part.15+0x88/0x98 [ 73.119969 ] __local_bh_enable_ip+0xb0/0xb8 [ 73.124156 ] napi_workfn+0x58/0x90 [ 73.127565 ] process_one_work+0x20c/0x478 [ 73.131579 ] worker_thread+0x50/0x4f0 [ 73.135249 ] kthread+0x124/0x128 [ 73.138484 ] ret_from_fork+0x10/0x1c Signed-off-by:
Xing Song <xing.song@mediatek.com> Tested-By:
Frank Wunderlich <frank-w@public-files.de> Link: https://lore.kernel.org/r/20211123033123.2684-1-xing.song@mediatek.com Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Felix Fietkau authored
Some drivers that do their own sequence number allocation (e.g. ath9k) rely on being able to modify params->ssn on starting tx ampdu sessions. This was broken by a change that modified it to use sta->tid_seq[tid] instead. Cc: stable@vger.kernel.org Fixes: 31d8bb4e ("mac80211: agg-tx: refactor sending addba") Reported-by:
Eneas U de Queiroz <cotequeiroz@gmail.com> Signed-off-by:
Felix Fietkau <nbd@nbd.name> Link: https://lore.kernel.org/r/20211124094024.43222-1-nbd@nbd.name Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Felix Fietkau authored
Since retransmission clears info->control, rate control needs to be called again, otherwise the driver might crash due to invalid rates. Cc: stable@vger.kernel.org # 5.14+ Reported-by:
Aaro Koskinen <aaro.koskinen@iki.fi> Reported-by:
Robert W <rwbugreport@lost-in-the-void.net> Fixes: 03c3911d ("mac80211: call ieee80211_tx_h_rate_ctrl() when dequeue") Signed-off-by:
Felix Fietkau <nbd@nbd.name> Tested-by:
Aaro Koskinen <aaro.koskinen@iki.fi> Link: https://lore.kernel.org/r/20211122204323.9787-1-nbd@nbd.name Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Johannes Berg authored
For admission control, obviously all of that only works for QoS data frames, otherwise we cannot even access the QoS field in the header. Syzbot reported (see below) an uninitialized value here due to a status of a non-QoS nullfunc packet, which isn't even long enough to contain the QoS header. Fix this to only do anything for QoS data packets. Reported-by:
<syzbot+614e82b88a1a4973e534@syzkaller.appspotmail.com> Fixes: 02219b3a ("mac80211: add WMM admission control support") Link: https://lore.kernel.org/r/20211122124737.dad29e65902a.Ieb04587afacb27c14e0de93ec1bfbefb238cc2a0@changeid Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Maxime Bizon authored
sta is NULL for mesh point (resolved later), so sk pacing parameters were not applied. Signed-off-by:
Maxime Bizon <mbizon@freebox.fr> Link: https://lore.kernel.org/r/66f51659416ac35d6b11a313bd3ffe8b8a43dd55.camel@freebox.fr Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Jakub Kicinski authored
Jakub Kicinski says: ==================== tls: splice_read fixes As I work my way to unlocked and zero-copy TLS Rx the obvious bugs in the splice_read implementation get harder and harder to ignore. This is to say the fixes here are discovered by code inspection, I'm not aware of anyone actually using splice_read. ==================== Link: https://lore.kernel.org/r/20211124232557.2039757-1-kuba@kernel.org Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
Previous patch fixes overriding callbacks incorrectly. Triggering the crash in sendpage_locked would be more spectacular but it's hard to get to, so take the easier path of proving this is broken and call getname. We're currently getting IPv4 socket info on an IPv6 socket. Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
We replace proto_ops whenever TLS is configured for RX. But our replacement also overrides sendpage_locked, which will crash unless TX is also configured. Similarly we plug both of those in for TLS_HW (NIC crypto offload) even tho TLS_HW has a completely different implementation for TX. Last but not least we always plug in something based on inet_stream_ops even though a few of the callbacks differ for IPv6 (getname, release, bind). Use a callback building method similar to what we do for struct proto. Fixes: c46234eb ("tls: RX path for ktls") Fixes: d4ffb02d ("net/tls: enable sk_msg redirect to tls socket egress") Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
Add tests for half-received and peeked records. Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
recvmsg() will put peek()ed and partially read records onto the rx_list. splice_read() needs to consult that list otherwise it may miss data. Align with recvmsg() and also put partially-read records onto rx_list. tls_sw_advance_skb() is pretty pointless now and will be removed in net-next. Fixes: 692d7b5d ("tls: Fix recvmsg() to be able to peek across multiple records") Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
Make sure we correctly reject splicing non-data records. Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
We don't support splicing control records. TLS 1.3 changes moved the record type check into the decrypt if(). The skb may already be decrypted and still be an alert. Note that decrypt_skb_update() is idempotent and updates ctx->decrypted so the if() is pointless. Reorder the check for decryption errors with the content type check while touching them. This part is not really a bug, because if decryption failed in TLS 1.3 content type will be DATA, and for TLS 1.2 it will be correct. Nevertheless its strange to touch output before checking if the function has failed. Fixes: fedf201e ("net: tls: Refactor control message handling on recv") Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
Test broken records. Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
Add helpers for sending and receiving special record types. Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
We have the same code 3 times, about to add a fourth copy. Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
- Nov 25, 2021
-
-
Dylan Hung authored
The issue happened randomly in runtime. The message "Link is Down" is popped but soon it recovered to "Link is Up". The "Link is Down" results from the incorrect read data for reading the PHY register via MDIO bus. The correct sequence for reading the data shall be: 1. fire the command 2. wait for command done (this step was missing) 3. wait for data idle 4. read data from data register Cc: stable@vger.kernel.org Fixes: f160e994 ("net: phy: Add mdio-aspeed") Reviewed-by:
Joel Stanley <joel@jms.id.au> Signed-off-by:
Dylan Hung <dylan_hung@aspeedtech.com> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Reviewed-by:
Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Link: https://lore.kernel.org/r/20211125024432.15809-1-dylan_hung@aspeedtech.com Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jesse Brandeburg authored
Oleksandr brought a bug report where netpoll causes trace messages in the log on igb. Danielle brought this back up as still occurring, so we'll try again. [22038.710800] ------------[ cut here ]------------ [22038.710801] igb_poll+0x0/0x1440 [igb] exceeded budget in poll [22038.710802] WARNING: CPU: 12 PID: 40362 at net/core/netpoll.c:155 netpoll_poll_dev+0x18a/0x1a0 As Alex suggested, change the driver to return work_done at the exit of napi_poll, which should be safe to do in this driver because it is not polling multiple queues in this single napi context (multiple queues attached to one MSI-X vector). Several other drivers contain the same simple sequence, so I hope this will not create new problems. Fixes: 16eb8815 ("igb: Refactor clean_rx_irq to reduce overhead and improve performance") Reported-by:
Oleksandr Natalenko <oleksandr@natalenko.name> Reported-by:
Danielle Ratson <danieller@nvidia.com> Suggested-by:
Alexander Duyck <alexander.duyck@gmail.com> Signed-off-by:
Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by:
Oleksandr Natalenko <oleksandr@natalenko.name> Tested-by:
Danielle Ratson <danieller@nvidia.com> Link: https://lore.kernel.org/r/20211123204000.1597971-1-jesse.brandeburg@intel.com Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
Karsten Graul says: ==================== net/smc: fixes 2021-11-24 Patch 1 from DaXing fixes a possible loop in smc_listen(). Patch 2 prevents a NULL pointer dereferencing while iterating over the lower network devices. ==================== Link: https://lore.kernel.org/r/20211124123238.471429-1-kgraul@linux.ibm.com Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Guo DaXing authored
The kernel_listen function in smc_listen will fail when all the available ports are occupied. At this point smc->clcsock->sk->sk_data_ready has been changed to smc_clcsock_data_ready. When we call smc_listen again, now both smc->clcsock->sk->sk_data_ready and smc->clcsk_data_ready point to the smc_clcsock_data_ready function. The smc_clcsock_data_ready() function calls lsmc->clcsk_data_ready which now points to itself resulting in an infinite loop. This patch restores smc->clcsock->sk->sk_data_ready with the old value. Fixes: a60a2b1e ("net/smc: reduce active tcp_listen workers") Signed-off-by:
Guo DaXing <guodaxing@huawei.com> Acked-by:
Tony Lu <tonylu@linux.alibaba.com> Signed-off-by:
Karsten Graul <kgraul@linux.ibm.com> Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Karsten Graul authored
Coverity reports a possible NULL dereferencing problem: in smc_vlan_by_tcpsk(): 6. returned_null: netdev_lower_get_next returns NULL (checked 29 out of 30 times). 7. var_assigned: Assigning: ndev = NULL return value from netdev_lower_get_next. 1623 ndev = (struct net_device *)netdev_lower_get_next(ndev, &lower); CID 1468509 (#1 of 1): Dereference null return value (NULL_RETURNS) 8. dereference: Dereferencing a pointer that might be NULL ndev when calling is_vlan_dev. 1624 if (is_vlan_dev(ndev)) { Remove the manual implementation and use netdev_walk_all_lower_dev() to iterate over the lower devices. While on it remove an obsolete function parameter comment. Fixes: cb9d43f6 ("net/smc: determine vlan_id of stacked net_device") Suggested-by:
Julian Wiedmann <jwi@linux.ibm.com> Signed-off-by:
Karsten Graul <kgraul@linux.ibm.com> Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
Marek Behún says: ==================== phylink resolve fixes With information from me and my nagging, Russell has produced two fixes for phylink, which add code that triggers another phylink_resolve() from phylink_resolve(), if certain conditions are met: interface is being changed or link is down and previous link was up These are needed because sometimes the PCS callbacks may provide stale values if link / speed / ... ==================== Link: https://lore.kernel.org/r/20211123154403.32051-1-kabel@kernel.org Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Russell King (Oracle) authored
On mv88e6xxx 1G/2.5G PCS, the SerDes register 4.2001.2 has the following description: This register bit indicates when link was lost since the last read. For the current link status, read this register back-to-back. Thus to get current link state, we need to read the register twice. But doing that in the link change interrupt handler would lead to potentially ignoring link down events, which we really want to avoid. Thus this needs to be solved in phylink's resolve, by retriggering another resolve in the event when PCS reports link down and previous link was up, and by re-reading PCS state if the previous link was down. The wrong value is read when phylink requests change from sgmii to 2500base-x mode, and link won't come up. This fixes the bug. Fixes: 9525ae83 ("phylink: add phylink infrastructure") Signed-off-by:
Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Signed-off-by:
Marek Behún <kabel@kernel.org> Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Russell King (Oracle) authored
On PHY state change the phylink_resolve() function can read stale information from the MAC and report incorrect link speed and duplex to the kernel message log. Example with a Marvell 88X3310 PHY connected to a SerDes port on Marvell 88E6393X switch: - PHY driver triggers state change due to PHY interface mode being changed from 10gbase-r to 2500base-x due to copper change in speed from 10Gbps to 2.5Gbps, but the PHY itself either hasn't yet changed its interface to the host, or the interrupt about loss of SerDes link hadn't arrived yet (there can be a delay of several milliseconds for this), so we still think that the 10gbase-r mode is up - phylink_resolve() - phylink_mac_pcs_get_state() - this fills in speed=10g link=up - interface mode is updated to 2500base-x but speed is left at 10Gbps - phylink_major_config() - interface is changed to 2500base-x - phylink_link_up() - mv88e6xxx_mac_link_up() - .port_set_speed_duplex() - speed is set to 10Gbps - reports "Link is Up - 10Gbps/Full" to dmesg Afterwards when the interrupt finally arrives for mv88e6xxx, another resolve is forced in which we get the correct speed from phylink_mac_pcs_get_state(), but since the interface is not being changed anymore, we don't call phylink_major_config() but only phylink_mac_config(), which does not set speed/duplex anymore. To fix this, we need to force the link down and trigger another resolve on PHY interface change event. Fixes: 9525ae83 ("phylink: add phylink infrastructure") Signed-off-by:
Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Signed-off-by:
Marek Behún <kabel@kernel.org> Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Heiner Kallweit authored
Usage of phy_ethtool_get_link_ksettings() in the link status change handler isn't needed, and in combination with the referenced change it results in a deadlock. Simply remove the call and replace it with direct access to phydev->speed. The duplex argument of lan743x_phy_update_flowcontrol() isn't used and can be removed. Fixes: c10a485c ("phy: phy_ethtool_ksettings_get: Lock the phy for consistency") Reported-by:
Alessandro B Maurici <abmaurici@gmail.com> Tested-by:
Alessandro B Maurici <abmaurici@gmail.com> Signed-off-by:
Heiner Kallweit <hkallweit1@gmail.com> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Link: https://lore.kernel.org/r/40e27f76-0ba3-dcef-ee32-a78b9df38b0f@gmail.com Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Eric Dumazet authored
While testing BIG TCP patch series, I was expecting that TCP_RR workloads with 80KB requests/answers would send one 80KB TSO packet, then being received as a single GRO packet. It turns out this was not happening, and the root cause was that cubic Hystart ACK train was triggering after a few (2 or 3) rounds of RPC. Hystart was wrongly setting CWND/SSTHRESH to 30, while my RPC needed a budget of ~20 segments. Ideally these TCP_RR flows should not exit slow start. Cubic Hystart should reset itself at each round, instead of assuming every TCP flow is a bulk one. Note that even after this patch, Hystart can still trigger, depending on scheduling artifacts, but at a higher CWND/SSTHRESH threshold, keeping optimal TSO packet sizes. Tested: ip link set dev eth0 gro_ipv6_max_size 131072 gso_ipv6_max_size 131072 nstat -n; netperf -H ... -t TCP_RR -l 5 -- -r 80000,80000 -K cubic; nstat|egrep "Ip6InReceives|Hystart|Ip6OutRequests" Before: 8605 Ip6InReceives 87541 0.0 Ip6OutRequests 129496 0.0 TcpExtTCPHystartTrainDetect 1 0.0 TcpExtTCPHystartTrainCwnd 30 0.0 After: 8760 Ip6InReceives 88514 0.0 Ip6OutRequests 87975 0.0 Fixes: ae27e98a ("[TCP] CUBIC v2.3") Co-developed-by:
Neal Cardwell <ncardwell@google.com> Signed-off-by:
Neal Cardwell <ncardwell@google.com> Signed-off-by:
Eric Dumazet <edumazet@google.com> Cc: Stephen Hemminger <stephen@networkplumber.org> Cc: Yuchung Cheng <ycheng@google.com> Cc: Soheil Hassas Yeganeh <soheil@google.com> Link: https://lore.kernel.org/r/20211123202535.1843771-1-eric.dumazet@gmail.com Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Florian Fainelli authored
Update the B53 Ethernet switch section to contain drivers/net/dsa/bcm_sf2*. Reported-by:
Russell King (Oracle) <linux@armlinux.org.uk> Signed-off-by:
Florian Fainelli <f.fainelli@gmail.com> Acked-by:
Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Link: https://lore.kernel.org/r/20211123222422.3745485-1-f.fainelli@gmail.com Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Jakub Kicinski authored
Merge tag 'ieee802154-for-net-2021-11-24' of git://git.kernel.org/pub/scm/linux/kernel/git/sschmidt/wpan Stefan Schmidt says: ==================== pull-request: ieee802154 for net 2021-11-24 A fix from Alexander which has been brought up various times found by automated checkers. Make sure values are in u32 range. * tag 'ieee802154-for-net-2021-11-24' of git://git.kernel.org/pub/scm/linux/kernel/git/sschmidt/wpan: net: ieee802154: handle iftypes as u32 ==================== Link: https://lore.kernel.org/r/20211124150934.3670248-1-stefan@datenfreihafen.org Signed-off-by:
Jakub Kicinski <kuba@kernel.org>
-