From 65161fb544aada499c912b6010a8f7d8e04f6130 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 17 Feb 2025 11:11:28 -0800 Subject: net: Fix dev_net(dev) race in unregister_netdevice_notifier_dev_net(). After the cited commit, dev_net(dev) is fetched before holding RTNL and passed to __unregister_netdevice_notifier_net(). However, dev_net(dev) might be different after holding RTNL. In the reported case [0], while removing a VF device, its netns was being dismantled and the VF was moved to init_net. So the following sequence is basically illegal when dev was fetched without lookup: net = dev_net(dev); rtnl_net_lock(net); Let's use a new helper rtnl_net_dev_lock() to fix the race. It fetches dev_net_rcu(dev), bumps its net->passive, and checks if dev_net_rcu(dev) is changed after rtnl_net_lock(). [0]: BUG: KASAN: slab-use-after-free in notifier_call_chain (kernel/notifier.c:75 (discriminator 2)) Read of size 8 at addr ffff88810cefb4c8 by task test-bridge-lag/21127 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 Call Trace: dump_stack_lvl (lib/dump_stack.c:123) print_report (mm/kasan/report.c:379 mm/kasan/report.c:489) kasan_report (mm/kasan/report.c:604) notifier_call_chain (kernel/notifier.c:75 (discriminator 2)) call_netdevice_notifiers_info (net/core/dev.c:2011) unregister_netdevice_many_notify (net/core/dev.c:11551) unregister_netdevice_queue (net/core/dev.c:11487) unregister_netdev (net/core/dev.c:11635) mlx5e_remove (drivers/net/ethernet/mellanox/mlx5/core/en_main.c:6552 drivers/net/ethernet/mellanox/mlx5/core/en_main.c:6579) mlx5_core auxiliary_bus_remove (drivers/base/auxiliary.c:230) device_release_driver_internal (drivers/base/dd.c:1275 drivers/base/dd.c:1296) bus_remove_device (./include/linux/kobject.h:193 drivers/base/base.h:73 drivers/base/bus.c:583) device_del (drivers/base/power/power.h:142 drivers/base/core.c:3855) mlx5_rescan_drivers_locked (./include/linux/auxiliary_bus.h:241 drivers/net/ethernet/mellanox/mlx5/core/dev.c:333 drivers/net/ethernet/mellanox/mlx5/core/dev.c:535 drivers/net/ethernet/mellanox/mlx5/core/dev.c:549) mlx5_core mlx5_unregister_device (drivers/net/ethernet/mellanox/mlx5/core/dev.c:468) mlx5_core mlx5_uninit_one (./include/linux/instrumented.h:68 ./include/asm-generic/bitops/instrumented-non-atomic.h:141 drivers/net/ethernet/mellanox/mlx5/core/main.c:1563) mlx5_core remove_one (drivers/net/ethernet/mellanox/mlx5/core/main.c:965 drivers/net/ethernet/mellanox/mlx5/core/main.c:2019) mlx5_core pci_device_remove (./include/linux/pm_runtime.h:129 drivers/pci/pci-driver.c:475) device_release_driver_internal (drivers/base/dd.c:1275 drivers/base/dd.c:1296) unbind_store (drivers/base/bus.c:245) kernfs_fop_write_iter (fs/kernfs/file.c:338) vfs_write (fs/read_write.c:587 (discriminator 1) fs/read_write.c:679 (discriminator 1)) ksys_write (fs/read_write.c:732) do_syscall_64 (arch/x86/entry/common.c:52 (discriminator 1) arch/x86/entry/common.c:83 (discriminator 1)) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130) RIP: 0033:0x7f6a4d5018b7 Fixes: 7fb1073300a2 ("net: Hold rtnl_net_lock() in (un)?register_netdevice_notifier_dev_net().") Reported-by: Yael Chemla Closes: https://lore.kernel.org/netdev/146eabfe-123c-4970-901e-e961b4c09bc3@nvidia.com/ Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20250217191129.19967-3-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/core/dev.c | 48 ++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 44 insertions(+), 4 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index b91658e8aedb..19e268568282 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2070,6 +2070,42 @@ static void __move_netdevice_notifier_net(struct net *src_net, __register_netdevice_notifier_net(dst_net, nb, true); } +static void rtnl_net_dev_lock(struct net_device *dev) +{ + bool again; + + do { + struct net *net; + + again = false; + + /* netns might be being dismantled. */ + rcu_read_lock(); + net = dev_net_rcu(dev); + net_passive_inc(net); + rcu_read_unlock(); + + rtnl_net_lock(net); + +#ifdef CONFIG_NET_NS + /* dev might have been moved to another netns. */ + if (!net_eq(net, rcu_access_pointer(dev->nd_net.net))) { + rtnl_net_unlock(net); + net_passive_dec(net); + again = true; + } +#endif + } while (again); +} + +static void rtnl_net_dev_unlock(struct net_device *dev) +{ + struct net *net = dev_net(dev); + + rtnl_net_unlock(net); + net_passive_dec(net); +} + int register_netdevice_notifier_dev_net(struct net_device *dev, struct notifier_block *nb, struct netdev_net_notifier *nn) @@ -2077,6 +2113,11 @@ int register_netdevice_notifier_dev_net(struct net_device *dev, struct net *net = dev_net(dev); int err; + /* rtnl_net_lock() assumes dev is not yet published by + * register_netdevice(). + */ + DEBUG_NET_WARN_ON_ONCE(!list_empty(&dev->dev_list)); + rtnl_net_lock(net); err = __register_netdevice_notifier_net(net, nb, false); if (!err) { @@ -2093,13 +2134,12 @@ int unregister_netdevice_notifier_dev_net(struct net_device *dev, struct notifier_block *nb, struct netdev_net_notifier *nn) { - struct net *net = dev_net(dev); int err; - rtnl_net_lock(net); + rtnl_net_dev_lock(dev); list_del(&nn->list); - err = __unregister_netdevice_notifier_net(net, nb); - rtnl_net_unlock(net); + err = __unregister_netdevice_notifier_net(dev_net(dev), nb); + rtnl_net_dev_unlock(dev); return err; } -- cgit v1.2.3 From d4c6bfc83936cb61fac99e9891c406fbdd40f964 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 17 Feb 2025 11:11:29 -0800 Subject: dev: Use rtnl_net_dev_lock() in unregister_netdev(). The following sequence is basically illegal when dev was fetched without lookup because dev_net(dev) might be different after holding rtnl_net_lock(): net = dev_net(dev); rtnl_net_lock(net); Let's use rtnl_net_dev_lock() in unregister_netdev(). Note that there is no real bug in unregister_netdev() for now because RTNL protects the scope even if dev_net(dev) is changed before/after RTNL. Fixes: 00fb9823939e ("dev: Hold per-netns RTNL in (un)?register_netdev().") Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20250217191129.19967-4-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/core/dev.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'net/core/dev.c') diff --git a/net/core/dev.c b/net/core/dev.c index 19e268568282..fafd2f4b5d5d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -11920,11 +11920,9 @@ EXPORT_SYMBOL(unregister_netdevice_many); */ void unregister_netdev(struct net_device *dev) { - struct net *net = dev_net(dev); - - rtnl_net_lock(net); + rtnl_net_dev_lock(dev); unregister_netdevice(dev); - rtnl_net_unlock(net); + rtnl_net_dev_unlock(dev); } EXPORT_SYMBOL(unregister_netdev); -- cgit v1.2.3 From 4b5a28b38c4a0106c64416a1b2042405166b26ce Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 18 Feb 2025 05:49:30 -0800 Subject: net: Add non-RCU dev_getbyhwaddr() helper Add dedicated helper for finding devices by hardware address when holding rtnl_lock, similar to existing dev_getbyhwaddr_rcu(). This prevents PROVE_LOCKING warnings when rtnl_lock is held but RCU read lock is not. Extract common address comparison logic into dev_addr_cmp(). The context about this change could be found in the following discussion: Link: https://lore.kernel.org/all/20250206-scarlet-ermine-of-improvement-1fcac5@leitao/ Cc: kuniyu@amazon.com Cc: ushankar@purestorage.com Suggested-by: Eric Dumazet Signed-off-by: Breno Leitao Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250218-arm_fix_selftest-v5-1-d3d6892db9e1@debian.org Signed-off-by: Jakub Kicinski --- include/linux/netdevice.h | 2 ++ net/core/dev.c | 37 ++++++++++++++++++++++++++++++++++--- 2 files changed, 36 insertions(+), 3 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c0a86afb85da..94b7d4eca003 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3275,6 +3275,8 @@ static inline struct net_device *first_net_device_rcu(struct net *net) } int netdev_boot_setup_check(struct net_device *dev); +struct net_device *dev_getbyhwaddr(struct net *net, unsigned short type, + const char *hwaddr); struct net_device *dev_getbyhwaddr_rcu(struct net *net, unsigned short type, const char *hwaddr); struct net_device *dev_getfirstbyhwtype(struct net *net, unsigned short type); diff --git a/net/core/dev.c b/net/core/dev.c index fafd2f4b5d5d..72459dd02f38 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1121,6 +1121,12 @@ out: return ret; } +static bool dev_addr_cmp(struct net_device *dev, unsigned short type, + const char *ha) +{ + return dev->type == type && !memcmp(dev->dev_addr, ha, dev->addr_len); +} + /** * dev_getbyhwaddr_rcu - find a device by its hardware address * @net: the applicable net namespace @@ -1129,7 +1135,7 @@ out: * * Search for an interface by MAC address. Returns NULL if the device * is not found or a pointer to the device. - * The caller must hold RCU or RTNL. + * The caller must hold RCU. * The returned device has not had its ref count increased * and the caller must therefore be careful about locking * @@ -1141,14 +1147,39 @@ struct net_device *dev_getbyhwaddr_rcu(struct net *net, unsigned short type, struct net_device *dev; for_each_netdev_rcu(net, dev) - if (dev->type == type && - !memcmp(dev->dev_addr, ha, dev->addr_len)) + if (dev_addr_cmp(dev, type, ha)) return dev; return NULL; } EXPORT_SYMBOL(dev_getbyhwaddr_rcu); +/** + * dev_getbyhwaddr() - find a device by its hardware address + * @net: the applicable net namespace + * @type: media type of device + * @ha: hardware address + * + * Similar to dev_getbyhwaddr_rcu(), but the owner needs to hold + * rtnl_lock. + * + * Context: rtnl_lock() must be held. + * Return: pointer to the net_device, or NULL if not found + */ +struct net_device *dev_getbyhwaddr(struct net *net, unsigned short type, + const char *ha) +{ + struct net_device *dev; + + ASSERT_RTNL(); + for_each_netdev(net, dev) + if (dev_addr_cmp(dev, type, ha)) + return dev; + + return NULL; +} +EXPORT_SYMBOL(dev_getbyhwaddr); + struct net_device *dev_getfirstbyhwtype(struct net *net, unsigned short type) { struct net_device *dev, *ret = NULL; -- cgit v1.2.3 From 6bc7e4eb0499562ccd291712fd7be0d1a5aad00a Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 18 Feb 2025 19:29:40 +0100 Subject: Revert "net: skb: introduce and use a single page frag cache" After the previous commit is finally safe to revert commit dbae2b062824 ("net: skb: introduce and use a single page frag cache"): do it here. The intended goal of such change was to counter a performance regression introduced by commit 3226b158e67c ("net: avoid 32 x truesize under-estimation for tiny skbs"). Unfortunately, the blamed commit introduces another regression for the virtio_net driver. Such a driver calls napi_alloc_skb() with a tiny size, so that the whole head frag could fit a 512-byte block. The single page frag cache uses a 1K fragment for such allocation, and the additional overhead, under small UDP packets flood, makes the page allocator a bottleneck. Thanks to commit bf9f1baa279f ("net: add dedicated kmem_cache for typical/small skb->head"), this revert does not re-introduce the original regression. Actually, in the relevant test on top of this revert, I measure a small but noticeable positive delta, just above noise level. The revert itself required some additional mangling due to recent updates in the affected code. Suggested-by: Eric Dumazet Fixes: dbae2b062824 ("net: skb: introduce and use a single page frag cache") Reviewed-by: Eric Dumazet Signed-off-by: Paolo Abeni --- include/linux/netdevice.h | 1 - net/core/dev.c | 17 ++++++++ net/core/skbuff.c | 104 +++------------------------------------------- 3 files changed, 22 insertions(+), 100 deletions(-) (limited to 'net/core/dev.c') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 94b7d4eca003..ab550a89b9bf 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -4117,7 +4117,6 @@ void netif_receive_skb_list(struct list_head *head); gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb); void napi_gro_flush(struct napi_struct *napi, bool flush_old); struct sk_buff *napi_get_frags(struct napi_struct *napi); -void napi_get_frags_check(struct napi_struct *napi); gro_result_t napi_gro_frags(struct napi_struct *napi); static inline void napi_free_frags(struct napi_struct *napi) diff --git a/net/core/dev.c b/net/core/dev.c index 72459dd02f38..1b252e9459fd 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6991,6 +6991,23 @@ netif_napi_dev_list_add(struct net_device *dev, struct napi_struct *napi) list_add_rcu(&napi->dev_list, higher); /* adds after higher */ } +/* Double check that napi_get_frags() allocates skbs with + * skb->head being backed by slab, not a page fragment. + * This is to make sure bug fixed in 3226b158e67c + * ("net: avoid 32 x truesize under-estimation for tiny skbs") + * does not accidentally come back. + */ +static void napi_get_frags_check(struct napi_struct *napi) +{ + struct sk_buff *skb; + + local_bh_disable(); + skb = napi_get_frags(napi); + WARN_ON_ONCE(skb && skb->head_frag); + napi_free_frags(napi); + local_bh_enable(); +} + void netif_napi_add_weight_locked(struct net_device *dev, struct napi_struct *napi, int (*poll)(struct napi_struct *, int), diff --git a/net/core/skbuff.c b/net/core/skbuff.c index f5a6d50570c4..7b03b64fdcb2 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -223,67 +223,9 @@ static void skb_under_panic(struct sk_buff *skb, unsigned int sz, void *addr) #define NAPI_SKB_CACHE_BULK 16 #define NAPI_SKB_CACHE_HALF (NAPI_SKB_CACHE_SIZE / 2) -#if PAGE_SIZE == SZ_4K - -#define NAPI_HAS_SMALL_PAGE_FRAG 1 -#define NAPI_SMALL_PAGE_PFMEMALLOC(nc) ((nc).pfmemalloc) - -/* specialized page frag allocator using a single order 0 page - * and slicing it into 1K sized fragment. Constrained to systems - * with a very limited amount of 1K fragments fitting a single - * page - to avoid excessive truesize underestimation - */ - -struct page_frag_1k { - void *va; - u16 offset; - bool pfmemalloc; -}; - -static void *page_frag_alloc_1k(struct page_frag_1k *nc, gfp_t gfp) -{ - struct page *page; - int offset; - - offset = nc->offset - SZ_1K; - if (likely(offset >= 0)) - goto use_frag; - - page = alloc_pages_node(NUMA_NO_NODE, gfp, 0); - if (!page) - return NULL; - - nc->va = page_address(page); - nc->pfmemalloc = page_is_pfmemalloc(page); - offset = PAGE_SIZE - SZ_1K; - page_ref_add(page, offset / SZ_1K); - -use_frag: - nc->offset = offset; - return nc->va + offset; -} -#else - -/* the small page is actually unused in this build; add dummy helpers - * to please the compiler and avoid later preprocessor's conditionals - */ -#define NAPI_HAS_SMALL_PAGE_FRAG 0 -#define NAPI_SMALL_PAGE_PFMEMALLOC(nc) false - -struct page_frag_1k { -}; - -static void *page_frag_alloc_1k(struct page_frag_1k *nc, gfp_t gfp_mask) -{ - return NULL; -} - -#endif - struct napi_alloc_cache { local_lock_t bh_lock; struct page_frag_cache page; - struct page_frag_1k page_small; unsigned int skb_count; void *skb_cache[NAPI_SKB_CACHE_SIZE]; }; @@ -293,23 +235,6 @@ static DEFINE_PER_CPU(struct napi_alloc_cache, napi_alloc_cache) = { .bh_lock = INIT_LOCAL_LOCK(bh_lock), }; -/* Double check that napi_get_frags() allocates skbs with - * skb->head being backed by slab, not a page fragment. - * This is to make sure bug fixed in 3226b158e67c - * ("net: avoid 32 x truesize under-estimation for tiny skbs") - * does not accidentally come back. - */ -void napi_get_frags_check(struct napi_struct *napi) -{ - struct sk_buff *skb; - - local_bh_disable(); - skb = napi_get_frags(napi); - WARN_ON_ONCE(!NAPI_HAS_SMALL_PAGE_FRAG && skb && skb->head_frag); - napi_free_frags(napi); - local_bh_enable(); -} - void *__napi_alloc_frag_align(unsigned int fragsz, unsigned int align_mask) { struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache); @@ -816,11 +741,8 @@ struct sk_buff *napi_alloc_skb(struct napi_struct *napi, unsigned int len) /* If requested length is either too small or too big, * we use kmalloc() for skb->head allocation. - * When the small frag allocator is available, prefer it over kmalloc - * for small fragments */ - if ((!NAPI_HAS_SMALL_PAGE_FRAG && - len <= SKB_WITH_OVERHEAD(SKB_SMALL_HEAD_CACHE_SIZE)) || + if (len <= SKB_WITH_OVERHEAD(SKB_SMALL_HEAD_CACHE_SIZE) || len > SKB_WITH_OVERHEAD(PAGE_SIZE) || (gfp_mask & (__GFP_DIRECT_RECLAIM | GFP_DMA))) { skb = __alloc_skb(len, gfp_mask, SKB_ALLOC_RX | SKB_ALLOC_NAPI, @@ -830,32 +752,16 @@ struct sk_buff *napi_alloc_skb(struct napi_struct *napi, unsigned int len) goto skb_success; } + len = SKB_HEAD_ALIGN(len); + if (sk_memalloc_socks()) gfp_mask |= __GFP_MEMALLOC; local_lock_nested_bh(&napi_alloc_cache.bh_lock); nc = this_cpu_ptr(&napi_alloc_cache); - if (NAPI_HAS_SMALL_PAGE_FRAG && len <= SKB_WITH_OVERHEAD(1024)) { - /* we are artificially inflating the allocation size, but - * that is not as bad as it may look like, as: - * - 'len' less than GRO_MAX_HEAD makes little sense - * - On most systems, larger 'len' values lead to fragment - * size above 512 bytes - * - kmalloc would use the kmalloc-1k slab for such values - * - Builds with smaller GRO_MAX_HEAD will very likely do - * little networking, as that implies no WiFi and no - * tunnels support, and 32 bits arches. - */ - len = SZ_1K; - data = page_frag_alloc_1k(&nc->page_small, gfp_mask); - pfmemalloc = NAPI_SMALL_PAGE_PFMEMALLOC(nc->page_small); - } else { - len = SKB_HEAD_ALIGN(len); - - data = page_frag_alloc(&nc->page, len, gfp_mask); - pfmemalloc = page_frag_cache_is_pfmemalloc(&nc->page); - } + data = page_frag_alloc(&nc->page, len, gfp_mask); + pfmemalloc = page_frag_cache_is_pfmemalloc(&nc->page); local_unlock_nested_bh(&napi_alloc_cache.bh_lock); if (unlikely(!data)) -- cgit v1.2.3