From efba01803c8570bab11d0d6188a630231d0ddccf Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 3 Dec 2008 00:36:15 -0800 Subject: bnx2: Add workaround to handle missed MSI. The bnx2 chips do not support per MSI vector masking. On 5706/5708, new MSI address/data are stored only when the MSI enable bit is toggled. As a result, SMP affinity no longer works in the latest kernel. A more serious problem is that the driver will no longer receive interrupts when the MSI receiving CPU goes offline. The workaround in this patch only addresses the problem of CPU going offline. When that happens, the driver's timer function will detect that it is making no forward progress on pending interrupt events and will recover from it. Eric Dumazet reported the problem. We also found that if an interrupt is internally asserted while MSI and INTA are disabled, the chip will end up in the same state after MSI is re-enabled. The same workaround is needed for this problem. Signed-off-by: Michael Chan Tested-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 35 ++++++++++++++++++++++++++++++++--- drivers/net/bnx2.h | 6 ++++++ 2 files changed, 38 insertions(+), 3 deletions(-) (limited to 'drivers') diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index d07e3f148951..a1a3d0e5d2b4 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -3144,6 +3144,28 @@ bnx2_has_work(struct bnx2_napi *bnapi) return 0; } +static void +bnx2_chk_missed_msi(struct bnx2 *bp) +{ + struct bnx2_napi *bnapi = &bp->bnx2_napi[0]; + u32 msi_ctrl; + + if (bnx2_has_work(bnapi)) { + msi_ctrl = REG_RD(bp, BNX2_PCICFG_MSI_CONTROL); + if (!(msi_ctrl & BNX2_PCICFG_MSI_CONTROL_ENABLE)) + return; + + if (bnapi->last_status_idx == bp->idle_chk_status_idx) { + REG_WR(bp, BNX2_PCICFG_MSI_CONTROL, msi_ctrl & + ~BNX2_PCICFG_MSI_CONTROL_ENABLE); + REG_WR(bp, BNX2_PCICFG_MSI_CONTROL, msi_ctrl); + bnx2_msi(bp->irq_tbl[0].vector, bnapi); + } + } + + bp->idle_chk_status_idx = bnapi->last_status_idx; +} + static void bnx2_poll_link(struct bnx2 *bp, struct bnx2_napi *bnapi) { struct status_block *sblk = bnapi->status_blk.msi; @@ -3218,14 +3240,15 @@ static int bnx2_poll(struct napi_struct *napi, int budget) work_done = bnx2_poll_work(bp, bnapi, work_done, budget); - if (unlikely(work_done >= budget)) - break; - /* bnapi->last_status_idx is used below to tell the hw how * much work has been processed, so we must read it before * checking for more work. */ bnapi->last_status_idx = sblk->status_idx; + + if (unlikely(work_done >= budget)) + break; + rmb(); if (likely(!bnx2_has_work(bnapi))) { netif_rx_complete(bp->dev, napi); @@ -4570,6 +4593,8 @@ bnx2_init_chip(struct bnx2 *bp) for (i = 0; i < BNX2_MAX_MSIX_VEC; i++) bp->bnx2_napi[i].last_status_idx = 0; + bp->idle_chk_status_idx = 0xffff; + bp->rx_mode = BNX2_EMAC_RX_MODE_SORT_MODE; /* Set up how to generate a link change interrupt. */ @@ -5718,6 +5743,10 @@ bnx2_timer(unsigned long data) if (atomic_read(&bp->intr_sem) != 0) goto bnx2_restart_timer; + if ((bp->flags & (BNX2_FLAG_USING_MSI | BNX2_FLAG_ONE_SHOT_MSI)) == + BNX2_FLAG_USING_MSI) + bnx2_chk_missed_msi(bp); + bnx2_send_heart_beat(bp); bp->stats_blk->stat_FwRxDrop = diff --git a/drivers/net/bnx2.h b/drivers/net/bnx2.h index 617d95340160..0b032c3c7b61 100644 --- a/drivers/net/bnx2.h +++ b/drivers/net/bnx2.h @@ -378,6 +378,9 @@ struct l2_fhdr { * pci_config_l definition * offset: 0000 */ +#define BNX2_PCICFG_MSI_CONTROL 0x00000058 +#define BNX2_PCICFG_MSI_CONTROL_ENABLE (1L<<16) + #define BNX2_PCICFG_MISC_CONFIG 0x00000068 #define BNX2_PCICFG_MISC_CONFIG_TARGET_BYTE_SWAP (1L<<2) #define BNX2_PCICFG_MISC_CONFIG_TARGET_MB_WORD_SWAP (1L<<3) @@ -6863,6 +6866,9 @@ struct bnx2 { u8 num_tx_rings; u8 num_rx_rings; + + u32 idle_chk_status_idx; + }; #define REG_RD(bp, offset) \ -- cgit v1.2.1 From 1c594c05a75770ab53a329fc4eb99c797a4bc7d7 Mon Sep 17 00:00:00 2001 From: Wilfried Klaebe Date: Wed, 3 Dec 2008 20:57:19 -0800 Subject: b1isa: fix b1isa_exit() to really remove registered capi controllers On "/etc/init.d/capiutils stop", this oops happened. The oops happens on reading /proc/capi/controllers because capi_ctrl->procinfo is called for the wrongly not unregistered controller, which points to b1isa_procinfo(), which was removed on module unload. b1isa_exit() did not call b1isa_remove() for its controllers because io[0] == 0 on module unload despite having been 0x340 on module load. Besides, just removing the controllers that where added on module load time and not those that were added later via b1isa_add_card() is wrong too - the place where all added cards are found is isa_dev[]. relevant dmesg lines: [ 0.000000] Linux version 2.6.27.4 (w@shubashi) (gcc version 4.3.2 (Debian 4.3.2-1) ) #3 Thu Oct 30 16:49:03 CET 2008 [ 67.403555] CAPI Subsystem Rev 1.1.2.8 [ 68.529154] capifs: Rev 1.1.2.3 [ 68.563292] capi20: Rev 1.1.2.7: started up with major 68 (middleware+capifs) [ 77.026936] b1: revision 1.1.2.2 [ 77.049992] b1isa: revision 1.1.2.3 [ 77.722655] kcapi: Controller [001]: b1isa-340 attached [ 77.722671] b1isa: AVM B1 ISA at i/o 0x340, irq 5, revision 255 [ 81.272669] b1isa-340: card 1 "B1" ready. [ 81.272683] b1isa-340: card 1 Protocol: DSS1 [ 81.272689] b1isa-340: card 1 Linetype: point to multipoint [ 81.272695] b1isa-340: B1-card (3.11-03) now active [ 81.272702] kcapi: card [001] "b1isa-340" ready. [ 153.721281] kcapi: card [001] down. [ 154.151889] BUG: unable to handle kernel paging request at e87af000 [ 154.152081] IP: [] [ 154.153292] *pde = 2655b067 *pte = 00000000 [ 154.153307] Oops: 0000 [#1] [ 154.153360] Modules linked in: rfcomm l2cap ppdev lp ipt_MASQUERADE tun capi capifs kernelcapi ac battery nfsd exportfs nfs lockd nfs_acl sunrpc sit tunnel4 bridge stp llc ipt_REJECT ipt_LOG xt_tcpudp xt_state iptable_filter iptable_mangle iptable_nat nf_nat nf_conntrack_ipv4 nf_conntrack ip_tables x_tables nls_utf8 isofs nls_base zlib_inflate loop ipv6 netconsole snd_via82xx dvb_usb_dib0700 gameport dib7000p dib7000m dvb_usb snd_ac97_codec ac97_bus dvb_core mt2266 snd_pcm tuner_xc2028 dib3000mc dibx000_common mt2060 dib0070 snd_page_alloc snd_mpu401_uart snd_seq_midi snd_seq_midi_event btusb snd_rawmidi bluetooth snd_seq snd_timer snd_seq_device snd via686a i2c_viapro soundcore i2c_core parport_pc parport button dm_mirror dm_log dm_snapshot floppy sg ohci1394 uhci_hcd ehci_hcd 8139too mii ieee1394 usbcore sr_mod cdrom sd_mod thermal processor fan [last unloaded: b1] [ 154.153360] [ 154.153360] Pid: 4132, comm: capiinit Not tainted (2.6.27.4 #3) [ 154.153360] EIP: 0060:[] EFLAGS: 00010286 CPU: 0 [ 154.153360] EIP is at 0xe87af000 [ 154.153360] EAX: e6b9ccc8 EBX: e6b9ccc8 ECX: e87a0c67 EDX: e87af000 [ 154.153360] ESI: e142bbc0 EDI: e87a56e0 EBP: e0505f0c ESP: e0505ee4 [ 154.153360] DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 0068 [ 154.153360] Process capiinit (pid: 4132, ti=e0504000 task=d1196cf0 task.ti=e0504000) [ 154.153360] Stack: e879f650 00000246 e0505ef4 c01472eb e0505f0c 00000246 e7001780 fffffff4 [ 154.153360] fffffff4 e142bbc0 e0505f48 c01a56c6 00000400 b805e000 d102dc80 e142bbe0 [ 154.153360] 00000000 e87a56e0 00000246 e12617ac 00000000 00000000 e1261760 fffffffb [ 154.153360] Call Trace: [ 154.153360] [] ? controller_show+0x20/0x90 [kernelcapi] [ 154.153360] [] ? trace_hardirqs_on+0xb/0x10 [ 154.153360] [] ? seq_read+0x126/0x2f0 [ 154.153360] [] ? seq_read+0x0/0x2f0 [ 154.153360] [] ? proc_reg_read+0x5c/0x90 [ 154.153360] [] ? vfs_read+0x99/0x140 [ 154.153360] [] ? proc_reg_read+0x0/0x90 [ 154.153360] [] ? sys_read+0x3d/0x70 [ 154.153360] [] ? sysenter_do_call+0x12/0x35 [ 154.153360] ======================= [ 154.153360] Code: Bad EIP value. [ 154.153360] EIP: [] 0xe87af000 SS:ESP 0068:e0505ee4 [ 154.153360] ---[ end trace 23750b6c2862de94 ]--- Signed-off-by: Wilfried Klaebe Signed-off-by: Andrew Morton Acked-by: Karsten Keil Signed-off-by: David S. Miller --- drivers/isdn/hardware/avm/b1isa.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'drivers') diff --git a/drivers/isdn/hardware/avm/b1isa.c b/drivers/isdn/hardware/avm/b1isa.c index 1e288eeb5e2a..6461a32bc838 100644 --- a/drivers/isdn/hardware/avm/b1isa.c +++ b/drivers/isdn/hardware/avm/b1isa.c @@ -233,10 +233,8 @@ static void __exit b1isa_exit(void) int i; for (i = 0; i < MAX_CARDS; i++) { - if (!io[i]) - break; - - b1isa_remove(&isa_dev[i]); + if (isa_dev[i].resource[0].start) + b1isa_remove(&isa_dev[i]); } unregister_capi_driver(&capi_driver_b1isa); } -- cgit v1.2.1 From bd0914104c61a852baf469b2d807322e5d0459b4 Mon Sep 17 00:00:00 2001 From: Pascal Terjan Date: Mon, 1 Dec 2008 12:24:25 +0000 Subject: hysdn: fix writing outside the field on 64 bits ifa_local is assumed to be unsigned long which lead to writing the address at dev->dev_addr-2 instead of +2 noticed thanks to gcc: drivers/isdn/hysdn/hysdn_net.c: In function `net_open': drivers/isdn/hysdn/hysdn_net.c:91: warning: array subscript is below array bounds Signed-off-by: Pascal Terjan Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- drivers/isdn/hysdn/hysdn_net.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers') diff --git a/drivers/isdn/hysdn/hysdn_net.c b/drivers/isdn/hysdn/hysdn_net.c index cfa8fa5e44ab..3f2a0a20c19b 100644 --- a/drivers/isdn/hysdn/hysdn_net.c +++ b/drivers/isdn/hysdn/hysdn_net.c @@ -83,12 +83,12 @@ net_open(struct net_device *dev) /* Fill in the MAC-level header (if not already set) */ if (!card->mac_addr[0]) { - for (i = 0; i < ETH_ALEN - sizeof(unsigned long); i++) + for (i = 0; i < ETH_ALEN; i++) dev->dev_addr[i] = 0xfc; if ((in_dev = dev->ip_ptr) != NULL) { struct in_ifaddr *ifa = in_dev->ifa_list; if (ifa != NULL) - memcpy(dev->dev_addr + (ETH_ALEN - sizeof(unsigned long)), &ifa->ifa_local, sizeof(unsigned long)); + memcpy(dev->dev_addr + (ETH_ALEN - sizeof(ifa->ifa_local)), &ifa->ifa_local, sizeof(ifa->ifa_local)); } } else memcpy(dev->dev_addr, card->mac_addr, ETH_ALEN); -- cgit v1.2.1 From 5176da7e5318669220e4d2fa856223054a3efc9f Mon Sep 17 00:00:00 2001 From: Baruch Siach Date: Tue, 2 Dec 2008 05:07:01 +0000 Subject: enc28j60: Fix sporadic packet loss (corrected again) Packet data read from the RX buffer the when the RSV is at the end of the RX buffer does not warp around. This causes packet loss, as the actual data is never read. Fix this by calculating the right packet data location. Thanks to Shachar Shemesh for suggesting the fix. Signed-off-by: Baruch Siach Acked-by: Claudio Lanconelli Signed-off-by: David S. Miller --- drivers/net/enc28j60.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) (limited to 'drivers') diff --git a/drivers/net/enc28j60.c b/drivers/net/enc28j60.c index e1b441effbbe..c414554ac321 100644 --- a/drivers/net/enc28j60.c +++ b/drivers/net/enc28j60.c @@ -568,6 +568,17 @@ static u16 erxrdpt_workaround(u16 next_packet_ptr, u16 start, u16 end) return erxrdpt; } +/* + * Calculate wrap around when reading beyond the end of the RX buffer + */ +static u16 rx_packet_start(u16 ptr) +{ + if (ptr + RSV_SIZE > RXEND_INIT) + return (ptr + RSV_SIZE) - (RXEND_INIT - RXSTART_INIT + 1); + else + return ptr + RSV_SIZE; +} + static void nolock_rxfifo_init(struct enc28j60_net *priv, u16 start, u16 end) { u16 erxrdpt; @@ -938,8 +949,9 @@ static void enc28j60_hw_rx(struct net_device *ndev) skb->dev = ndev; skb_reserve(skb, NET_IP_ALIGN); /* copy the packet from the receive buffer */ - enc28j60_mem_read(priv, priv->next_pk_ptr + sizeof(rsv), - len, skb_put(skb, len)); + enc28j60_mem_read(priv, + rx_packet_start(priv->next_pk_ptr), + len, skb_put(skb, len)); if (netif_msg_pktdata(priv)) dump_packet(__func__, skb->len, skb->data); skb->protocol = eth_type_trans(skb, ndev); -- cgit v1.2.1 From 2cc002c4bbce4d918ab94b494d61c6991c907d5e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Wed, 3 Dec 2008 22:18:59 -0800 Subject: netx-eth: initialize per device spinlock MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The spinlock used in the netx-eth driver was never properly initialized. This was noticed using CONFIG_DEBUG_SPINLOCK=y Signed-off-by: Uwe Kleine-König Acked-by: Sascha Hauer Signed-off-by: David S. Miller --- drivers/net/netx-eth.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers') diff --git a/drivers/net/netx-eth.c b/drivers/net/netx-eth.c index b9bed82e1d21..b289a0a2b945 100644 --- a/drivers/net/netx-eth.c +++ b/drivers/net/netx-eth.c @@ -401,6 +401,8 @@ static int netx_eth_drv_probe(struct platform_device *pdev) priv->xmac_base = priv->xc->xmac_base; priv->sram_base = priv->xc->sram_base; + spin_lock_init(&priv->lock); + ret = pfifo_request(PFIFO_MASK(priv->id)); if (ret) { printk("unable to request PFIFO\n"); -- cgit v1.2.1 From 40a9a8299116297429298e8fcee08235134883f7 Mon Sep 17 00:00:00 2001 From: Tomas Winkler Date: Tue, 25 Nov 2008 23:29:03 +0200 Subject: iwlwifi: clean key table in iwl_clear_stations_table function This patch cleans uCode key table bit map iwl_clear_stations_table since all stations are cleared also the key table must be. Since the keys are not removed properly on suspend by mac80211 this may result in exhausting key table on resume leading to memory corruption during removal This patch also fixes a memory corruption problem reported in http://marc.info/?l=linux-wireless&m=122641417231586&w=2 and tracked in http://bugzilla.kernel.org/show_bug.cgi?id=12040. When the key is removed a second time the offset is set to 255 - this index is not valid for the ucode_key_table and corrupts the eeprom pointer (which is 255 bits from ucode_key_table). Signed-off-by: Tomas Winkler Signed-off-by: Zhu Yi Reported-by: Carlos R. Mafra Reported-by: Lukas Hejtmanek Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl-core.c | 3 +++ drivers/net/wireless/iwlwifi/iwl-sta.c | 24 +++++++++++++++++++++--- 2 files changed, 24 insertions(+), 3 deletions(-) (limited to 'drivers') diff --git a/drivers/net/wireless/iwlwifi/iwl-core.c b/drivers/net/wireless/iwlwifi/iwl-core.c index 4c312c55f90c..01a845851338 100644 --- a/drivers/net/wireless/iwlwifi/iwl-core.c +++ b/drivers/net/wireless/iwlwifi/iwl-core.c @@ -290,6 +290,9 @@ void iwl_clear_stations_table(struct iwl_priv *priv) priv->num_stations = 0; memset(priv->stations, 0, sizeof(priv->stations)); + /* clean ucode key table bit map */ + priv->ucode_key_table = 0; + spin_unlock_irqrestore(&priv->sta_lock, flags); } EXPORT_SYMBOL(iwl_clear_stations_table); diff --git a/drivers/net/wireless/iwlwifi/iwl-sta.c b/drivers/net/wireless/iwlwifi/iwl-sta.c index 61797f3f8d5c..26f7084d3011 100644 --- a/drivers/net/wireless/iwlwifi/iwl-sta.c +++ b/drivers/net/wireless/iwlwifi/iwl-sta.c @@ -475,7 +475,7 @@ static int iwl_get_free_ucode_key_index(struct iwl_priv *priv) if (!test_and_set_bit(i, &priv->ucode_key_table)) return i; - return -1; + return WEP_INVALID_OFFSET; } int iwl_send_static_wepkey_cmd(struct iwl_priv *priv, u8 send_if_empty) @@ -620,6 +620,9 @@ static int iwl_set_wep_dynamic_key_info(struct iwl_priv *priv, /* else, we are overriding an existing key => no need to allocated room * in uCode. */ + WARN(priv->stations[sta_id].sta.key.key_offset == WEP_INVALID_OFFSET, + "no space for new kew"); + priv->stations[sta_id].sta.key.key_flags = key_flags; priv->stations[sta_id].sta.sta.modify_mask = STA_MODIFY_KEY_MASK; priv->stations[sta_id].sta.mode = STA_CONTROL_MODIFY_MSK; @@ -637,6 +640,7 @@ static int iwl_set_ccmp_dynamic_key_info(struct iwl_priv *priv, { unsigned long flags; __le16 key_flags = 0; + int ret; key_flags |= (STA_KEY_FLG_CCMP | STA_KEY_FLG_MAP_KEY_MSK); key_flags |= cpu_to_le16(keyconf->keyidx << STA_KEY_FLG_KEYID_POS); @@ -664,14 +668,18 @@ static int iwl_set_ccmp_dynamic_key_info(struct iwl_priv *priv, /* else, we are overriding an existing key => no need to allocated room * in uCode. */ + WARN(priv->stations[sta_id].sta.key.key_offset == WEP_INVALID_OFFSET, + "no space for new kew"); + priv->stations[sta_id].sta.key.key_flags = key_flags; priv->stations[sta_id].sta.sta.modify_mask = STA_MODIFY_KEY_MASK; priv->stations[sta_id].sta.mode = STA_CONTROL_MODIFY_MSK; + ret = iwl_send_add_sta(priv, &priv->stations[sta_id].sta, CMD_ASYNC); + spin_unlock_irqrestore(&priv->sta_lock, flags); - IWL_DEBUG_INFO("hwcrypto: modify ucode station key info\n"); - return iwl_send_add_sta(priv, &priv->stations[sta_id].sta, CMD_ASYNC); + return ret; } static int iwl_set_tkip_dynamic_key_info(struct iwl_priv *priv, @@ -696,6 +704,9 @@ static int iwl_set_tkip_dynamic_key_info(struct iwl_priv *priv, /* else, we are overriding an existing key => no need to allocated room * in uCode. */ + WARN(priv->stations[sta_id].sta.key.key_offset == WEP_INVALID_OFFSET, + "no space for new kew"); + /* This copy is acutally not needed: we get the key with each TX */ memcpy(priv->stations[sta_id].keyinfo.key, keyconf->key, 16); @@ -734,6 +745,13 @@ int iwl_remove_dynamic_key(struct iwl_priv *priv, return 0; } + if (priv->stations[sta_id].sta.key.key_offset == WEP_INVALID_OFFSET) { + IWL_WARNING("Removing wrong key %d 0x%x\n", + keyconf->keyidx, key_flags); + spin_unlock_irqrestore(&priv->sta_lock, flags); + return 0; + } + if (!test_and_clear_bit(priv->stations[sta_id].sta.key.key_offset, &priv->ucode_key_table)) IWL_ERROR("index %d not used in uCode key table.\n", -- cgit v1.2.1 From b8ddafd759e356c839774a8b87011b768deb53b3 Mon Sep 17 00:00:00 2001 From: Zhu Yi Date: Thu, 27 Nov 2008 13:42:20 +0800 Subject: ipw2200: fix netif_*_queue() removal regression In "ipw2200: Call netif_*_queue() interfaces properly", netif_stop_queue() and netif_wake_queue() were removed with the reason "netif_carrier_{on,off}() handles starting and stopping packet flow into the driver". The patch also removes a valid condition check that ipw_tx_skb() cannot be called if device is not in STATUS_ASSOCIATED state. But netif_carrier_off() doesn't guarantee netdev->hard_start_xmit won't be called because linkwatch event is handled in a delayed workqueue. This caused a kernel oops reported by Frank Seidel: https://bugzilla.novell.com/show_bug.cgi?id=397390 This patch fixes the problem by moving the STATUS_ASSOCIATED check back to ipw_tx_skb(). It also adds a missing netif_carrier_off() call to ipw_disassociate(). Signed-off-by: Zhu Yi Signed-off-by: Chatre, Reinette Tested-by: Frank Seidel Signed-off-by: John W. Linville --- drivers/net/wireless/ipw2200.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'drivers') diff --git a/drivers/net/wireless/ipw2200.c b/drivers/net/wireless/ipw2200.c index dcce3542d5a7..7a9f901d4ff6 100644 --- a/drivers/net/wireless/ipw2200.c +++ b/drivers/net/wireless/ipw2200.c @@ -3897,6 +3897,7 @@ static int ipw_disassociate(void *data) if (!(priv->status & (STATUS_ASSOCIATED | STATUS_ASSOCIATING))) return 0; ipw_send_disassociate(data, 0); + netif_carrier_off(priv->net_dev); return 1; } @@ -10190,6 +10191,9 @@ static int ipw_tx_skb(struct ipw_priv *priv, struct ieee80211_txb *txb, u16 remaining_bytes; int fc; + if (!(priv->status & STATUS_ASSOCIATED)) + goto drop; + hdr_len = ieee80211_get_hdrlen(le16_to_cpu(hdr->frame_ctl)); switch (priv->ieee->iw_mode) { case IW_MODE_ADHOC: -- cgit v1.2.1 From cde6901b7b69557a6f90f3183f76eda581af015e Mon Sep 17 00:00:00 2001 From: Shaddy Baddah Date: Fri, 28 Nov 2008 17:10:45 +1100 Subject: zd1211rw: use unaligned safe memcmp() in-place of compare_ether_addr() Under my 2.6.28-rc6 sparc64, when associating to an AP through my zd1211rw device, I was seeing kernel log messages like (not exact output): Kernel unaligned access at TPC[10129b68] zd_mac_rx+0x144/0x32c [zd1211rw] For the zd1211rw module, on RX, the 80211 packet will be located after the PLCP header in the skb data buffer. The PLCP header being 5 bytes long, the 80211 header will start unaligned from an aligned skb buffer. As per Documentation/unaligned-memory-access.txt, we must replace the not unaligned() safe compare_ether_addr() with memcmp() to protect architectures that require alignment. Signed-off-by: Shaddy Baddah Signed-off-by: John W. Linville --- drivers/net/wireless/zd1211rw/zd_mac.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c index fe1867b25ff7..cac732f4047f 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.c +++ b/drivers/net/wireless/zd1211rw/zd_mac.c @@ -615,7 +615,7 @@ static int filter_ack(struct ieee80211_hw *hw, struct ieee80211_hdr *rx_hdr, struct ieee80211_hdr *tx_hdr; tx_hdr = (struct ieee80211_hdr *)skb->data; - if (likely(!compare_ether_addr(tx_hdr->addr2, rx_hdr->addr1))) + if (likely(!memcmp(tx_hdr->addr2, rx_hdr->addr1, ETH_ALEN))) { __skb_unlink(skb, q); tx_status(hw, skb, IEEE80211_TX_STAT_ACK, stats->signal, 1); -- cgit v1.2.1