summaryrefslogtreecommitdiffstats
path: root/net
Commit message (Collapse)AuthorAgeFilesLines
* mac80211: FILS AEAD protection for station mode association framesJouni Malinen2016-10-277-4/+403
| | | | | | | | | This adds support for encrypting (Re)Association Request frame and decryption (Re)Association Response frame when using FILS in station mode. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: Add FILS auth alg mappingJouni Malinen2016-10-271-0/+12
| | | | | Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: Add KEK/nonces for FILS association framesJouni Malinen2016-10-271-0/+12
| | | | | | | | | The new nl80211 attributes can be used to provide KEK and nonces to allow the driver to encrypt and decrypt FILS (Re)Association Request/Response frames in station mode. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: Add Fast Initial Link Setup (FILS) auth algsJouni Malinen2016-10-271-2/+19
| | | | | | | This defines authentication algorithms for FILS (IEEE 802.11ai). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: Allow AUTH_DATA to be used for FILSJouni Malinen2016-10-271-3/+6
| | | | | | | | | The special SAE case should be limited only for SAE since the more generic AUTH_DATA can now be used with other authentication algorithms as well. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: Rename SAE_DATA to more generic AUTH_DATAJouni Malinen2016-10-274-19/+19
| | | | | | | | | | | | | | | | | This adds defines and nl80211 extensions to allow FILS Authentication to be implemented similarly to SAE. FILS does not need the special rules for the Authentication transaction number and Status code fields, but it does need to add non-IE fields. The previously used NL80211_ATTR_SAE_DATA can be reused for this to avoid having to duplicate that implementation. Rename that attribute to more generic NL80211_ATTR_AUTH_DATA (with backwards compatibility define for NL80211_SAE_DATA). Also document the special rules related to the Authentication transaction number and Status code fiels. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* nl80211: use nla_parse_nested() instead of nla_parse()Johannes Berg2016-10-271-48/+37Star
| | | | | | | It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* nl80211: move unsplit command advertising to a separate functionJohannes Berg2016-10-271-75/+92
| | | | | | | | | | | | | | | | When we split the wiphy dump because it got too large, I added a comment and asked that all new command advertising be done only for userspace clients capable of receiving split data, in order to not break older ones (which can't use the new commands anyway) This mostly worked, and we haven't added many new commands, but I occasionally get patches that modify the wrong place. Make this easier to detect and understand by splitting out the old commands to a separate function that makes it more clear it should never be modified again. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: validate new interface's beacon intervalsJohannes Berg2016-10-272-1/+12
| | | | | | | | | | | As part of interface combination checking, verify any new interface's beacon intervals. In fact, just always add the beacon interval since that's harmless. With this patch, mac80211 is prepared for drivers that set the min_beacon_int_gcd parameter in interface combinations. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: validate beacon int as part of iface combinationsJohannes Berg2016-10-271-31/+52
| | | | | | | | | | | | | | Remove the pointless checking against interface combinations in the initial basic beacon interval validation, that currently isn't taking into account radar detection or channels properly. Instead, just validate the basic range there, and then delay real checking to the interface combination validation that drivers must do. This means that drivers wanting to use the beacon_int_min_gcd will now have to pass the new_beacon_int when validating the AP/mesh start. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: disallow beacon_int_min_gcd with IBSSJohannes Berg2016-10-271-0/+15
| | | | | | | | | | | | | | This can't really be supported right now, because the IBSS interface may change its beacon interval at any time due to joining another network; thus, there's already "support" for different beacon intervals here, implicitly. Until we figure out how we should handle this case (continue to allow it to arbitrarily join? Join only if compatible?) disallow advertising that different beacon intervals are supported if IBSS is allowed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: mesh: track (and thus validate) beacon intervalJohannes Berg2016-10-271-0/+2
| | | | | | | | This is needed for beacon interval validation; if we don't store it, then new interfaces added won't validate that the beacon interval is the same as existing ones. Fix this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: fix beacon interval in interface combination iterationJohannes Berg2016-10-271-4/+2Star
| | | | | | | | | We shouldn't abort the iteration with an error when one of the potential combinations can't accomodate the beacon interval request, we should just skip that particular combination. Fix the code to do so. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: add generic helper to check interface is runningArend Van Spriel2016-10-272-41/+20Star
| | | | | | | | | | | Add a helper using wdev to check if interface is running. This deals with both non-netdev and netdev interfaces. In struct wireless_dev replace 'p2p_started' and 'nan_started' by 'is_running' as those are mutually exclusive anyway, and unify all the code to use wdev_running(). Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* wireless: deprecate WDS and disable by defaultJohannes Berg2016-10-271-0/+10
| | | | | | | | | | | | | | | | | | | | The old WDS 4-addr frame support is very limited, e.g. * no encryption is possible on such links * it cannot support rate/HT/VHT negotiation * management APIs are very restricted These make the WDS legacy mode useless in practice. All of these are resolved by the 4-addr AP/client support, so there's also no reason to improve WDS in the future. Therefore, add a Kconfig option to disable legacy WDS. This gives people an "emergency valve" while they migrate to the better-supported 4-addr AP/client option; we plan to remove it (and the associated cfg80211/mac80211 code, which is the ultimate goal) in the future. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: allow vendor commands to be sent to nan interfaceAndrei Otcheretianski2016-10-191-1/+2
| | | | | | | | | Allow vendor commands that require WIPHY_VENDOR_CMD_NEED_RUNNING flag to be sent to NAN interface. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: allow aborting in-progress connection atttemptsIlan Peer2016-10-191-1/+1
| | | | | | | | | | | | | | | On a disconnect request from userspace, cfg80211 currently calls called rdev_disconnect() only in case that 'current_bss' was set, i.e. connection had been established. Change this to allow the userspace call to succeed and call the driver's disconnect() method also while the connection attempt is in progress, to be able to abort attempts. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> [change commit subject/message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: uapsd_queues is in QoS IE orderEmmanuel Grumbach2016-10-195-7/+20
| | | | | | | | | | | | The uapsd_queue field is in QoS IE order and not in IEEE80211_AC_*'s order. This means that mac80211 would get confused between BK and BE which is certainly not such a big deal but needs to be fixed. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: add a HW flag for supporting HW TX fragmentationSara Sharon2016-10-194-3/+8
| | | | | | | | | | | | | | | Currently mac80211 determines whether HW does fragmentation by checking whether the set_frag_threshold callback is set or not. However, some drivers may want to set the HW fragmentation capability depending on HW generation. Allow this by checking a HW flag instead of checking the callback. Signed-off-by: Sara Sharon <sara.sharon@intel.com> [added the flag to ath10k and wlcore] Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: allow the driver not to pass the tid to ieee80211_sta_uapsd_triggerEmmanuel Grumbach2016-10-191-2/+3
| | | | | | | | | | | iwlwifi will check internally that the tid maps to an AC that is trigger enabled, but can't know what tid exactly. Allow the driver to pass a generic tid and make mac80211 assume that a trigger frame was received. Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: improve RX aggregation data in debugfsJohannes Berg2016-10-191-2/+7
| | | | | | | | | | When the driver sets the SUPPORTS_REORDERING_BUFFER hardware flag, the debugfs data for RX aggregation sessions won't even indicate that a session is open. Since the previous fix to store the dialog token separately, we can indicate that it's open and add the token so that there's at least some data (ssn is not available.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: fix tid_agg_rx NULL dereferenceJohannes Berg2016-10-193-9/+5Star
| | | | | | | | | | | | | | | On drivers setting the SUPPORTS_REORDERING_BUFFER hardware flag, we crash when the peer sends an AddBA request while we already have a session open on the seame TID; this is because on those drivers, the tid_agg_rx is left NULL even though the session is valid, and the agg_session_valid bit is set. To fix this, store the dialog tokens outside the tid_agg_rx to be able to compare them to the received AddBA request. Fixes: f89e07d4cf26 ("mac80211: agg-rx: refuse ADDBA Request with timeout update") Reported-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: fix possible memory leak in cfg80211_iter_combinations()Wei Yongjun2016-10-181-1/+3
| | | | | | | | | | 'limits' is malloced in cfg80211_iter_combinations() and should be freed before leaving from the error handling cases, otherwise it will cause memory leak. Fixes: 0c317a02ca98 ("cfg80211: support virtual interfaces with different beacon intervals") Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: fix A-MSDU outer SA/DAMichael Braun2016-10-172-1/+24
| | | | | | | | | According to IEEE 802.11-2012 section 8.3.2 table 8-19, the outer SA/DA of A-MSDU frames need to be changed depending on FromDS/ToDS values. Signed-off-by: Michael Braun <michael-dev@fami-braun.de> [use ether_addr_copy and add alignment annotations] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: avoid extra memcpy in A-MSDU head creationMichael Braun2016-10-171-10/+13
| | | | | Signed-off-by: Michael Braun <michael-dev@fami-braun.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* nl80211: ifdef WoWLAN related policiesJohannes Berg2016-10-171-0/+2
| | | | | | | | To avoid unused variable warnings when CONFIG_PM isn't set, add the appropriate ifdef to the policies that are only used for WoWLAN, which can only be invoked when CONFIG_PM is set. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* nl80211: correctly use nl80211_nan_srf_policyJohannes Berg2016-10-171-1/+2
| | | | | | | This was clearly intended to be used in the attribute parsing, so do that instead of leaving the attribute policy unused. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: support virtual interfaces with different beacon intervalsPurushottam Kushwaha2016-10-133-10/+49
| | | | | | | | | | | | | | | | | | | | This commit provides a mechanism for the host drivers to advertise the support for different beacon intervals among the respective interface combinations in a group, through NL80211_IFACE_COMB_BI_MIN_GCD (u32). This value will be compared against GCD of all beaconing interfaces of matching combinations. If the driver doesn't advertise this value, the old behaviour where all beacon intervals must be identical is retained. If it is specified, then any beacon interval for an interface in the interface combination as well as the GCD of all active beacon intervals in the combination must be greater or equal to this value. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> [change commit message, some variable names, small other things] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* cfg80211: pass struct to interface combination check/iterPurushottam Kushwaha2016-10-132-40/+32Star
| | | | | | | | | | Move the growing parameter list to a structure for the interface combination check and iteration functions in cfg80211 and mac80211 to make the code easier to understand. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> [edit commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: preserve more bits when building QoS headerJohannes Berg2016-10-121-7/+16
| | | | | | | | | | | | Michael Braun reported that when trying to inject A-MSDUs over monitor interfaces, the frame doesn't come out right since the QoS header A-MSDU bit is overwritten. Rather than adding that bit specifically simply preserve those bits that we don't set here, since we typically get here with a zeroed-out QoS header anyway. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: filter multicast data packets on AP / AP_VLANMichael Braun2016-10-127-27/+58
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch adds filtering for multicast data packets on AP_VLAN interfaces that have no authorized station connected and changes filtering on AP interfaces to not count stations assigned to AP_VLAN interfaces. This saves airtime and avoids waking up other stations currently authorized in this BSS. When using WPA, the packets dropped could not be decrypted by any station. The behaviour when there are no AP_VLAN interfaces is left unchanged. When there are AP_VLAN interfaces, this patch 1. adds filtering multicast data packets sent on AP_VLAN interfaces that have no authorized station connected. No filtering happens on 4addr AP_VLAN interfaces. 2. makes filtering of multicast data packets sent on AP interfaces depend on the number of authorized stations in this bss not assigned to an AP_VLAN interface. Therefore, a new num_mcast_sta counter is added for AP_VLAN interfaces. The existing one for AP interfaces is altered to not track stations assigned to an AP_VLAN interface. The new counter is exposed in debugfs. Signed-off-by: Michael Braun <michael-dev@fami-braun.de> [reformat commit message a bit, unline ieee80211_vif_{inc,dec}_num_mcast] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: remove unnecessary num_mcast_sta checkMichael Braun2016-10-121-7/+0Star
| | | | | | | | | | | | Checking for num_mcast_sta in __ieee80211_request_smps_ap() is unnecessary as sta list will be empty in this case anyway, so the list iteration will just exit immediately. Since this isn't a "hot" code path, it doesn't really matter, and the next patch will redefine num_mcast_sta to make this check invalid. Signed-off-by: Michael Braun <michael-dev@fami-braun.de> [change commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* mac80211: remove unnecessary mesh checkJohannes Berg2016-10-121-4/+1Star
| | | | | | | | | sta_info_get_bss() is equivalent to sta_info_get() in the mesh case, since sta->sdata->bss will be NULL (it's only set for AP/AP_VLAN interfaces.) Thus, the mesh check here isn't actually needed - remove it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
* treewide: remove redundant #include <linux/kconfig.h>Masahiro Yamada2016-10-121-2/+0Star
| | | | | | | | | | | | | | | | | | | Kernel source files need not include <linux/kconfig.h> explicitly because the top Makefile forces to include it with: -include $(srctree)/include/linux/kconfig.h This commit removes explicit includes except the following: * arch/s390/include/asm/facilities_src.h * tools/testing/radix-tree/linux/kernel.h These two are used for host programs. Link: http://lkml.kernel.org/r/1473656164-11929-1-git-send-email-yamada.masahiro@socionext.com Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/netLinus Torvalds2016-10-1118-240/+357
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull networking fixes from David Miller: 1) Netfilter list handling fix, from Linus. 2) RXRPC/AFS bug fixes from David Howells (oops on call to serviceless endpoints, build warnings, missing notifications, etc.) From David Howells. 3) Kernel log message missing newlines, from Colin Ian King. 4) Don't enter direct reclaim in netlink dumps, the idea is to use a high order allocation first and fallback quickly to a 0-order allocation if such a high-order one cannot be done cheaply and without reclaim. From Eric Dumazet. 5) Fix firmware download errors in btusb bluetooth driver, from Ethan Hsieh. 6) Missing Kconfig deps for QCOM_EMAC, from Geert Uytterhoeven. 7) Fix MDIO_XGENE dup Kconfig entry. From Laura Abbott. 8) Constrain ipv6 rtr_solicits sysctl values properly, from Maciej Żenczykowski. * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (65 commits) netfilter: Fix slab corruption. be2net: Enable VF link state setting for BE3 be2net: Fix TX stats for TSO packets be2net: Update Copyright string in be_hw.h be2net: NCSI FW section should be properly updated with ethtool for BE3 be2net: Provide an alternate way to read pf_num for BEx chips wan/fsl_ucc_hdlc: Fix size used in dma_free_coherent() net: macb: NULL out phydev after removing mdio bus xen-netback: make sure that hashes are not send to unaware frontends Fixing a bug in team driver due to incorrect 'unsigned int' to 'int' conversion MAINTAINERS: add myself as a maintainer of xen-netback ipv6 addrconf: disallow rtr_solicits < -1 Bluetooth: btusb: Fix atheros firmware download error drivers: net: phy: Correct duplicate MDIO_XGENE entry ethernet: qualcomm: QCOM_EMAC should depend on HAS_DMA and HAS_IOMEM net: ethernet: mediatek: remove hwlro property in the device tree net: ethernet: mediatek: get hw lro capability by the chip id instead of by the dtsi net: ethernet: mediatek: get the chip id by ETHDMASYS registers net: bgmac: Fix errant feature flag check netlink: do not enter direct reclaim from netlink_dump() ...
| * netfilter: Fix slab corruption.Linus Torvalds2016-10-111-75/+33Star
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Use the correct pattern for singly linked list insertion and deletion. We can also calculate the list head outside of the mutex. Fixes: e3b37f11e6e4 ("netfilter: replace list_head with single linked list") Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Reviewed-by: Aaron Conole <aconole@bytheb.org> Signed-off-by: David S. Miller <davem@davemloft.net> net/netfilter/core.c | 108 ++++++++++++++++----------------------------------- 1 file changed, 33 insertions(+), 75 deletions(-)
| * Merge branch 'for-upstream' of ↵David S. Miller2016-10-083-45/+65
| |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth Johan Hedberg says: ==================== pull request: bluetooth 2016-10-08 Here are a couple of Bluetooth fixes for the 4.9 kernel: - Firmware download fix for Atheros controllers - Fixes to the content of LE scan response - New USB ID for a Marvell chipset Please let me know if there are any issues pulling. Thanks. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
| | * Bluetooth: Refactor append name and appearanceMichał Narajowski2016-10-063-50/+38Star
| | | | | | | | | | | | | | | | | | | | | Use eir_append_data to remove code duplication. Signed-off-by: Michał Narajowski <michal.narajowski@codecoup.pl> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| | * Bluetooth: Add appearance to default scan rsp dataMichał Narajowski2016-10-061-1/+10
| | | | | | | | | | | | | | | | | | | | | | | | Add appearance value to beginning of scan rsp data for default advertising instance if the value is not 0. Signed-off-by: Michał Narajowski <michal.narajowski@codecoup.pl> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| | * Bluetooth: Fix local name in scan rspMichał Narajowski2016-10-061-12/+35
| | | | | | | | | | | | | | | | | | | | | | | | | | | Use complete name if it fits. If not and there is short name check if it fits. If not then use shortened name as prefix of complete name. Signed-off-by: Michał Narajowski <michal.narajowski@codecoup.pl> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
| * | ipv6 addrconf: disallow rtr_solicits < -1Maciej Żenczykowski2016-10-081-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This disallows setting /proc/sys/net/ipv6/conf/*/router_solicitations to values below -1. -1 continues to mean an unlimited number of retransmits. Note: this depends on 'ipv6 addrconf: remove addrconf_sysctl_hop_limit()' Signed-off-by: Maciej Żenczykowski <maze@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | Merge tag 'rxrpc-rewrite-20161004' of ↵David S. Miller2016-10-0711-116/+251
| |\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs David Howells says: ==================== rxrpc: Fixes This set of patches contains a bunch of fixes: (1) Fix an oops on incoming call to a local endpoint without a bound service. (2) Only ping for a lost reply in a client call (this is inapplicable to service calls). (3) Fix maybe uninitialised variable warnings in the ACK/ABORT sending function by splitting it. (4) Fix loss of PING RESPONSE ACKs due to them being subsumed by PING ACK generation. (5) OpenAFS improperly terminates calls it makes as a client under some circumstances by not fully hard-ACK'ing the last DATA packets. This is alleviated by a new call appearing on the same channel implicitly completing the previous call on that channel. Handle this implicit completion. (6) Properly handle expiry of service calls due to the aforementioned improper termination with no follow up call to implicitly complete it: (a) The call's background processor needs to be queued to complete the call, send an abort and notify the socket. (b) The call's background processor needs to notify the socket (or the kernel service) when it has completed the call. (c) A negative error code must thence be returned to the kernel service so that it knows the call died. (d) The AFS filesystem must detect the fatal error and end the call. (7) Must produce a DELAY ACK when the actual service operation takes a while to process and must cancel the ACK when the reply is ready. (8) Don't request an ACK on the last DATA packet of the Tx phase as this confuses OpenAFS. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
| | * | rxrpc: Don't request an ACK on the last DATA packet of a call's Tx phaseDavid Howells2016-10-061-5/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Don't request an ACK on the last DATA packet of a call's Tx phase as for a client there will be a reply packet or some sort of ACK to shift phase. If the ACK is requested, OpenAFS sends a REQUESTED-ACK ACK with soft-ACKs in it and doesn't follow up with a hard-ACK. If we don't set the flag, OpenAFS will send a DELAY ACK that hard-ACKs the reply data, thereby allowing the call to terminate cleanly. Signed-off-by: David Howells <dhowells@redhat.com>
| | * | rxrpc: Need to produce an ACK for service op if op takes a long timeDavid Howells2016-10-065-6/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We need to generate a DELAY ACK from the service end of an operation if we start doing the actual operation work and it takes longer than expected. This will hard-ACK the request data and allow the client to release its resources. To make this work: (1) We have to set the ack timer and propose an ACK when the call moves to the RXRPC_CALL_SERVER_ACK_REQUEST and clear the pending ACK and cancel the timer when we start transmitting the reply (the first DATA packet of the reply implicitly ACKs the request phase). (2) It must be possible to set the timer when the caller is holding call->state_lock, so split the lock-getting part of the timer function out. (3) Add trace notes for the ACK we're requesting and the timer we clear. Signed-off-by: David Howells <dhowells@redhat.com>
| | * | rxrpc: Return negative error code to kernel serviceDavid Howells2016-10-061-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In rxrpc_kernel_recv_data(), when we return the error number incurred by a failed call, we must negate it before returning it as it's stored as positive (that's what we have to pass back to userspace). Signed-off-by: David Howells <dhowells@redhat.com>
| | * | rxrpc: Add missing notificationDavid Howells2016-10-061-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The call's background processor work item needs to notify the socket when it completes a call so that recvmsg() or the AFS fs can deal with it. Without this, call expiry isn't handled. Signed-off-by: David Howells <dhowells@redhat.com>
| | * | rxrpc: Queue the call on expiryDavid Howells2016-10-061-4/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When a call expires, it must be queued for the background processor to deal with otherwise a service call that is improperly terminated will just sit there awaiting an ACK and won't expire. Signed-off-by: David Howells <dhowells@redhat.com>
| | * | rxrpc: Partially handle OpenAFS's improper termination of callsDavid Howells2016-10-061-0/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | OpenAFS doesn't always correctly terminate client calls that it makes - this includes calls the OpenAFS servers make to the cache manager service. It should end the client call with either: (1) An ACK that has firstPacket set to one greater than the seq number of the reply DATA packet with the LAST_PACKET flag set (thereby hard-ACK'ing all packets). nAcks should be 0 and acks[] should be empty (ie. no soft-ACKs). (2) An ACKALL packet. OpenAFS, though, may send an ACK packet with firstPacket set to the last seq number or less and soft-ACKs listed for all packets up to and including the last DATA packet. The transmitter, however, is obliged to keep the call live and the soft-ACK'd DATA packets around until they're hard-ACK'd as the receiver is permitted to drop any merely soft-ACK'd packet and request retransmission by sending an ACK packet with a NACK in it. Further, OpenAFS will also terminate a client call by beginning the next client call on the same connection channel. This implicitly completes the previous call. This patch handles implicit ACK of a call on a channel by the reception of the first packet of the next call on that channel. If another call doesn't come along to implicitly ACK a call, then we have to time the call out. There are some bugs there that will be addressed in subsequent patches. Signed-off-by: David Howells <dhowells@redhat.com>
| | * | rxrpc: Fix loss of PING RESPONSE ACK production due to PING ACKsDavid Howells2016-10-068-29/+82
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Separate the output of PING ACKs from the output of other sorts of ACK so that if we receive a PING ACK and schedule transmission of a PING RESPONSE ACK, the response doesn't get cancelled by a PING ACK we happen to be scheduling transmission of at the same time. If a PING RESPONSE gets lost, the other side might just sit there waiting for it and refuse to proceed otherwise. Signed-off-by: David Howells <dhowells@redhat.com>
| | * | rxrpc: Fix warning by splitting rxrpc_send_call_packet()David Howells2016-10-068-84/+102
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Split rxrpc_send_data_packet() to separate ACK generation (which is more complicated) from ABORT generation. This simplifies the code a bit and fixes the following warning: In file included from ../net/rxrpc/output.c:20:0: net/rxrpc/output.c: In function 'rxrpc_send_call_packet': net/rxrpc/ar-internal.h:1187:27: error: 'top' may be used uninitialized in this function [-Werror=maybe-uninitialized] net/rxrpc/output.c:103:24: note: 'top' was declared here net/rxrpc/output.c:225:25: error: 'hard_ack' may be used uninitialized in this function [-Werror=maybe-uninitialized] Reported-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: David Howells <dhowells@redhat.com>
generated by cgit v1.2.3-55-g7522 (git 2.39.0) at 2024-07-12 12:29:55 +0200