- 13 Jul, 2018 13 commits
-
-
Antoine Tenart authored
This patch adds an extra indirection when setting the indirection table into the RSS hardware table to improve the packets distribution across CPUs. For example, if 2 queues are used on a multi-core system this new indirection will choose two queues on two different CPUs instead of the two first queues which are on the same first CPU. Signed-off-by:
Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by:
Maxime Chevallier <maxime.chevallier@bootlin.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Antoine Tenart authored
This patch adds the RSS indirection table support, allowing to use the ethtool -x and -X options to dump and set this table. Signed-off-by:
-
Maxime Chevallier authored
PPv2 Controller has 8 RSS Tables, of 32 entries each. A lookup in the RXQ2RSS_TABLE is performed for each incoming packet, and the RSS Table to be used is chosen according to the default rx queue that would be used for the packet. This default rx queue is set in the Lookup_id Table (also called Decoding Table), and is equal to the port->first_rxq. Since the Classifier itself isn't active at any time for the moment, this doesn't have a direct effect, the default rx queue at the moment is the one where all packets end-up into. Signed-off-by:
-
Maxime Chevallier authored
There is no RSS_TABLE register in PPv2 Controller. The register 0x1510 which was specified is actually named "RSS_HASH_SEL", but isn't used by this driver at all. Based on how this register was used, it should have been the RXQ2RSS_TABLE register, which allows to select the RSS table that will be used for the incoming packet. The RSS_TABLE_POINTER is actually a field of this RXQ2RSS_TABLE register. Since RSS tables are actually not used by the driver for now, this commit does not fix a runtime bug. Signed-off-by:
-
Antoine Tenart authored
Cosmetic patch fixing a typo in one of the RSS comments. Signed-off-by:
-
Maxime Chevallier authored
The number of receive queue per port is : - MVPP2_DEFAULT_RXQ if in single queue mode - MVPP2_DEFAULT_RXQ * num_possible_cpus if in multi queue mode with MVPP2_DEFAULT_RXQ = 4. However, we don't use the extra rx queues at the moment, we really only need one per port per CPU, until some more advanced classification rules are implemented. Suggested-by:
Stefan Chulski <stefanc@marvell.com> Signed-off-by:
-
Maxime Chevallier authored
There's a dedicated #define that indicates the number of rx queues per port per cpu, this commit removes a harcoded use of that value This doesn't fix any runtime bugs since the harcoded value matches the expected value. Signed-off-by:
-
Yan Markman authored
Since RSS only applies when we have per-cpu rx queues, it should only be enabled when the driver is configured to make use of multi-queue mode. Signed-off-by:
Yan Markman <ymarkman@marvell.com> [Maxime: Commit message] Signed-off-by:
-
Maxime Chevallier authored
The multi queue mode is needed to have RSS available, and offers some nice advantages, being able to have one rx queue vector per CPU. This mode has been usable through the use of a module parameter, this commit makes it the default value. Signed-off-by:
-
Maxime Chevallier authored
The PPv2 driver defines 2 "queue_modes" : - QDIST_SINGLE_MODE, where each port share one rx queue vector between all CPUs - QDIST_MULTI_MODE, where each port has one rx queue vector per CPU. Multi queue mode isn't available on PPv2.1, make sure we fallback to single mode when running on this revision. Signed-off-by:
-
Maxime Chevallier authored
The size of the the RSS indirection tables should be defined in mvpp2.h, so that we can use it in all files of the PPv2 driver. This commit moves the define in mvpp2.h, and adds the missing #include in mvpp2_cls.h. Signed-off-by:
-
Maxime Chevallier authored
Include guards should be put before #includes. This doesn't fix any bug, but prevent future compilation issues when adding new files in the mvpp2 driver The Header Parser init function needs the platform_device definition, and with the fixed include guards we need to add the missing include. Signed-off-by:
-
Prashant Bhole authored
Following crash occurs in validate_xmit_skb_list() when same skb is iterated multiple times in the loop and consume_skb() is called. The root cause is calling list_del_init(&skb->list) and not clearing skb->next in d4546c25. list_del_init(&skb->list) sets skb->next to point to skb itself. skb->next needs to be cleared because other parts of network stack uses another kind of SKB lists. validate_xmit_skb_list() uses such list. A similar type of bugfix was reported by Jesper Dangaard Brouer. https://patchwork.ozlabs.org/patch/942541/ This patch clears skb->next and changes list_del_init() to list_del() so that list->prev will maintain the list poison. [ 148.185511] ================================================================== [ 148.187865] BUG: KASAN: use-after-free in validate_xmit_skb_list+0x4b/0xa0 [ 148.190158] Read of size 8 at addr ffff8801e52eefc0 by task swapper/1/0 [ 148.192940] [ 148.193642] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 4.18.0-rc3+ #25 [ 148.195423] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS ?-20180531_142017-buildhw-08.phx2.fedoraproject.org-1.fc28 04/01/2014 [ 148.199129] Call Trace: [ 148.200565] <IRQ> [ 148.201911] dump_stack+0xc6/0x14c [ 148.203572] ? dump_stack_print_info.cold.1+0x2f/0x2f [ 148.205083] ? kmsg_dump_rewind_nolock+0x59/0x59 [ 148.206307] ? validate_xmit_skb+0x2c6/0x560 [ 148.207432] ? debug_show_held_locks+0x30/0x30 [ 148.208571] ? validate_xmit_skb_list+0x4b/0xa0 [ 148.211144] print_address_description+0x6c/0x23c [ 148.212601] ? validate_xmit_skb_list+0x4b/0xa0 [ 148.213782] kasan_report.cold.6+0x241/0x2fd [ 148.214958] validate_xmit_skb_list+0x4b/0xa0 [ 148.216494] sch_direct_xmit+0x1b0/0x680 [ 148.217601] ? dev_watchdog+0x4e0/0x4e0 [ 148.218675] ? do_raw_spin_trylock+0x10/0x120 [ 148.219818] ? do_raw_spin_lock+0xe0/0xe0 [ 148.221032] __dev_queue_xmit+0x1167/0x1810 [ 148.222155] ? sched_clock+0x5/0x10 [...] [ 148.474257] Allocated by task 0: [ 148.475363] kasan_kmalloc+0xbf/0xe0 [ 148.476503] kmem_cache_alloc+0xb4/0x1b0 [ 148.477654] __build_skb+0x91/0x250 [ 148.478677] build_skb+0x67/0x180 [ 148.479657] e1000_clean_rx_irq+0x542/0x8a0 [ 148.480757] e1000_clean+0x652/0xd10 [ 148.481772] net_rx_action+0x4ea/0xc20 [ 148.482808] __do_softirq+0x1f9/0x574 [ 148.483831] [ 148.484575] Freed by task 0: [ 148.485504] __kasan_slab_free+0x12e/0x180 [ 148.486589] kmem_cache_free+0xb4/0x240 [ 148.487634] kfree_skbmem+0xed/0x150 [ 148.488648] consume_skb+0x146/0x250 [ 148.489665] validate_xmit_skb+0x2b7/0x560 [ 148.490754] validate_xmit_skb_list+0x70/0xa0 [ 148.491897] sch_direct_xmit+0x1b0/0x680 [ 148.493949] __dev_queue_xmit+0x1167/0x1810 [ 148.495103] br_dev_queue_push_xmit+0xce/0x250 [ 148.496196] br_forward_finish+0x276/0x280 [ 148.497234] __br_forward+0x44f/0x520 [ 148.498260] br_forward+0x19f/0x1b0 [ 148.499264] br_handle_frame_finish+0x65e/0x980 [ 148.500398] NF_HOOK.constprop.10+0x290/0x2a0 [ 148.501522] br_handle_frame+0x417/0x640 [ 148.502582] __netif_receive_skb_core+0xaac/0x18f0 [ 148.503753] __netif_receive_skb_one_core+0x98/0x120 [ 148.504958] netif_receive_skb_internal+0xe3/0x330 [ 148.506154] napi_gro_complete+0x190/0x2a0 [ 148.507243] dev_gro_receive+0x9f7/0x1100 [ 148.508316] napi_gro_receive+0xcb/0x260 [ 148.509387] e1000_clean_rx_irq+0x2fc/0x8a0 [ 148.510501] e1000_clean+0x652/0xd10 [ 148.511523] net_rx_action+0x4ea/0xc20 [ 148.512566] __do_softirq+0x1f9/0x574 [ 148.513598] [ 148.514346] The buggy address belongs to the object at ffff8801e52eefc0 [ 148.514346] which belongs to the cache skbuff_head_cache of size 232 [ 148.517047] The buggy address is located 0 bytes inside of [ 148.517047] 232-byte region [ffff8801e52eefc0, ffff8801e52ef0a8) [ 148.519549] The buggy address belongs to the page: [ 148.520726] page:ffffea000794bb00 count:1 mapcount:0 mapping:ffff880106f4dfc0 index:0xffff8801e52ee840 compound_mapcount: 0 [ 148.524325] flags: 0x17ffffc0008100(slab|head) [ 148.525481] raw: 0017ffffc0008100 ffff880106b938d0 ffff880106b938d0 ffff880106f4dfc0 [ 148.527503] raw: ffff8801e52ee840 0000000000190011 00000001ffffffff 0000000000000000 [ 148.529547] page dumped because: kasan: bad access detected Fixes: d4546c25 ("net: Convert GRO SKB handling to list_head.") Signed-off-by:
Prashant Bhole <bhole_prashant_q7@lab.ntt.co.jp> Reported-by:
Tyler Hicks <tyhicks@canonical.com> Tested-by:
-
- 12 Jul, 2018 27 commits
-
-
David S. Miller authored
Julian Wiedmann says: ==================== s390/qeth: updates 2018-07-11 please apply this first batch of qeth patches for net-next. It brings the usual cleanups, and some performance improvements to the transmit paths. ==================== Signed-off-by: -
Julian Wiedmann authored
Move the xmit of offload-eligible (ie IPv4) traffic on OSA over to the new, copy-free path. As with L2, we'll need to preserve the skb_orphan() behaviour of the old code path until TX completion is sufficiently fast. Signed-off-by:
Julian Wiedmann <jwi@linux.ibm.com> Signed-off-by:
-
Julian Wiedmann authored
This implements a new xmit path for L3 HiperSockets, which carves the HW header from skb headroom instead of allocating it from the hdr cache. It also adds NETIF_F_SG support. The delta in qeth_l3_xmit() is all just removal of IQD-specific code and some minor consolidation. Signed-off-by:
-
Julian Wiedmann authored
In preparation for future work, move the high-level xmit work into a separate wrapper. This matches the L2 xmit code. Signed-off-by:
-
Julian Wiedmann authored
When a L3 device doesn't offer TSO, allow the stack to build full-size GSO skbs. Signed-off-by:
-
Julian Wiedmann authored
Remove some redundant EXPORTs. While at it, also move some L2-only prototypes into the proper header file. Signed-off-by:
-
Julian Wiedmann authored
Reshuffle the code a bit so that everything is in one place. Signed-off-by:
-
Julian Wiedmann authored
Consolidate duplicated code, fix the misuse of RTN_UNSPEC and simplify the handling of non-unicast traffic on IQD devices. Signed-off-by:
-
Julian Wiedmann authored
Changing a device's address lists (or its promisc mode) already triggers an RX modeset, there's no need to do it manually from the L2 driver's ndo_vlan_rx_kill_vid() hook. Also when setting a device online, dev_open() already calls dev_set_rx_mode(). So a manual modeset is only necessary from the recovery path. Signed-off-by:
-
Julian Wiedmann authored
Except for tracing, the pointer is not used. At the same time, accessing it from qeth_qdio_output_handler() is racy: whenever qeth_qdio_cq_handler() gets control, its call to qeth_qdio_handle_aob() frees the AOB. So the AOB pointer that qeth_qdio_output_handler() stores into 'buffer' can go stale at any time, and trigger a use-after-free. Signed-off-by:
-
Julian Wiedmann authored
Use the new qeth_scrub_qdio_buffer() helper, remove an extra parameter from qeth_clear_output_buffer(), init the bufstates.user field just once (in qeth_flush_buffers()) and remove some noisy trace messages. Signed-off-by:
-
Jesper Dangaard Brouer authored
In commit 5fa12739 ("net: ipv4: listify ip_rcv_finish") calling dst_input(skb) was split-out. The ip_sublist_rcv_finish() just calls dst_input(skb) in a loop. The problem is that ip_sublist_rcv_finish() forgot to remove the SKB from the list before invoking dst_input(). Further more we need to clear skb->next as other parts of the network stack use another kind of SKB lists for xmit_more (see dev_hard_start_xmit). A crash occurs if e.g. dst_input() invoke ip_forward(), which calls dst_output()/ip_output() that eventually calls __dev_queue_xmit() + sch_direct_xmit(), and a crash occurs in validate_xmit_skb_list(). This patch only fixes the crash, but there is a huge potential for a performance boost if we can pass an SKB-list through to ip_forward. Fixes: 5fa12739 ("net: ipv4: listify ip_rcv_finish") Signed-off-by:
Jesper Dangaard Brouer <brouer@redhat.com> Acked-by:
Edward Cree <ecree@solarflare.com> Signed-off-by:
-
Arnd Bergmann authored
getnstimeofday64 is deprecated in favor of the ktime_get() family of functions. The direct replacement would be ktime_get_real_ts64(), but I'm picking the basic ktime_get() instead: - using a ktime_t simplifies the code compared to timespec64 - using monotonic time instead of real time avoids issues caused by a concurrent settimeofday() or during a leap second adjustment. Acked-by:
Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
Arnd Bergmann authored
The two do the same thing, but we want to have a consistent naming in the kernel. Signed-off-by:
Felix Manlunas <felix.manlunas@cavium.com> Signed-off-by:
-
David S. Miller authored
Davide Caratti says: ==================== net/sched: act_skbedit: lockless data path the data path of act_skbedit can be faster if we avoid using spinlocks: - patch 1 converts act_skbedit statistics to use per-cpu counters - patch 2 lets act_skbedit use RCU to read/update its configuration test procedure (using pktgen from https://github.com/netoptimizer): # ip link add name eth1 type dummy # ip link set dev eth1 up # tc qdisc add dev eth1 clsact # tc filter add dev eth1 egress matchall action skbedit priority c1a0:c1a0 # for c in 1 2 4 ; do > ./pktgen_bench_xmit_mode_queue_xmit.sh -v -s 64 -t $c -n 5000000 -i eth1 > done test results (avg. pps/thread) $c | before patch | after patch | improvement ----+--------------+--------------+------------ 1 | 3917464 ± 3% | 4000458 ± 3% | irrelevant 2 | 3455367 ± 4% | 3953076 ± 1% | +14% 4 | 2496594 ± 2% | 3801123 ± 3% | +52% v2: rebased on latest net-next ==================== Signed-off-by:
-
Davide Caratti authored
use RCU instead of spin_{,un}lock_bh, to protect concurrent read/write on act_skbedit configuration. This reduces the effects of contention in the data path, in case multiple readers are present. Signed-off-by:Davide Caratti <dcaratti@redhat.com> Signed-off-by:
-
Davide Caratti authored
use per-CPU counters, instead of sharing a single set of stats with all cores: this removes the need of spinlocks when stats are read/updated. Signed-off-by:
-
Arnd Bergmann authored
Using get_seconds() for timestamps is deprecated since it can lead to overflows on 32-bit systems. While the interface generally doesn't overflow until year 2106, the specific implementation of the TCP PAWS algorithm breaks in 2038 when the intermediate signed 32-bit timestamps overflow. A related problem is that the local timestamps in CLOCK_REALTIME form lead to unexpected behavior when settimeofday is called to set the system clock backwards or forwards by more than 24 days. While the first problem could be solved by using an overflow-safe method of comparing the timestamps, a nicer solution is to use a monotonic clocksource with ktime_get_seconds() that simply doesn't overflow (at least not until 136 years after boot) and that doesn't change during settimeofday(). To make 32-bit and 64-bit architectures behave the same way here, and also save a few bytes in the tcp_options_received structure, I'm changing the type to a 32-bit integer, which is now safe on all architectures. Finally, the ts_recent_stamp field also (confusingly) gets used to store a jiffies value in tcp_synq_overflow()/tcp_synq_no_recent_overflow(). This is currently safe, but changing the type to 32-bit requires some small changes there to keep it working. Signed-off-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
-
Vakul Garg authored
Instead of kzalloc/free for aead_request allocation and free, use functions aead_request_alloc(), aead_request_free(). It ensures that any sensitive crypto material held in crypto transforms is securely erased from memory. Signed-off-by:
Vakul Garg <vakul.garg@nxp.com> Acked-by:
Dave Watson <davejwatson@fb.com> Signed-off-by:
-
Pieter Jansen van Vuuren authored
Extend tc tunnel_key action unit tests with geneve options. Tests include testing single and multiple geneve options, as well as testing geneve options that are expected to fail. Signed-off-by:
Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Acked-by:
Lucas Bates <lucasb@mojatatu.com> Signed-off-by:
-
David S. Miller authored
Ivan Vecera says: ==================== be2net: small structures clean-up The series: - removes unused / unneccessary fields in several be2net structures - re-order fields in some structures to eliminate holes, cache-lines crosses - as result reduces size of main struct be_adapter by 4kB ==================== Signed-off-by: -
Ivan Vecera authored
The current position of .rss_flags field in struct rss_info causes that fields .rsstable and .rssqueue (both 128 bytes long) crosses cache-line boundaries. Moving it at the end properly align all fields. Before patch: struct rss_info { u64 rss_flags; /* 0 8 */ u8 rsstable[128]; /* 8 128 */ /* --- cacheline 2 boundary (128 bytes) was 8 bytes ago --- */ u8 rss_queue[128]; /* 136 128 */ /* --- cacheline 4 boundary (256 bytes) was 8 bytes ago --- */ u8 rss_hkey[40]; /* 264 40 */ }; After patch: struct rss_info { u8 rsstable[128]; /* 0 128 */ /* --- cacheline 2 boundary (128 bytes) --- */ u8 rss_queue[128]; /* 128 128 */ /* --- cacheline 4 boundary (256 bytes) --- */ u8 rss_hkey[40]; /* 256 40 */ u64 rss_flags; /* 296 8 */ }; Signed-off-by:Ivan Vecera <cera@cera.cz> Signed-off-by:
-
Ivan Vecera authored
- Unionize two u8 fields where only one of them is used depending on NIC chipset. - Move recovery_supported field after that union These changes eliminate 7-bytes hole in the struct and makes it smaller by 8 bytes. Signed-off-by:
-
Ivan Vecera authored
Signed-off-by:
-
Ivan Vecera authored
Before patch: struct be_tx_obj { u32 db_offset; /* 0 4 */ /* XXX 4 bytes hole, try to pack */ struct be_queue_info q; /* 8 56 */ /* --- cacheline 1 boundary (64 bytes) --- */ struct be_queue_info cq; /* 64 56 */ struct be_tx_compl_info txcp; /* 120 4 */ /* XXX 4 bytes hole, try to pack */ /* --- cacheline 2 boundary (128 bytes) --- */ struct sk_buff * sent_skb_list[2048]; /* 128 16384 */ ... }: After patch: struct be_tx_obj { u32 db_offset; /* 0 4 */ struct be_tx_compl_info txcp; /* 4 4 */ struct be_queue_info q; /* 8 56 */ /* --- cacheline 1 boundary (64 bytes) --- */ struct be_queue_info cq; /* 64 56 */ struct sk_buff * sent_skb_list[2048]; /* 120 16384 */ ... }; Signed-off-by: -
Ivan Vecera authored
Re-order fields in struct be_eq_obj to ensure that .napi field begins at start of cache-line. Also the .adapter field is moved to the first cache-line next to .q field and 3 fields (idx,msi_idx,spurious_intr) and the 4-bytes hole to 3rd cache-line. Signed-off-by:
-
Ivan Vecera authored
The event queue description (be_eq_obj.desc) field is used only to format string for IRQ name and it is not really needed to hold this value. Remove it and use local variable to format string for IRQ name. Signed-off-by:
-
