Re: PROBLEM: iwlagn kernel 2.6.32.3 ooops

Re: PROBLEM: iwlagn kernel 2.6.32.3 ooops

[Andrew Morton]

On Sat, 6 Feb 2010 18:03:50 +0100
lkml@Think-Future.de wrote:

>   Hi, 

Suitable cc's added, more below...

> This ooops happens on kernel 2.6.32.3. On a .32.7 it didn't 
> happen yet. Maybe fixed, maybe not.
> 
> Keywords (i.e., modules, networking, kernel):
> iwlagn, kernel 2.6.32.3
> 
> 
> Below two msgs:
> The first is the ooops that occurred when using the box as usual (IRC: after 
> this crash no rmmod && modprobe have been possible).
> 
> The second - IRC - when the iwlagn module crashed, rebooted, crashed differently, 
> was then removed after the different ooops and re-modprobed.
> 
> 
> [250420.677157] ip: page allocation failure. order:3, mode:0x8020

This one isn't an "oops" - it's a warning that a memory allocation
attempt failed.  Which is hardly surprising: an order-3 GFP_ATOMIC
allocation is highly unreliable and the driver just shouldn't be
attempting it.

> [250420.677168] Pid: 11584, comm: ip Not tainted 2.6.32.3 #5
> [250420.677173] Call Trace:
> [250420.677190]  [<c10a53f8>] ? __alloc_pages_nodemask+0x518/0x5f0
> [250420.677204]  [<c1007c20>] ? dma_generic_alloc_coherent+0x0/0x100
> [250420.677214]  [<c1007c92>] ? dma_generic_alloc_coherent+0x72/0x100
> [250420.677224]  [<c1007c20>] ? dma_generic_alloc_coherent+0x0/0x100
> [250420.677241]  [<fca1c9d5>] ? iwl_tx_queue_init+0x285/0x380 [iwlcore]
> [250420.677252]  [<c171d7f9>] ? _spin_lock_irqsave+0x19/0x40
> [250420.678452]  [<fca1cceb>] ? iwl_txq_ctx_reset+0x21b/0x5d0 [iwlcore]
> [250420.678462]  [<c171d7f9>] ? _spin_lock_irqsave+0x19/0x40
> [250420.678478]  [<c171d9ee>] ? _spin_unlock_irqrestore+0xe/0x30
> [250420.678494]  [<fca15821>] ? iwl_hw_nic_init+0xc1/0x130 [iwlcore]
> [250420.678505]  [<fca4d8eb>] ? iwl_prepare_card_hw+0xb/0xa0 [iwlagn]
> [250420.678515]  [<fca4e6ac>] ? __iwl_up+0x9c/0x350 [iwlagn]
> [250420.678526]  [<fca4ee5f>] ? iwl_mac_start+0x4ff/0xb50 [iwlagn]
> [250420.678537]  [<c15e560c>] ? netlink_broadcast+0x22c/0x370
> [250420.678547]  [<c15e5c32>] ? nlmsg_notify+0x42/0xb0
> [250420.678556]  [<c162298c>] ? inetdev_event+0x1c/0x460
> [250420.678565]  [<c171d7f9>] ? _spin_lock_irqsave+0x19/0x40
> [250420.678573]  [<c171d9ee>] ? _spin_unlock_irqrestore+0xe/0x30
> [250420.678586]  [<c16ed1c7>] ? ieee80211_open+0x3e7/0x7a0
> [250420.678596]  [<c15b06d2>] ? dev_open+0xc2/0x100
> [250420.678606]  [<c104864a>] ? local_bh_disable+0xa/0x10
> [250420.678615]  [<c15afb8b>] ? dev_change_flags+0x8b/0x1b0
> [250420.678626]  [<c15b8642>] ? do_setlink+0x1e2/0x370
> [250420.678636]  [<c15b9d6e>] ? rtnl_newlink+0x3fe/0x4c0
> [250420.678648]  [<c1069800>] ? tick_do_broadcast+0x20/0x70
> [250420.678657]  [<c1069922>] ? tick_handle_oneshot_broadcast+0xd2/0x110
> [250420.678667]  [<c15b8d77>] ? rtnl_fill_ifinfo+0x2e7/0x4f0
> [250420.678675]  [<c171da7c>] ? _spin_unlock+0xc/0x30
> [250420.678684]  [<c1005df9>] ? timer_interrupt+0x39/0x40
> [250420.678693]  [<c108421f>] ? handle_IRQ_event+0x7f/0x1a0
> [250420.678702]  [<c171d63d>] ? _spin_lock+0xd/0x30
> [250420.678711]  [<c15b9970>] ? rtnl_newlink+0x0/0x4c0
> [250420.678720]  [<c15b98cb>] ? rtnetlink_rcv_msg+0x16b/0x210
> [250420.678729]  [<c15b9760>] ? rtnetlink_rcv_msg+0x0/0x210
> [250420.678738]  [<c15e5f06>] ? netlink_rcv_skb+0x66/0x90
> [250420.678746]  [<c15b9759>] ? rtnetlink_rcv+0x19/0x20
> [250420.678755]  [<c15e5be4>] ? netlink_unicast+0x274/0x280
> [250420.678764]  [<c15e63e9>] ? netlink_sendmsg+0x1d9/0x2e0
> [250420.678777]  [<c159f311>] ? sock_sendmsg+0x111/0x130
> [250420.678788]  [<c105c2a0>] ? autoremove_wake_function+0x0/0x50
> [250420.678797]  [<c105c2a0>] ? autoremove_wake_function+0x0/0x50
> [250420.678808]  [<c1027975>] ? kunmap_atomic+0x35/0x80
> [250420.678817]  [<c10a4b5f>] ? get_page_from_freelist+0x26f/0x500
> [250420.678828]  [<c159f4a4>] ? sys_sendmsg+0x174/0x270
> [250420.678837]  [<c15a0756>] ? sys_recvmsg+0x1a6/0x230
> [250420.678847]  [<c10bfaa1>] ? anon_vma_prepare+0x91/0xe0
> [250420.678857]  [<c10b2c8b>] ? __inc_zone_state+0x1b/0x90
> [250420.678866]  [<c10a7a37>] ? __lru_cache_add+0x67/0xb0
> [250420.678874]  [<c171da7c>] ? _spin_unlock+0xc/0x30
> [250420.678883]  [<c10b6569>] ? do_wp_page+0xf9/0x700
> [250420.678894]  [<c10b80e4>] ? handle_mm_fault+0x514/0x7c0
> [250420.678905]  [<c15a0f3e>] ? sys_socketcall+0xee/0x2c0
> [250420.678914]  [<c1002ef0>] ? sysenter_do_call+0x12/0x22
> [250420.678921] Mem-Info:
> [250420.678925] DMA per-cpu:
> [250420.678930] CPU    0: hi:    0, btch:   1 usd:   0
> [250420.678936] CPU    1: hi:    0, btch:   1 usd:   0
> [250420.678940] Normal per-cpu:
> [250420.678945] CPU    0: hi:  186, btch:  31 usd: 173
> [250420.678951] CPU    1: hi:  186, btch:  31 usd: 151
> [250420.678956] HighMem per-cpu:
> [250420.678960] CPU    0: hi:  186, btch:  31 usd:  95
> [250420.678966] CPU    1: hi:  186, btch:  31 usd: 149
> [250420.678977] active_anon:206883 inactive_anon:48519 isolated_anon:0
> [250420.678980]  active_file:185862 inactive_file:166683 isolated_file:33
> [250420.678983]  unevictable:831 dirty:5805 writeback:0 unstable:0
> [250420.678986]  free:367782 slab_reclaimable:19779 slab_unreclaimable:13076
> [250420.678989]  mapped:26361 shmem:25241 pagetables:2094 bounce:0
> [250420.679004] DMA free:3552kB min:64kB low:80kB high:96kB active_anon:0kB inactive_anon:0kB active_file:1472kB inactive_file:5152kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:15844kB mlocked:0kB dirty:0kB writeback:0kB mapped:0kB shmem:0kB slab_reclaimable:572kB slab_unreclaimable:4988kB kernel_stack:104kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
> [250420.679017] lowmem_reserve[]: 0 867 3967 3967
> [250420.679038] Normal free:155480kB min:3732kB low:4664kB high:5596kB active_anon:652kB inactive_anon:14312kB active_file:154244kB inactive_file:371780kB unevictable:0kB isolated(anon):0kB isolated(file):132kB present:887976kB mlocked:0kB dirty:432kB writeback:0kB mapped:364kB shmem:400kB slab_reclaimable:78544kB slab_unreclaimable:47316kB kernel_stack:2624kB pagetables:8376kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
> [250420.679053] lowmem_reserve[]: 0 0 24801 24801
> [250420.679074] HighMem free:1312220kB min:512kB low:3848kB high:7184kB active_anon:826880kB inactive_anon:179764kB active_file:587732kB inactive_file:289800kB unevictable:3324kB isolated(anon):0kB isolated(file):0kB present:3174548kB mlocked:0kB dirty:22788kB writeback:0kB mapped:105080kB shmem:100564kB slab_reclaimable:0kB slab_unreclaimable:0kB kernel_stack:0kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
> [250420.679089] lowmem_reserve[]: 0 0 0 0
> [250420.679099] DMA: 726*4kB 61*8kB 10*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 3552kB
> [250420.679125] Normal: 38018*4kB 208*8kB 106*16kB 1*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 155464kB
> [250420.679150] HighMem: 12315*4kB 40702*8kB 32102*16kB 9425*32kB 1468*64kB 194*128kB 13*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1312220kB
> [250420.679177] 379888 total pagecache pages
> [250420.679181] 1260 pages in swap cache
> [250420.679186] Swap cache stats: add 1305, delete 45, find 8532/8534
> [250420.679191] Free swap  = 1946704kB
> [250420.679195] Total swap = 1951856kB
> [250420.731629] 1294336 pages RAM
> [250420.731635] 1066498 pages HighMem
> [250420.731639] 277784 pages reserved
> [250420.731642] 327537 pages shared
> [250420.731646] 426568 pages non-shared
> [250420.731655] iwlagn 0000:03:00.0: pci_alloc_consistent(32768) failed
> [250420.731707] iwlagn 0000:03:00.0: Tx 0 queue init failed
> [250420.731719] iwlagn 0000:03:00.0: Unable to init nic
> [250452.043227] e1000e: eth0 NIC Link is Down
> [251624.072132] usb 1-3: USB disconnect, address 20
> 
> 
> 
> 
> [252109.496953] iwlagn: Intel(R) Wireless WiFi Link AGN driver for Linux, 1.3.27ks
> [252109.496961] iwlagn: Copyright(c) 2003-2009 Intel Corporation
> [252109.497071] iwlagn 0000:03:00.0: PCI INT A -> GSI 17 (level, low) -> IRQ 17
> [252109.497094] iwlagn 0000:03:00.0: setting latency timer to 64
> [252109.497156] iwlagn 0000:03:00.0: Detected Intel Wireless WiFi Link 4965AGN REV=0x4
> [252109.547060] iwlagn 0000:03:00.0: Tunable channels: 13 802.11bg, 19 802.11a channels
> [252109.547202] iwlagn 0000:03:00.0: irq 29 for MSI/MSI-X
> [252109.548104] phy14: Selected rate control algorithm 'iwl-agn-rs'
> [252109.921801] udev: renamed network interface wlan0 to eth1
> [252110.693015] iwlagn 0000:03:00.0: firmware: requesting iwlwifi-4965-2.ucode
> [252110.810198] iwlagn 0000:03:00.0: loaded firmware version 228.61.2.24
> [252110.989598] ------------[ cut here ]------------
> [252110.989620] WARNING: at drivers/net/wireless/iwlwifi/iwl-tx.c:1151 iwl_tx_cmd_complete+0x2ed/0x330 [iwlcore]()

This one is a warning, not an oops.  It could be a consequence of the
memory allocation failure.  If so, the code should be changed to
prevent this warning from coming out.  Because a GFP_ATOMIC memory
allocation failure is common, and expected (especially when it's
order-3!).  The driver should handle it without blurting big warnings.


> [252110.989626] Hardware name: 7666B4G
> [252110.989632] wrong command queue 0, sequence 0xE0 readp=0 writep=0
> [252110.989636] Modules linked in: iwlagn iwlcore btusb xt_TCPMSS uvcvideo nf_conntrack_netlink iptable_nat nf_nat iptable_mangle ipt_LOG xt_tcpudp xt_state xt_multiport iptable_filter ip_tables x_tables e1000e rtc_cmos [last unloaded: iwlcore]
> [252110.989681] Pid: 5684, comm: Xorg Not tainted 2.6.32.3 #5
> [252110.989686] Call Trace:
> [252110.989700]  [<fc000aed>] ? iwl_tx_cmd_complete+0x2ed/0x330 [iwlcore]
> [252110.989712]  [<fc000aed>] ? iwl_tx_cmd_complete+0x2ed/0x330 [iwlcore]
> [252110.989725]  [<c104256f>] ? warn_slowpath_common+0x6f/0xd0
> [252110.989737]  [<fc000aed>] ? iwl_tx_cmd_complete+0x2ed/0x330 [iwlcore]
> [252110.989746]  [<c104261b>] ? warn_slowpath_fmt+0x2b/0x30
> [252110.989758]  [<fc000aed>] ? iwl_tx_cmd_complete+0x2ed/0x330 [iwlcore]
> [252110.989770]  [<c1031e28>] ? select_task_rq_fair+0x368/0x9a0
> [252110.989781]  [<fc03cd15>] ? iwl_rx_handle+0xf5/0x2a0 [iwlagn]
> [252110.989792]  [<fc03d056>] ? iwl_irq_tasklet_legacy+0x196/0x440 [iwlagn]
> [252110.989803]  [<c10473b9>] ? tasklet_action+0x69/0xe0
> [252110.989812]  [<c1048737>] ? __do_softirq+0xe7/0x1e0
> [252110.989821]  [<c108421f>] ? handle_IRQ_event+0x7f/0x1a0
> [252110.989829]  [<c104885d>] ? do_softirq+0x2d/0x40
> [252110.989837]  [<c1048a45>] ? irq_exit+0x65/0x80
> [252110.989845]  [<c1004ea5>] ? do_IRQ+0x55/0xc0
> [252110.989854]  [<c10035a9>] ? common_interrupt+0x29/0x30
> [252110.989866]  [<c165005e>] ? unix_stream_sendmsg+0x2be/0x390
> [252110.989879]  [<c159e99c>] ? sock_aio_write+0x11c/0x130
> [252110.989890]  [<c159e880>] ? sock_aio_write+0x0/0x130
> [252110.989899]  [<c10cf34e>] ? do_sync_readv_writev+0xce/0x110
> [252110.989909]  [<c105c2a0>] ? autoremove_wake_function+0x0/0x50
> [252110.989920]  [<c12a863c>] ? security_file_permission+0xc/0x10
> [252110.989928]  [<c10cf62a>] ? rw_verify_area+0x5a/0xd0
> [252110.989936]  [<c10cfaae>] ? do_readv_writev+0x9e/0x1b0
> [252110.989945]  [<c159e880>] ? sock_aio_write+0x0/0x130
> [252110.989954]  [<c1060560>] ? hrtimer_start+0x20/0x30
> [252110.989962]  [<c12a863c>] ? security_file_permission+0xc/0x10
> [252110.989970]  [<c10cf62a>] ? rw_verify_area+0x5a/0xd0
> [252110.989979]  [<c10cfbfe>] ? vfs_writev+0x3e/0x60
> [252110.989987]  [<c10cfd07>] ? sys_writev+0x47/0x90
> [252110.989995]  [<c1002ef0>] ? sysenter_do_call+0x12/0x22
> [252110.990002] ---[ end trace bd50f16bd2f33949 ]---
> [252110.990010] iwl data: 00000000: 02 00 04 00 e5 e4 e0 00 01 00 00 00 01 00 00 00  ................
> [252110.990017] iwl data: 00000010: 01 00 0a 00 e7 e4 e0 00 0b 01 00 00 00 01 e6 00  ................
> [252114.817112] iwlagn 0000:03:00.0: START_ALIVE timeout after 4000ms.
> [252122.146120] iwlagn 0000:03:00.0: START_ALIVE timeout after 4000ms.

Re: PROBLEM: iwlagn kernel 2.6.32.3 ooops

[reinette chatre]

On Mon, 2010-02-08 at 14:31 -0800, Andrew Morton wrote:
> On Sat, 6 Feb 2010 18:03:50 +0100
> lkml@Think-Future.de wrote:
> > 
> > 
> > [250420.677157] ip: page allocation failure. order:3, mode:0x8020
> 
> This one isn't an "oops" - it's a warning that a memory allocation
> attempt failed.  Which is hardly surprising: an order-3 GFP_ATOMIC
> allocation is highly unreliable and the driver just shouldn't be
> attempting it.

This is the familiar http://bugzilla.kernel.org/show_bug.cgi?id=14141
which is not specific to iwlwifi.

Even so, from 2.6.33 iwlwifi will be doing paged RX which reduces the
amount of atomic memory needed. 

> > [252109.496953] iwlagn: Intel(R) Wireless WiFi Link AGN driver for Linux, 1.3.27ks
> > [252109.496961] iwlagn: Copyright(c) 2003-2009 Intel Corporation
> > [252109.497071] iwlagn 0000:03:00.0: PCI INT A -> GSI 17 (level, low) -> IRQ 17
> > [252109.497094] iwlagn 0000:03:00.0: setting latency timer to 64
> > [252109.497156] iwlagn 0000:03:00.0: Detected Intel Wireless WiFi Link 4965AGN REV=0x4
> > [252109.547060] iwlagn 0000:03:00.0: Tunable channels: 13 802.11bg, 19 802.11a channels
> > [252109.547202] iwlagn 0000:03:00.0: irq 29 for MSI/MSI-X
> > [252109.548104] phy14: Selected rate control algorithm 'iwl-agn-rs'
> > [252109.921801] udev: renamed network interface wlan0 to eth1
> > [252110.693015] iwlagn 0000:03:00.0: firmware: requesting iwlwifi-4965-2.ucode
> > [252110.810198] iwlagn 0000:03:00.0: loaded firmware version 228.61.2.24
> > [252110.989598] ------------[ cut here ]------------
> > [252110.989620] WARNING: at drivers/net/wireless/iwlwifi/iwl-tx.c:1151 iwl_tx_cmd_complete+0x2ed/0x330 [iwlcore]()
> 
> This one is a warning, not an oops.  It could be a consequence of the
> memory allocation failure.

Based on the preceding messages the driver was reloaded at this point so
the previous memory allocations should have no effect.

>   If so, the code should be changed to
> prevent this warning from coming out.  Because a GFP_ATOMIC memory
> allocation failure is common, and expected (especially when it's
> order-3!).  The driver should handle it without blurting big warnings.

The warning is not related to the memory allocation failure. This is a
different problem. While I dig a bit into the trace, could you please
share the steps to reproduce? Could you please activate some debugging
to see what else is going on at the time? 

Please compile driver with CONFIG_IWLWIFI_DEBUG and load module with
"modprobe iwlagn debug=0x1043fff"

Thank you

Reinette

Re: [Ilw] Re: PROBLEM: iwlagn kernel 2.6.32.3 ooops

[reinette chatre]

On Mon, 2010-02-08 at 15:00 -0800, Chatre, Reinette wrote:
> 
> On Mon, 2010-02-08 at 14:31 -0800, Andrew Morton wrote:
> > On Sat, 6 Feb 2010 18:03:50 +0100
> > lkml@Think-Future.de wrote:

> > > [252109.496953] iwlagn: Intel(R) Wireless WiFi Link AGN driver for Linux, 1.3.27ks
> > > [252109.496961] iwlagn: Copyright(c) 2003-2009 Intel Corporation
> > > [252109.497071] iwlagn 0000:03:00.0: PCI INT A -> GSI 17 (level, low) -> IRQ 17
> > > [252109.497094] iwlagn 0000:03:00.0: setting latency timer to 64
> > > [252109.497156] iwlagn 0000:03:00.0: Detected Intel Wireless WiFi Link 4965AGN REV=0x4
> > > [252109.547060] iwlagn 0000:03:00.0: Tunable channels: 13 802.11bg, 19 802.11a channels
> > > [252109.547202] iwlagn 0000:03:00.0: irq 29 for MSI/MSI-X
> > > [252109.548104] phy14: Selected rate control algorithm 'iwl-agn-rs'
> > > [252109.921801] udev: renamed network interface wlan0 to eth1
> > > [252110.693015] iwlagn 0000:03:00.0: firmware: requesting iwlwifi-4965-2.ucode
> > > [252110.810198] iwlagn 0000:03:00.0: loaded firmware version 228.61.2.24
> > > [252110.989598] ------------[ cut here ]------------
> > > [252110.989620] WARNING: at drivers/net/wireless/iwlwifi/iwl-tx.c:1151 iwl_tx_cmd_complete+0x2ed/0x330 [iwlcore]()
> > 

Could you please run with the patch below and check if the new error
message appears?

diff --git a/drivers/net/wireless/iwlwifi/iwl-agn.c b/drivers/net/wireless/iwlwifi/iwl-agn.c
index 921dc4a..6911995 100644
--- a/drivers/net/wireless/iwlwifi/iwl-agn.c
+++ b/drivers/net/wireless/iwlwifi/iwl-agn.c
@@ -759,6 +759,8 @@ void iwl_rx_handle(struct iwl_priv *priv)
 	u8 fill_rx = 0;
 	u32 count = 8;
 	int total_empty;
+	bool valid;
+	int len;
 
 	/* uCode's read index (stored in shared DRAM) indicates the last Rx
 	 * buffer that the driver may process (last buffer filled by ucode). */
@@ -791,6 +793,11 @@ void iwl_rx_handle(struct iwl_priv *priv)
 				 priv->hw_params.rx_buf_size + 256,
 				 PCI_DMA_FROMDEVICE);
 		pkt = (struct iwl_rx_packet *)rxb->skb->data;
+		len = le32_to_cpu(pkt->len_n_flags) & FH_RSCSR_FRAME_SIZE_MSK;
+		valid = (len >= sizeof(__le32) + sizeof(struct iwl_cmd_header));
+		/* Make sure we can at least parse the header */
+		if (!valid)
+			IWL_ERR(priv, "Invalid packet from device.\n");
 
 		/* Reclaim a command buffer only if this packet is a response
 		 *   to a (driver-originated) command.
@@ -809,7 +816,7 @@ void iwl_rx_handle(struct iwl_priv *priv)
 		/* Based on type of command response or notification,
 		 *   handle those that need handling via function in
 		 *   rx_handlers table.  See iwl_setup_rx_handlers() */
-		if (priv->rx_handlers[pkt->hdr.cmd]) {
+		if (valid && priv->rx_handlers[pkt->hdr.cmd]) {
 			IWL_DEBUG_RX(priv, "r = %d, i = %d, %s, 0x%02x\n", r,
 				i, get_cmd_string(pkt->hdr.cmd), pkt->hdr.cmd);
 			priv->rx_handlers[pkt->hdr.cmd] (priv, rxb);
@@ -822,7 +829,7 @@ void iwl_rx_handle(struct iwl_priv *priv)
 				pkt->hdr.cmd);
 		}
 
-		if (reclaim) {
+		if (valid && reclaim) {
 			/* Invoke any callbacks, transfer the skb to caller, and
 			 * fire off the (possibly) blocking iwl_send_cmd()
 			 * as we reclaim the driver command queue */

Re: PROBLEM: iwlagn kernel 2.6.32.3 ooops

[Stanislaw Gruszka]

On Mon, Feb 08, 2010 at 02:31:18PM -0800, Andrew Morton wrote:
> On Sat, 6 Feb 2010 18:03:50 +0100
> lkml@Think-Future.de wrote:
> 
> >   Hi, 
> 
> Suitable cc's added, more below...
> 
> > This ooops happens on kernel 2.6.32.3. On a .32.7 it didn't 
> > happen yet. Maybe fixed, maybe not.
> > 
> > Keywords (i.e., modules, networking, kernel):
> > iwlagn, kernel 2.6.32.3
> > 
> > 
> > Below two msgs:
> > The first is the ooops that occurred when using the box as usual (IRC: after 
> > this crash no rmmod && modprobe have been possible).
> > 
> > The second - IRC - when the iwlagn module crashed, rebooted, crashed differently, 
> > was then removed after the different ooops and re-modprobed.
> > 
> > 
> > [250420.677157] ip: page allocation failure. order:3, mode:0x8020
> 
> This one isn't an "oops" - it's a warning that a memory allocation
> attempt failed.  Which is hardly surprising: an order-3 GFP_ATOMIC
> allocation is highly unreliable and the driver just shouldn't be
> attempting it.
> 
> > [250420.677168] Pid: 11584, comm: ip Not tainted 2.6.32.3 #5
> > [250420.677173] Call Trace:
> > [250420.677190]  [<c10a53f8>] ? __alloc_pages_nodemask+0x518/0x5f0
> > [250420.677204]  [<c1007c20>] ? dma_generic_alloc_coherent+0x0/0x100
> > [250420.677214]  [<c1007c92>] ? dma_generic_alloc_coherent+0x72/0x100
> > [250420.677224]  [<c1007c20>] ? dma_generic_alloc_coherent+0x0/0x100
> > [250420.677241]  [<fca1c9d5>] ? iwl_tx_queue_init+0x285/0x380 [iwlcore]

That is at initialization time. To avoid atomic allocation we can just
use dma_alloc_coherent instead of pci_alloc_consistent. I'm going to
prepare a patch.

Stanislaw

Re: PROBLEM: iwlagn kernel 2.6.32.3 ooops

[reinette chatre]

Hi Stanislaw,

On Tue, 2010-02-09 at 08:52 -0800, Stanislaw Gruszka wrote:
>  
> > > [250420.677168] Pid: 11584, comm: ip Not tainted 2.6.32.3 #5
> > > [250420.677173] Call Trace:
> > > [250420.677190]  [<c10a53f8>] ? __alloc_pages_nodemask+0x518/0x5f0
> > > [250420.677204]  [<c1007c20>] ? dma_generic_alloc_coherent+0x0/0x100
> > > [250420.677214]  [<c1007c92>] ? dma_generic_alloc_coherent+0x72/0x100
> > > [250420.677224]  [<c1007c20>] ? dma_generic_alloc_coherent+0x0/0x100
> > > [250420.677241]  [<fca1c9d5>] ? iwl_tx_queue_init+0x285/0x380 [iwlcore]
> 
> That is at initialization time. To avoid atomic allocation we can just
> use dma_alloc_coherent instead of pci_alloc_consistent. I'm going to
> prepare a patch.

I do not understand your goal here ... as seen in this trace
pci_alloc_consistent itself already calls dma_alloc_coherent right away.

Reinette

Re: PROBLEM: iwlagn kernel 2.6.32.3 ooops

[Stanislaw Gruszka]

Hi

On Tue, Feb 09, 2010 at 09:04:52AM -0800, reinette chatre wrote:
> I do not understand your goal here ... as seen in this trace
> pci_alloc_consistent itself already calls dma_alloc_coherent right away.

We can call dma_alloc_coherent with GFP_KERNEL, pci_alloc_consistent is
more or less equal to dma_alloc_coherent with GFP_ATOMIC flag.

Stanislaw

[PATCH] iwlwifi: use dma_alloc_coherent

[Stanislaw Gruszka]

Change pci_alloc_consistent() to dma_alloc_coherent() so we can use
GFP_KERNEL flag.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
---
Tested with 3945 and 5300

 drivers/net/wireless/iwlwifi/iwl-3945.c     |    8 +++-----
 drivers/net/wireless/iwlwifi/iwl-core.c     |   12 ++++++------
 drivers/net/wireless/iwlwifi/iwl-helpers.h  |    7 ++++---
 drivers/net/wireless/iwlwifi/iwl-rx.c       |   21 +++++++++++----------
 drivers/net/wireless/iwlwifi/iwl-tx.c       |   23 ++++++++++++-----------
 drivers/net/wireless/iwlwifi/iwl3945-base.c |   16 ++++++++--------
 6 files changed, 44 insertions(+), 43 deletions(-)

diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.c b/drivers/net/wireless/iwlwifi/iwl-3945.c
index 234891d..e955515 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.c
@@ -2474,11 +2474,9 @@ int iwl3945_hw_set_hw_params(struct iwl_priv *priv)
 	memset((void *)&priv->hw_params, 0,
 	       sizeof(struct iwl_hw_params));
 
-	priv->shared_virt =
-	    pci_alloc_consistent(priv->pci_dev,
-				 sizeof(struct iwl3945_shared),
-				 &priv->shared_phys);
-
+	priv->shared_virt = dma_alloc_coherent(&priv->pci_dev->dev,
+					       sizeof(struct iwl3945_shared),
+					       &priv->shared_phys, GFP_KERNEL);
 	if (!priv->shared_virt) {
 		IWL_ERR(priv, "failed to allocate pci memory\n");
 		mutex_unlock(&priv->mutex);
diff --git a/drivers/net/wireless/iwlwifi/iwl-core.c b/drivers/net/wireless/iwlwifi/iwl-core.c
index 5461f10..f6a7866 100644
--- a/drivers/net/wireless/iwlwifi/iwl-core.c
+++ b/drivers/net/wireless/iwlwifi/iwl-core.c
@@ -1658,9 +1658,9 @@ EXPORT_SYMBOL(iwl_set_tx_power);
 void iwl_free_isr_ict(struct iwl_priv *priv)
 {
 	if (priv->ict_tbl_vir) {
-		pci_free_consistent(priv->pci_dev, (sizeof(u32) * ICT_COUNT) +
-					PAGE_SIZE, priv->ict_tbl_vir,
-					priv->ict_tbl_dma);
+		dma_free_coherent(&priv->pci_dev->dev,
+				  (sizeof(u32) * ICT_COUNT) + PAGE_SIZE,
+				  priv->ict_tbl_vir, priv->ict_tbl_dma);
 		priv->ict_tbl_vir = NULL;
 	}
 }
@@ -1676,9 +1676,9 @@ int iwl_alloc_isr_ict(struct iwl_priv *priv)
 	if (priv->cfg->use_isr_legacy)
 		return 0;
 	/* allocate shrared data table */
-	priv->ict_tbl_vir = pci_alloc_consistent(priv->pci_dev, (sizeof(u32) *
-						  ICT_COUNT) + PAGE_SIZE,
-						  &priv->ict_tbl_dma);
+	priv->ict_tbl_vir = dma_alloc_coherent(&priv->pci_dev->dev,
+					(sizeof(u32) * ICT_COUNT) + PAGE_SIZE,
+					&priv->ict_tbl_dma, GFP_KERNEL);
 	if (!priv->ict_tbl_vir)
 		return -ENOMEM;
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-helpers.h b/drivers/net/wireless/iwlwifi/iwl-helpers.h
index bd0b12e..f8481e8 100644
--- a/drivers/net/wireless/iwlwifi/iwl-helpers.h
+++ b/drivers/net/wireless/iwlwifi/iwl-helpers.h
@@ -80,8 +80,8 @@ static inline void iwl_free_fw_desc(struct pci_dev *pci_dev,
 				    struct fw_desc *desc)
 {
 	if (desc->v_addr)
-		pci_free_consistent(pci_dev, desc->len,
-				    desc->v_addr, desc->p_addr);
+		dma_free_coherent(&pci_dev->dev, desc->len,
+				  desc->v_addr, desc->p_addr);
 	desc->v_addr = NULL;
 	desc->len = 0;
 }
@@ -89,7 +89,8 @@ static inline void iwl_free_fw_desc(struct pci_dev *pci_dev,
 static inline int iwl_alloc_fw_desc(struct pci_dev *pci_dev,
 				    struct fw_desc *desc)
 {
-	desc->v_addr = pci_alloc_consistent(pci_dev, desc->len, &desc->p_addr);
+	desc->v_addr = dma_alloc_coherent(&pci_dev->dev, desc->len,
+					  &desc->p_addr, GFP_KERNEL);
 	return (desc->v_addr != NULL) ? 0 : -ENOMEM;
 }
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-rx.c b/drivers/net/wireless/iwlwifi/iwl-rx.c
index 6f36b6e..cfcb3c1 100644
--- a/drivers/net/wireless/iwlwifi/iwl-rx.c
+++ b/drivers/net/wireless/iwlwifi/iwl-rx.c
@@ -350,10 +350,10 @@ void iwl_rx_queue_free(struct iwl_priv *priv, struct iwl_rx_queue *rxq)
 		}
 	}
 
-	pci_free_consistent(priv->pci_dev, 4 * RX_QUEUE_SIZE, rxq->bd,
-			    rxq->dma_addr);
-	pci_free_consistent(priv->pci_dev, sizeof(struct iwl_rb_status),
-			    rxq->rb_stts, rxq->rb_stts_dma);
+	dma_free_coherent(&priv->pci_dev->dev, 4 * RX_QUEUE_SIZE, rxq->bd,
+			  rxq->dma_addr);
+	dma_free_coherent(&priv->pci_dev->dev, sizeof(struct iwl_rb_status),
+			  rxq->rb_stts, rxq->rb_stts_dma);
 	rxq->bd = NULL;
 	rxq->rb_stts  = NULL;
 }
@@ -362,7 +362,7 @@ EXPORT_SYMBOL(iwl_rx_queue_free);
 int iwl_rx_queue_alloc(struct iwl_priv *priv)
 {
 	struct iwl_rx_queue *rxq = &priv->rxq;
-	struct pci_dev *dev = priv->pci_dev;
+	struct device *dev = &priv->pci_dev->dev;
 	int i;
 
 	spin_lock_init(&rxq->lock);
@@ -370,12 +370,13 @@ int iwl_rx_queue_alloc(struct iwl_priv *priv)
 	INIT_LIST_HEAD(&rxq->rx_used);
 
 	/* Alloc the circular buffer of Read Buffer Descriptors (RBDs) */
-	rxq->bd = pci_alloc_consistent(dev, 4 * RX_QUEUE_SIZE, &rxq->dma_addr);
+	rxq->bd = dma_alloc_coherent(dev, 4 * RX_QUEUE_SIZE, &rxq->dma_addr,
+				     GFP_KERNEL);
 	if (!rxq->bd)
 		goto err_bd;
 
-	rxq->rb_stts = pci_alloc_consistent(dev, sizeof(struct iwl_rb_status),
-					&rxq->rb_stts_dma);
+	rxq->rb_stts = dma_alloc_coherent(dev, sizeof(struct iwl_rb_status),
+					  &rxq->rb_stts_dma, GFP_KERNEL);
 	if (!rxq->rb_stts)
 		goto err_rb;
 
@@ -392,8 +393,8 @@ int iwl_rx_queue_alloc(struct iwl_priv *priv)
 	return 0;
 
 err_rb:
-	pci_free_consistent(priv->pci_dev, 4 * RX_QUEUE_SIZE, rxq->bd,
-			    rxq->dma_addr);
+	dma_free_coherent(&priv->pci_dev->dev, 4 * RX_QUEUE_SIZE, rxq->bd,
+			  rxq->dma_addr);
 err_bd:
 	return -ENOMEM;
 }
diff --git a/drivers/net/wireless/iwlwifi/iwl-tx.c b/drivers/net/wireless/iwlwifi/iwl-tx.c
index 87ce2bd..281d318 100644
--- a/drivers/net/wireless/iwlwifi/iwl-tx.c
+++ b/drivers/net/wireless/iwlwifi/iwl-tx.c
@@ -60,7 +60,8 @@ static const u16 default_tid_to_tx_fifo[] = {
 static inline int iwl_alloc_dma_ptr(struct iwl_priv *priv,
 				    struct iwl_dma_ptr *ptr, size_t size)
 {
-	ptr->addr = pci_alloc_consistent(priv->pci_dev, size, &ptr->dma);
+	ptr->addr = dma_alloc_coherent(&priv->pci_dev->dev, size, &ptr->dma,
+				       GFP_KERNEL);
 	if (!ptr->addr)
 		return -ENOMEM;
 	ptr->size = size;
@@ -73,7 +74,7 @@ static inline void iwl_free_dma_ptr(struct iwl_priv *priv,
 	if (unlikely(!ptr->addr))
 		return;
 
-	pci_free_consistent(priv->pci_dev, ptr->size, ptr->addr, ptr->dma);
+	dma_free_coherent(&priv->pci_dev->dev, ptr->size, ptr->addr, ptr->dma);
 	memset(ptr, 0, sizeof(*ptr));
 }
 
@@ -132,7 +133,7 @@ void iwl_tx_queue_free(struct iwl_priv *priv, int txq_id)
 {
 	struct iwl_tx_queue *txq = &priv->txq[txq_id];
 	struct iwl_queue *q = &txq->q;
-	struct pci_dev *dev = priv->pci_dev;
+	struct device *dev = &priv->pci_dev->dev;
 	int i;
 
 	if (q->n_bd == 0)
@@ -149,8 +150,8 @@ void iwl_tx_queue_free(struct iwl_priv *priv, int txq_id)
 
 	/* De-alloc circular buffer of TFDs */
 	if (txq->q.n_bd)
-		pci_free_consistent(dev, priv->hw_params.tfd_size *
-				    txq->q.n_bd, txq->tfds, txq->q.dma_addr);
+		dma_free_coherent(dev, priv->hw_params.tfd_size *
+				  txq->q.n_bd, txq->tfds, txq->q.dma_addr);
 
 	/* De-alloc array of per-TFD driver data */
 	kfree(txq->txb);
@@ -179,7 +180,7 @@ void iwl_cmd_queue_free(struct iwl_priv *priv)
 {
 	struct iwl_tx_queue *txq = &priv->txq[IWL_CMD_QUEUE_NUM];
 	struct iwl_queue *q = &txq->q;
-	struct pci_dev *dev = priv->pci_dev;
+	struct device *dev = &priv->pci_dev->dev;
 	int i;
 
 	if (q->n_bd == 0)
@@ -191,8 +192,8 @@ void iwl_cmd_queue_free(struct iwl_priv *priv)
 
 	/* De-alloc circular buffer of TFDs */
 	if (txq->q.n_bd)
-		pci_free_consistent(dev, priv->hw_params.tfd_size *
-				    txq->q.n_bd, txq->tfds, txq->q.dma_addr);
+		dma_free_coherent(dev, priv->hw_params.tfd_size * txq->q.n_bd,
+				  txq->tfds, txq->q.dma_addr);
 
 	/* deallocate arrays */
 	kfree(txq->cmd);
@@ -283,7 +284,7 @@ static int iwl_queue_init(struct iwl_priv *priv, struct iwl_queue *q,
 static int iwl_tx_queue_alloc(struct iwl_priv *priv,
 			      struct iwl_tx_queue *txq, u32 id)
 {
-	struct pci_dev *dev = priv->pci_dev;
+	struct device *dev = &priv->pci_dev->dev;
 	size_t tfd_sz = priv->hw_params.tfd_size * TFD_QUEUE_SIZE_MAX;
 
 	/* Driver private data, only for Tx (not command) queues,
@@ -302,8 +303,8 @@ static int iwl_tx_queue_alloc(struct iwl_priv *priv,
 
 	/* Circular buffer of transmit frame descriptors (TFDs),
 	 * shared with device */
-	txq->tfds = pci_alloc_consistent(dev, tfd_sz, &txq->q.dma_addr);
-
+	txq->tfds = dma_alloc_coherent(dev, tfd_sz, &txq->q.dma_addr,
+				       GFP_KERNEL);
 	if (!txq->tfds) {
 		IWL_ERR(priv, "pci_alloc_consistent(%zd) failed\n", tfd_sz);
 		goto error;
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index f8e4e4b..f297865 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -352,10 +352,10 @@ static int iwl3945_send_beacon_cmd(struct iwl_priv *priv)
 static void iwl3945_unset_hw_params(struct iwl_priv *priv)
 {
 	if (priv->shared_virt)
-		pci_free_consistent(priv->pci_dev,
-				    sizeof(struct iwl3945_shared),
-				    priv->shared_virt,
-				    priv->shared_phys);
+		dma_free_coherent(&priv->pci_dev->dev,
+				  sizeof(struct iwl3945_shared),
+				  priv->shared_virt,
+				  priv->shared_phys);
 }
 
 static void iwl3945_build_tx_cmd_hwcrypto(struct iwl_priv *priv,
@@ -1253,10 +1253,10 @@ static void iwl3945_rx_queue_free(struct iwl_priv *priv, struct iwl_rx_queue *rx
 		}
 	}
 
-	pci_free_consistent(priv->pci_dev, 4 * RX_QUEUE_SIZE, rxq->bd,
-			    rxq->dma_addr);
-	pci_free_consistent(priv->pci_dev, sizeof(struct iwl_rb_status),
-			    rxq->rb_stts, rxq->rb_stts_dma);
+	dma_free_coherent(&priv->pci_dev->dev, 4 * RX_QUEUE_SIZE, rxq->bd,
+			  rxq->dma_addr);
+	dma_free_coherent(&priv->pci_dev->dev, sizeof(struct iwl_rb_status),
+			  rxq->rb_stts, rxq->rb_stts_dma);
 	rxq->bd = NULL;
 	rxq->rb_stts  = NULL;
 }
-- 
1.6.2.5

Re: [PATCH] iwlwifi: use dma_alloc_coherent

[reinette chatre]

Hi Stanislaw,

On Wed, 2010-02-10 at 05:07 -0800, Stanislaw Gruszka wrote:
> Change pci_alloc_consistent() to dma_alloc_coherent() so we can use
> GFP_KERNEL flag.
> 
> Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
> ---

This looks good - thank you very much. I am going to include it in our
testing for a bit and then I will send it upstream for you with our
other pending patches. 

Reinette

Re: [PATCH] iwlwifi: use dma_alloc_coherent

[lkml]

  Hi,

Ignorant when this patch was included, the reported problems ceased to show up.

Running 2.6.32.7 since 6th of february.

reboot   system boot  2.6.32.7         Sat Feb  6 16:53 - 13:22 (11+20:29)

I'll report back if needed.

Thank you for your work and time!

Nils

@Andrew: Thanks for clarifying the oops/warning diff. I was probably too lazy too care about that 
  when it just made me loosing my connectivity.. sorta user-experience-drownded stuff.. ;)