From 0856655a25476d4431005e39d606e349050066b0 Mon Sep 17 00:00:00 2001 From: Loic Poulain Date: Mon, 11 Dec 2017 09:52:22 +0100 Subject: [PATCH 01/27] wcn36xx: Fix dynamic power saving Since driver does not report hardware dynamic power saving cap, this is up to the mac80211 to manage power saving timeout and state machine, using the ieee80211 config callback to report PS changes. This patch enables/disables PS mode according to the new configuration. Remove old behaviour enabling PS mode in a static way, this make the device unusable when power save is enabled since device is forced to PS regardless RX/TX traffic. Acked-by: Bjorn Andersson Signed-off-by: Loic Poulain Signed-off-by: Kalle Valo --- drivers/net/wireless/ath/wcn36xx/main.c | 23 ++++++++++++----------- drivers/net/wireless/ath/wcn36xx/pmc.c | 6 ++++-- 2 files changed, 16 insertions(+), 13 deletions(-) diff --git a/drivers/net/wireless/ath/wcn36xx/main.c b/drivers/net/wireless/ath/wcn36xx/main.c index f7d228b5ba93..987f1252a3cf 100644 --- a/drivers/net/wireless/ath/wcn36xx/main.c +++ b/drivers/net/wireless/ath/wcn36xx/main.c @@ -384,6 +384,18 @@ static int wcn36xx_config(struct ieee80211_hw *hw, u32 changed) } } + if (changed & IEEE80211_CONF_CHANGE_PS) { + list_for_each_entry(tmp, &wcn->vif_list, list) { + vif = wcn36xx_priv_to_vif(tmp); + if (hw->conf.flags & IEEE80211_CONF_PS) { + if (vif->bss_conf.ps) /* ps allowed ? */ + wcn36xx_pmc_enter_bmps_state(wcn, vif); + } else { + wcn36xx_pmc_exit_bmps_state(wcn, vif); + } + } + } + mutex_unlock(&wcn->conf_mutex); return 0; @@ -747,17 +759,6 @@ static void wcn36xx_bss_info_changed(struct ieee80211_hw *hw, vif_priv->dtim_period = bss_conf->dtim_period; } - if (changed & BSS_CHANGED_PS) { - wcn36xx_dbg(WCN36XX_DBG_MAC, - "mac bss PS set %d\n", - bss_conf->ps); - if (bss_conf->ps) { - wcn36xx_pmc_enter_bmps_state(wcn, vif); - } else { - wcn36xx_pmc_exit_bmps_state(wcn, vif); - } - } - if (changed & BSS_CHANGED_BSSID) { wcn36xx_dbg(WCN36XX_DBG_MAC, "mac bss changed_bssid %pM\n", bss_conf->bssid); diff --git a/drivers/net/wireless/ath/wcn36xx/pmc.c b/drivers/net/wireless/ath/wcn36xx/pmc.c index 589fe5f70971..1976b80c235f 100644 --- a/drivers/net/wireless/ath/wcn36xx/pmc.c +++ b/drivers/net/wireless/ath/wcn36xx/pmc.c @@ -45,8 +45,10 @@ int wcn36xx_pmc_exit_bmps_state(struct wcn36xx *wcn, struct wcn36xx_vif *vif_priv = wcn36xx_vif_to_priv(vif); if (WCN36XX_BMPS != vif_priv->pw_state) { - wcn36xx_err("Not in BMPS mode, no need to exit from BMPS mode!\n"); - return -EINVAL; + /* Unbalanced call or last BMPS enter failed */ + wcn36xx_dbg(WCN36XX_DBG_PMC, + "Not in BMPS mode, no need to exit\n"); + return -EALREADY; } wcn36xx_smd_exit_bmps(wcn, vif); vif_priv->pw_state = WCN36XX_FULL_POWER; From 943309d4aad6732b905f3f500e6e17e33c211494 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Thu, 4 Jan 2018 09:19:13 +0200 Subject: [PATCH 02/27] iwlwifi: pcie: fix DMA memory mapping / unmapping 22000 devices (previously referenced as A000) can support short transmit queues. This means that we have less DMA descriptors (TFD) for those shorter queues. Previous devices must still have 256 TFDs for each queue even if those 256 TFDs point to fewer buffers. When I introduced support for the short queues for 22000 I broke older devices by assuming that they can also have less TFDs in their queues. This led to several problems: 1) the payload of the commands weren't unmapped properly which caused the SWIOTLB to complain at some point. 2) the hardware could get confused and we get hardware crashes. The corresponding bugzilla entries are: https://bugzilla.kernel.org/show_bug.cgi?id=198201 https://bugzilla.kernel.org/show_bug.cgi?id=198265 Cc: stable@vger.kernel.org # 4.14+ Fixes: 4ecab5616023 ("iwlwifi: pcie: support short Tx queues for A000 device family") Reviewed-by: Sharon, Sara Signed-off-by: Emmanuel Grumbach Signed-off-by: Kalle Valo --- drivers/net/wireless/intel/iwlwifi/pcie/internal.h | 10 +++++++--- drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c | 11 +++-------- drivers/net/wireless/intel/iwlwifi/pcie/tx.c | 8 ++++---- 3 files changed, 14 insertions(+), 15 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h index d749abeca3ae..403e65c309d0 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h +++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h @@ -670,11 +670,15 @@ static inline u8 iwl_pcie_get_cmd_index(struct iwl_txq *q, u32 index) return index & (q->n_window - 1); } -static inline void *iwl_pcie_get_tfd(struct iwl_trans_pcie *trans_pcie, +static inline void *iwl_pcie_get_tfd(struct iwl_trans *trans, struct iwl_txq *txq, int idx) { - return txq->tfds + trans_pcie->tfd_size * iwl_pcie_get_cmd_index(txq, - idx); + struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); + + if (trans->cfg->use_tfh) + idx = iwl_pcie_get_cmd_index(txq, idx); + + return txq->tfds + trans_pcie->tfd_size * idx; } static inline void iwl_enable_rfkill_int(struct iwl_trans *trans) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c index 16b345f54ff0..6d0a907d5ba5 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c @@ -171,8 +171,6 @@ static void iwl_pcie_gen2_tfd_unmap(struct iwl_trans *trans, static void iwl_pcie_gen2_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq) { - struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); - /* rd_ptr is bounded by TFD_QUEUE_SIZE_MAX and * idx is bounded by n_window */ @@ -181,7 +179,7 @@ static void iwl_pcie_gen2_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq) lockdep_assert_held(&txq->lock); iwl_pcie_gen2_tfd_unmap(trans, &txq->entries[idx].meta, - iwl_pcie_get_tfd(trans_pcie, txq, idx)); + iwl_pcie_get_tfd(trans, txq, idx)); /* free SKB */ if (txq->entries) { @@ -364,11 +362,9 @@ struct iwl_tfh_tfd *iwl_pcie_gen2_build_tfd(struct iwl_trans *trans, struct sk_buff *skb, struct iwl_cmd_meta *out_meta) { - struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; int idx = iwl_pcie_get_cmd_index(txq, txq->write_ptr); - struct iwl_tfh_tfd *tfd = - iwl_pcie_get_tfd(trans_pcie, txq, idx); + struct iwl_tfh_tfd *tfd = iwl_pcie_get_tfd(trans, txq, idx); dma_addr_t tb_phys; bool amsdu; int i, len, tb1_len, tb2_len, hdr_len; @@ -565,8 +561,7 @@ static int iwl_pcie_gen2_enqueue_hcmd(struct iwl_trans *trans, u8 group_id = iwl_cmd_groupid(cmd->id); const u8 *cmddata[IWL_MAX_CMD_TBS_PER_TFD]; u16 cmdlen[IWL_MAX_CMD_TBS_PER_TFD]; - struct iwl_tfh_tfd *tfd = - iwl_pcie_get_tfd(trans_pcie, txq, txq->write_ptr); + struct iwl_tfh_tfd *tfd = iwl_pcie_get_tfd(trans, txq, txq->write_ptr); memset(tfd, 0, sizeof(*tfd)); diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c index fed6d842a5e1..3f85713c41dc 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c @@ -373,7 +373,7 @@ static void iwl_pcie_tfd_unmap(struct iwl_trans *trans, { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); int i, num_tbs; - void *tfd = iwl_pcie_get_tfd(trans_pcie, txq, index); + void *tfd = iwl_pcie_get_tfd(trans, txq, index); /* Sanity check on number of chunks */ num_tbs = iwl_pcie_tfd_get_num_tbs(trans, tfd); @@ -2018,7 +2018,7 @@ static int iwl_fill_data_tbs(struct iwl_trans *trans, struct sk_buff *skb, } trace_iwlwifi_dev_tx(trans->dev, skb, - iwl_pcie_get_tfd(trans_pcie, txq, txq->write_ptr), + iwl_pcie_get_tfd(trans, txq, txq->write_ptr), trans_pcie->tfd_size, &dev_cmd->hdr, IWL_FIRST_TB_SIZE + tb1_len, hdr_len); @@ -2092,7 +2092,7 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb, IEEE80211_CCMP_HDR_LEN : 0; trace_iwlwifi_dev_tx(trans->dev, skb, - iwl_pcie_get_tfd(trans_pcie, txq, txq->write_ptr), + iwl_pcie_get_tfd(trans, txq, txq->write_ptr), trans_pcie->tfd_size, &dev_cmd->hdr, IWL_FIRST_TB_SIZE + tb1_len, 0); @@ -2425,7 +2425,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb, memcpy(&txq->first_tb_bufs[txq->write_ptr], &dev_cmd->hdr, IWL_FIRST_TB_SIZE); - tfd = iwl_pcie_get_tfd(trans_pcie, txq, txq->write_ptr); + tfd = iwl_pcie_get_tfd(trans, txq, txq->write_ptr); /* Set up entry for this TFD in Tx byte-count array */ iwl_pcie_txq_update_byte_cnt_tbl(trans, txq, le16_to_cpu(tx_cmd->len), iwl_pcie_tfd_get_num_tbs(trans, tfd)); From 5731a879d03bdaa00265f8ebc32dfd0e65d25276 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Thu, 4 Jan 2018 20:02:09 -0800 Subject: [PATCH 03/27] bpf: sockmap missing NULL psock check Add psock NULL check to handle a racing sock event that can get the sk_callback_lock before this case but after xchg happens causing the refcnt to hit zero and sock user data (psock) to be null and queued for garbage collection. Also add a comment in the code because this is a bit subtle and not obvious in my opinion. Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann --- kernel/bpf/sockmap.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c index 5ee2e41893d9..1712d319c2d8 100644 --- a/kernel/bpf/sockmap.c +++ b/kernel/bpf/sockmap.c @@ -591,8 +591,15 @@ static void sock_map_free(struct bpf_map *map) write_lock_bh(&sock->sk_callback_lock); psock = smap_psock_sk(sock); - smap_list_remove(psock, &stab->sock_map[i]); - smap_release_sock(psock, sock); + /* This check handles a racing sock event that can get the + * sk_callback_lock before this case but after xchg happens + * causing the refcnt to hit zero and sock user data (psock) + * to be null and queued for garbage collection. + */ + if (likely(psock)) { + smap_list_remove(psock, &stab->sock_map[i]); + smap_release_sock(psock, sock); + } write_unlock_bh(&sock->sk_callback_lock); } rcu_read_unlock(); From 2b36047e7889b7efee22c11e17f035f721855731 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Fri, 5 Jan 2018 15:02:00 -0800 Subject: [PATCH 04/27] selftests/bpf: fix test_align since commit 82abbf8d2fc4 the verifier rejects the bit-wise arithmetic on pointers earlier. The test 'dubious pointer arithmetic' now has less output to match on. Adjust it. Fixes: 82abbf8d2fc4 ("bpf: do not allow root to mangle valid pointers") Reported-by: kernel test robot Signed-off-by: Alexei Starovoitov Signed-off-by: Daniel Borkmann --- tools/testing/selftests/bpf/test_align.c | 22 +--------------------- 1 file changed, 1 insertion(+), 21 deletions(-) diff --git a/tools/testing/selftests/bpf/test_align.c b/tools/testing/selftests/bpf/test_align.c index 8591c89c0828..471bbbdb94db 100644 --- a/tools/testing/selftests/bpf/test_align.c +++ b/tools/testing/selftests/bpf/test_align.c @@ -474,27 +474,7 @@ static struct bpf_align_test tests[] = { .result = REJECT, .matches = { {4, "R5=pkt(id=0,off=0,r=0,imm=0)"}, - /* ptr & 0x40 == either 0 or 0x40 */ - {5, "R5=inv(id=0,umax_value=64,var_off=(0x0; 0x40))"}, - /* ptr << 2 == unknown, (4n) */ - {7, "R5=inv(id=0,smax_value=9223372036854775804,umax_value=18446744073709551612,var_off=(0x0; 0xfffffffffffffffc))"}, - /* (4n) + 14 == (4n+2). We blow our bounds, because - * the add could overflow. - */ - {8, "R5=inv(id=0,var_off=(0x2; 0xfffffffffffffffc))"}, - /* Checked s>=0 */ - {10, "R5=inv(id=0,umin_value=2,umax_value=9223372036854775806,var_off=(0x2; 0x7ffffffffffffffc))"}, - /* packet pointer + nonnegative (4n+2) */ - {12, "R6=pkt(id=1,off=0,r=0,umin_value=2,umax_value=9223372036854775806,var_off=(0x2; 0x7ffffffffffffffc))"}, - {14, "R4=pkt(id=1,off=4,r=0,umin_value=2,umax_value=9223372036854775806,var_off=(0x2; 0x7ffffffffffffffc))"}, - /* NET_IP_ALIGN + (4n+2) == (4n), alignment is fine. - * We checked the bounds, but it might have been able - * to overflow if the packet pointer started in the - * upper half of the address space. - * So we did not get a 'range' on R6, and the access - * attempt will fail. - */ - {16, "R6=pkt(id=1,off=0,r=0,umin_value=2,umax_value=9223372036854775806,var_off=(0x2; 0x7ffffffffffffffc))"}, + /* R5 bitwise operator &= on pointer prohibited */ } }, { From b2157399cc9898260d6031c5bfe45fe137c1fbe7 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Sun, 7 Jan 2018 17:33:02 -0800 Subject: [PATCH 05/27] bpf: prevent out-of-bounds speculation Under speculation, CPUs may mis-predict branches in bounds checks. Thus, memory accesses under a bounds check may be speculated even if the bounds check fails, providing a primitive for building a side channel. To avoid leaking kernel data round up array-based maps and mask the index after bounds check, so speculated load with out of bounds index will load either valid value from the array or zero from the padded area. Unconditionally mask index for all array types even when max_entries are not rounded to power of 2 for root user. When map is created by unpriv user generate a sequence of bpf insns that includes AND operation to make sure that JITed code includes the same 'index & index_mask' operation. If prog_array map is created by unpriv user replace bpf_tail_call(ctx, map, index); with if (index >= max_entries) { index &= map->index_mask; bpf_tail_call(ctx, map, index); } (along with roundup to power 2) to prevent out-of-bounds speculation. There is secondary redundant 'if (index >= max_entries)' in the interpreter and in all JITs, but they can be optimized later if necessary. Other array-like maps (cpumap, devmap, sockmap, perf_event_array, cgroup_array) cannot be used by unpriv, so no changes there. That fixes bpf side of "Variant 1: bounds check bypass (CVE-2017-5753)" on all architectures with and without JIT. v2->v3: Daniel noticed that attack potentially can be crafted via syscall commands without loading the program, so add masking to those paths as well. Signed-off-by: Alexei Starovoitov Acked-by: John Fastabend Signed-off-by: Daniel Borkmann --- include/linux/bpf.h | 2 ++ kernel/bpf/arraymap.c | 47 +++++++++++++++++++++++++++++++++---------- kernel/bpf/verifier.c | 36 +++++++++++++++++++++++++++++++++ 3 files changed, 74 insertions(+), 11 deletions(-) diff --git a/include/linux/bpf.h b/include/linux/bpf.h index e55e4255a210..1b985ca4ffbe 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -52,6 +52,7 @@ struct bpf_map { u32 pages; u32 id; int numa_node; + bool unpriv_array; struct user_struct *user; const struct bpf_map_ops *ops; struct work_struct work; @@ -221,6 +222,7 @@ struct bpf_prog_aux { struct bpf_array { struct bpf_map map; u32 elem_size; + u32 index_mask; /* 'ownership' of prog_array is claimed by the first program that * is going to use this map or by the first program which FD is stored * in the map to make sure that all callers and callees have the same diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 7c25426d3cf5..aaa319848e7d 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -53,9 +53,10 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) { bool percpu = attr->map_type == BPF_MAP_TYPE_PERCPU_ARRAY; int numa_node = bpf_map_attr_numa_node(attr); + u32 elem_size, index_mask, max_entries; + bool unpriv = !capable(CAP_SYS_ADMIN); struct bpf_array *array; u64 array_size; - u32 elem_size; /* check sanity of attributes */ if (attr->max_entries == 0 || attr->key_size != 4 || @@ -72,11 +73,20 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) elem_size = round_up(attr->value_size, 8); + max_entries = attr->max_entries; + index_mask = roundup_pow_of_two(max_entries) - 1; + + if (unpriv) + /* round up array size to nearest power of 2, + * since cpu will speculate within index_mask limits + */ + max_entries = index_mask + 1; + array_size = sizeof(*array); if (percpu) - array_size += (u64) attr->max_entries * sizeof(void *); + array_size += (u64) max_entries * sizeof(void *); else - array_size += (u64) attr->max_entries * elem_size; + array_size += (u64) max_entries * elem_size; /* make sure there is no u32 overflow later in round_up() */ if (array_size >= U32_MAX - PAGE_SIZE) @@ -86,6 +96,8 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) array = bpf_map_area_alloc(array_size, numa_node); if (!array) return ERR_PTR(-ENOMEM); + array->index_mask = index_mask; + array->map.unpriv_array = unpriv; /* copy mandatory map attributes */ array->map.map_type = attr->map_type; @@ -121,12 +133,13 @@ static void *array_map_lookup_elem(struct bpf_map *map, void *key) if (unlikely(index >= array->map.max_entries)) return NULL; - return array->value + array->elem_size * index; + return array->value + array->elem_size * (index & array->index_mask); } /* emit BPF instructions equivalent to C code of array_map_lookup_elem() */ static u32 array_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) { + struct bpf_array *array = container_of(map, struct bpf_array, map); struct bpf_insn *insn = insn_buf; u32 elem_size = round_up(map->value_size, 8); const int ret = BPF_REG_0; @@ -135,7 +148,12 @@ static u32 array_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) *insn++ = BPF_ALU64_IMM(BPF_ADD, map_ptr, offsetof(struct bpf_array, value)); *insn++ = BPF_LDX_MEM(BPF_W, ret, index, 0); - *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 3); + if (map->unpriv_array) { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 4); + *insn++ = BPF_ALU32_IMM(BPF_AND, ret, array->index_mask); + } else { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 3); + } if (is_power_of_2(elem_size)) { *insn++ = BPF_ALU64_IMM(BPF_LSH, ret, ilog2(elem_size)); @@ -157,7 +175,7 @@ static void *percpu_array_map_lookup_elem(struct bpf_map *map, void *key) if (unlikely(index >= array->map.max_entries)) return NULL; - return this_cpu_ptr(array->pptrs[index]); + return this_cpu_ptr(array->pptrs[index & array->index_mask]); } int bpf_percpu_array_copy(struct bpf_map *map, void *key, void *value) @@ -177,7 +195,7 @@ int bpf_percpu_array_copy(struct bpf_map *map, void *key, void *value) */ size = round_up(map->value_size, 8); rcu_read_lock(); - pptr = array->pptrs[index]; + pptr = array->pptrs[index & array->index_mask]; for_each_possible_cpu(cpu) { bpf_long_memcpy(value + off, per_cpu_ptr(pptr, cpu), size); off += size; @@ -225,10 +243,11 @@ static int array_map_update_elem(struct bpf_map *map, void *key, void *value, return -EEXIST; if (array->map.map_type == BPF_MAP_TYPE_PERCPU_ARRAY) - memcpy(this_cpu_ptr(array->pptrs[index]), + memcpy(this_cpu_ptr(array->pptrs[index & array->index_mask]), value, map->value_size); else - memcpy(array->value + array->elem_size * index, + memcpy(array->value + + array->elem_size * (index & array->index_mask), value, map->value_size); return 0; } @@ -262,7 +281,7 @@ int bpf_percpu_array_update(struct bpf_map *map, void *key, void *value, */ size = round_up(map->value_size, 8); rcu_read_lock(); - pptr = array->pptrs[index]; + pptr = array->pptrs[index & array->index_mask]; for_each_possible_cpu(cpu) { bpf_long_memcpy(per_cpu_ptr(pptr, cpu), value + off, size); off += size; @@ -613,6 +632,7 @@ static void *array_of_map_lookup_elem(struct bpf_map *map, void *key) static u32 array_of_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) { + struct bpf_array *array = container_of(map, struct bpf_array, map); u32 elem_size = round_up(map->value_size, 8); struct bpf_insn *insn = insn_buf; const int ret = BPF_REG_0; @@ -621,7 +641,12 @@ static u32 array_of_map_gen_lookup(struct bpf_map *map, *insn++ = BPF_ALU64_IMM(BPF_ADD, map_ptr, offsetof(struct bpf_array, value)); *insn++ = BPF_LDX_MEM(BPF_W, ret, index, 0); - *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 5); + if (map->unpriv_array) { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 6); + *insn++ = BPF_ALU32_IMM(BPF_AND, ret, array->index_mask); + } else { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 5); + } if (is_power_of_2(elem_size)) *insn++ = BPF_ALU64_IMM(BPF_LSH, ret, ilog2(elem_size)); else diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 04b24876cd23..b414d6b2d470 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1729,6 +1729,13 @@ static int check_call(struct bpf_verifier_env *env, int func_id, int insn_idx) err = check_func_arg(env, BPF_REG_2, fn->arg2_type, &meta); if (err) return err; + if (func_id == BPF_FUNC_tail_call) { + if (meta.map_ptr == NULL) { + verbose(env, "verifier bug\n"); + return -EINVAL; + } + env->insn_aux_data[insn_idx].map_ptr = meta.map_ptr; + } err = check_func_arg(env, BPF_REG_3, fn->arg3_type, &meta); if (err) return err; @@ -4456,6 +4463,35 @@ static int fixup_bpf_calls(struct bpf_verifier_env *env) */ insn->imm = 0; insn->code = BPF_JMP | BPF_TAIL_CALL; + + /* instead of changing every JIT dealing with tail_call + * emit two extra insns: + * if (index >= max_entries) goto out; + * index &= array->index_mask; + * to avoid out-of-bounds cpu speculation + */ + map_ptr = env->insn_aux_data[i + delta].map_ptr; + if (map_ptr == BPF_MAP_PTR_POISON) { + verbose(env, "tail_call obusing map_ptr\n"); + return -EINVAL; + } + if (!map_ptr->unpriv_array) + continue; + insn_buf[0] = BPF_JMP_IMM(BPF_JGE, BPF_REG_3, + map_ptr->max_entries, 2); + insn_buf[1] = BPF_ALU32_IMM(BPF_AND, BPF_REG_3, + container_of(map_ptr, + struct bpf_array, + map)->index_mask); + insn_buf[2] = *insn; + cnt = 3; + new_prog = bpf_patch_insn_data(env, i + delta, insn_buf, cnt); + if (!new_prog) + return -ENOMEM; + + delta += cnt - 1; + env->prog = prog = new_prog; + insn = new_prog->insnsi + i + delta; continue; } From b8fd0823e0770c2d5fdbd865bccf0d5e058e5287 Mon Sep 17 00:00:00 2001 From: Andrii Vladyka Date: Thu, 4 Jan 2018 13:09:17 +0200 Subject: [PATCH 06/27] net: core: fix module type in sock_diag_bind Use AF_INET6 instead of AF_INET in IPv6-related code path Signed-off-by: Andrii Vladyka Signed-off-by: David S. Miller --- net/core/sock_diag.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/sock_diag.c b/net/core/sock_diag.c index 217f4e3b82f6..146b50e30659 100644 --- a/net/core/sock_diag.c +++ b/net/core/sock_diag.c @@ -288,7 +288,7 @@ static int sock_diag_bind(struct net *net, int group) case SKNLGRP_INET6_UDP_DESTROY: if (!sock_diag_handlers[AF_INET6]) request_module("net-pf-%d-proto-%d-type-%d", PF_NETLINK, - NETLINK_SOCK_DIAG, AF_INET); + NETLINK_SOCK_DIAG, AF_INET6); break; } return 0; From 3dc2fa47549aca71773afdd12a78d31802bb22b4 Mon Sep 17 00:00:00 2001 From: Xiongfeng Wang Date: Mon, 8 Jan 2018 19:43:00 +0800 Subject: [PATCH 07/27] net: caif: use strlcpy() instead of strncpy() gcc-8 reports net/caif/caif_dev.c: In function 'caif_enroll_dev': ./include/linux/string.h:245:9: warning: '__builtin_strncpy' output may be truncated copying 15 bytes from a string of length 15 [-Wstringop-truncation] net/caif/cfctrl.c: In function 'cfctrl_linkup_request': ./include/linux/string.h:245:9: warning: '__builtin_strncpy' output may be truncated copying 15 bytes from a string of length 15 [-Wstringop-truncation] net/caif/cfcnfg.c: In function 'caif_connect_client': ./include/linux/string.h:245:9: warning: '__builtin_strncpy' output may be truncated copying 15 bytes from a string of length 15 [-Wstringop-truncation] The compiler require that the input param 'len' of strncpy() should be greater than the length of the src string, so that '\0' is copied as well. We can just use strlcpy() to avoid this warning. Signed-off-by: Xiongfeng Wang Signed-off-by: David S. Miller --- net/caif/caif_dev.c | 5 ++--- net/caif/cfcnfg.c | 10 ++++------ net/caif/cfctrl.c | 4 ++-- 3 files changed, 8 insertions(+), 11 deletions(-) diff --git a/net/caif/caif_dev.c b/net/caif/caif_dev.c index 2d38b6e34203..e0adcd123f48 100644 --- a/net/caif/caif_dev.c +++ b/net/caif/caif_dev.c @@ -334,9 +334,8 @@ void caif_enroll_dev(struct net_device *dev, struct caif_dev_common *caifdev, mutex_lock(&caifdevs->lock); list_add_rcu(&caifd->list, &caifdevs->list); - strncpy(caifd->layer.name, dev->name, - sizeof(caifd->layer.name) - 1); - caifd->layer.name[sizeof(caifd->layer.name) - 1] = 0; + strlcpy(caifd->layer.name, dev->name, + sizeof(caifd->layer.name)); caifd->layer.transmit = transmit; cfcnfg_add_phy_layer(cfg, dev, diff --git a/net/caif/cfcnfg.c b/net/caif/cfcnfg.c index 273cb07f57d8..8f00bea093b9 100644 --- a/net/caif/cfcnfg.c +++ b/net/caif/cfcnfg.c @@ -268,17 +268,15 @@ static int caif_connect_req_to_link_param(struct cfcnfg *cnfg, case CAIFPROTO_RFM: l->linktype = CFCTRL_SRV_RFM; l->u.datagram.connid = s->sockaddr.u.rfm.connection_id; - strncpy(l->u.rfm.volume, s->sockaddr.u.rfm.volume, - sizeof(l->u.rfm.volume)-1); - l->u.rfm.volume[sizeof(l->u.rfm.volume)-1] = 0; + strlcpy(l->u.rfm.volume, s->sockaddr.u.rfm.volume, + sizeof(l->u.rfm.volume)); break; case CAIFPROTO_UTIL: l->linktype = CFCTRL_SRV_UTIL; l->endpoint = 0x00; l->chtype = 0x00; - strncpy(l->u.utility.name, s->sockaddr.u.util.service, - sizeof(l->u.utility.name)-1); - l->u.utility.name[sizeof(l->u.utility.name)-1] = 0; + strlcpy(l->u.utility.name, s->sockaddr.u.util.service, + sizeof(l->u.utility.name)); caif_assert(sizeof(l->u.utility.name) > 10); l->u.utility.paramlen = s->param.size; if (l->u.utility.paramlen > sizeof(l->u.utility.params)) diff --git a/net/caif/cfctrl.c b/net/caif/cfctrl.c index f5afda1abc76..655ed7032150 100644 --- a/net/caif/cfctrl.c +++ b/net/caif/cfctrl.c @@ -258,8 +258,8 @@ int cfctrl_linkup_request(struct cflayer *layer, tmp16 = cpu_to_le16(param->u.utility.fifosize_bufs); cfpkt_add_body(pkt, &tmp16, 2); memset(utility_name, 0, sizeof(utility_name)); - strncpy(utility_name, param->u.utility.name, - UTILITY_NAME_LENGTH - 1); + strlcpy(utility_name, param->u.utility.name, + UTILITY_NAME_LENGTH); cfpkt_add_body(pkt, utility_name, UTILITY_NAME_LENGTH); tmp8 = param->u.utility.paramlen; cfpkt_add_body(pkt, &tmp8, 1); From 20b50d79974ea3192e8c3ab7faf4e536e5f14d8f Mon Sep 17 00:00:00 2001 From: Nicolai Stange Date: Mon, 8 Jan 2018 15:54:44 +0100 Subject: [PATCH 08/27] net: ipv4: emulate READ_ONCE() on ->hdrincl bit-field in raw_sendmsg() Commit 8f659a03a0ba ("net: ipv4: fix for a race condition in raw_sendmsg") fixed the issue of possibly inconsistent ->hdrincl handling due to concurrent updates by reading this bit-field member into a local variable and using the thus stabilized value in subsequent tests. However, aforementioned commit also adds the (correct) comment that /* hdrincl should be READ_ONCE(inet->hdrincl) * but READ_ONCE() doesn't work with bit fields */ because as it stands, the compiler is free to shortcut or even eliminate the local variable at its will. Note that I have not seen anything like this happening in reality and thus, the concern is a theoretical one. However, in order to be on the safe side, emulate a READ_ONCE() on the bit-field by doing it on the local 'hdrincl' variable itself: int hdrincl = inet->hdrincl; hdrincl = READ_ONCE(hdrincl); This breaks the chain in the sense that the compiler is not allowed to replace subsequent reads from hdrincl with reloads from inet->hdrincl. Fixes: 8f659a03a0ba ("net: ipv4: fix for a race condition in raw_sendmsg") Signed-off-by: Nicolai Stange Reviewed-by: Stefano Brivio Signed-off-by: David S. Miller --- net/ipv4/raw.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 125c1eab3eaa..5e570aa9e43b 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -520,9 +520,11 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) goto out; /* hdrincl should be READ_ONCE(inet->hdrincl) - * but READ_ONCE() doesn't work with bit fields + * but READ_ONCE() doesn't work with bit fields. + * Doing this indirectly yields the same result. */ hdrincl = inet->hdrincl; + hdrincl = READ_ONCE(hdrincl); /* * Check the flags. */ From 2fdd18118dad86bf5e7880d8d02ea27be23e3671 Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Mon, 8 Jan 2018 08:50:17 +0200 Subject: [PATCH 09/27] docs-rst: networking: wire up msg_zerocopy Fix the following 'make htmldocs' complaint: Documentation/networking/msg_zerocopy.rst:: WARNING: document isn't included in any toctree. Signed-off-by: Mike Rapoport Signed-off-by: David S. Miller --- Documentation/networking/index.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/networking/index.rst b/Documentation/networking/index.rst index 66e620866245..7d4b15977d61 100644 --- a/Documentation/networking/index.rst +++ b/Documentation/networking/index.rst @@ -9,6 +9,7 @@ Contents: batman-adv kapi z8530book + msg_zerocopy .. only:: subproject @@ -16,4 +17,3 @@ Contents: ======= * :ref:`genindex` - From 195e2addbce09e5afbc766efc1e6567c9ce840d3 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Sat, 6 Jan 2018 21:53:26 +0300 Subject: [PATCH 10/27] SolutionEngine771x: fix Ether platform data The 'sh_eth' driver's probe() method would fail on the SolutionEngine7710 board and crash on SolutionEngine7712 board as the platform code is hopelessly behind the driver's platform data -- it passes the PHY address instead of 'struct sh_eth_plat_data *'; pass the latter to the driver in order to fix the bug... Fixes: 71557a37adb5 ("[netdrvr] sh_eth: Add SH7619 support") Signed-off-by: Sergei Shtylyov Signed-off-by: David S. Miller --- arch/sh/boards/mach-se/770x/setup.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/arch/sh/boards/mach-se/770x/setup.c b/arch/sh/boards/mach-se/770x/setup.c index 77c35350ee77..b7fa7a87e946 100644 --- a/arch/sh/boards/mach-se/770x/setup.c +++ b/arch/sh/boards/mach-se/770x/setup.c @@ -9,6 +9,7 @@ */ #include #include +#include #include #include #include @@ -115,6 +116,11 @@ static struct platform_device heartbeat_device = { #if defined(CONFIG_CPU_SUBTYPE_SH7710) ||\ defined(CONFIG_CPU_SUBTYPE_SH7712) /* SH771X Ethernet driver */ +static struct sh_eth_plat_data sh_eth_plat = { + .phy = PHY_ID, + .phy_interface = PHY_INTERFACE_MODE_MII, +}; + static struct resource sh_eth0_resources[] = { [0] = { .start = SH_ETH0_BASE, @@ -132,7 +138,7 @@ static struct platform_device sh_eth0_device = { .name = "sh771x-ether", .id = 0, .dev = { - .platform_data = PHY_ID, + .platform_data = &sh_eth_plat, }, .num_resources = ARRAY_SIZE(sh_eth0_resources), .resource = sh_eth0_resources, @@ -155,7 +161,7 @@ static struct platform_device sh_eth1_device = { .name = "sh771x-ether", .id = 1, .dev = { - .platform_data = PHY_ID, + .platform_data = &sh_eth_plat, }, .num_resources = ARRAY_SIZE(sh_eth1_resources), .resource = sh_eth1_resources, From f9a531d6731d74f1e24298d9641c2dc1fef2631b Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Sat, 6 Jan 2018 21:53:27 +0300 Subject: [PATCH 11/27] SolutionEngine771x: add Ether TSU resource After the Ether platform data is fixed, the driver probe() method would still fail since the 'struct sh_eth_cpu_data' corresponding to SH771x indicates the presence of TSU but the memory resource for it is absent. Add the missing TSU resource to both Ether devices and fix the harmless off-by-one error in the main memory resources, while at it... Fixes: 4986b996882d ("net: sh_eth: remove the SH_TSU_ADDR") Signed-off-by: Sergei Shtylyov Signed-off-by: David S. Miller --- arch/sh/boards/mach-se/770x/setup.c | 14 ++++++++++++-- arch/sh/include/mach-se/mach/se.h | 1 + 2 files changed, 13 insertions(+), 2 deletions(-) diff --git a/arch/sh/boards/mach-se/770x/setup.c b/arch/sh/boards/mach-se/770x/setup.c index b7fa7a87e946..412326d59e6f 100644 --- a/arch/sh/boards/mach-se/770x/setup.c +++ b/arch/sh/boards/mach-se/770x/setup.c @@ -124,10 +124,15 @@ static struct sh_eth_plat_data sh_eth_plat = { static struct resource sh_eth0_resources[] = { [0] = { .start = SH_ETH0_BASE, - .end = SH_ETH0_BASE + 0x1B8, + .end = SH_ETH0_BASE + 0x1B8 - 1, .flags = IORESOURCE_MEM, }, [1] = { + .start = SH_TSU_BASE, + .end = SH_TSU_BASE + 0x200 - 1, + .flags = IORESOURCE_MEM, + }, + [2] = { .start = SH_ETH0_IRQ, .end = SH_ETH0_IRQ, .flags = IORESOURCE_IRQ, @@ -147,10 +152,15 @@ static struct platform_device sh_eth0_device = { static struct resource sh_eth1_resources[] = { [0] = { .start = SH_ETH1_BASE, - .end = SH_ETH1_BASE + 0x1B8, + .end = SH_ETH1_BASE + 0x1B8 - 1, .flags = IORESOURCE_MEM, }, [1] = { + .start = SH_TSU_BASE, + .end = SH_TSU_BASE + 0x200 - 1, + .flags = IORESOURCE_MEM, + }, + [2] = { .start = SH_ETH1_IRQ, .end = SH_ETH1_IRQ, .flags = IORESOURCE_IRQ, diff --git a/arch/sh/include/mach-se/mach/se.h b/arch/sh/include/mach-se/mach/se.h index 4246ef9b07a3..aa83fe1ff0b1 100644 --- a/arch/sh/include/mach-se/mach/se.h +++ b/arch/sh/include/mach-se/mach/se.h @@ -100,6 +100,7 @@ /* Base address */ #define SH_ETH0_BASE 0xA7000000 #define SH_ETH1_BASE 0xA7000400 +#define SH_TSU_BASE 0xA7000800 /* PHY ID */ #if defined(CONFIG_CPU_SUBTYPE_SH7710) # define PHY_ID 0x00 From 4512c43eac7e007d982e7ea45152ea6f3f4d1921 Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Mon, 8 Jan 2018 10:34:00 -0800 Subject: [PATCH 12/27] ipv6: remove null_entry before adding default route In the current code, when creating a new fib6 table, tb6_root.leaf gets initialized to net->ipv6.ip6_null_entry. If a default route is being added with rt->rt6i_metric = 0xffffffff, fib6_add() will add this route after net->ipv6.ip6_null_entry. As null_entry is shared, it could cause problem. In order to fix it, set fn->leaf to NULL before calling fib6_add_rt2node() when trying to add the first default route. And reset fn->leaf to null_entry when adding fails or when deleting the last default route. syzkaller reported the following issue which is fixed by this commit: WARNING: suspicious RCU usage 4.15.0-rc5+ #171 Not tainted ----------------------------- net/ipv6/ip6_fib.c:1702 suspicious rcu_dereference_protected() usage! other info that might help us debug this: rcu_scheduler_active = 2, debug_locks = 1 4 locks held by swapper/0/0: #0: ((&net->ipv6.ip6_fib_timer)){+.-.}, at: [<00000000d43f631b>] lockdep_copy_map include/linux/lockdep.h:178 [inline] #0: ((&net->ipv6.ip6_fib_timer)){+.-.}, at: [<00000000d43f631b>] call_timer_fn+0x1c6/0x820 kernel/time/timer.c:1310 #1: (&(&net->ipv6.fib6_gc_lock)->rlock){+.-.}, at: [<000000002ff9d65c>] spin_lock_bh include/linux/spinlock.h:315 [inline] #1: (&(&net->ipv6.fib6_gc_lock)->rlock){+.-.}, at: [<000000002ff9d65c>] fib6_run_gc+0x9d/0x3c0 net/ipv6/ip6_fib.c:2007 #2: (rcu_read_lock){....}, at: [<0000000091db762d>] __fib6_clean_all+0x0/0x3a0 net/ipv6/ip6_fib.c:1560 #3: (&(&tb->tb6_lock)->rlock){+.-.}, at: [<000000009e503581>] spin_lock_bh include/linux/spinlock.h:315 [inline] #3: (&(&tb->tb6_lock)->rlock){+.-.}, at: [<000000009e503581>] __fib6_clean_all+0x1d0/0x3a0 net/ipv6/ip6_fib.c:1948 stack backtrace: CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.15.0-rc5+ #171 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:17 [inline] dump_stack+0x194/0x257 lib/dump_stack.c:53 lockdep_rcu_suspicious+0x123/0x170 kernel/locking/lockdep.c:4585 fib6_del+0xcaa/0x11b0 net/ipv6/ip6_fib.c:1701 fib6_clean_node+0x3aa/0x4f0 net/ipv6/ip6_fib.c:1892 fib6_walk_continue+0x46c/0x8a0 net/ipv6/ip6_fib.c:1815 fib6_walk+0x91/0xf0 net/ipv6/ip6_fib.c:1863 fib6_clean_tree+0x1e6/0x340 net/ipv6/ip6_fib.c:1933 __fib6_clean_all+0x1f4/0x3a0 net/ipv6/ip6_fib.c:1949 fib6_clean_all net/ipv6/ip6_fib.c:1960 [inline] fib6_run_gc+0x16b/0x3c0 net/ipv6/ip6_fib.c:2016 fib6_gc_timer_cb+0x20/0x30 net/ipv6/ip6_fib.c:2033 call_timer_fn+0x228/0x820 kernel/time/timer.c:1320 expire_timers kernel/time/timer.c:1357 [inline] __run_timers+0x7ee/0xb70 kernel/time/timer.c:1660 run_timer_softirq+0x4c/0xb0 kernel/time/timer.c:1686 __do_softirq+0x2d7/0xb85 kernel/softirq.c:285 invoke_softirq kernel/softirq.c:365 [inline] irq_exit+0x1cc/0x200 kernel/softirq.c:405 exiting_irq arch/x86/include/asm/apic.h:540 [inline] smp_apic_timer_interrupt+0x16b/0x700 arch/x86/kernel/apic/apic.c:1052 apic_timer_interrupt+0xa9/0xb0 arch/x86/entry/entry_64.S:904 Reported-by: syzbot Fixes: 66f5d6ce53e6 ("ipv6: replace rwlock with rcu and spinlock in fib6_table") Signed-off-by: Wei Wang Acked-by: Martin KaFai Lau Signed-off-by: David S. Miller --- net/ipv6/ip6_fib.c | 38 +++++++++++++++++++++++++++++--------- 1 file changed, 29 insertions(+), 9 deletions(-) diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index d11a5578e4f8..9dcc3924a975 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -640,6 +640,11 @@ static struct fib6_node *fib6_add_1(struct net *net, if (!(fn->fn_flags & RTN_RTINFO)) { RCU_INIT_POINTER(fn->leaf, NULL); rt6_release(leaf); + /* remove null_entry in the root node */ + } else if (fn->fn_flags & RTN_TL_ROOT && + rcu_access_pointer(fn->leaf) == + net->ipv6.ip6_null_entry) { + RCU_INIT_POINTER(fn->leaf, NULL); } return fn; @@ -1270,13 +1275,17 @@ out: return err; failure: - /* fn->leaf could be NULL if fn is an intermediate node and we - * failed to add the new route to it in both subtree creation - * failure and fib6_add_rt2node() failure case. - * In both cases, fib6_repair_tree() should be called to fix - * fn->leaf. + /* fn->leaf could be NULL and fib6_repair_tree() needs to be called if: + * 1. fn is an intermediate node and we failed to add the new + * route to it in both subtree creation failure and fib6_add_rt2node() + * failure case. + * 2. fn is the root node in the table and we fail to add the first + * default route to it. */ - if (fn && !(fn->fn_flags & (RTN_RTINFO|RTN_ROOT))) + if (fn && + (!(fn->fn_flags & (RTN_RTINFO|RTN_ROOT)) || + (fn->fn_flags & RTN_TL_ROOT && + !rcu_access_pointer(fn->leaf)))) fib6_repair_tree(info->nl_net, table, fn); /* Always release dst as dst->__refcnt is guaranteed * to be taken before entering this function @@ -1531,6 +1540,12 @@ static struct fib6_node *fib6_repair_tree(struct net *net, struct fib6_walker *w; int iter = 0; + /* Set fn->leaf to null_entry for root node. */ + if (fn->fn_flags & RTN_TL_ROOT) { + rcu_assign_pointer(fn->leaf, net->ipv6.ip6_null_entry); + return fn; + } + for (;;) { struct fib6_node *fn_r = rcu_dereference_protected(fn->right, lockdep_is_held(&table->tb6_lock)); @@ -1685,10 +1700,15 @@ static void fib6_del_route(struct fib6_table *table, struct fib6_node *fn, } read_unlock(&net->ipv6.fib6_walker_lock); - /* If it was last route, expunge its radix tree node */ + /* If it was last route, call fib6_repair_tree() to: + * 1. For root node, put back null_entry as how the table was created. + * 2. For other nodes, expunge its radix tree node. + */ if (!rcu_access_pointer(fn->leaf)) { - fn->fn_flags &= ~RTN_RTINFO; - net->ipv6.rt6_stats->fib_route_nodes--; + if (!(fn->fn_flags & RTN_TL_ROOT)) { + fn->fn_flags &= ~RTN_RTINFO; + net->ipv6.rt6_stats->fib_route_nodes--; + } fn = fib6_repair_tree(net, table, fn); } From be95a845cc4402272994ce290e3ad928aff06cb9 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 9 Jan 2018 13:17:44 +0100 Subject: [PATCH 13/27] bpf: avoid false sharing of map refcount with max_entries In addition to commit b2157399cc98 ("bpf: prevent out-of-bounds speculation") also change the layout of struct bpf_map such that false sharing of fast-path members like max_entries is avoided when the maps reference counter is altered. Therefore enforce them to be placed into separate cachelines. pahole dump after change: struct bpf_map { const struct bpf_map_ops * ops; /* 0 8 */ struct bpf_map * inner_map_meta; /* 8 8 */ void * security; /* 16 8 */ enum bpf_map_type map_type; /* 24 4 */ u32 key_size; /* 28 4 */ u32 value_size; /* 32 4 */ u32 max_entries; /* 36 4 */ u32 map_flags; /* 40 4 */ u32 pages; /* 44 4 */ u32 id; /* 48 4 */ int numa_node; /* 52 4 */ bool unpriv_array; /* 56 1 */ /* XXX 7 bytes hole, try to pack */ /* --- cacheline 1 boundary (64 bytes) --- */ struct user_struct * user; /* 64 8 */ atomic_t refcnt; /* 72 4 */ atomic_t usercnt; /* 76 4 */ struct work_struct work; /* 80 32 */ char name[16]; /* 112 16 */ /* --- cacheline 2 boundary (128 bytes) --- */ /* size: 128, cachelines: 2, members: 17 */ /* sum members: 121, holes: 1, sum holes: 7 */ }; Now all entries in the first cacheline are read only throughout the life time of the map, set up once during map creation. Overall struct size and number of cachelines doesn't change from the reordering. struct bpf_map is usually first member and embedded in map structs in specific map implementations, so also avoid those members to sit at the end where it could potentially share the cacheline with first map values e.g. in the array since remote CPUs could trigger map updates just as well for those (easily dirtying members like max_entries intentionally as well) while having subsequent values in cache. Quoting from Google's Project Zero blog [1]: Additionally, at least on the Intel machine on which this was tested, bouncing modified cache lines between cores is slow, apparently because the MESI protocol is used for cache coherence [8]. Changing the reference counter of an eBPF array on one physical CPU core causes the cache line containing the reference counter to be bounced over to that CPU core, making reads of the reference counter on all other CPU cores slow until the changed reference counter has been written back to memory. Because the length and the reference counter of an eBPF array are stored in the same cache line, this also means that changing the reference counter on one physical CPU core causes reads of the eBPF array's length to be slow on other physical CPU cores (intentional false sharing). While this doesn't 'control' the out-of-bounds speculation through masking the index as in commit b2157399cc98, triggering a manipulation of the map's reference counter is really trivial, so lets not allow to easily affect max_entries from it. Splitting to separate cachelines also generally makes sense from a performance perspective anyway in that fast-path won't have a cache miss if the map gets pinned, reused in other progs, etc out of control path, thus also avoids unintentional false sharing. [1] https://googleprojectzero.blogspot.ch/2018/01/reading-privileged-memory-with-side.html Signed-off-by: Daniel Borkmann Signed-off-by: Alexei Starovoitov --- include/linux/bpf.h | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 1b985ca4ffbe..fe2cb7c398e3 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -43,7 +43,14 @@ struct bpf_map_ops { }; struct bpf_map { - atomic_t refcnt; + /* 1st cacheline with read-mostly members of which some + * are also accessed in fast-path (e.g. ops, max_entries). + */ + const struct bpf_map_ops *ops ____cacheline_aligned; + struct bpf_map *inner_map_meta; +#ifdef CONFIG_SECURITY + void *security; +#endif enum bpf_map_type map_type; u32 key_size; u32 value_size; @@ -53,15 +60,16 @@ struct bpf_map { u32 id; int numa_node; bool unpriv_array; - struct user_struct *user; - const struct bpf_map_ops *ops; - struct work_struct work; + /* 7 bytes hole */ + + /* 2nd cacheline with misc members to avoid false sharing + * particularly with refcounting. + */ + struct user_struct *user ____cacheline_aligned; + atomic_t refcnt; atomic_t usercnt; - struct bpf_map *inner_map_meta; + struct work_struct work; char name[BPF_OBJ_NAME_LEN]; -#ifdef CONFIG_SECURITY - void *security; -#endif }; /* function argument constraints */ From 290af86629b25ffd1ed6232c4e9107da031705cb Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Tue, 9 Jan 2018 10:04:29 -0800 Subject: [PATCH 14/27] bpf: introduce BPF_JIT_ALWAYS_ON config The BPF interpreter has been used as part of the spectre 2 attack CVE-2017-5715. A quote from goolge project zero blog: "At this point, it would normally be necessary to locate gadgets in the host kernel code that can be used to actually leak data by reading from an attacker-controlled location, shifting and masking the result appropriately and then using the result of that as offset to an attacker-controlled address for a load. But piecing gadgets together and figuring out which ones work in a speculation context seems annoying. So instead, we decided to use the eBPF interpreter, which is built into the host kernel - while there is no legitimate way to invoke it from inside a VM, the presence of the code in the host kernel's text section is sufficient to make it usable for the attack, just like with ordinary ROP gadgets." To make attacker job harder introduce BPF_JIT_ALWAYS_ON config option that removes interpreter from the kernel in favor of JIT-only mode. So far eBPF JIT is supported by: x64, arm64, arm32, sparc64, s390, powerpc64, mips64 The start of JITed program is randomized and code page is marked as read-only. In addition "constant blinding" can be turned on with net.core.bpf_jit_harden v2->v3: - move __bpf_prog_ret0 under ifdef (Daniel) v1->v2: - fix init order, test_bpf and cBPF (Daniel's feedback) - fix offloaded bpf (Jakub's feedback) - add 'return 0' dummy in case something can invoke prog->bpf_func - retarget bpf tree. For bpf-next the patch would need one extra hunk. It will be sent when the trees are merged back to net-next Considered doing: int bpf_jit_enable __read_mostly = BPF_EBPF_JIT_DEFAULT; but it seems better to land the patch as-is and in bpf-next remove bpf_jit_enable global variable from all JITs, consolidate in one place and remove this jit_init() function. Signed-off-by: Alexei Starovoitov Signed-off-by: Daniel Borkmann --- init/Kconfig | 7 +++++++ kernel/bpf/core.c | 19 +++++++++++++++++++ lib/test_bpf.c | 11 +++++++---- net/core/filter.c | 6 ++---- net/core/sysctl_net_core.c | 6 ++++++ net/socket.c | 9 +++++++++ 6 files changed, 50 insertions(+), 8 deletions(-) diff --git a/init/Kconfig b/init/Kconfig index 2934249fba46..5e2a4a391ba9 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -1392,6 +1392,13 @@ config BPF_SYSCALL Enable the bpf() system call that allows to manipulate eBPF programs and maps via file descriptors. +config BPF_JIT_ALWAYS_ON + bool "Permanently enable BPF JIT and remove BPF interpreter" + depends on BPF_SYSCALL && HAVE_EBPF_JIT && BPF_JIT + help + Enables BPF JIT and removes BPF interpreter to avoid + speculative execution of BPF instructions by the interpreter + config USERFAULTFD bool "Enable userfaultfd() system call" select ANON_INODES diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index 86b50aa26ee8..51ec2dda7f08 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -767,6 +767,7 @@ noinline u64 __bpf_call_base(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) } EXPORT_SYMBOL_GPL(__bpf_call_base); +#ifndef CONFIG_BPF_JIT_ALWAYS_ON /** * __bpf_prog_run - run eBPF program on a given context * @ctx: is the data we are operating on @@ -1317,6 +1318,14 @@ EVAL6(PROG_NAME_LIST, 224, 256, 288, 320, 352, 384) EVAL4(PROG_NAME_LIST, 416, 448, 480, 512) }; +#else +static unsigned int __bpf_prog_ret0(const void *ctx, + const struct bpf_insn *insn) +{ + return 0; +} +#endif + bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp) { @@ -1364,9 +1373,13 @@ static int bpf_check_tail_call(const struct bpf_prog *fp) */ struct bpf_prog *bpf_prog_select_runtime(struct bpf_prog *fp, int *err) { +#ifndef CONFIG_BPF_JIT_ALWAYS_ON u32 stack_depth = max_t(u32, fp->aux->stack_depth, 1); fp->bpf_func = interpreters[(round_up(stack_depth, 32) / 32) - 1]; +#else + fp->bpf_func = __bpf_prog_ret0; +#endif /* eBPF JITs can rewrite the program in case constant * blinding is active. However, in case of error during @@ -1376,6 +1389,12 @@ struct bpf_prog *bpf_prog_select_runtime(struct bpf_prog *fp, int *err) */ if (!bpf_prog_is_dev_bound(fp->aux)) { fp = bpf_int_jit_compile(fp); +#ifdef CONFIG_BPF_JIT_ALWAYS_ON + if (!fp->jited) { + *err = -ENOTSUPP; + return fp; + } +#endif } else { *err = bpf_prog_offload_compile(fp); if (*err) diff --git a/lib/test_bpf.c b/lib/test_bpf.c index 9e9748089270..f369889e521d 100644 --- a/lib/test_bpf.c +++ b/lib/test_bpf.c @@ -6250,9 +6250,8 @@ static struct bpf_prog *generate_filter(int which, int *err) return NULL; } } - /* We don't expect to fail. */ if (*err) { - pr_cont("FAIL to attach err=%d len=%d\n", + pr_cont("FAIL to prog_create err=%d len=%d\n", *err, fprog.len); return NULL; } @@ -6276,6 +6275,10 @@ static struct bpf_prog *generate_filter(int which, int *err) * checks. */ fp = bpf_prog_select_runtime(fp, err); + if (*err) { + pr_cont("FAIL to select_runtime err=%d\n", *err); + return NULL; + } break; } @@ -6461,8 +6464,8 @@ static __init int test_bpf(void) pass_cnt++; continue; } - - return err; + err_cnt++; + continue; } pr_cont("jited:%u ", fp->jited); diff --git a/net/core/filter.c b/net/core/filter.c index 6a85e67fafce..d339ef170df6 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1054,11 +1054,9 @@ static struct bpf_prog *bpf_migrate_filter(struct bpf_prog *fp) */ goto out_err_free; - /* We are guaranteed to never error here with cBPF to eBPF - * transitions, since there's no issue with type compatibility - * checks on program arrays. - */ fp = bpf_prog_select_runtime(fp, &err); + if (err) + goto out_err_free; kfree(old_prog); return fp; diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index cbc3dde4cfcc..a47ad6cd41c0 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -325,7 +325,13 @@ static struct ctl_table net_core_table[] = { .data = &bpf_jit_enable, .maxlen = sizeof(int), .mode = 0644, +#ifndef CONFIG_BPF_JIT_ALWAYS_ON .proc_handler = proc_dointvec +#else + .proc_handler = proc_dointvec_minmax, + .extra1 = &one, + .extra2 = &one, +#endif }, # ifdef CONFIG_HAVE_EBPF_JIT { diff --git a/net/socket.c b/net/socket.c index 05f361faec45..78acd6ce74c7 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2619,6 +2619,15 @@ out_fs: core_initcall(sock_init); /* early initcall */ +static int __init jit_init(void) +{ +#ifdef CONFIG_BPF_JIT_ALWAYS_ON + bpf_jit_enable = 1; +#endif + return 0; +} +pure_initcall(jit_init); + #ifdef CONFIG_PROC_FS void socket_seq_show(struct seq_file *seq) { From 2e83acb970684008baee471427270c029a76ddbd Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Mon, 8 Jan 2018 19:02:27 -0200 Subject: [PATCH 15/27] sctp: GFP_ATOMIC is not needed in sctp_setsockopt_events So replace it with GFP_USER and also add __GFP_NOWARN. Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- net/sctp/socket.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index b4fb6e4886d2..54c046783a89 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -2277,7 +2277,7 @@ static int sctp_setsockopt_events(struct sock *sk, char __user *optval, if (asoc && sctp_outq_is_empty(&asoc->outqueue)) { event = sctp_ulpevent_make_sender_dry_event(asoc, - GFP_ATOMIC); + GFP_USER | __GFP_NOWARN); if (!event) return -ENOMEM; From 5960cefab9df76600a1a7d4ff592c59e14616e88 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Mon, 8 Jan 2018 19:02:28 -0200 Subject: [PATCH 16/27] sctp: add a ceiling to optlen in some sockopts Hangbin Liu reported that some sockopt calls could cause the kernel to log a warning on memory allocation failure if the user supplied a large optlen value. That is because some of them called memdup_user() without a ceiling on optlen, allowing it to try to allocate really large buffers. This patch adds a ceiling by limiting optlen to the maximum allowed that would still make sense for these sockopt. Reported-by: Hangbin Liu Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- net/sctp/socket.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 54c046783a89..022b94f11fd8 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3498,6 +3498,8 @@ static int sctp_setsockopt_hmac_ident(struct sock *sk, if (optlen < sizeof(struct sctp_hmacalgo)) return -EINVAL; + optlen = min_t(unsigned int, optlen, sizeof(struct sctp_hmacalgo) + + SCTP_AUTH_NUM_HMACS * sizeof(u16)); hmacs = memdup_user(optval, optlen); if (IS_ERR(hmacs)) @@ -3536,6 +3538,11 @@ static int sctp_setsockopt_auth_key(struct sock *sk, if (optlen <= sizeof(struct sctp_authkey)) return -EINVAL; + /* authkey->sca_keylength is u16, so optlen can't be bigger than + * this. + */ + optlen = min_t(unsigned int, optlen, USHRT_MAX + + sizeof(struct sctp_authkey)); authkey = memdup_user(optval, optlen); if (IS_ERR(authkey)) @@ -3893,6 +3900,9 @@ static int sctp_setsockopt_reset_streams(struct sock *sk, if (optlen < sizeof(*params)) return -EINVAL; + /* srs_number_streams is u16, so optlen can't be bigger than this. */ + optlen = min_t(unsigned int, optlen, USHRT_MAX + + sizeof(__u16) * sizeof(*params)); params = memdup_user(optval, optlen); if (IS_ERR(params)) From c76f97c99ae6d26d14c7f0e50e074382bfbc9f98 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Mon, 8 Jan 2018 19:02:29 -0200 Subject: [PATCH 17/27] sctp: make use of pre-calculated len Some sockopt handling functions were calculating the length of the buffer to be written to userspace and then calculating it again when actually writing the buffer, which could lead to some write not using an up-to-date length. This patch updates such places to just make use of the len variable. Also, replace some sizeof(type) to sizeof(var). Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- net/sctp/socket.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 022b94f11fd8..9b01e994f661 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -5025,7 +5025,7 @@ static int sctp_getsockopt_autoclose(struct sock *sk, int len, char __user *optv len = sizeof(int); if (put_user(len, optlen)) return -EFAULT; - if (copy_to_user(optval, &sctp_sk(sk)->autoclose, sizeof(int))) + if (copy_to_user(optval, &sctp_sk(sk)->autoclose, len)) return -EFAULT; return 0; } @@ -5655,6 +5655,9 @@ copy_getaddrs: err = -EFAULT; goto out; } + /* XXX: We should have accounted for sizeof(struct sctp_getaddrs) too, + * but we can't change it anymore. + */ if (put_user(bytes_copied, optlen)) err = -EFAULT; out: @@ -6091,7 +6094,7 @@ static int sctp_getsockopt_maxseg(struct sock *sk, int len, params.assoc_id = 0; } else if (len >= sizeof(struct sctp_assoc_value)) { len = sizeof(struct sctp_assoc_value); - if (copy_from_user(¶ms, optval, sizeof(params))) + if (copy_from_user(¶ms, optval, len)) return -EFAULT; } else return -EINVAL; @@ -6261,7 +6264,9 @@ static int sctp_getsockopt_active_key(struct sock *sk, int len, if (len < sizeof(struct sctp_authkeyid)) return -EINVAL; - if (copy_from_user(&val, optval, sizeof(struct sctp_authkeyid))) + + len = sizeof(struct sctp_authkeyid); + if (copy_from_user(&val, optval, len)) return -EFAULT; asoc = sctp_id2assoc(sk, val.scact_assoc_id); @@ -6273,7 +6278,6 @@ static int sctp_getsockopt_active_key(struct sock *sk, int len, else val.scact_keynumber = ep->active_key_id; - len = sizeof(struct sctp_authkeyid); if (put_user(len, optlen)) return -EFAULT; if (copy_to_user(optval, &val, len)) @@ -6299,7 +6303,7 @@ static int sctp_getsockopt_peer_auth_chunks(struct sock *sk, int len, if (len < sizeof(struct sctp_authchunks)) return -EINVAL; - if (copy_from_user(&val, optval, sizeof(struct sctp_authchunks))) + if (copy_from_user(&val, optval, sizeof(val))) return -EFAULT; to = p->gauth_chunks; @@ -6344,7 +6348,7 @@ static int sctp_getsockopt_local_auth_chunks(struct sock *sk, int len, if (len < sizeof(struct sctp_authchunks)) return -EINVAL; - if (copy_from_user(&val, optval, sizeof(struct sctp_authchunks))) + if (copy_from_user(&val, optval, sizeof(val))) return -EFAULT; to = p->gauth_chunks; From 11d827a993a969c3c6ec56758ff63a44ba19b466 Mon Sep 17 00:00:00 2001 From: Yangbo Lu Date: Tue, 9 Jan 2018 11:02:33 +0800 Subject: [PATCH 18/27] net: gianfar_ptp: move set_fipers() to spinlock protecting area set_fipers() calling should be protected by spinlock in case that any interrupt breaks related registers setting and the function we expect. This patch is to move set_fipers() to spinlock protecting area in ptp_gianfar_adjtime(). Signed-off-by: Yangbo Lu Acked-by: Richard Cochran Reviewed-by: Fabio Estevam Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/gianfar_ptp.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/ethernet/freescale/gianfar_ptp.c b/drivers/net/ethernet/freescale/gianfar_ptp.c index 544114281ea7..9f8d4f8e57e3 100644 --- a/drivers/net/ethernet/freescale/gianfar_ptp.c +++ b/drivers/net/ethernet/freescale/gianfar_ptp.c @@ -319,11 +319,10 @@ static int ptp_gianfar_adjtime(struct ptp_clock_info *ptp, s64 delta) now = tmr_cnt_read(etsects); now += delta; tmr_cnt_write(etsects, now); + set_fipers(etsects); spin_unlock_irqrestore(&etsects->lock, flags); - set_fipers(etsects); - return 0; } From af60d61fa846725566f4a876ae04f891bdff1c7a Mon Sep 17 00:00:00 2001 From: Kornilios Kourtis Date: Tue, 9 Jan 2018 09:52:22 +0100 Subject: [PATCH 19/27] doc: clarification about setting SO_ZEROCOPY Signed-off-by: Kornilios Kourtis Acked-by: Willem de Bruijn Signed-off-by: David S. Miller --- Documentation/networking/msg_zerocopy.rst | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/Documentation/networking/msg_zerocopy.rst b/Documentation/networking/msg_zerocopy.rst index 77f6d7e25cfd..291a01264967 100644 --- a/Documentation/networking/msg_zerocopy.rst +++ b/Documentation/networking/msg_zerocopy.rst @@ -72,6 +72,10 @@ this flag, a process must first signal intent by setting a socket option: if (setsockopt(fd, SOL_SOCKET, SO_ZEROCOPY, &one, sizeof(one))) error(1, errno, "setsockopt zerocopy"); +Setting the socket option only works when the socket is in its initial +(TCP_CLOSED) state. Trying to set the option for a socket returned by accept(), +for example, will lead to an EBUSY error. In this case, the option should be set +to the listening socket and it will be inherited by the accepted sockets. Transmission ------------ From b0d55b5bc77755501be9de2c935d106ff8dba9ac Mon Sep 17 00:00:00 2001 From: Xiongfeng Wang Date: Tue, 9 Jan 2018 19:58:18 +0800 Subject: [PATCH 20/27] caif_usb: use strlcpy() instead of strncpy() gcc-8 reports net/caif/caif_usb.c: In function 'cfusbl_device_notify': ./include/linux/string.h:245:9: warning: '__builtin_strncpy' output may be truncated copying 15 bytes from a string of length 15 [-Wstringop-truncation] The compiler require that the input param 'len' of strncpy() should be greater than the length of the src string, so that '\0' is copied as well. We can just use strlcpy() to avoid this warning. Signed-off-by: Xiongfeng Wang Signed-off-by: David S. Miller --- net/caif/caif_usb.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/net/caif/caif_usb.c b/net/caif/caif_usb.c index 5cd44f001f64..1a082a946045 100644 --- a/net/caif/caif_usb.c +++ b/net/caif/caif_usb.c @@ -176,9 +176,7 @@ static int cfusbl_device_notify(struct notifier_block *me, unsigned long what, dev_add_pack(&caif_usb_type); pack_added = true; - strncpy(layer->name, dev->name, - sizeof(layer->name) - 1); - layer->name[sizeof(layer->name) - 1] = 0; + strlcpy(layer->name, dev->name, sizeof(layer->name)); return 0; } From 95f566de0269a0c59fd6a737a147731302136429 Mon Sep 17 00:00:00 2001 From: Madalin Bucur Date: Tue, 9 Jan 2018 14:43:34 +0200 Subject: [PATCH 21/27] of_mdio: avoid MDIO bus removal when a PHY is missing If one of the child devices is missing the of_mdiobus_register_phy() call will return -ENODEV. When a missing device is encountered the registration of the remaining PHYs is stopped and the MDIO bus will fail to register. Propagate all errors except ENODEV to avoid it. Signed-off-by: Madalin Bucur Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/of/of_mdio.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index 3481e69738b5..a327be1d264b 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -231,7 +231,12 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np) rc = of_mdiobus_register_phy(mdio, child, addr); else rc = of_mdiobus_register_device(mdio, child, addr); - if (rc) + + if (rc == -ENODEV) + dev_err(&mdio->dev, + "MDIO device at address %d is missing.\n", + addr); + else if (rc) goto unregister; } @@ -255,7 +260,7 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np) if (of_mdiobus_child_is_phy(child)) { rc = of_mdiobus_register_phy(mdio, child, addr); - if (rc) + if (rc && rc != -ENODEV) goto unregister; } } From 78bbb15f2239bc8e663aa20bbe1987c91a0b75f6 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Tue, 9 Jan 2018 13:40:41 -0800 Subject: [PATCH 22/27] 8021q: fix a memory leak for VLAN 0 device A vlan device with vid 0 is allow to creat by not able to be fully cleaned up by unregister_vlan_dev() which checks for vlan_id!=0. Also, VLAN 0 is probably not a valid number and it is kinda "reserved" for HW accelerating devices, but it is probably too late to reject it from creation even if makes sense. Instead, just remove the check in unregister_vlan_dev(). Reported-by: Dmitry Vyukov Fixes: ad1afb003939 ("vlan_dev: VLAN 0 should be treated as "no vlan tag" (802.1p packet)") Cc: Vlad Yasevich Cc: Ben Hutchings Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- net/8021q/vlan.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c index 8dfdd94e430f..bad01b14a4ad 100644 --- a/net/8021q/vlan.c +++ b/net/8021q/vlan.c @@ -111,12 +111,7 @@ void unregister_vlan_dev(struct net_device *dev, struct list_head *head) vlan_gvrp_uninit_applicant(real_dev); } - /* Take it out of our own structures, but be sure to interlock with - * HW accelerating devices or SW vlan input packet processing if - * VLAN is not 0 (leave it there for 802.1p). - */ - if (vlan_id) - vlan_vid_del(real_dev, vlan->vlan_proto, vlan_id); + vlan_vid_del(real_dev, vlan->vlan_proto, vlan_id); /* Get rid of the vlan's reference to real_dev */ dev_put(real_dev); From fc2336505fb49a8b932a0a67a9745c408b79992c Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 9 Jan 2018 18:14:28 -0800 Subject: [PATCH 23/27] nfp: always unmask aux interrupts at init The link state and exception interrupts may be masked when we probe. The firmware should in theory prevent sending (and automasking) those interrupts if the device is disabled, but if my reading of the FW code is correct there are firmwares out there with race conditions in this area. The interrupt may also be masked if previous driver which used the device was malfunctioning and we didn't load the FW (there is no other good way to comprehensively reset the PF). Note that FW unmasks the data interrupts by itself when vNIC is enabled, such helpful operation is not performed for LSC/EXN interrupts. Always unmask the auxiliary interrupts after request_irq(). On the remove path add missing PCI write flush before free_irq(). Fixes: 4c3523623dc0 ("net: add driver for Netronome NFP4000/NFP6000 NIC VFs") Signed-off-by: Jakub Kicinski Reviewed-by: Dirk van der Merwe Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index 1a603fdd9e80..99b0487b6d82 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -568,6 +568,7 @@ nfp_net_aux_irq_request(struct nfp_net *nn, u32 ctrl_offset, return err; } nn_writeb(nn, ctrl_offset, entry->entry); + nfp_net_irq_unmask(nn, entry->entry); return 0; } @@ -582,6 +583,7 @@ static void nfp_net_aux_irq_free(struct nfp_net *nn, u32 ctrl_offset, unsigned int vector_idx) { nn_writeb(nn, ctrl_offset, 0xff); + nn_pci_flush(nn); free_irq(nn->irq_entries[vector_idx].vector, nn); } From 8e033a93b37f37aa9fca71a370a895155320af60 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Wed, 10 Jan 2018 11:42:43 +0100 Subject: [PATCH 24/27] mlxsw: pci: Wait after reset before accessing HW After performing reset driver polls on HW indication until learning that the reset is done, but immediately after reset the device becomes unresponsive which might lead to completion timeout on the first read. Wait for 100ms before starting the polling. Fixes: 233fa44bd67a ("mlxsw: pci: Implement reset done check") Signed-off-by: Yuval Mintz Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/pci.c | 7 ++++++- drivers/net/ethernet/mellanox/mlxsw/pci_hw.h | 1 + 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c index 23f7d828cf67..6ef20e5cc77d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci.c +++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c @@ -1643,7 +1643,12 @@ static int mlxsw_pci_sw_reset(struct mlxsw_pci *mlxsw_pci, return 0; } - wmb(); /* reset needs to be written before we read control register */ + /* Reset needs to be written before we read control register, and + * we must wait for the HW to become responsive once again + */ + wmb(); + msleep(MLXSW_PCI_SW_RESET_WAIT_MSECS); + end = jiffies + msecs_to_jiffies(MLXSW_PCI_SW_RESET_TIMEOUT_MSECS); do { u32 val = mlxsw_pci_read32(mlxsw_pci, FW_READY); diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h b/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h index a6441208e9d9..fb082ad21b00 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h +++ b/drivers/net/ethernet/mellanox/mlxsw/pci_hw.h @@ -59,6 +59,7 @@ #define MLXSW_PCI_SW_RESET 0xF0010 #define MLXSW_PCI_SW_RESET_RST_BIT BIT(0) #define MLXSW_PCI_SW_RESET_TIMEOUT_MSECS 5000 +#define MLXSW_PCI_SW_RESET_WAIT_MSECS 100 #define MLXSW_PCI_FW_READY 0xA1844 #define MLXSW_PCI_FW_READY_MASK 0xFFFF #define MLXSW_PCI_FW_READY_MAGIC 0x5E From db84924c4fc3be1ef0c965d5ece5f6d785c77c5f Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Wed, 10 Jan 2018 11:42:44 +0100 Subject: [PATCH 25/27] mlxsw: spectrum_qdisc: Don't use variable array in mlxsw_sp_tclass_congestion_enable Resolve the sparse warning: "sparse: Variable length array is used." Use 2 arrays for 2 PRM register accesses. Fixes: 96f17e0776c2 ("mlxsw: spectrum: Support RED qdisc offload") Signed-off-by: Jiri Pirko Reviewed-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c index c33beac5def0..b5397da94d7f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c @@ -46,7 +46,8 @@ mlxsw_sp_tclass_congestion_enable(struct mlxsw_sp_port *mlxsw_sp_port, int tclass_num, u32 min, u32 max, u32 probability, bool is_ecn) { - char cwtp_cmd[max_t(u8, MLXSW_REG_CWTP_LEN, MLXSW_REG_CWTPM_LEN)]; + char cwtpm_cmd[MLXSW_REG_CWTPM_LEN]; + char cwtp_cmd[MLXSW_REG_CWTP_LEN]; struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; int err; @@ -60,10 +61,10 @@ mlxsw_sp_tclass_congestion_enable(struct mlxsw_sp_port *mlxsw_sp_port, if (err) return err; - mlxsw_reg_cwtpm_pack(cwtp_cmd, mlxsw_sp_port->local_port, tclass_num, + mlxsw_reg_cwtpm_pack(cwtpm_cmd, mlxsw_sp_port->local_port, tclass_num, MLXSW_REG_CWTP_DEFAULT_PROFILE, true, is_ecn); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(cwtpm), cwtp_cmd); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(cwtpm), cwtpm_cmd); } static int From 862c03ee1deb7e19e0f9931682e0294ecd1fcaf9 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 10 Jan 2018 03:45:49 -0800 Subject: [PATCH 26/27] ipv6: fix possible mem leaks in ipv6_make_skb() ip6_setup_cork() might return an error, while memory allocations have been done and must be rolled back. Fixes: 6422398c2ab0 ("ipv6: introduce ipv6_make_skb") Signed-off-by: Eric Dumazet Cc: Vlad Yasevich Reported-by: Mike Maloney Acked-by: Mike Maloney Signed-off-by: David S. Miller --- net/ipv6/ip6_output.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index f7dd51c42314..688ba5f7516b 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -1735,9 +1735,10 @@ struct sk_buff *ip6_make_skb(struct sock *sk, cork.base.opt = NULL; v6_cork.opt = NULL; err = ip6_setup_cork(sk, &cork, &v6_cork, ipc6, rt, fl6); - if (err) + if (err) { + ip6_cork_release(&cork, &v6_cork); return ERR_PTR(err); - + } if (ipc6->dontfrag < 0) ipc6->dontfrag = inet6_sk(sk)->dontfrag; From ccc12b11c5332c84442ef120dcd631523be75089 Mon Sep 17 00:00:00 2001 From: Mathieu Xhonneux Date: Wed, 10 Jan 2018 13:35:49 +0000 Subject: [PATCH 27/27] ipv6: sr: fix TLVs not being copied using setsockopt Function ipv6_push_rthdr4 allows to add an IPv6 Segment Routing Header to a socket through setsockopt, but the current implementation doesn't copy possible TLVs at the end of the SRH received from userspace. Therefore, the execution of the following branch if (sr_has_hmac(sr_phdr)) { ... } will never complete since the len and type fields of a possible HMAC TLV are not copied, hence seg6_get_tlv_hmac will return an error, and the HMAC will not be computed. This commit adds a memcpy in case TLVs have been appended to the SRH. Fixes: a149e7c7ce81 ("ipv6: sr: add support for SRH injection through setsockopt") Acked-by: David Lebrun Signed-off-by: Mathieu Xhonneux Signed-off-by: David S. Miller --- net/ipv6/exthdrs.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 83bd75713535..bc68eb661970 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -925,6 +925,15 @@ static void ipv6_push_rthdr4(struct sk_buff *skb, u8 *proto, sr_phdr->segments[0] = **addr_p; *addr_p = &sr_ihdr->segments[sr_ihdr->segments_left]; + if (sr_ihdr->hdrlen > hops * 2) { + int tlvs_offset, tlvs_length; + + tlvs_offset = (1 + hops * 2) << 3; + tlvs_length = (sr_ihdr->hdrlen - hops * 2) << 3; + memcpy((char *)sr_phdr + tlvs_offset, + (char *)sr_ihdr + tlvs_offset, tlvs_length); + } + #ifdef CONFIG_IPV6_SEG6_HMAC if (sr_has_hmac(sr_phdr)) { struct net *net = NULL;