From ef1c75593e770aff8749e902aa0deb6855a3f485 Mon Sep 17 00:00:00 2001 From: Rohit Maheshwari Date: Wed, 3 Jun 2020 09:58:13 +0530 Subject: crypto/chcr: error seen if CONFIG_CHELSIO_TLS_DEVICE isn't set cxgb4_uld_in_use() is used only by cxgb4_ktls_det_feature() which is under CONFIG_CHELSIO_TLS_DEVICE macro. Fixes: a3ac249a1ab5 ("cxgb4/chcr: Enable ktls settings at run time") Signed-off-by: Rohit Maheshwari Signed-off-by: David S. Miller --- drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c index 0307e9c69a47..08439e215efe 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c @@ -663,6 +663,7 @@ static int uld_attach(struct adapter *adap, unsigned int uld) return 0; } +#ifdef CONFIG_CHELSIO_TLS_DEVICE static bool cxgb4_uld_in_use(struct adapter *adap) { const struct tid_info *t = &adap->tids; @@ -670,7 +671,6 @@ static bool cxgb4_uld_in_use(struct adapter *adap) return (atomic_read(&t->conns_in_use) || t->stids_in_use); } -#ifdef CONFIG_CHELSIO_TLS_DEVICE /* cxgb4_set_ktls_feature: request FW to enable/disable ktls settings. * @adap: adapter info * @enable: 1 to enable / 0 to disable ktls settings. -- cgit v1.2.3 From c36f05559104b66bcd7f617e931e38c680227b74 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Tue, 2 Jun 2020 21:49:10 -0700 Subject: genetlink: fix memory leaks in genl_family_rcv_msg_dumpit() There are two kinds of memory leaks in genl_family_rcv_msg_dumpit(): 1. Before we call ops->start(), whenever an error happens, we forget to free the memory allocated in genl_family_rcv_msg_dumpit(). 2. When ops->start() fails, the 'info' has been already installed on the per socket control block, so we should not free it here. More importantly, nlk->cb_running is still false at this point, so netlink_sock_destruct() cannot free it either. The first kind of memory leaks is easier to resolve, but the second one requires some deeper thoughts. After reviewing how netfilter handles this, the most elegant solution I find is just to use a similar way to allocate the memory, that is, moving memory allocations from caller into ops->start(). With this, we can solve both kinds of memory leaks: for 1), no memory allocation happens before ops->start(); for 2), ops->start() handles its own failures and 'info' is installed to the socket control block only when success. The only ugliness here is we have to pass all local variables on stack via a struct, but this is not hard to understand. Alternatively, we can introduce a ops->free() to solve this too, but it is overkill as only genetlink has this problem so far. Fixes: 1927f41a22a0 ("net: genetlink: introduce dump info struct to be available during dumpit op") Reported-by: syzbot+21f04f481f449c8db840@syzkaller.appspotmail.com Cc: "Jason A. Donenfeld" Cc: Florian Westphal Cc: Pablo Neira Ayuso Cc: Jiri Pirko Cc: YueHaibing Cc: Shaochun Chen Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- net/netlink/genetlink.c | 94 ++++++++++++++++++++++++++++++------------------- 1 file changed, 58 insertions(+), 36 deletions(-) diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index 2f049692e012..6c19b91bbb86 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -513,15 +513,58 @@ static void genl_family_rcv_msg_attrs_free(const struct genl_family *family, kfree(attrbuf); } -static int genl_lock_start(struct netlink_callback *cb) +struct genl_start_context { + const struct genl_family *family; + struct nlmsghdr *nlh; + struct netlink_ext_ack *extack; + const struct genl_ops *ops; + int hdrlen; +}; + +static int genl_start(struct netlink_callback *cb) { - const struct genl_ops *ops = genl_dumpit_info(cb)->ops; + struct genl_start_context *ctx = cb->data; + const struct genl_ops *ops = ctx->ops; + struct genl_dumpit_info *info; + struct nlattr **attrs = NULL; int rc = 0; + if (ops->validate & GENL_DONT_VALIDATE_DUMP) + goto no_attrs; + + if (ctx->nlh->nlmsg_len < nlmsg_msg_size(ctx->hdrlen)) + return -EINVAL; + + attrs = genl_family_rcv_msg_attrs_parse(ctx->family, ctx->nlh, ctx->extack, + ops, ctx->hdrlen, + GENL_DONT_VALIDATE_DUMP_STRICT, + true); + if (IS_ERR(attrs)) + return PTR_ERR(attrs); + +no_attrs: + info = genl_dumpit_info_alloc(); + if (!info) { + kfree(attrs); + return -ENOMEM; + } + info->family = ctx->family; + info->ops = ops; + info->attrs = attrs; + + cb->data = info; if (ops->start) { - genl_lock(); + if (!ctx->family->parallel_ops) + genl_lock(); rc = ops->start(cb); - genl_unlock(); + if (!ctx->family->parallel_ops) + genl_unlock(); + } + + if (rc) { + kfree(attrs); + genl_dumpit_info_free(info); + cb->data = NULL; } return rc; } @@ -548,7 +591,7 @@ static int genl_lock_done(struct netlink_callback *cb) rc = ops->done(cb); genl_unlock(); } - genl_family_rcv_msg_attrs_free(info->family, info->attrs, true); + genl_family_rcv_msg_attrs_free(info->family, info->attrs, false); genl_dumpit_info_free(info); return rc; } @@ -573,43 +616,23 @@ static int genl_family_rcv_msg_dumpit(const struct genl_family *family, const struct genl_ops *ops, int hdrlen, struct net *net) { - struct genl_dumpit_info *info; - struct nlattr **attrs = NULL; + struct genl_start_context ctx; int err; if (!ops->dumpit) return -EOPNOTSUPP; - if (ops->validate & GENL_DONT_VALIDATE_DUMP) - goto no_attrs; - - if (nlh->nlmsg_len < nlmsg_msg_size(hdrlen)) - return -EINVAL; - - attrs = genl_family_rcv_msg_attrs_parse(family, nlh, extack, - ops, hdrlen, - GENL_DONT_VALIDATE_DUMP_STRICT, - true); - if (IS_ERR(attrs)) - return PTR_ERR(attrs); - -no_attrs: - /* Allocate dumpit info. It is going to be freed by done() callback. */ - info = genl_dumpit_info_alloc(); - if (!info) { - genl_family_rcv_msg_attrs_free(family, attrs, true); - return -ENOMEM; - } - - info->family = family; - info->ops = ops; - info->attrs = attrs; + ctx.family = family; + ctx.nlh = nlh; + ctx.extack = extack; + ctx.ops = ops; + ctx.hdrlen = hdrlen; if (!family->parallel_ops) { struct netlink_dump_control c = { .module = family->module, - .data = info, - .start = genl_lock_start, + .data = &ctx, + .start = genl_start, .dump = genl_lock_dumpit, .done = genl_lock_done, }; @@ -617,12 +640,11 @@ no_attrs: genl_unlock(); err = __netlink_dump_start(net->genl_sock, skb, nlh, &c); genl_lock(); - } else { struct netlink_dump_control c = { .module = family->module, - .data = info, - .start = ops->start, + .data = &ctx, + .start = genl_start, .dump = ops->dumpit, .done = genl_parallel_done, }; -- cgit v1.2.3 From 5e9eeccc58f3e6bcc99b929670665d2ce047e9c9 Mon Sep 17 00:00:00 2001 From: Tuong Lien Date: Wed, 3 Jun 2020 12:06:01 +0700 Subject: tipc: fix NULL pointer dereference in streaming syzbot found the following crash: general protection fault, probably for non-canonical address 0xdffffc0000000019: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x00000000000000c8-0x00000000000000cf] CPU: 1 PID: 7060 Comm: syz-executor394 Not tainted 5.7.0-rc6-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__tipc_sendstream+0xbde/0x11f0 net/tipc/socket.c:1591 Code: 00 00 00 00 48 39 5c 24 28 48 0f 44 d8 e8 fa 3e db f9 48 b8 00 00 00 00 00 fc ff df 48 8d bb c8 00 00 00 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 e2 04 00 00 48 8b 9b c8 00 00 00 48 b8 00 00 00 RSP: 0018:ffffc90003ef7818 EFLAGS: 00010202 RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffffff8797fd9d RDX: 0000000000000019 RSI: ffffffff8797fde6 RDI: 00000000000000c8 RBP: ffff888099848040 R08: ffff88809a5f6440 R09: fffffbfff1860b4c R10: ffffffff8c305a5f R11: fffffbfff1860b4b R12: ffff88809984857e R13: 0000000000000000 R14: ffff888086aa4000 R15: 0000000000000000 FS: 00000000009b4880(0000) GS:ffff8880ae700000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020000140 CR3: 00000000a7fdf000 CR4: 00000000001406e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: tipc_sendstream+0x4c/0x70 net/tipc/socket.c:1533 sock_sendmsg_nosec net/socket.c:652 [inline] sock_sendmsg+0xcf/0x120 net/socket.c:672 ____sys_sendmsg+0x32f/0x810 net/socket.c:2352 ___sys_sendmsg+0x100/0x170 net/socket.c:2406 __sys_sendmmsg+0x195/0x480 net/socket.c:2496 __do_sys_sendmmsg net/socket.c:2525 [inline] __se_sys_sendmmsg net/socket.c:2522 [inline] __x64_sys_sendmmsg+0x99/0x100 net/socket.c:2522 do_syscall_64+0xf6/0x7d0 arch/x86/entry/common.c:295 entry_SYSCALL_64_after_hwframe+0x49/0xb3 RIP: 0033:0x440199 ... This bug was bisected to commit 0a3e060f340d ("tipc: add test for Nagle algorithm effectiveness"). However, it is not the case, the trouble was from the base in the case of zero data length message sending, we would unexpectedly make an empty 'txq' queue after the 'tipc_msg_append()' in Nagle mode. A similar crash can be generated even without the bisected patch but at the link layer when it accesses the empty queue. We solve the issues by building at least one buffer to go with socket's header and an optional data section that may be empty like what we had with the 'tipc_msg_build()'. Note: the previous commit 4c21daae3dbc ("tipc: Fix NULL pointer dereference in __tipc_sendstream()") is obsoleted by this one since the 'txq' will be never empty and the check of 'skb != NULL' is unnecessary but it is safe anyway. Reported-by: syzbot+8eac6d030e7807c21d32@syzkaller.appspotmail.com Fixes: c0bceb97db9e ("tipc: add smart nagle feature") Acked-by: Jon Maloy Signed-off-by: Tuong Lien Signed-off-by: David S. Miller --- net/tipc/msg.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/tipc/msg.c b/net/tipc/msg.c index c0afcd627c5e..046e4cb3acea 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -221,7 +221,7 @@ int tipc_msg_append(struct tipc_msg *_hdr, struct msghdr *m, int dlen, accounted = skb ? msg_blocks(buf_msg(skb)) : 0; total = accounted; - while (rem) { + do { if (!skb || skb->len >= mss) { skb = tipc_buf_acquire(mss, GFP_KERNEL); if (unlikely(!skb)) @@ -245,7 +245,7 @@ int tipc_msg_append(struct tipc_msg *_hdr, struct msghdr *m, int dlen, skb_put(skb, cpy); rem -= cpy; total += msg_blocks(hdr) - curr; - } + } while (rem); return total - accounted; } -- cgit v1.2.3 From bb986a50421a11bf31a81afb15b9b8f45a4a3a11 Mon Sep 17 00:00:00 2001 From: Ahmed Abdelsalam Date: Wed, 3 Jun 2020 06:54:42 +0000 Subject: seg6: fix seg6_validate_srh() to avoid slab-out-of-bounds MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The seg6_validate_srh() is used to validate SRH for three cases: case1: SRH of data-plane SRv6 packets to be processed by the Linux kernel. Case2: SRH of the netlink message received from user-space (iproute2) Case3: SRH injected into packets through setsockopt In case1, the SRH can be encoded in the Reduced way (i.e., first SID is carried in DA only and not represented as SID in the SRH) and the seg6_validate_srh() now handles this case correctly. In case2 and case3, the SRH shouldn’t be encoded in the Reduced way otherwise we lose the first segment (i.e., the first hop). The current implementation of the seg6_validate_srh() allow SRH of case2 and case3 to be encoded in the Reduced way. This leads a slab-out-of-bounds problem. This patch verifies SRH of case1, case2 and case3. Allowing case1 to be reduced while preventing SRH of case2 and case3 from being reduced . Reported-by: syzbot+e8c028b62439eac42073@syzkaller.appspotmail.com Reported-by: YueHaibing Fixes: 0cb7498f234e ("seg6: fix SRH processing to comply with RFC8754") Signed-off-by: Ahmed Abdelsalam Signed-off-by: David S. Miller --- include/net/seg6.h | 2 +- net/core/filter.c | 2 +- net/ipv6/ipv6_sockglue.c | 2 +- net/ipv6/seg6.c | 16 ++++++++++------ net/ipv6/seg6_iptunnel.c | 2 +- net/ipv6/seg6_local.c | 6 +++--- 6 files changed, 17 insertions(+), 13 deletions(-) diff --git a/include/net/seg6.h b/include/net/seg6.h index 640724b35273..9d19c15e8545 100644 --- a/include/net/seg6.h +++ b/include/net/seg6.h @@ -57,7 +57,7 @@ extern void seg6_iptunnel_exit(void); extern int seg6_local_init(void); extern void seg6_local_exit(void); -extern bool seg6_validate_srh(struct ipv6_sr_hdr *srh, int len); +extern bool seg6_validate_srh(struct ipv6_sr_hdr *srh, int len, bool reduced); extern int seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh, int proto); extern int seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh); diff --git a/net/core/filter.c b/net/core/filter.c index d01a244b5087..209482a4eaa2 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -5050,7 +5050,7 @@ static int bpf_push_seg6_encap(struct sk_buff *skb, u32 type, void *hdr, u32 len int err; struct ipv6_sr_hdr *srh = (struct ipv6_sr_hdr *)hdr; - if (!seg6_validate_srh(srh, len)) + if (!seg6_validate_srh(srh, len, false)) return -EINVAL; switch (type) { diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 2c843ff5e3a9..20576e87a5f7 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -493,7 +493,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, struct ipv6_sr_hdr *srh = (struct ipv6_sr_hdr *) opt->srcrt; - if (!seg6_validate_srh(srh, optlen)) + if (!seg6_validate_srh(srh, optlen, false)) goto sticky_done; break; } diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c index 37b434293bda..d2f8138e5a73 100644 --- a/net/ipv6/seg6.c +++ b/net/ipv6/seg6.c @@ -25,7 +25,7 @@ #include #endif -bool seg6_validate_srh(struct ipv6_sr_hdr *srh, int len) +bool seg6_validate_srh(struct ipv6_sr_hdr *srh, int len, bool reduced) { unsigned int tlv_offset; int max_last_entry; @@ -37,13 +37,17 @@ bool seg6_validate_srh(struct ipv6_sr_hdr *srh, int len) if (((srh->hdrlen + 1) << 3) != len) return false; - max_last_entry = (srh->hdrlen / 2) - 1; - - if (srh->first_segment > max_last_entry) + if (!reduced && srh->segments_left > srh->first_segment) { return false; + } else { + max_last_entry = (srh->hdrlen / 2) - 1; - if (srh->segments_left > srh->first_segment + 1) - return false; + if (srh->first_segment > max_last_entry) + return false; + + if (srh->segments_left > srh->first_segment + 1) + return false; + } tlv_offset = sizeof(*srh) + ((srh->first_segment + 1) << 4); diff --git a/net/ipv6/seg6_iptunnel.c b/net/ipv6/seg6_iptunnel.c index c7cbfeae94f5..e0e9f48ab14f 100644 --- a/net/ipv6/seg6_iptunnel.c +++ b/net/ipv6/seg6_iptunnel.c @@ -426,7 +426,7 @@ static int seg6_build_state(struct net *net, struct nlattr *nla, } /* verify that SRH is consistent */ - if (!seg6_validate_srh(tuninfo->srh, tuninfo_len - sizeof(*tuninfo))) + if (!seg6_validate_srh(tuninfo->srh, tuninfo_len - sizeof(*tuninfo), false)) return -EINVAL; newts = lwtunnel_state_alloc(tuninfo_len + sizeof(*slwt)); diff --git a/net/ipv6/seg6_local.c b/net/ipv6/seg6_local.c index 52493423f329..eba23279912d 100644 --- a/net/ipv6/seg6_local.c +++ b/net/ipv6/seg6_local.c @@ -87,7 +87,7 @@ static struct ipv6_sr_hdr *get_srh(struct sk_buff *skb) */ srh = (struct ipv6_sr_hdr *)(skb->data + srhoff); - if (!seg6_validate_srh(srh, len)) + if (!seg6_validate_srh(srh, len, true)) return NULL; return srh; @@ -495,7 +495,7 @@ bool seg6_bpf_has_valid_srh(struct sk_buff *skb) return false; srh->hdrlen = (u8)(srh_state->hdrlen >> 3); - if (!seg6_validate_srh(srh, (srh->hdrlen + 1) << 3)) + if (!seg6_validate_srh(srh, (srh->hdrlen + 1) << 3, true)) return false; srh_state->valid = true; @@ -670,7 +670,7 @@ static int parse_nla_srh(struct nlattr **attrs, struct seg6_local_lwt *slwt) if (len < sizeof(*srh) + sizeof(struct in6_addr)) return -EINVAL; - if (!seg6_validate_srh(srh, len)) + if (!seg6_validate_srh(srh, len, false)) return -EINVAL; slwt->srh = kmemdup(srh, len, GFP_KERNEL); -- cgit v1.2.3 From cd07ecccba13b8bd5023ffe7be57363d07e3105f Mon Sep 17 00:00:00 2001 From: Sameeh Jubran Date: Wed, 3 Jun 2020 08:50:22 +0000 Subject: net: ena: xdp: XDP_TX: fix memory leak When sending very high packet rate, the XDP tx queues can get full and start dropping packets. In this case we don't free the pages which results in ena driver draining the system memory. Fix: Simply free the pages when necessary. Fixes: 548c4940b9f1 ("net: ena: Implement XDP_TX action") Signed-off-by: Sameeh Jubran Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_netdev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index a0af74c93971..e101fc934f83 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -355,7 +355,7 @@ error_unmap_dma: ena_unmap_tx_buff(xdp_ring, tx_info); tx_info->xdpf = NULL; error_drop_packet: - + __free_page(tx_info->xdp_rx_page); return NETDEV_TX_OK; } -- cgit v1.2.3 From 3921a81c31df6057183aeb7f7d204003bf699d6f Mon Sep 17 00:00:00 2001 From: Sameeh Jubran Date: Wed, 3 Jun 2020 08:50:23 +0000 Subject: net: ena: xdp: update napi budget for DROP and ABORTED This patch fixes two issues with XDP: 1. If the XDP verdict is XDP_ABORTED we break the loop, which results in us handling one buffer per napi cycle instead of the total budget (usually 64). To overcome this simply change the xdp_verdict check to != XDP_PASS. When the verdict is XDP_PASS, the skb is not expected to be NULL. 2. Update the residual budget for XDP_DROP and XDP_ABORTED, since packets are handled in these cases. Fixes: 548c4940b9f1 ("net: ena: Implement XDP_TX action") Signed-off-by: Sameeh Jubran Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_netdev.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index e101fc934f83..dda4b8fc9525 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -1646,11 +1646,9 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, &next_to_clean); if (unlikely(!skb)) { - if (xdp_verdict == XDP_TX) { + if (xdp_verdict == XDP_TX) ena_free_rx_page(rx_ring, &rx_ring->rx_buffer_info[rx_ring->ena_bufs[0].req_id]); - res_budget--; - } for (i = 0; i < ena_rx_ctx.descs; i++) { rx_ring->free_ids[next_to_clean] = rx_ring->ena_bufs[i].req_id; @@ -1658,8 +1656,10 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, ENA_RX_RING_IDX_NEXT(next_to_clean, rx_ring->ring_size); } - if (xdp_verdict == XDP_TX || xdp_verdict == XDP_DROP) + if (xdp_verdict != XDP_PASS) { + res_budget--; continue; + } break; } -- cgit v1.2.3 From 9d149045b3c0e44c049cdbce8a64e19415290017 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Wed, 3 Jun 2020 11:12:14 +0200 Subject: geneve: change from tx_error to tx_dropped on missing metadata If the geneve interface is in collect_md (external) mode, it can't send any packets submitted directly to its net interface, as such packets won't have metadata attached. This is expected. However, the kernel itself sends some packets to the interface, most notably, IPv6 DAD, IPv6 multicast listener reports, etc. This is not wrong, as tunnel metadata can be specified in routing table (although technically, that has never worked for IPv6, but hopefully will be fixed eventually) and then the interface must correctly participate in IPv6 housekeeping. The problem is that any such attempt increases the tx_error counter. Just bringing up a geneve interface with IPv6 enabled is enough to see a number of tx_errors. That causes confusion among users, prompting them to find a network error where there is none. Change the counter used to tx_dropped. That better conveys the meaning (there's nothing wrong going on, just some packets are getting dropped) and hopefully will make admins panic less. Signed-off-by: Jiri Benc Signed-off-by: David S. Miller --- drivers/net/geneve.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index 6b461be1820b..75266580b586 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -987,9 +987,10 @@ static netdev_tx_t geneve_xmit(struct sk_buff *skb, struct net_device *dev) if (geneve->collect_md) { info = skb_tunnel_info(skb); if (unlikely(!info || !(info->mode & IP_TUNNEL_INFO_TX))) { - err = -EINVAL; netdev_dbg(dev, "no tunnel metadata\n"); - goto tx_error; + dev_kfree_skb(skb); + dev->stats.tx_dropped++; + return NETDEV_TX_OK; } } else { info = &geneve->info; @@ -1006,7 +1007,7 @@ static netdev_tx_t geneve_xmit(struct sk_buff *skb, struct net_device *dev) if (likely(!err)) return NETDEV_TX_OK; -tx_error: + dev_kfree_skb(skb); if (err == -ELOOP) -- cgit v1.2.3 From 67122a7910bf2135dc7f7ececfcf16a5bdb362c1 Mon Sep 17 00:00:00 2001 From: Michal Vokáč Date: Wed, 3 Jun 2020 13:31:39 +0200 Subject: net: dsa: qca8k: Fix "Unexpected gfp" kernel exception MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 7e99e3470172 ("net: dsa: remove dsa_switch_alloc helper") replaced the dsa_switch_alloc helper by devm_kzalloc in all DSA drivers. Unfortunately it introduced a typo in qca8k.c driver and wrong argument is passed to the devm_kzalloc function. This fix mitigates the following kernel exception: Unexpected gfp: 0x6 (__GFP_HIGHMEM|GFP_DMA32). Fixing up to gfp: 0x101 (GFP_DMA|__GFP_ZERO). Fix your code! CPU: 1 PID: 44 Comm: kworker/1:1 Not tainted 5.5.9-yocto-ua #1 Hardware name: Freescale i.MX6 Quad/DualLite (Device Tree) Workqueue: events deferred_probe_work_func [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [] (show_stack) from [] (dump_stack+0x90/0xa4) [] (dump_stack) from [] (new_slab+0x20c/0x214) [] (new_slab) from [] (___slab_alloc.constprop.0+0x1b8/0x540) [] (___slab_alloc.constprop.0) from [] (__slab_alloc.constprop.0+0x1c/0x24) [] (__slab_alloc.constprop.0) from [] (__kmalloc_track_caller+0x1b0/0x298) [] (__kmalloc_track_caller) from [] (devm_kmalloc+0x24/0x70) [] (devm_kmalloc) from [] (qca8k_sw_probe+0x94/0x1ac) [] (qca8k_sw_probe) from [] (mdio_probe+0x30/0x54) [] (mdio_probe) from [] (really_probe+0x1e0/0x348) [] (really_probe) from [] (driver_probe_device+0x60/0x16c) [] (driver_probe_device) from [] (bus_for_each_drv+0x70/0x94) [] (bus_for_each_drv) from [] (__device_attach+0xb4/0x11c) [] (__device_attach) from [] (bus_probe_device+0x84/0x8c) [] (bus_probe_device) from [] (deferred_probe_work_func+0x64/0x90) [] (deferred_probe_work_func) from [] (process_one_work+0x1d4/0x41c) [] (process_one_work) from [] (worker_thread+0x248/0x528) [] (worker_thread) from [] (kthread+0x124/0x150) [] (kthread) from [] (ret_from_fork+0x14/0x3c) Exception stack(0xee1b5fb0 to 0xee1b5ff8) 5fa0: 00000000 00000000 00000000 00000000 5fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 5fe0: 00000000 00000000 00000000 00000000 00000013 00000000 qca8k 2188000.ethernet-1:0a: Using legacy PHYLIB callbacks. Please migrate to PHYLINK! qca8k 2188000.ethernet-1:0a eth2 (uninitialized): PHY [2188000.ethernet-1:01] driver [Generic PHY] qca8k 2188000.ethernet-1:0a eth1 (uninitialized): PHY [2188000.ethernet-1:02] driver [Generic PHY] Fixes: 7e99e3470172 ("net: dsa: remove dsa_switch_alloc helper") Signed-off-by: Michal Vokáč Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/dsa/qca8k.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/dsa/qca8k.c b/drivers/net/dsa/qca8k.c index 9f4205b4439b..d2b5ab403e06 100644 --- a/drivers/net/dsa/qca8k.c +++ b/drivers/net/dsa/qca8k.c @@ -1079,8 +1079,7 @@ qca8k_sw_probe(struct mdio_device *mdiodev) if (id != QCA8K_ID_QCA8337) return -ENODEV; - priv->ds = devm_kzalloc(&mdiodev->dev, sizeof(*priv->ds), - QCA8K_NUM_PORTS); + priv->ds = devm_kzalloc(&mdiodev->dev, sizeof(*priv->ds), GFP_KERNEL); if (!priv->ds) return -ENOMEM; -- cgit v1.2.3 From 11d6011c2cf29f7c8181ebde6c8bc0c4d83adcd7 Mon Sep 17 00:00:00 2001 From: "Ahmed S. Darwish" Date: Wed, 3 Jun 2020 16:49:44 +0200 Subject: net: core: device_rename: Use rwsem instead of a seqcount Sequence counters write paths are critical sections that must never be preempted, and blocking, even for CONFIG_PREEMPTION=n, is not allowed. Commit 5dbe7c178d3f ("net: fix kernel deadlock with interface rename and netdev name retrieval.") handled a deadlock, observed with CONFIG_PREEMPTION=n, where the devnet_rename seqcount read side was infinitely spinning: it got scheduled after the seqcount write side blocked inside its own critical section. To fix that deadlock, among other issues, the commit added a cond_resched() inside the read side section. While this will get the non-preemptible kernel eventually unstuck, the seqcount reader is fully exhausting its slice just spinning -- until TIF_NEED_RESCHED is set. The fix is also still broken: if the seqcount reader belongs to a real-time scheduling policy, it can spin forever and the kernel will livelock. Disabling preemption over the seqcount write side critical section will not work: inside it are a number of GFP_KERNEL allocations and mutex locking through the drivers/base/ :: device_rename() call chain. >From all the above, replace the seqcount with a rwsem. Fixes: 5dbe7c178d3f (net: fix kernel deadlock with interface rename and netdev name retrieval.) Fixes: 30e6c9fa93cf (net: devnet_rename_seq should be a seqcount) Fixes: c91f6df2db49 (sockopt: Change getsockopt() of SO_BINDTODEVICE to return an interface name) Cc: Reported-by: kbuild test robot [ v1 missing up_read() on error exit ] Reported-by: Dan Carpenter [ v1 missing up_read() on error exit ] Signed-off-by: Ahmed S. Darwish Reviewed-by: Sebastian Andrzej Siewior Signed-off-by: David S. Miller --- net/core/dev.c | 40 ++++++++++++++++++---------------------- 1 file changed, 18 insertions(+), 22 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index 10684833f864..061496a1f640 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -79,6 +79,7 @@ #include #include #include +#include #include #include #include @@ -194,7 +195,7 @@ static DEFINE_SPINLOCK(napi_hash_lock); static unsigned int napi_gen_id = NR_CPUS; static DEFINE_READ_MOSTLY_HASHTABLE(napi_hash, 8); -static seqcount_t devnet_rename_seq; +static DECLARE_RWSEM(devnet_rename_sem); static inline void dev_base_seq_inc(struct net *net) { @@ -998,33 +999,28 @@ EXPORT_SYMBOL(dev_get_by_napi_id); * @net: network namespace * @name: a pointer to the buffer where the name will be stored. * @ifindex: the ifindex of the interface to get the name from. - * - * The use of raw_seqcount_begin() and cond_resched() before - * retrying is required as we want to give the writers a chance - * to complete when CONFIG_PREEMPTION is not set. */ int netdev_get_name(struct net *net, char *name, int ifindex) { struct net_device *dev; - unsigned int seq; + int ret; -retry: - seq = raw_seqcount_begin(&devnet_rename_seq); + down_read(&devnet_rename_sem); rcu_read_lock(); + dev = dev_get_by_index_rcu(net, ifindex); if (!dev) { - rcu_read_unlock(); - return -ENODEV; + ret = -ENODEV; + goto out; } strcpy(name, dev->name); - rcu_read_unlock(); - if (read_seqcount_retry(&devnet_rename_seq, seq)) { - cond_resched(); - goto retry; - } - return 0; + ret = 0; +out: + rcu_read_unlock(); + up_read(&devnet_rename_sem); + return ret; } /** @@ -1296,10 +1292,10 @@ int dev_change_name(struct net_device *dev, const char *newname) likely(!(dev->priv_flags & IFF_LIVE_RENAME_OK))) return -EBUSY; - write_seqcount_begin(&devnet_rename_seq); + down_write(&devnet_rename_sem); if (strncmp(newname, dev->name, IFNAMSIZ) == 0) { - write_seqcount_end(&devnet_rename_seq); + up_write(&devnet_rename_sem); return 0; } @@ -1307,7 +1303,7 @@ int dev_change_name(struct net_device *dev, const char *newname) err = dev_get_valid_name(net, dev, newname); if (err < 0) { - write_seqcount_end(&devnet_rename_seq); + up_write(&devnet_rename_sem); return err; } @@ -1322,11 +1318,11 @@ rollback: if (ret) { memcpy(dev->name, oldname, IFNAMSIZ); dev->name_assign_type = old_assign_type; - write_seqcount_end(&devnet_rename_seq); + up_write(&devnet_rename_sem); return ret; } - write_seqcount_end(&devnet_rename_seq); + up_write(&devnet_rename_sem); netdev_adjacent_rename_links(dev, oldname); @@ -1347,7 +1343,7 @@ rollback: /* err >= 0 after dev_alloc_name() or stores the first errno */ if (err >= 0) { err = ret; - write_seqcount_begin(&devnet_rename_seq); + down_write(&devnet_rename_sem); memcpy(dev->name, oldname, IFNAMSIZ); memcpy(oldname, newname, IFNAMSIZ); dev->name_assign_type = old_assign_type; -- cgit v1.2.3 From 79cbb6bc3332da7162c2581e151659ab8ebaa528 Mon Sep 17 00:00:00 2001 From: "Ahmed S. Darwish" Date: Wed, 3 Jun 2020 16:49:45 +0200 Subject: net: phy: fixed_phy: Remove unused seqcount Commit bf7afb29d545 ("phy: improve safety of fixed-phy MII register reading") protected the fixed PHY status with a sequence counter. Two years later, commit d2b977939b18 ("net: phy: fixed-phy: remove fixed_phy_update_state()") removed the sequence counter's write side critical section -- neutralizing its read side retry loop. Remove the unused seqcount. Signed-off-by: Ahmed S. Darwish Reviewed-by: Sebastian Andrzej Siewior Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/phy/fixed_phy.c | 28 +++++++++++----------------- 1 file changed, 11 insertions(+), 17 deletions(-) diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index 4a3d34f40cb9..c4641b1704d6 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -19,7 +19,6 @@ #include #include #include -#include #include #include #include @@ -34,7 +33,6 @@ struct fixed_mdio_bus { struct fixed_phy { int addr; struct phy_device *phydev; - seqcount_t seqcount; struct fixed_phy_status status; bool no_carrier; int (*link_update)(struct net_device *, struct fixed_phy_status *); @@ -80,19 +78,17 @@ static int fixed_mdio_read(struct mii_bus *bus, int phy_addr, int reg_num) list_for_each_entry(fp, &fmb->phys, node) { if (fp->addr == phy_addr) { struct fixed_phy_status state; - int s; - - do { - s = read_seqcount_begin(&fp->seqcount); - fp->status.link = !fp->no_carrier; - /* Issue callback if user registered it. */ - if (fp->link_update) - fp->link_update(fp->phydev->attached_dev, - &fp->status); - /* Check the GPIO for change in status */ - fixed_phy_update(fp); - state = fp->status; - } while (read_seqcount_retry(&fp->seqcount, s)); + + fp->status.link = !fp->no_carrier; + + /* Issue callback if user registered it. */ + if (fp->link_update) + fp->link_update(fp->phydev->attached_dev, + &fp->status); + + /* Check the GPIO for change in status */ + fixed_phy_update(fp); + state = fp->status; return swphy_read_reg(reg_num, &state); } @@ -150,8 +146,6 @@ static int fixed_phy_add_gpiod(unsigned int irq, int phy_addr, if (!fp) return -ENOMEM; - seqcount_init(&fp->seqcount); - if (irq != PHY_POLL) fmb->mii_bus->irq[phy_addr] = irq; -- cgit v1.2.3 From 6501bf87602f799b7e502014f8bc0aa58b868277 Mon Sep 17 00:00:00 2001 From: "Ahmed S. Darwish" Date: Wed, 3 Jun 2020 16:49:46 +0200 Subject: u64_stats: Document writer non-preemptibility requirement MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The u64_stats mechanism uses sequence counters to protect against 64-bit values tearing on 32-bit architectures. Updating such statistics is a sequence counter write side critical section. Preemption must be disabled before entering this seqcount write critical section. Failing to do so, the seqcount read side can preempt the write side section and spin for the entire scheduler tick. If that reader belongs to a real-time scheduling class, it can spin forever and the kernel will livelock. Document this statistics update side non-preemptibility requirement. Reword the introductory paragraph to highlight u64_stats raison d'être: 64-bit values tearing protection on 32-bit architectures. Divide documentation on a basis of internal design vs. usage constraints. Reword the u64_stats header file top comment to always mention "Reader" or "Writer" at the start of each bullet point, making it easier to follow which side each point is actually for. Clarify the statement "whole thing is a NOOP on 64bit arches or UP kernels". For 32-bit UP kernels, preemption is always disabled for the statistics read side section. Signed-off-by: Ahmed S. Darwish Reviewed-by: Sebastian Andrzej Siewior Signed-off-by: David S. Miller --- include/linux/u64_stats_sync.h | 43 ++++++++++++++++++++++-------------------- 1 file changed, 23 insertions(+), 20 deletions(-) diff --git a/include/linux/u64_stats_sync.h b/include/linux/u64_stats_sync.h index 9de5c10293f5..c6abb79501b3 100644 --- a/include/linux/u64_stats_sync.h +++ b/include/linux/u64_stats_sync.h @@ -3,33 +3,36 @@ #define _LINUX_U64_STATS_SYNC_H /* - * To properly implement 64bits network statistics on 32bit and 64bit hosts, - * we provide a synchronization point, that is a noop on 64bit or UP kernels. + * Protect against 64-bit values tearing on 32-bit architectures. This is + * typically used for statistics read/update in different subsystems. * * Key points : - * 1) Use a seqcount on SMP 32bits, with low overhead. - * 2) Whole thing is a noop on 64bit arches or UP kernels. - * 3) Write side must ensure mutual exclusion or one seqcount update could + * + * - Use a seqcount on 32-bit SMP, only disable preemption for 32-bit UP. + * - The whole thing is a no-op on 64-bit architectures. + * + * Usage constraints: + * + * 1) Write side must ensure mutual exclusion, or one seqcount update could * be lost, thus blocking readers forever. - * If this synchronization point is not a mutex, but a spinlock or - * spinlock_bh() or disable_bh() : - * 3.1) Write side should not sleep. - * 3.2) Write side should not allow preemption. - * 3.3) If applicable, interrupts should be disabled. * - * 4) If reader fetches several counters, there is no guarantee the whole values - * are consistent (remember point 1) : this is a noop on 64bit arches anyway) + * 2) Write side must disable preemption, or a seqcount reader can preempt the + * writer and also spin forever. + * + * 3) Write side must use the _irqsave() variant if other writers, or a reader, + * can be invoked from an IRQ context. * - * 5) readers are allowed to sleep or be preempted/interrupted : They perform - * pure reads. But if they have to fetch many values, it's better to not allow - * preemptions/interruptions to avoid many retries. + * 4) If reader fetches several counters, there is no guarantee the whole values + * are consistent w.r.t. each other (remember point #2: seqcounts are not + * used for 64bit architectures). * - * 6) If counter might be written by an interrupt, readers should block interrupts. - * (On UP, there is no seqcount_t protection, a reader allowing interrupts could - * read partial values) + * 5) Readers are allowed to sleep or be preempted/interrupted: they perform + * pure reads. * - * 7) For irq and softirq uses, readers can use u64_stats_fetch_begin_irq() and - * u64_stats_fetch_retry_irq() helpers + * 6) Readers must use both u64_stats_fetch_{begin,retry}_irq() if the stats + * might be updated from a hardirq or softirq context (remember point #1: + * seqcounts are not used for UP kernels). 32-bit UP stat readers could read + * corrupted 64-bit values otherwise. * * Usage : * -- cgit v1.2.3 From c7e261d81783387a0502878cd229327e7c54322e Mon Sep 17 00:00:00 2001 From: "Ahmed S. Darwish" Date: Wed, 3 Jun 2020 16:49:47 +0200 Subject: net: mdiobus: Disable preemption upon u64_stats update The u64_stats mechanism uses sequence counters to protect against 64-bit values tearing on 32-bit architectures. Updating u64_stats is thus a sequence counter write side critical section where preemption must be disabled. For mdiobus_stats_acct(), disable preemption upon the u64_stats update. It is called from process context through mdiobus_read() and mdiobus_write(). Reported-by: kernel test robot Signed-off-by: Ahmed S. Darwish Reviewed-by: Sebastian Andrzej Siewior Signed-off-by: David S. Miller --- drivers/net/phy/mdio_bus.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c index 255fdfcc13a6..6ceee82b2839 100644 --- a/drivers/net/phy/mdio_bus.c +++ b/drivers/net/phy/mdio_bus.c @@ -764,6 +764,7 @@ EXPORT_SYMBOL(mdiobus_scan); static void mdiobus_stats_acct(struct mdio_bus_stats *stats, bool op, int ret) { + preempt_disable(); u64_stats_update_begin(&stats->syncp); u64_stats_inc(&stats->transfers); @@ -778,6 +779,7 @@ static void mdiobus_stats_acct(struct mdio_bus_stats *stats, bool op, int ret) u64_stats_inc(&stats->writes); out: u64_stats_update_end(&stats->syncp); + preempt_enable(); } /** -- cgit v1.2.3 From f6c1fb0a76d97447ea7f928ee6a113ee15318df1 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 3 Jun 2020 20:50:25 +0300 Subject: net: ethernet: dwmac: Fix an error code in imx_dwmac_probe() The code is return PTR_ERR(NULL) which is zero or success. We should return -ENOMEM instead. Fixes: 94abdad6974a5 ("net: ethernet: dwmac: add ethernet glue logic for NXP imx8 chip") Signed-off-by: Dan Carpenter Acked-by: Fugang Duan Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c index 5010af7dab4a..3c5df5eeed6c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c @@ -225,7 +225,7 @@ static int imx_dwmac_probe(struct platform_device *pdev) dwmac = devm_kzalloc(&pdev->dev, sizeof(*dwmac), GFP_KERNEL); if (!dwmac) - return PTR_ERR(dwmac); + return -ENOMEM; plat_dat = stmmac_probe_config_dt(pdev, &stmmac_res.mac); if (IS_ERR(plat_dat)) -- cgit v1.2.3 From 120068481405f9e3d60a95bd464496dff6b54669 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Wed, 3 Jun 2020 22:29:06 +0200 Subject: r8169: fix failing WoL Th referenced change added an extra hw reset to rtl8169_net_suspend() what makes WoL fail on few chip versions. Therefore skip the extra reset if we're going down and WoL is enabled. In rtl_shutdown() rtl8169_hw_reset() is called by rtl8169_net_suspend() already if needed, therefore avoid issues issue by removing the extra call. The fix was tested on a system with RTL8168g. Meanwhile rtl8169_hw_reset() does more than a hw reset and should be renamed. But that's net-next material. Fixes: 8ac8e8c64b53 ("r8169: make rtl8169_down central chip quiesce function") Signed-off-by: Heiner Kallweit Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/r8169_main.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 4d2ec9742cee..dad84ecf5a77 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -3928,7 +3928,7 @@ static void rtl8169_tx_clear(struct rtl8169_private *tp) netdev_reset_queue(tp->dev); } -static void rtl8169_hw_reset(struct rtl8169_private *tp) +static void rtl8169_hw_reset(struct rtl8169_private *tp, bool going_down) { /* Give a racing hard_start_xmit a few cycles to complete. */ synchronize_rcu(); @@ -3938,6 +3938,9 @@ static void rtl8169_hw_reset(struct rtl8169_private *tp) rtl_rx_close(tp); + if (going_down && tp->dev->wol_enabled) + goto no_reset; + switch (tp->mac_version) { case RTL_GIGA_MAC_VER_27: case RTL_GIGA_MAC_VER_28: @@ -3959,7 +3962,7 @@ static void rtl8169_hw_reset(struct rtl8169_private *tp) } rtl_hw_reset(tp); - +no_reset: rtl8169_tx_clear(tp); rtl8169_init_ring_indexes(tp); } @@ -3972,7 +3975,7 @@ static void rtl_reset_work(struct rtl8169_private *tp) napi_disable(&tp->napi); netif_stop_queue(dev); - rtl8169_hw_reset(tp); + rtl8169_hw_reset(tp, false); for (i = 0; i < NUM_RX_DESC; i++) rtl8169_mark_to_asic(tp->RxDescArray + i); @@ -4637,7 +4640,7 @@ static void rtl8169_down(struct rtl8169_private *tp) phy_stop(tp->phydev); napi_disable(&tp->napi); - rtl8169_hw_reset(tp); + rtl8169_hw_reset(tp, true); rtl_pll_power_down(tp); @@ -4942,8 +4945,6 @@ static void rtl_shutdown(struct pci_dev *pdev) /* Restore original MAC address */ rtl_rar_set(tp, tp->dev->perm_addr); - rtl8169_hw_reset(tp); - if (system_state == SYSTEM_POWER_OFF) { if (tp->saved_wolopts) { rtl_wol_suspend_quirk(tp); -- cgit v1.2.3 From 09820ce88b4d36c452b23e05042542f87ff122d0 Mon Sep 17 00:00:00 2001 From: Valentin Longchamp Date: Wed, 3 Jun 2020 23:28:23 +0200 Subject: net: ethernet: freescale: remove unneeded include for ucc_geth net/sch_generic.h does not need to be included, remove it. Signed-off-by: Valentin Longchamp Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/ucc_geth.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/ethernet/freescale/ucc_geth.c b/drivers/net/ethernet/freescale/ucc_geth.c index 552e7554a9f8..db791f60b884 100644 --- a/drivers/net/ethernet/freescale/ucc_geth.c +++ b/drivers/net/ethernet/freescale/ucc_geth.c @@ -42,7 +42,6 @@ #include #include #include -#include #include "ucc_geth.h" -- cgit v1.2.3 From 7cdee28c4eeda8242c9fdbb02e315dbf35064afe Mon Sep 17 00:00:00 2001 From: Roelof Berg Date: Wed, 3 Jun 2020 23:54:14 +0200 Subject: lan743x: Use correct MAC_CR configuration for 1 GBit speed Corrected the MAC_CR configuration bits for 1 GBit operation. The data sheet allows MAC_CR(2:1) to be 10 and also 11 for 1 GBit/s speed, but only 10 works correctly. Devices tested: Microchip Lan7431, fixed-phy mode Microchip Lan7430, normal phy mode Fixes: 6f197fb63850 ("lan743x: Added fixed link and RGMII support") Signed-off-by: Roelof Berg Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/ethernet/microchip/lan743x_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c index 36624e3c633b..c5c5c688b7e2 100644 --- a/drivers/net/ethernet/microchip/lan743x_main.c +++ b/drivers/net/ethernet/microchip/lan743x_main.c @@ -985,7 +985,7 @@ static void lan743x_phy_link_status_change(struct net_device *netdev) break; case SPEED_1000: data |= MAC_CR_CFG_H_; - data |= MAC_CR_CFG_L_; + data &= ~MAC_CR_CFG_L_; break; } lan743x_csr_write(adapter, MAC_CR, data); -- cgit v1.2.3 From 98749b7188affbf2900c2aab704a8853901d1139 Mon Sep 17 00:00:00 2001 From: Wang Hai Date: Thu, 4 Jun 2020 20:18:51 +0800 Subject: yam: fix possible memory leak in yam_init_driver If register_netdev(dev) fails, free_netdev(dev) needs to be called, otherwise a memory leak will occur. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: Hulk Robot Signed-off-by: Wang Hai Signed-off-by: David S. Miller --- drivers/net/hamradio/yam.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/hamradio/yam.c b/drivers/net/hamradio/yam.c index 71cdef9fb56b..5ab53e9942f3 100644 --- a/drivers/net/hamradio/yam.c +++ b/drivers/net/hamradio/yam.c @@ -1133,6 +1133,7 @@ static int __init yam_init_driver(void) err = register_netdev(dev); if (err) { printk(KERN_WARNING "yam: cannot register net device %s\n", dev->name); + free_netdev(dev); goto error; } yam_devs[i] = dev; -- cgit v1.2.3 From 6761893eeaa378321198f06194ef2e1e4e8a4ad4 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Thu, 4 Jun 2020 18:55:45 +0200 Subject: inet_connection_sock: clear inet_num out of destroy helper Clearing the 'inet_num' field is necessary and safe if and only if the socket is not bound. The MPTCP protocol calls the destroy helper on bound sockets, as tcp_v{4,6}_syn_recv_sock completed successfully. Move the clearing of such field out of the common code, otherwise the MPTCP MP_JOIN error path will find the wrong 'inet_num' value on socket disposal, __inet_put_port() will acquire the wrong lock and bind_node removal could race with other modifiers possibly corrupting the bind hash table. Reported-and-tested-by: Christoph Paasch Fixes: 729cd6436f35 ("mptcp: cope better with MP_JOIN failure") Signed-off-by: Paolo Abeni Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/inet_connection_sock.h | 1 - net/ipv4/inet_connection_sock.c | 1 + 2 files changed, 1 insertion(+), 1 deletion(-) diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 2f1f8c3efb26..e5b388f5fa20 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -292,7 +292,6 @@ static inline void inet_csk_prepare_for_destroy_sock(struct sock *sk) /* The below has to be done to allow calling inet_csk_destroy_sock */ sock_set_flag(sk, SOCK_DEAD); percpu_counter_inc(sk->sk_prot->orphan_count); - inet_sk(sk)->inet_num = 0; } void inet_csk_destroy_sock(struct sock *sk); diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index f40b1b72f979..afaf582a5aa9 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -902,6 +902,7 @@ void inet_csk_prepare_forced_close(struct sock *sk) bh_unlock_sock(sk); sock_put(sk); inet_csk_prepare_for_destroy_sock(sk); + inet_sk(sk)->inet_num = 0; } EXPORT_SYMBOL(inet_csk_prepare_forced_close); -- cgit v1.2.3 From a624a86510adaeefd33aac224751e89348596d2a Mon Sep 17 00:00:00 2001 From: Vinay Kumar Yadav Date: Fri, 5 Jun 2020 01:53:44 +0530 Subject: crypto/chtls:Fix compile error when CONFIG_IPV6 is disabled Fix compile errors,warnings when CONFIG_IPV6 is disabled and inconsistent indenting. v1->v2: - Corrected errors/warnings reported when used newer gcc version, unused array. Fixes: 6abde0b24122 ("crypto/chtls: IPv6 support for inline TLS") Signed-off-by: Vinay Kumar Yadav Signed-off-by: David S. Miller --- drivers/crypto/chelsio/chcr_algo.h | 4 --- drivers/crypto/chelsio/chtls/chtls_cm.c | 46 +++++++++++++++++++++++-------- drivers/crypto/chelsio/chtls/chtls_main.c | 2 ++ 3 files changed, 36 insertions(+), 16 deletions(-) diff --git a/drivers/crypto/chelsio/chcr_algo.h b/drivers/crypto/chelsio/chcr_algo.h index f58c2b5c7fc5..d4f6e010dc79 100644 --- a/drivers/crypto/chelsio/chcr_algo.h +++ b/drivers/crypto/chelsio/chcr_algo.h @@ -389,10 +389,6 @@ static inline void copy_hash_init_values(char *key, int digestsize) } } -static const u8 sgl_lengths[20] = { - 0, 1, 2, 3, 4, 4, 5, 6, 7, 7, 8, 9, 10, 10, 11, 12, 13, 13, 14, 15 -}; - /* Number of len fields(8) * size of one addr field */ #define PHYSDSGL_MAX_LEN_SIZE 16 diff --git a/drivers/crypto/chelsio/chtls/chtls_cm.c b/drivers/crypto/chelsio/chtls/chtls_cm.c index 9a642c79a657..f200fae6f7cb 100644 --- a/drivers/crypto/chelsio/chtls/chtls_cm.c +++ b/drivers/crypto/chelsio/chtls/chtls_cm.c @@ -93,8 +93,10 @@ static struct net_device *chtls_find_netdev(struct chtls_dev *cdev, struct sock *sk) { struct net_device *ndev = cdev->ports[0]; +#if IS_ENABLED(CONFIG_IPV6) struct net_device *temp; int addr_type; +#endif switch (sk->sk_family) { case PF_INET: @@ -102,19 +104,21 @@ static struct net_device *chtls_find_netdev(struct chtls_dev *cdev, return ndev; ndev = ip_dev_find(&init_net, inet_sk(sk)->inet_rcv_saddr); break; +#if IS_ENABLED(CONFIG_IPV6) case PF_INET6: addr_type = ipv6_addr_type(&sk->sk_v6_rcv_saddr); if (likely(addr_type == IPV6_ADDR_ANY)) return ndev; - for_each_netdev_rcu(&init_net, temp) { - if (ipv6_chk_addr(&init_net, (struct in6_addr *) - &sk->sk_v6_rcv_saddr, temp, 1)) { - ndev = temp; - break; + for_each_netdev_rcu(&init_net, temp) { + if (ipv6_chk_addr(&init_net, (struct in6_addr *) + &sk->sk_v6_rcv_saddr, temp, 1)) { + ndev = temp; + break; + } } - } break; +#endif default: return NULL; } @@ -476,8 +480,10 @@ void chtls_destroy_sock(struct sock *sk) csk->cdev = NULL; if (sk->sk_family == AF_INET) sk->sk_prot = &tcp_prot; +#if IS_ENABLED(CONFIG_IPV6) else sk->sk_prot = &tcpv6_prot; +#endif sk->sk_prot->destroy(sk); } @@ -629,14 +635,15 @@ static void chtls_reset_synq(struct listen_ctx *listen_ctx) int chtls_listen_start(struct chtls_dev *cdev, struct sock *sk) { struct net_device *ndev; +#if IS_ENABLED(CONFIG_IPV6) + bool clip_valid = false; +#endif struct listen_ctx *ctx; struct adapter *adap; struct port_info *pi; - bool clip_valid; + int ret = 0; int stid; - int ret; - clip_valid = false; rcu_read_lock(); ndev = chtls_find_netdev(cdev, sk); rcu_read_unlock(); @@ -674,6 +681,7 @@ int chtls_listen_start(struct chtls_dev *cdev, struct sock *sk) inet_sk(sk)->inet_rcv_saddr, inet_sk(sk)->inet_sport, 0, cdev->lldi->rxq_ids[0]); +#if IS_ENABLED(CONFIG_IPV6) } else { int addr_type; @@ -689,6 +697,7 @@ int chtls_listen_start(struct chtls_dev *cdev, struct sock *sk) &sk->sk_v6_rcv_saddr, inet_sk(sk)->inet_sport, cdev->lldi->rxq_ids[0]); +#endif } if (ret > 0) ret = net_xmit_errno(ret); @@ -696,8 +705,10 @@ int chtls_listen_start(struct chtls_dev *cdev, struct sock *sk) goto del_hash; return 0; del_hash: +#if IS_ENABLED(CONFIG_IPV6) if (clip_valid) cxgb4_clip_release(ndev, (const u32 *)&sk->sk_v6_rcv_saddr, 1); +#endif listen_hash_del(cdev, sk); free_stid: cxgb4_free_stid(cdev->tids, stid, sk->sk_family); @@ -711,8 +722,6 @@ free_ctx: void chtls_listen_stop(struct chtls_dev *cdev, struct sock *sk) { struct listen_ctx *listen_ctx; - struct chtls_sock *csk; - int addr_type = 0; int stid; stid = listen_hash_del(cdev, sk); @@ -725,7 +734,11 @@ void chtls_listen_stop(struct chtls_dev *cdev, struct sock *sk) cxgb4_remove_server(cdev->lldi->ports[0], stid, cdev->lldi->rxq_ids[0], sk->sk_family == PF_INET6); +#if IS_ENABLED(CONFIG_IPV6) if (sk->sk_family == PF_INET6) { + struct chtls_sock *csk; + int addr_type = 0; + csk = rcu_dereference_sk_user_data(sk); addr_type = ipv6_addr_type((const struct in6_addr *) &sk->sk_v6_rcv_saddr); @@ -733,6 +746,7 @@ void chtls_listen_stop(struct chtls_dev *cdev, struct sock *sk) cxgb4_clip_release(csk->egress_dev, (const u32 *) &sk->sk_v6_rcv_saddr, 1); } +#endif chtls_disconnect_acceptq(sk); } @@ -941,9 +955,11 @@ static unsigned int chtls_select_mss(const struct chtls_sock *csk, tp = tcp_sk(sk); tcpoptsz = 0; +#if IS_ENABLED(CONFIG_IPV6) if (sk->sk_family == AF_INET6) iphdrsz = sizeof(struct ipv6hdr) + sizeof(struct tcphdr); else +#endif iphdrsz = sizeof(struct iphdr) + sizeof(struct tcphdr); if (req->tcpopt.tstamp) tcpoptsz += round_up(TCPOLEN_TIMESTAMP, 4); @@ -1091,13 +1107,13 @@ static struct sock *chtls_recv_sock(struct sock *lsk, const struct cpl_pass_accept_req *req, struct chtls_dev *cdev) { + struct neighbour *n = NULL; struct inet_sock *newinet; const struct iphdr *iph; struct tls_context *ctx; struct net_device *ndev; struct chtls_sock *csk; struct dst_entry *dst; - struct neighbour *n; struct tcp_sock *tp; struct sock *newsk; u16 port_id; @@ -1115,6 +1131,7 @@ static struct sock *chtls_recv_sock(struct sock *lsk, goto free_sk; n = dst_neigh_lookup(dst, &iph->saddr); +#if IS_ENABLED(CONFIG_IPV6) } else { const struct ipv6hdr *ip6h; struct flowi6 fl6; @@ -1131,6 +1148,7 @@ static struct sock *chtls_recv_sock(struct sock *lsk, if (IS_ERR(dst)) goto free_sk; n = dst_neigh_lookup(dst, &ip6h->saddr); +#endif } if (!n) goto free_sk; @@ -1158,6 +1176,7 @@ static struct sock *chtls_recv_sock(struct sock *lsk, newinet->inet_daddr = iph->saddr; newinet->inet_rcv_saddr = iph->daddr; newinet->inet_saddr = iph->daddr; +#if IS_ENABLED(CONFIG_IPV6) } else { struct tcp6_sock *newtcp6sk = (struct tcp6_sock *)newsk; struct inet_request_sock *treq = inet_rsk(oreq); @@ -1175,6 +1194,7 @@ static struct sock *chtls_recv_sock(struct sock *lsk, newinet->inet_opt = NULL; newinet->inet_daddr = LOOPBACK4_IPV6; newinet->inet_saddr = LOOPBACK4_IPV6; +#endif } oreq->ts_recent = PASS_OPEN_TID_G(ntohl(req->tos_stid)); @@ -1337,10 +1357,12 @@ static void chtls_pass_accept_request(struct sock *sk, if (iph->version == 0x4) { chtls_set_req_addr(oreq, iph->daddr, iph->saddr); ip_dsfield = ipv4_get_dsfield(iph); +#if IS_ENABLED(CONFIG_IPV6) } else { inet_rsk(oreq)->ir_v6_rmt_addr = ipv6_hdr(skb)->saddr; inet_rsk(oreq)->ir_v6_loc_addr = ipv6_hdr(skb)->daddr; ip_dsfield = ipv6_get_dsfield(ipv6_hdr(skb)); +#endif } if (req->tcpopt.wsf <= 14 && sock_net(sk)->ipv4.sysctl_tcp_window_scaling) { diff --git a/drivers/crypto/chelsio/chtls/chtls_main.c b/drivers/crypto/chelsio/chtls/chtls_main.c index 7dfffdde9593..d98b89d0fa6e 100644 --- a/drivers/crypto/chelsio/chtls/chtls_main.c +++ b/drivers/crypto/chelsio/chtls/chtls_main.c @@ -608,9 +608,11 @@ static void __init chtls_init_ulp_ops(void) chtls_cpl_prot.recvmsg = chtls_recvmsg; chtls_cpl_prot.setsockopt = chtls_setsockopt; chtls_cpl_prot.getsockopt = chtls_getsockopt; +#if IS_ENABLED(CONFIG_IPV6) chtls_cpl_protv6 = chtls_cpl_prot; chtls_init_rsk_ops(&chtls_cpl_protv6, &chtls_rsk_opsv6, &tcpv6_prot, PF_INET6); +#endif } static int __init chtls_register(void) -- cgit v1.2.3 From 7d877c35ca84cfa634fd63c2b64bf7b6ae9c71cb Mon Sep 17 00:00:00 2001 From: Pavel Machek Date: Thu, 4 Jun 2020 23:42:59 +0200 Subject: net/xdp: use shift instead of 64 bit division 64bit division is kind of expensive, and shift should do the job here. Signed-off-by: Pavel Machek (CIP) Signed-off-by: David S. Miller --- net/xdp/xdp_umem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c index 1bbaf1747e4f..a0d2b757807f 100644 --- a/net/xdp/xdp_umem.c +++ b/net/xdp/xdp_umem.c @@ -336,7 +336,7 @@ static int xdp_umem_reg(struct xdp_umem *umem, struct xdp_umem_reg *mr) if ((addr + size) < addr) return -EINVAL; - npgs = div_u64(size, PAGE_SIZE); + npgs = size >> PAGE_SHIFT; if (npgs > U32_MAX) return -EINVAL; -- cgit v1.2.3 From 178f67b1288b6952117fdc4e5ffbd4c4bd4e4a7f Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 5 Jun 2020 14:04:13 +0300 Subject: ethtool: linkinfo: remove an unnecessary NULL check This code generates a Smatch warning: net/ethtool/linkinfo.c:143 ethnl_set_linkinfo() warn: variable dereferenced before check 'info' (see line 119) Fortunately, the "info" pointer is never NULL so the check can be removed. Signed-off-by: Dan Carpenter Reviewed-by: Michal Kubecek Signed-off-by: David S. Miller --- net/ethtool/linkinfo.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/ethtool/linkinfo.c b/net/ethtool/linkinfo.c index 677068deb68c..5eaf173eaaca 100644 --- a/net/ethtool/linkinfo.c +++ b/net/ethtool/linkinfo.c @@ -140,8 +140,7 @@ int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info) ret = __ethtool_get_link_ksettings(dev, &ksettings); if (ret < 0) { - if (info) - GENL_SET_ERR_MSG(info, "failed to retrieve link settings"); + GENL_SET_ERR_MSG(info, "failed to retrieve link settings"); goto out_ops; } lsettings = &ksettings.base; -- cgit v1.2.3 From e22437207683a9099d5fb3a1b5cc8f5c7b24a583 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Fri, 5 Jun 2020 15:53:24 +0300 Subject: net: ethernet: mvneta: fix MVNETA_SKB_HEADROOM alignment Commit ca23cb0bc50f ("mvneta: MVNETA_SKB_HEADROOM set last 3 bits to zero") added headroom alignment check against 8. Hovewer (if we imagine that NET_SKB_PAD or XDP_PACKET_HEADROOM is not aligned to cacheline size), it actually aligns headroom down, while skb/xdp_buff headroom should be *at least* equal to one of the values (depending on XDP prog presence). So, fix the check to align the value up. This satisfies both hardware/driver and network stack requirements. Fixes: ca23cb0bc50f ("mvneta: MVNETA_SKB_HEADROOM set last 3 bits to zero") Signed-off-by: Alexander Lobakin Signed-off-by: David S. Miller --- drivers/net/ethernet/marvell/mvneta.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 011cd26953d9..4cc9abd61c43 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -325,7 +325,7 @@ cache_line_size()) /* Driver assumes that the last 3 bits are 0 */ -#define MVNETA_SKB_HEADROOM (max(XDP_PACKET_HEADROOM, NET_SKB_PAD) & ~0x7) +#define MVNETA_SKB_HEADROOM ALIGN(max(NET_SKB_PAD, XDP_PACKET_HEADROOM), 8) #define MVNETA_SKB_PAD (SKB_DATA_ALIGN(sizeof(struct skb_shared_info) + \ MVNETA_SKB_HEADROOM)) #define MVNETA_SKB_SIZE(len) (SKB_DATA_ALIGN(len) + MVNETA_SKB_PAD) -- cgit v1.2.3 From e9293c982d88b2d859e26ec42b72c06f20235315 Mon Sep 17 00:00:00 2001 From: Dan Murphy Date: Fri, 5 Jun 2020 09:01:04 -0500 Subject: net: dp83869: Fix OF_MDIO config check When CONFIG_OF_MDIO is set to be a module the code block is not compiled. Use the IS_ENABLED macro that checks for both built in as well as module. Fixes: 01db923e83779 ("net: phy: dp83869: Add TI dp83869 phy") Signed-off-by: Dan Murphy Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/dp83869.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/dp83869.c b/drivers/net/phy/dp83869.c index cfb22a21a2e6..df85ae5b79e4 100644 --- a/drivers/net/phy/dp83869.c +++ b/drivers/net/phy/dp83869.c @@ -176,7 +176,7 @@ static int dp83869_set_strapped_mode(struct phy_device *phydev) return 0; } -#ifdef CONFIG_OF_MDIO +#if IS_ENABLED(CONFIG_OF_MDIO) static int dp83869_of_init(struct phy_device *phydev) { struct dp83869_private *dp83869 = phydev->priv; -- cgit v1.2.3 From 506de00677b84dfc6718cbbd3495b1d90df5d098 Mon Sep 17 00:00:00 2001 From: Dan Murphy Date: Fri, 5 Jun 2020 09:01:05 -0500 Subject: net: dp83867: Fix OF_MDIO config check When CONFIG_OF_MDIO is set to be a module the code block is not compiled. Use the IS_ENABLED macro that checks for both built in as well as module. Fixes: 2a10154abcb75 ("net: phy: dp83867: Add TI dp83867 phy") Signed-off-by: Dan Murphy Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/dp83867.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c index 4017ae1692d8..f3c04981b8da 100644 --- a/drivers/net/phy/dp83867.c +++ b/drivers/net/phy/dp83867.c @@ -488,7 +488,7 @@ static int dp83867_verify_rgmii_cfg(struct phy_device *phydev) return 0; } -#ifdef CONFIG_OF_MDIO +#if IS_ENABLED(CONFIG_OF_MDIO) static int dp83867_of_init(struct phy_device *phydev) { struct dp83867_private *dp83867 = phydev->priv; -- cgit v1.2.3 From 5cd119d9a05f1c1a08778a7305b4ca0f16bc1e20 Mon Sep 17 00:00:00 2001 From: Dan Murphy Date: Fri, 5 Jun 2020 09:01:06 -0500 Subject: net: marvell: Fix OF_MDIO config check When CONFIG_OF_MDIO is set to be a module the code block is not compiled. Use the IS_ENABLED macro that checks for both built in as well as module. Fixes: cf41a51db8985 ("of/phylib: Use device tree properties to initialize Marvell PHYs.") Signed-off-by: Dan Murphy Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/marvell.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index 4ea226566cec..c9ecf3c8c3fd 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -429,7 +429,7 @@ static int m88e1101_config_aneg(struct phy_device *phydev) return marvell_config_aneg(phydev); } -#ifdef CONFIG_OF_MDIO +#if IS_ENABLED(CONFIG_OF_MDIO) /* Set and/or override some configuration registers based on the * marvell,reg-init property stored in the of_node for the phydev. * -- cgit v1.2.3 From ae602786407fef34e1d66b3c8f278a10ed37197e Mon Sep 17 00:00:00 2001 From: Dan Murphy Date: Fri, 5 Jun 2020 09:01:07 -0500 Subject: net: mscc: Fix OF_MDIO config check When CONFIG_OF_MDIO is set to be a module the code block is not compiled. Use the IS_ENABLED macro that checks for both built in as well as module. Fixes: 4f58e6dceb0e4 ("net: phy: Cleanup the Edge-Rate feature in Microsemi PHYs.") Signed-off-by: Dan Murphy Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/mscc/mscc.h | 2 +- drivers/net/phy/mscc/mscc_main.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/phy/mscc/mscc.h b/drivers/net/phy/mscc/mscc.h index f828c917b9f7..fbcee5fce7b2 100644 --- a/drivers/net/phy/mscc/mscc.h +++ b/drivers/net/phy/mscc/mscc.h @@ -374,7 +374,7 @@ struct vsc8531_private { #endif }; -#ifdef CONFIG_OF_MDIO +#if IS_ENABLED(CONFIG_OF_MDIO) struct vsc8531_edge_rate_table { u32 vddmac; u32 slowdown[8]; diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c index 7ed0285206d0..b9d60bc8b6b2 100644 --- a/drivers/net/phy/mscc/mscc_main.c +++ b/drivers/net/phy/mscc/mscc_main.c @@ -98,7 +98,7 @@ static const struct vsc85xx_hw_stat vsc8584_hw_stats[] = { }, }; -#ifdef CONFIG_OF_MDIO +#if IS_ENABLED(CONFIG_OF_MDIO) static const struct vsc8531_edge_rate_table edge_table[] = { {MSCC_VDDMAC_3300, { 0, 2, 4, 7, 10, 17, 29, 53} }, {MSCC_VDDMAC_2500, { 0, 3, 6, 10, 14, 23, 37, 63} }, @@ -382,7 +382,7 @@ out_unlock: mutex_unlock(&phydev->lock); } -#ifdef CONFIG_OF_MDIO +#if IS_ENABLED(CONFIG_OF_MDIO) static int vsc85xx_edge_rate_magic_get(struct phy_device *phydev) { u32 vdd, sd; -- cgit v1.2.3 From 49113d5e0c3f3fbf409075e69c2d1e157c387ac5 Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Fri, 5 Jun 2020 16:00:09 +0200 Subject: net: phy: mscc: fix Serdes configuration in vsc8584_config_init When converting the MSCC PHY driver to shared PHY packages, the Serdes configuration in vsc8584_config_init was modified to use 'base_addr' instead of 'base' as the port number. But 'base_addr' isn't equal to 'addr' for all PHYs inside the package, which leads to the Serdes still being enabled on those ports. This patch fixes it. Fixes: deb04e9c0ff2 ("net: phy: mscc: use phy_package_shared") Signed-off-by: Antoine Tenart Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/mscc/mscc_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c index b9d60bc8b6b2..5ddc44f87eaf 100644 --- a/drivers/net/phy/mscc/mscc_main.c +++ b/drivers/net/phy/mscc/mscc_main.c @@ -1396,7 +1396,7 @@ static int vsc8584_config_init(struct phy_device *phydev) /* Disable SerDes for 100Base-FX */ ret = vsc8584_cmd(phydev, PROC_CMD_FIBER_MEDIA_CONF | - PROC_CMD_FIBER_PORT(vsc8531->base_addr) | + PROC_CMD_FIBER_PORT(vsc8531->addr) | PROC_CMD_FIBER_DISABLE | PROC_CMD_READ_MOD_WRITE_PORT | PROC_CMD_RST_CONF_PORT | PROC_CMD_FIBER_100BASE_FX); @@ -1405,7 +1405,7 @@ static int vsc8584_config_init(struct phy_device *phydev) /* Disable SerDes for 1000Base-X */ ret = vsc8584_cmd(phydev, PROC_CMD_FIBER_MEDIA_CONF | - PROC_CMD_FIBER_PORT(vsc8531->base_addr) | + PROC_CMD_FIBER_PORT(vsc8531->addr) | PROC_CMD_FIBER_DISABLE | PROC_CMD_READ_MOD_WRITE_PORT | PROC_CMD_RST_CONF_PORT | PROC_CMD_FIBER_1000BASE_X); -- cgit v1.2.3 From 90c56b3877995d948dc382fe833a1c5eeaaee2ad Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 5 Jun 2020 07:21:22 -0700 Subject: net: ethtool: Fix comment mentioning typo in IS_ENABLED() This has no code changes, but it's a typo noticed in other clean-ups, so we might as well fix it. IS_ENABLED() takes full names, and should have the "CONFIG_" prefix. Reported-by: Joe Perches Link: https://lore.kernel.org/lkml/b08611018fdb6d88757c6008a5c02fa0e07b32fb.camel@perches.com Signed-off-by: Kees Cook Signed-off-by: David S. Miller --- include/linux/ethtool_netlink.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/ethtool_netlink.h b/include/linux/ethtool_netlink.h index 8fbe4f97ffad..1e7bf78cb382 100644 --- a/include/linux/ethtool_netlink.h +++ b/include/linux/ethtool_netlink.h @@ -67,5 +67,5 @@ static inline int ethnl_cable_test_step(struct phy_device *phydev, u32 first, { return -EOPNOTSUPP; } -#endif /* IS_ENABLED(ETHTOOL_NETLINK) */ +#endif /* IS_ENABLED(CONFIG_ETHTOOL_NETLINK) */ #endif /* _LINUX_ETHTOOL_NETLINK_H_ */ -- cgit v1.2.3 From fdb4276aae1100aa13f955eff6c944a8200a15ac Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Fri, 5 Jun 2020 17:12:41 +0200 Subject: vsock/vmci: make vmci_vsock_transport_cb() static MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix the following gcc-9.3 warning when building with 'make W=1': net/vmw_vsock/vmci_transport.c:2058:6: warning: no previous prototype for ‘vmci_vsock_transport_cb’ [-Wmissing-prototypes] 2058 | void vmci_vsock_transport_cb(bool is_host) | ^~~~~~~~~~~~~~~~~~~~~~~ Fixes: b1bba80a4376 ("vsock/vmci: register vmci_transport only when VMCI guest/host are active") Reported-by: kernel test robot Signed-off-by: Stefano Garzarella Signed-off-by: David S. Miller --- net/vmw_vsock/vmci_transport.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c index 4b8b1150a738..8b65323207db 100644 --- a/net/vmw_vsock/vmci_transport.c +++ b/net/vmw_vsock/vmci_transport.c @@ -2055,7 +2055,7 @@ static bool vmci_check_transport(struct vsock_sock *vsk) return vsk->transport == &vmci_transport; } -void vmci_vsock_transport_cb(bool is_host) +static void vmci_vsock_transport_cb(bool is_host) { int features; -- cgit v1.2.3 From 6da95b52b8ea4944ed1e596ee6db176ee83ddbeb Mon Sep 17 00:00:00 2001 From: Alok Prasad Date: Fri, 5 Jun 2020 16:30:34 +0000 Subject: net: qed: fixes crash while running driver in kdump kernel This fixes a crash introduced by recent is_kdump_kernel() check. The source of the crash is that kdump kernel can be loaded on a system with already created VFs. But for such VFs, it will follow a logic path of PF and eventually crash. Thus, we are partially reverting back previous changes and instead use is_kdump_kernel is a single init point of PF init, where we disable SRIOV explicitly. Fixes: 37d4f8a6b41f ("net: qed: Disable SRIOV functionality inside kdump kernel") Cc: Bhupesh Sharma Signed-off-by: Igor Russkikh Signed-off-by: Alok Prasad Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_sriov.c | 4 ++++ drivers/net/ethernet/qlogic/qed/qed_sriov.h | 10 +++------- drivers/net/ethernet/qlogic/qede/qede_main.c | 2 +- 3 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index 66876af814c4..20679fd4204b 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include "qed_cxt.h" #include "qed_hsi.h" @@ -607,6 +608,9 @@ int qed_iov_hw_info(struct qed_hwfn *p_hwfn) int pos; int rc; + if (is_kdump_kernel()) + return 0; + if (IS_VF(p_hwfn->cdev)) return 0; diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.h b/drivers/net/ethernet/qlogic/qed/qed_sriov.h index aabeaf03135e..368e88565783 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.h +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.h @@ -32,7 +32,6 @@ #ifndef _QED_SRIOV_H #define _QED_SRIOV_H -#include #include #include "qed_vf.h" @@ -41,12 +40,9 @@ #define QED_VF_ARRAY_LENGTH (3) #ifdef CONFIG_QED_SRIOV -#define IS_VF(cdev) (is_kdump_kernel() ? \ - (0) : ((cdev)->b_is_vf)) -#define IS_PF(cdev) (is_kdump_kernel() ? \ - (1) : !((cdev)->b_is_vf)) -#define IS_PF_SRIOV(p_hwfn) (is_kdump_kernel() ? \ - (0) : !!((p_hwfn)->cdev->p_iov_info)) +#define IS_VF(cdev) ((cdev)->b_is_vf) +#define IS_PF(cdev) (!((cdev)->b_is_vf)) +#define IS_PF_SRIOV(p_hwfn) (!!((p_hwfn)->cdev->p_iov_info)) #else #define IS_VF(cdev) (0) #define IS_PF(cdev) (1) diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index b2d154258b07..756c05eb96f3 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -1265,7 +1265,7 @@ static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id) case QEDE_PRIVATE_VF: if (debug & QED_LOG_VERBOSE_MASK) dev_err(&pdev->dev, "Probing a VF\n"); - is_vf = is_kdump_kernel() ? false : true; + is_vf = true; break; default: if (debug & QED_LOG_VERBOSE_MASK) -- cgit v1.2.3 From 7f89cc07d22a4152f3c649a89aa8fe240effd24a Mon Sep 17 00:00:00 2001 From: Denis Efremov Date: Fri, 5 Jun 2020 22:11:44 +0300 Subject: cxgb4: Use kfree() instead kvfree() where appropriate Use kfree(buf) in blocked_fl_read() because the memory is allocated with kzalloc(). Use kfree(t) in blocked_fl_write() because the memory is allocated with kcalloc(). Signed-off-by: Denis Efremov Signed-off-by: David S. Miller --- drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c index 41315712deb8..828499256004 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c @@ -3357,7 +3357,7 @@ static ssize_t blocked_fl_read(struct file *filp, char __user *ubuf, adap->sge.egr_sz, adap->sge.blocked_fl); len += sprintf(buf + len, "\n"); size = simple_read_from_buffer(ubuf, count, ppos, buf, len); - kvfree(buf); + kfree(buf); return size; } @@ -3374,12 +3374,12 @@ static ssize_t blocked_fl_write(struct file *filp, const char __user *ubuf, err = bitmap_parse_user(ubuf, count, t, adap->sge.egr_sz); if (err) { - kvfree(t); + kfree(t); return err; } bitmap_copy(adap->sge.blocked_fl, t, adap->sge.egr_sz); - kvfree(t); + kfree(t); return count; } -- cgit v1.2.3 From 4a3084aaa88e70a0af0e38b72ca29af32ec6b9c5 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 3 Jun 2020 18:12:43 +1000 Subject: rhashtable: Drop raw RCU deref in nested_table_free This patch replaces some unnecessary uses of rcu_dereference_raw in the rhashtable code with rcu_dereference_protected. The top-level nested table entry is only marked as RCU because it shares the same type as the tree entries underneath it. So it doesn't need any RCU protection. We also don't need RCU protection when we're freeing a nested RCU table because by this stage we've long passed a memory barrier when anyone could change the nested table. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- lib/rhashtable.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/lib/rhashtable.c b/lib/rhashtable.c index bdb7e4cadf05..9f6890aedd1a 100644 --- a/lib/rhashtable.c +++ b/lib/rhashtable.c @@ -63,13 +63,22 @@ EXPORT_SYMBOL_GPL(lockdep_rht_bucket_is_held); #define ASSERT_RHT_MUTEX(HT) #endif +static inline union nested_table *nested_table_top( + const struct bucket_table *tbl) +{ + /* The top-level bucket entry does not need RCU protection + * because it's set at the same time as tbl->nest. + */ + return (void *)rcu_dereference_protected(tbl->buckets[0], 1); +} + static void nested_table_free(union nested_table *ntbl, unsigned int size) { const unsigned int shift = PAGE_SHIFT - ilog2(sizeof(void *)); const unsigned int len = 1 << shift; unsigned int i; - ntbl = rcu_dereference_raw(ntbl->table); + ntbl = rcu_dereference_protected(ntbl->table, 1); if (!ntbl) return; @@ -89,7 +98,7 @@ static void nested_bucket_table_free(const struct bucket_table *tbl) union nested_table *ntbl; unsigned int i; - ntbl = (union nested_table *)rcu_dereference_raw(tbl->buckets[0]); + ntbl = nested_table_top(tbl); for (i = 0; i < len; i++) nested_table_free(ntbl + i, size); @@ -1173,7 +1182,7 @@ struct rhash_lock_head **__rht_bucket_nested(const struct bucket_table *tbl, unsigned int subhash = hash; union nested_table *ntbl; - ntbl = (union nested_table *)rcu_dereference_raw(tbl->buckets[0]); + ntbl = nested_table_top(tbl); ntbl = rht_dereference_bucket_rcu(ntbl[index].table, tbl, hash); subhash >>= tbl->nest; @@ -1213,7 +1222,7 @@ struct rhash_lock_head **rht_bucket_nested_insert(struct rhashtable *ht, unsigned int size = tbl->size >> tbl->nest; union nested_table *ntbl; - ntbl = (union nested_table *)rcu_dereference_raw(tbl->buckets[0]); + ntbl = nested_table_top(tbl); hash >>= tbl->nest; ntbl = nested_table_alloc(ht, &ntbl[index].table, size <= (1 << shift)); -- cgit v1.2.3 From 4e2905adac9f0fdc25154ac83719a986c2367a14 Mon Sep 17 00:00:00 2001 From: Dan Murphy Date: Fri, 5 Jun 2020 15:51:03 -0500 Subject: net: dp83869: Reset return variable if PHY strap is read When the PHY's strap register is read to determine if lane swapping is needed the phy_read_mmd returns the value back into the ret variable. If the call to read the strap fails the failed value is returned. If the call to read the strap is successful then ret is possibly set to a non-zero positive number. Without reseting the ret value to 0 this will cause the parse DT function to return a failure. Fixes: c4566aec6e808 ("net: phy: dp83869: Update port-mirroring to read straps") Signed-off-by: Dan Murphy Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/dp83869.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/phy/dp83869.c b/drivers/net/phy/dp83869.c index df85ae5b79e4..53ed3abc26c9 100644 --- a/drivers/net/phy/dp83869.c +++ b/drivers/net/phy/dp83869.c @@ -218,10 +218,13 @@ static int dp83869_of_init(struct phy_device *phydev) ret = phy_read_mmd(phydev, DP83869_DEVADDR, DP83869_STRAP_STS1); if (ret < 0) return ret; + if (ret & DP83869_STRAP_MIRROR_ENABLED) dp83869->port_mirroring = DP83869_PORT_MIRRORING_EN; else dp83869->port_mirroring = DP83869_PORT_MIRRORING_DIS; + + ret = 0; } if (of_property_read_u32(of_node, "rx-fifo-depth", -- cgit v1.2.3 From 2dc2f760052da4925482ecdcdc5c94d4a599153c Mon Sep 17 00:00:00 2001 From: Vadim Pasternak Date: Sun, 7 Jun 2020 11:10:27 +0300 Subject: mlxsw: core: Use different get_trend() callbacks for different thermal zones The driver registers three different types of thermal zones: For the ASIC itself, for port modules and for gearboxes. Currently, all three types use the same get_trend() callback which does not work correctly for the ASIC thermal zone. The callback assumes that the device data is of type 'struct mlxsw_thermal_module', whereas for the ASIC thermal zone 'struct mlxsw_thermal' is passed as device data. Fix this by using one get_trend() callback for the ASIC thermal zone and another for the other two types. Fixes: 6f73862fabd9 ("mlxsw: core: Add the hottest thermal zone detection") Signed-off-by: Vadim Pasternak Reviewed-by: Jiri Pirko Signed-off-by: Ido Schimmel Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/core_thermal.c | 23 ++++++++++++++++++---- 1 file changed, 19 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c index ce0a6837daa3..05f8d5a92862 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c @@ -391,8 +391,7 @@ static int mlxsw_thermal_set_trip_hyst(struct thermal_zone_device *tzdev, static int mlxsw_thermal_trend_get(struct thermal_zone_device *tzdev, int trip, enum thermal_trend *trend) { - struct mlxsw_thermal_module *tz = tzdev->devdata; - struct mlxsw_thermal *thermal = tz->parent; + struct mlxsw_thermal *thermal = tzdev->devdata; if (trip < 0 || trip >= MLXSW_THERMAL_NUM_TRIPS) return -EINVAL; @@ -593,6 +592,22 @@ mlxsw_thermal_module_trip_hyst_set(struct thermal_zone_device *tzdev, int trip, return 0; } +static int mlxsw_thermal_module_trend_get(struct thermal_zone_device *tzdev, + int trip, enum thermal_trend *trend) +{ + struct mlxsw_thermal_module *tz = tzdev->devdata; + struct mlxsw_thermal *thermal = tz->parent; + + if (trip < 0 || trip >= MLXSW_THERMAL_NUM_TRIPS) + return -EINVAL; + + if (tzdev == thermal->tz_highest_dev) + return 1; + + *trend = THERMAL_TREND_STABLE; + return 0; +} + static struct thermal_zone_device_ops mlxsw_thermal_module_ops = { .bind = mlxsw_thermal_module_bind, .unbind = mlxsw_thermal_module_unbind, @@ -604,7 +619,7 @@ static struct thermal_zone_device_ops mlxsw_thermal_module_ops = { .set_trip_temp = mlxsw_thermal_module_trip_temp_set, .get_trip_hyst = mlxsw_thermal_module_trip_hyst_get, .set_trip_hyst = mlxsw_thermal_module_trip_hyst_set, - .get_trend = mlxsw_thermal_trend_get, + .get_trend = mlxsw_thermal_module_trend_get, }; static int mlxsw_thermal_gearbox_temp_get(struct thermal_zone_device *tzdev, @@ -643,7 +658,7 @@ static struct thermal_zone_device_ops mlxsw_thermal_gearbox_ops = { .set_trip_temp = mlxsw_thermal_module_trip_temp_set, .get_trip_hyst = mlxsw_thermal_module_trip_hyst_get, .set_trip_hyst = mlxsw_thermal_module_trip_hyst_set, - .get_trend = mlxsw_thermal_trend_get, + .get_trend = mlxsw_thermal_module_trend_get, }; static int mlxsw_thermal_get_max_state(struct thermal_cooling_device *cdev, -- cgit v1.2.3 From 4d3da2d8d91f66988a829a18a0ce59945e8ae4fb Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Sun, 7 Jun 2020 15:02:58 +0200 Subject: net: dsa: lantiq_gswip: fix and improve the unsupported interface error While trying to use the lantiq_gswip driver on one of my boards I made a mistake when specifying the phy-mode (because the out-of-tree driver wants phy-mode "gmii" or "mii" for the internal PHYs). In this case the following error is printed multiple times: Unsupported interface: 3 While it gives at least a hint at what may be wrong it is not very user friendly. Print the human readable phy-mode and also which port is configured incorrectly (this hardware supports ports 0..6) to improve the cases where someone made a mistake. Fixes: 14fceff4771e51 ("net: dsa: Add Lantiq / Intel DSA driver for vrx200") Signed-off-by: Martin Blumenstingl Acked-by: Hauke Mehrtens Signed-off-by: David S. Miller --- drivers/net/dsa/lantiq_gswip.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c index cf6fa8fede33..521ebc072903 100644 --- a/drivers/net/dsa/lantiq_gswip.c +++ b/drivers/net/dsa/lantiq_gswip.c @@ -1452,7 +1452,8 @@ static void gswip_phylink_validate(struct dsa_switch *ds, int port, unsupported: bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS); - dev_err(ds->dev, "Unsupported interface: %d\n", state->interface); + dev_err(ds->dev, "Unsupported interface '%s' for port %d\n", + phy_modes(state->interface), port); return; } -- cgit v1.2.3