rk3188 emac stuck at high load

rk3188 emac stuck at high load

[Shuyu Wei]

Hi,
I'm testing Radxa Rock Pro board with mainline kernel 4.4.0-rc6. 
The problem can be reproduced by simultaneously wget two large files at high speed to /dev/null.

$wget https://192.168.1.1/file.bin -O /dev/null -q &
$wget https://192.168.1.1/file.bin -O /dev/null

or open two sftp session to download two large files to /dev/null.

Some addtional information that might help:
    Downloading from encrypted https links or sftp server triggers the problem much easier.
    Disabling SMP in kernel seems to fix the problem.
    Sometimes the emac just stops sending and receiving frames, sometimes it produces kernel panics like below.

[ 2191.975729] BUG: Bad page state in process swapper/0  pfn:8d960
[ 2191.982354] page:ef463180 count:-1 mapcount:0 mapping:  (null) index:0x0
[ 2191.989047] flags: 0x0()
[ 2191.991594] page dumped because: nonzero _count
[ 2191.996127] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.4.0-rc6 #114
[ 2192.002475] Hardware name: Rockchip (Device Tree)
[ 2192.007174] Backtrace: 
[ 2192.009658] [<c00134d4>] (dump_backtrace) from [<c0013680>] (show_stack+0x18/0x1c)
[ 2192.017220]  r7:c051c4f8 r6:ef463180 r5:c05b7000 r4:00000000
[ 2192.022948] [<c0013668>] (show_stack) from [<c0219d90>] (dump_stack+0x90/0xa0)
[ 2192.030176] [<c0219d00>] (dump_stack) from [<c00b2cd4>] (bad_page+0xdc/0x12c)
[ 2192.037302]  r5:c059a100 r4:c05f430c
[ 2192.040913] [<c00b2bf8>] (bad_page) from [<c00b606c>] (get_page_from_freelist+0x388/0x95c)
[ 2192.049166]  r9:00000008 r8:ef463180 r7:c051c4d0 r6:00000000 r5:00000000 r4:c051c4e4
[ 2192.056982] [<c00b5ce4>] (get_page_from_freelist) from [<c00b6880>] (__alloc_pages_nodemask+0xd8/0x8e8)
[ 2192.066362]  r10:c001b068 r9:00000000 r8:ee0b02b0 r7:60000113 r6:00000003 r5:02095220
[ 2192.074254]  r4:c05ca1c0
[ 2192.076809] [<c00b67a8>] (__alloc_pages_nodemask) from [<c00b7140>] (__alloc_page_frag+0xb0/0x160)
[ 2192.085757]  r10:c001b068 r9:00000000 r8:ee0b02b0 r7:60000113 r6:02080020 r5:00000740
[ 2192.093650]  r4:eedbc884
[ 2192.096207] [<c00b7090>] (__alloc_page_frag) from [<c03273b4>] (__netdev_alloc_skb+0xa0/0x104)
[ 2192.104806]  r7:60000113 r6:eedbc884 r5:ee0b0000 r4:00000740
[ 2192.110525] [<c0327314>] (__netdev_alloc_skb) from [<c02aac00>] (arc_emac_poll+0x318/0x57c)
[ 2192.118865]  r9:00000000 r8:ee0b02b0 r7:0000019c r6:ee163780 r5:00000670 r4:ee0b0000
[ 2192.126683] [<c02aa8e8>] (arc_emac_poll) from [<c0339ed8>] (net_rx_action+0x1f0/0x2ec)
[ 2192.134590]  r10:c0599df8 r9:c059a100 r8:00073760 r7:0000012c r6:00000028 r5:c02aa8e8
[ 2192.142483]  r4:ee0b04e0
[ 2192.145040] [<c0339ce8>] (net_rx_action) from [<c0026f5c>] (__do_softirq+0x134/0x258)
[ 2192.152860]  r10:c059a080 r9:40000003 r8:00000003 r7:00000100 r6:c0598000 r5:c059a08c
[ 2192.160751]  r4:00000000
[ 2192.163303] [<c0026e28>] (__do_softirq) from [<c0027344>] (irq_exit+0xb8/0x120)
[ 2192.170604]  r10:c04818a4 r9:c059a4a0 r8:ee808000 r7:00000000 r6:00000000 r5:0000001a
[ 2192.178496]  r4:c05943bc
[ 2192.181051] [<c002728c>] (irq_exit) from [<c00656a4>] (__handle_domain_irq+0x68/0xbc)
[ 2192.188871]  r5:0000001a r4:c05943bc
[ 2192.192479] [<c006563c>] (__handle_domain_irq) from [<c0009438>] (gic_handle_irq+0x40/0x78)
[ 2192.200818]  r9:c059a4a0 r8:f0803100 r7:f0802100 r6:c0599f00 r5:f080210c r4:c059a87c
[ 2192.208630] [<c00093f8>] (gic_handle_irq) from [<c0014214>] (__irq_svc+0x54/0x70)
[ 2192.216105] Exception stack(0xc0599f00 to 0xc0599f48)
[ 2192.221161] 9f00: 00000001 00000000 00000000 c001e940 c0598000 c059a454 00000000 c05ca291
[ 2192.229336] 9f20: 00000000 c059a4a0 c04818a4 c0599f5c c0599f60 c0599f50 c0010208 c001020c
[ 2192.237505] 9f40: 60000013 ffffffff
[ 2192.240990]  r9:c059a4a0 r8:00000000 r7:c0599f34 r6:ffffffff r5:60000013 r4:c001020c
[ 2192.248810] [<c00101cc>] (arch_cpu_idle) from [<c005bbec>] (default_idle_call+0x28/0x34)
[ 2192.256899] [<c005bbc4>] (default_idle_call) from [<c005be60>] (cpu_startup_entry+0x214/0x270)
[ 2192.265515] [<c005bc4c>] (cpu_startup_entry) from [<c047a6f4>] (rest_init+0x80/0x84)
[ 2192.273248]  r7:ffffffff
[ 2192.275808] [<c047a674>] (rest_init) from [<c055fca0>] (start_kernel+0x348/0x354)
[ 2192.283288] [<c055f958>] (start_kernel) from [<60008078>] (0x60008078)
[ 2192.289798] Disabling lock debugging due to kernel taint
[ 2192.300779] rockchip_emac 10204000.ethernet eth0: BUG! Tx Ring full when queue awake!

I'm glad to provide more information if needed.

Shuyu Wei
 		 	   		  

Re: rk3188 emac stuck at high load

[Caesar Wang]

Hi Shuyu,

Thanks your report!  Do you have fixed this issue?
I guess that caused by the cpufreq.

if you run 'cat sys/kernel/debug/clk/clk_summary' to check the emac 
parent clock, Is it the APLL?
Maybe you can change the parent clock to fix this issue if it's.

Fox example:
uboot：
https://github.com/rockchip-linux/u-boot/commit/d64ef6b272d84c92bd02a7925f500880633c8599

kernel:
https://patchwork.kernel.org/patch/8108231/


I try to install many somethings with ubuntu os on rk3036. (that's also 
working with the emac..)

.....
Get:35 http://ports.ubuntu.com/ubuntu-ports/ wily/main 
libpython2.7-minimal armhf 2.7.10-4ubuntu1 [338 kB]
Get:36 http://ports.ubuntu.com/ubuntu-ports/ wily/main 
libpython2.7-stdlib armhf 2.7.10-4ubuntu1 [1750 kB]
Get:37 http://ports.ubuntu.com/ubuntu-ports/ wily/main libpython2.7 
armhf 2.7.10-4ubuntu1 [935 kB]
Get:38 http://ports.ubuntu.com/ubuntu-ports/ wily/main vim-runtime all 
2:7.4.712-2ubuntu4 [4984 kB]
85% [38 vim-runtime 4405 kB/4984 kB 88%] 195 kB/s 11s[  314.685641] BUG: 
Bad page state in process swapper/0 pfn:7e0
[  314.691607] page:dffcee00 count:-1 mapcount:0 mapping:  (null) index:0x0
[  314.698311] flags: 0x0()
[  314.700866] page dumped because: nonzero _count
88% [38 vim-runtime 4738 kB/4984 kB 95%] 195 kB/s 9s[  316.096013] 
skbuff: skb_over_panic: text:c04ae19c len:2878 p>
[  316.108471] ------------[ cut here ]------------
[  316.113109] kernel BUG at 
/home/wxt/work/android/brillo/kernel/net/core/skbuff.c:102!
[  316.120938] Internal error: Oops - BUG: 0 [#1] SMP ARM
[  316.126079] Modules linked in:
[  316.129166] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G B           
4.4.0-rc8+ #35
[  316.136732] Hardware name: Rockchip (Device Tree)
[  316.141444] task: c0f0c0b0 ti: c0f00000 task.ti: c0f00000
[  316.146862] PC is at skb_panic+0x60/0x6c
[  316.150799] LR is at _raw_spin_unlock_irqrestore+0x1c/0x24
[  316.156291] pc : [<c093f694>]    lr : [<c0947fb0>] psr: 60000113
[  316.156291] sp : c0f01d88  ip : c0f01c08  fp : c0f01dbc
[  316.167761] r10: df718540  r9 : deabbcc0  r8 : 000003a0
[  316.172989] r7 : c0c3ba0a  r6 : debb0c00  r5 : debb0c64  r4 : debb17a2
[  316.179517] r3 : c0e933b8  r2 : 00000000  r1 : 60000113  r0 : 0000007d
...
[  316.372994] [<c093f694>] (skb_panic) from [<c07353b8>] 
(skb_put+0x54/0x60)
[  316.379888] [<c07353b8>] (skb_put) from [<c04ae19c>] 
(arc_emac_poll+0x26c/0x4bc)
[  316.387299] [<c04ae19c>] (arc_emac_poll) from [<c07456b4>] 
(net_rx_action+0xf4/0x308)
[  316.395146] [<c07456b4>] (net_rx_action) from [<c012365c>] 
(__do_softirq+0x10c/0x2d8)
[  316.402991] [<c012365c>] (__do_softirq) from [<c0123adc>] 
(irq_exit+0x94/0x104)
[  316.410313] [<c0123adc>] (irq_exit) from [<c016a778>] 
(__handle_domain_irq+0x98/0xbc)
[  316.418156] [<c016a778>] (__handle_domain_irq) from [<c010143c>] 
(gic_handle_irq+0x58/0xa0)
[  316.426514] [<c010143c>] (gic_handle_irq) from [<c010be94>] 
(__irq_svc+0x54/

在 2015年12月24日 23:52, Shuyu Wei 写道:
> Hi,
> I'm testing Radxa Rock Pro board with mainline kernel 4.4.0-rc6.
> The problem can be reproduced by simultaneously wget two large files at high speed to /dev/null.
>
> $wget https://192.168.1.1/file.bin -O /dev/null -q &
> $wget https://192.168.1.1/file.bin -O /dev/null
>
> or open two sftp session to download two large files to /dev/null.
>
> Some addtional information that might help:
>      Downloading from encrypted https links or sftp server triggers the problem much easier.
>      Disabling SMP in kernel seems to fix the problem.
>      Sometimes the emac just stops sending and receiving frames, sometimes it produces kernel panics like below.
>
> [ 2191.975729] BUG: Bad page state in process swapper/0  pfn:8d960
> [ 2191.982354] page:ef463180 count:-1 mapcount:0 mapping:  (null) index:0x0
> [ 2191.989047] flags: 0x0()
> [ 2191.991594] page dumped because: nonzero _count
> [ 2191.996127] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.4.0-rc6 #114
> [ 2192.002475] Hardware name: Rockchip (Device Tree)
> [ 2192.007174] Backtrace:
> [ 2192.009658] [<c00134d4>] (dump_backtrace) from [<c0013680>] (show_stack+0x18/0x1c)
> [ 2192.017220]  r7:c051c4f8 r6:ef463180 r5:c05b7000 r4:00000000
> [ 2192.022948] [<c0013668>] (show_stack) from [<c0219d90>] (dump_stack+0x90/0xa0)
> [ 2192.030176] [<c0219d00>] (dump_stack) from [<c00b2cd4>] (bad_page+0xdc/0x12c)
> [ 2192.037302]  r5:c059a100 r4:c05f430c
> [ 2192.040913] [<c00b2bf8>] (bad_page) from [<c00b606c>] (get_page_from_freelist+0x388/0x95c)
> [ 2192.049166]  r9:00000008 r8:ef463180 r7:c051c4d0 r6:00000000 r5:00000000 r4:c051c4e4
> [ 2192.056982] [<c00b5ce4>] (get_page_from_freelist) from [<c00b6880>] (__alloc_pages_nodemask+0xd8/0x8e8)
> [ 2192.066362]  r10:c001b068 r9:00000000 r8:ee0b02b0 r7:60000113 r6:00000003 r5:02095220
> [ 2192.074254]  r4:c05ca1c0
> [ 2192.076809] [<c00b67a8>] (__alloc_pages_nodemask) from [<c00b7140>] (__alloc_page_frag+0xb0/0x160)
> [ 2192.085757]  r10:c001b068 r9:00000000 r8:ee0b02b0 r7:60000113 r6:02080020 r5:00000740
> [ 2192.093650]  r4:eedbc884
> [ 2192.096207] [<c00b7090>] (__alloc_page_frag) from [<c03273b4>] (__netdev_alloc_skb+0xa0/0x104)
> [ 2192.104806]  r7:60000113 r6:eedbc884 r5:ee0b0000 r4:00000740
> [ 2192.110525] [<c0327314>] (__netdev_alloc_skb) from [<c02aac00>] (arc_emac_poll+0x318/0x57c)
> [ 2192.118865]  r9:00000000 r8:ee0b02b0 r7:0000019c r6:ee163780 r5:00000670 r4:ee0b0000
> [ 2192.126683] [<c02aa8e8>] (arc_emac_poll) from [<c0339ed8>] (net_rx_action+0x1f0/0x2ec)
> [ 2192.134590]  r10:c0599df8 r9:c059a100 r8:00073760 r7:0000012c r6:00000028 r5:c02aa8e8
> [ 2192.142483]  r4:ee0b04e0
> [ 2192.145040] [<c0339ce8>] (net_rx_action) from [<c0026f5c>] (__do_softirq+0x134/0x258)
> [ 2192.152860]  r10:c059a080 r9:40000003 r8:00000003 r7:00000100 r6:c0598000 r5:c059a08c
> [ 2192.160751]  r4:00000000
> [ 2192.163303] [<c0026e28>] (__do_softirq) from [<c0027344>] (irq_exit+0xb8/0x120)
> [ 2192.170604]  r10:c04818a4 r9:c059a4a0 r8:ee808000 r7:00000000 r6:00000000 r5:0000001a
> [ 2192.178496]  r4:c05943bc
> [ 2192.181051] [<c002728c>] (irq_exit) from [<c00656a4>] (__handle_domain_irq+0x68/0xbc)
> [ 2192.188871]  r5:0000001a r4:c05943bc
> [ 2192.192479] [<c006563c>] (__handle_domain_irq) from [<c0009438>] (gic_handle_irq+0x40/0x78)
> [ 2192.200818]  r9:c059a4a0 r8:f0803100 r7:f0802100 r6:c0599f00 r5:f080210c r4:c059a87c
> [ 2192.208630] [<c00093f8>] (gic_handle_irq) from [<c0014214>] (__irq_svc+0x54/0x70)
> [ 2192.216105] Exception stack(0xc0599f00 to 0xc0599f48)
> [ 2192.221161] 9f00: 00000001 00000000 00000000 c001e940 c0598000 c059a454 00000000 c05ca291
> [ 2192.229336] 9f20: 00000000 c059a4a0 c04818a4 c0599f5c c0599f60 c0599f50 c0010208 c001020c
> [ 2192.237505] 9f40: 60000013 ffffffff
> [ 2192.240990]  r9:c059a4a0 r8:00000000 r7:c0599f34 r6:ffffffff r5:60000013 r4:c001020c
> [ 2192.248810] [<c00101cc>] (arch_cpu_idle) from [<c005bbec>] (default_idle_call+0x28/0x34)
> [ 2192.256899] [<c005bbc4>] (default_idle_call) from [<c005be60>] (cpu_startup_entry+0x214/0x270)
> [ 2192.265515] [<c005bc4c>] (cpu_startup_entry) from [<c047a6f4>] (rest_init+0x80/0x84)
> [ 2192.273248]  r7:ffffffff
> [ 2192.275808] [<c047a674>] (rest_init) from [<c055fca0>] (start_kernel+0x348/0x354)
> [ 2192.283288] [<c055f958>] (start_kernel) from [<60008078>] (0x60008078)
> [ 2192.289798] Disabling lock debugging due to kernel taint
> [ 2192.300779] rockchip_emac 10204000.ethernet eth0: BUG! Tx Ring full when queue awake!
>
> I'm glad to provide more information if needed.
>
> Shuyu Wei
>   		 	   		
> _______________________________________________
> Linux-rockchip mailing list
> Linux-rockchip@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-rockchip


-- 
Thanks,
Caesar


_______________________________________________
Linux-rockchip mailing list
Linux-rockchip@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-rockchip

RE: rk3188 emac stuck at high load

[Shuyu Wei]

Hi Caesar,

Thanks for your reply, I thought I have fixed the issue by this patch:
http://lists.infradead.org/pipermail/linux-rockchip/2016-January/006412.html
Maybe I sent it to the wrong maintainer so it didn't get noticed.
I'll try your solution and report as soon as possible. 
----------------------------------------
> Subject: Re: rk3188 emac stuck at high load
> To: shuyu.w-1ViLX0X+lBJBDgjK7y7TUQ@public.gmane.org
> CC: linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org; heiko-4mtYJXux2i+zQB+pC5nmwQ@public.gmane.org
> From: caesar.upstream-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org
> Date: Mon, 25 Jan 2016 20:49:36 +0800
>
> Hi Shuyu,
>
> Thanks your report! Do you have fixed this issue?
> I guess that caused by the cpufreq.
>
> if you run 'cat sys/kernel/debug/clk/clk_summary' to check the emac
> parent clock, Is it the APLL?
> Maybe you can change the parent clock to fix this issue if it's.
>
> Fox example:
> uboot：
> https://github.com/rockchip-linux/u-boot/commit/d64ef6b272d84c92bd02a7925f500880633c8599
>
> kernel:
> https://patchwork.kernel.org/patch/8108231/
>
>
> I try to install many somethings with ubuntu os on rk3036. (that's also
> working with the emac..)
>
> .....
> Get:35 http://ports.ubuntu.com/ubuntu-ports/ wily/main
> libpython2.7-minimal armhf 2.7.10-4ubuntu1 [338 kB]
> Get:36 http://ports.ubuntu.com/ubuntu-ports/ wily/main
> libpython2.7-stdlib armhf 2.7.10-4ubuntu1 [1750 kB]
> Get:37 http://ports.ubuntu.com/ubuntu-ports/ wily/main libpython2.7
> armhf 2.7.10-4ubuntu1 [935 kB]
> Get:38 http://ports.ubuntu.com/ubuntu-ports/ wily/main vim-runtime all
> 2:7.4.712-2ubuntu4 [4984 kB]
> 85% [38 vim-runtime 4405 kB/4984 kB 88%] 195 kB/s 11s[ 314.685641] BUG:
> Bad page state in process swapper/0 pfn:7e0
> [ 314.691607] page:dffcee00 count:-1 mapcount:0 mapping: (null) index:0x0
> [ 314.698311] flags: 0x0()
> [ 314.700866] page dumped because: nonzero _count
> 88% [38 vim-runtime 4738 kB/4984 kB 95%] 195 kB/s 9s[ 316.096013]
> skbuff: skb_over_panic: text:c04ae19c len:2878 p>
> [ 316.108471] ------------[ cut here ]------------
> [ 316.113109] kernel BUG at
> /home/wxt/work/android/brillo/kernel/net/core/skbuff.c:102!
> [ 316.120938] Internal error: Oops - BUG: 0 [#1] SMP ARM
> [ 316.126079] Modules linked in:
> [ 316.129166] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G B
> 4.4.0-rc8+ #35
> [ 316.136732] Hardware name: Rockchip (Device Tree)
> [ 316.141444] task: c0f0c0b0 ti: c0f00000 task.ti: c0f00000
> [ 316.146862] PC is at skb_panic+0x60/0x6c
> [ 316.150799] LR is at _raw_spin_unlock_irqrestore+0x1c/0x24
> [ 316.156291] pc : [<c093f694>] lr : [<c0947fb0>] psr: 60000113
> [ 316.156291] sp : c0f01d88 ip : c0f01c08 fp : c0f01dbc
> [ 316.167761] r10: df718540 r9 : deabbcc0 r8 : 000003a0
> [ 316.172989] r7 : c0c3ba0a r6 : debb0c00 r5 : debb0c64 r4 : debb17a2
> [ 316.179517] r3 : c0e933b8 r2 : 00000000 r1 : 60000113 r0 : 0000007d
> ...
> [ 316.372994] [<c093f694>] (skb_panic) from [<c07353b8>]
> (skb_put+0x54/0x60)
> [ 316.379888] [<c07353b8>] (skb_put) from [<c04ae19c>]
> (arc_emac_poll+0x26c/0x4bc)
> [ 316.387299] [<c04ae19c>] (arc_emac_poll) from [<c07456b4>]
> (net_rx_action+0xf4/0x308)
> [ 316.395146] [<c07456b4>] (net_rx_action) from [<c012365c>]
> (__do_softirq+0x10c/0x2d8)
> [ 316.402991] [<c012365c>] (__do_softirq) from [<c0123adc>]
> (irq_exit+0x94/0x104)
> [ 316.410313] [<c0123adc>] (irq_exit) from [<c016a778>]
> (__handle_domain_irq+0x98/0xbc)
> [ 316.418156] [<c016a778>] (__handle_domain_irq) from [<c010143c>]
> (gic_handle_irq+0x58/0xa0)
> [ 316.426514] [<c010143c>] (gic_handle_irq) from [<c010be94>]
> (__irq_svc+0x54/
>
> 在 2015年12月24日 23:52, Shuyu Wei 写道:
>> Hi,
>> I'm testing Radxa Rock Pro board with mainline kernel 4.4.0-rc6.
>> The problem can be reproduced by simultaneously wget two large files at high speed to /dev/null.
>>
>> $wget https://192.168.1.1/file.bin -O /dev/null -q &
>> $wget https://192.168.1.1/file.bin -O /dev/null
>>
>> or open two sftp session to download two large files to /dev/null.
>>
>> Some addtional information that might help:
>> Downloading from encrypted https links or sftp server triggers the problem much easier.
>> Disabling SMP in kernel seems to fix the problem.
>> Sometimes the emac just stops sending and receiving frames, sometimes it produces kernel panics like below.
>>
>> [ 2191.975729] BUG: Bad page state in process swapper/0 pfn:8d960
>> [ 2191.982354] page:ef463180 count:-1 mapcount:0 mapping: (null) index:0x0
>> [ 2191.989047] flags: 0x0()
>> [ 2191.991594] page dumped because: nonzero _count
>> [ 2191.996127] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.4.0-rc6 #114
>> [ 2192.002475] Hardware name: Rockchip (Device Tree)
>> [ 2192.007174] Backtrace:
>> [ 2192.009658] [<c00134d4>] (dump_backtrace) from [<c0013680>] (show_stack+0x18/0x1c)
>> [ 2192.017220] r7:c051c4f8 r6:ef463180 r5:c05b7000 r4:00000000
>> [ 2192.022948] [<c0013668>] (show_stack) from [<c0219d90>] (dump_stack+0x90/0xa0)
>> [ 2192.030176] [<c0219d00>] (dump_stack) from [<c00b2cd4>] (bad_page+0xdc/0x12c)
>> [ 2192.037302] r5:c059a100 r4:c05f430c
>> [ 2192.040913] [<c00b2bf8>] (bad_page) from [<c00b606c>] (get_page_from_freelist+0x388/0x95c)
>> [ 2192.049166] r9:00000008 r8:ef463180 r7:c051c4d0 r6:00000000 r5:00000000 r4:c051c4e4
>> [ 2192.056982] [<c00b5ce4>] (get_page_from_freelist) from [<c00b6880>] (__alloc_pages_nodemask+0xd8/0x8e8)
>> [ 2192.066362] r10:c001b068 r9:00000000 r8:ee0b02b0 r7:60000113 r6:00000003 r5:02095220
>> [ 2192.074254] r4:c05ca1c0
>> [ 2192.076809] [<c00b67a8>] (__alloc_pages_nodemask) from [<c00b7140>] (__alloc_page_frag+0xb0/0x160)
>> [ 2192.085757] r10:c001b068 r9:00000000 r8:ee0b02b0 r7:60000113 r6:02080020 r5:00000740
>> [ 2192.093650] r4:eedbc884
>> [ 2192.096207] [<c00b7090>] (__alloc_page_frag) from [<c03273b4>] (__netdev_alloc_skb+0xa0/0x104)
>> [ 2192.104806] r7:60000113 r6:eedbc884 r5:ee0b0000 r4:00000740
>> [ 2192.110525] [<c0327314>] (__netdev_alloc_skb) from [<c02aac00>] (arc_emac_poll+0x318/0x57c)
>> [ 2192.118865] r9:00000000 r8:ee0b02b0 r7:0000019c r6:ee163780 r5:00000670 r4:ee0b0000
>> [ 2192.126683] [<c02aa8e8>] (arc_emac_poll) from [<c0339ed8>] (net_rx_action+0x1f0/0x2ec)
>> [ 2192.134590] r10:c0599df8 r9:c059a100 r8:00073760 r7:0000012c r6:00000028 r5:c02aa8e8
>> [ 2192.142483] r4:ee0b04e0
>> [ 2192.145040] [<c0339ce8>] (net_rx_action) from [<c0026f5c>] (__do_softirq+0x134/0x258)
>> [ 2192.152860] r10:c059a080 r9:40000003 r8:00000003 r7:00000100 r6:c0598000 r5:c059a08c
>> [ 2192.160751] r4:00000000
>> [ 2192.163303] [<c0026e28>] (__do_softirq) from [<c0027344>] (irq_exit+0xb8/0x120)
>> [ 2192.170604] r10:c04818a4 r9:c059a4a0 r8:ee808000 r7:00000000 r6:00000000 r5:0000001a
>> [ 2192.178496] r4:c05943bc
>> [ 2192.181051] [<c002728c>] (irq_exit) from [<c00656a4>] (__handle_domain_irq+0x68/0xbc)
>> [ 2192.188871] r5:0000001a r4:c05943bc
>> [ 2192.192479] [<c006563c>] (__handle_domain_irq) from [<c0009438>] (gic_handle_irq+0x40/0x78)
>> [ 2192.200818] r9:c059a4a0 r8:f0803100 r7:f0802100 r6:c0599f00 r5:f080210c r4:c059a87c
>> [ 2192.208630] [<c00093f8>] (gic_handle_irq) from [<c0014214>] (__irq_svc+0x54/0x70)
>> [ 2192.216105] Exception stack(0xc0599f00 to 0xc0599f48)
>> [ 2192.221161] 9f00: 00000001 00000000 00000000 c001e940 c0598000 c059a454 00000000 c05ca291
>> [ 2192.229336] 9f20: 00000000 c059a4a0 c04818a4 c0599f5c c0599f60 c0599f50 c0010208 c001020c
>> [ 2192.237505] 9f40: 60000013 ffffffff
>> [ 2192.240990] r9:c059a4a0 r8:00000000 r7:c0599f34 r6:ffffffff r5:60000013 r4:c001020c
>> [ 2192.248810] [<c00101cc>] (arch_cpu_idle) from [<c005bbec>] (default_idle_call+0x28/0x34)
>> [ 2192.256899] [<c005bbc4>] (default_idle_call) from [<c005be60>] (cpu_startup_entry+0x214/0x270)
>> [ 2192.265515] [<c005bc4c>] (cpu_startup_entry) from [<c047a6f4>] (rest_init+0x80/0x84)
>> [ 2192.273248] r7:ffffffff
>> [ 2192.275808] [<c047a674>] (rest_init) from [<c055fca0>] (start_kernel+0x348/0x354)
>> [ 2192.283288] [<c055f958>] (start_kernel) from [<60008078>] (0x60008078)
>> [ 2192.289798] Disabling lock debugging due to kernel taint
>> [ 2192.300779] rockchip_emac 10204000.ethernet eth0: BUG! Tx Ring full when queue awake!
>>
>> I'm glad to provide more information if needed.
>>
>> Shuyu Wei
>>
>> _______________________________________________
>> Linux-rockchip mailing list
>> Linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org
>> http://lists.infradead.org/mailman/listinfo/linux-rockchip
>
>
> --
> Thanks,
> Caesar
>
 		 	   		  

RE: rk3188 emac stuck at high load

[Shuyu Wei]

Hi Caesar,
I checked /sys/kernel/debug/clk/clk_summary, sclk_macref is under dpll.
Besides, the problem exists even if I build the kernel with cpufreq disabled.
Maybe my previous patch is right. Shall I resend it to get it merged?

----------------------------------------
> From: shuyu.w-1ViLX0X+lBJBDgjK7y7TUQ@public.gmane.org
> To: caesar.upstream-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org
> CC: linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org; heiko-4mtYJXux2i+zQB+pC5nmwQ@public.gmane.org
> Subject: RE: rk3188 emac stuck at high load
> Date: Mon, 25 Jan 2016 13:33:02 +0000
>
> Hi Caesar,
>
> Thanks for your reply, I thought I have fixed the issue by this patch:
> http://lists.infradead.org/pipermail/linux-rockchip/2016-January/006412.html
> Maybe I sent it to the wrong maintainer so it didn't get noticed.
> I'll try your solution and report as soon as possible.
> ----------------------------------------
>> Subject: Re: rk3188 emac stuck at high load
>> To: shuyu.w-1ViLX0X+lBJBDgjK7y7TUQ@public.gmane.org
>> CC: linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org; heiko-4mtYJXux2i+zQB+pC5nmwQ@public.gmane.org
>> From: caesar.upstream-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org
>> Date: Mon, 25 Jan 2016 20:49:36 +0800
>>
>> Hi Shuyu,
>>
>> Thanks your report! Do you have fixed this issue?
>> I guess that caused by the cpufreq.
>>
>> if you run 'cat sys/kernel/debug/clk/clk_summary' to check the emac
>> parent clock, Is it the APLL?
>> Maybe you can change the parent clock to fix this issue if it's.
>>
>> Fox example:
>> uboot：
>> https://github.com/rockchip-linux/u-boot/commit/d64ef6b272d84c92bd02a7925f500880633c8599
>>
>> kernel:
>> https://patchwork.kernel.org/patch/8108231/
>>
>>
>> I try to install many somethings with ubuntu os on rk3036. (that's also
>> working with the emac..)
>>
>> .....
>> Get:35 http://ports.ubuntu.com/ubuntu-ports/ wily/main
>> libpython2.7-minimal armhf 2.7.10-4ubuntu1 [338 kB]
>> Get:36 http://ports.ubuntu.com/ubuntu-ports/ wily/main
>> libpython2.7-stdlib armhf 2.7.10-4ubuntu1 [1750 kB]
>> Get:37 http://ports.ubuntu.com/ubuntu-ports/ wily/main libpython2.7
>> armhf 2.7.10-4ubuntu1 [935 kB]
>> Get:38 http://ports.ubuntu.com/ubuntu-ports/ wily/main vim-runtime all
>> 2:7.4.712-2ubuntu4 [4984 kB]
>> 85% [38 vim-runtime 4405 kB/4984 kB 88%] 195 kB/s 11s[ 314.685641] BUG:
>> Bad page state in process swapper/0 pfn:7e0
>> [ 314.691607] page:dffcee00 count:-1 mapcount:0 mapping: (null) index:0x0
>> [ 314.698311] flags: 0x0()
>> [ 314.700866] page dumped because: nonzero _count
>> 88% [38 vim-runtime 4738 kB/4984 kB 95%] 195 kB/s 9s[ 316.096013]
>> skbuff: skb_over_panic: text:c04ae19c len:2878 p>
>> [ 316.108471] ------------[ cut here ]------------
>> [ 316.113109] kernel BUG at
>> /home/wxt/work/android/brillo/kernel/net/core/skbuff.c:102!
>> [ 316.120938] Internal error: Oops - BUG: 0 [#1] SMP ARM
>> [ 316.126079] Modules linked in:
>> [ 316.129166] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G B
>> 4.4.0-rc8+ #35
>> [ 316.136732] Hardware name: Rockchip (Device Tree)
>> [ 316.141444] task: c0f0c0b0 ti: c0f00000 task.ti: c0f00000
>> [ 316.146862] PC is at skb_panic+0x60/0x6c
>> [ 316.150799] LR is at _raw_spin_unlock_irqrestore+0x1c/0x24
>> [ 316.156291] pc : [<c093f694>] lr : [<c0947fb0>] psr: 60000113
>> [ 316.156291] sp : c0f01d88 ip : c0f01c08 fp : c0f01dbc
>> [ 316.167761] r10: df718540 r9 : deabbcc0 r8 : 000003a0
>> [ 316.172989] r7 : c0c3ba0a r6 : debb0c00 r5 : debb0c64 r4 : debb17a2
>> [ 316.179517] r3 : c0e933b8 r2 : 00000000 r1 : 60000113 r0 : 0000007d
>> ...
>> [ 316.372994] [<c093f694>] (skb_panic) from [<c07353b8>]
>> (skb_put+0x54/0x60)
>> [ 316.379888] [<c07353b8>] (skb_put) from [<c04ae19c>]
>> (arc_emac_poll+0x26c/0x4bc)
>> [ 316.387299] [<c04ae19c>] (arc_emac_poll) from [<c07456b4>]
>> (net_rx_action+0xf4/0x308)
>> [ 316.395146] [<c07456b4>] (net_rx_action) from [<c012365c>]
>> (__do_softirq+0x10c/0x2d8)
>> [ 316.402991] [<c012365c>] (__do_softirq) from [<c0123adc>]
>> (irq_exit+0x94/0x104)
>> [ 316.410313] [<c0123adc>] (irq_exit) from [<c016a778>]
>> (__handle_domain_irq+0x98/0xbc)
>> [ 316.418156] [<c016a778>] (__handle_domain_irq) from [<c010143c>]
>> (gic_handle_irq+0x58/0xa0)
>> [ 316.426514] [<c010143c>] (gic_handle_irq) from [<c010be94>]
>> (__irq_svc+0x54/
>>
>> 在 2015年12月24日 23:52, Shuyu Wei 写道:
>>> Hi,
>>> I'm testing Radxa Rock Pro board with mainline kernel 4.4.0-rc6.
>>> The problem can be reproduced by simultaneously wget two large files at high speed to /dev/null.
>>>
>>> $wget https://192.168.1.1/file.bin -O /dev/null -q &
>>> $wget https://192.168.1.1/file.bin -O /dev/null
>>>
>>> or open two sftp session to download two large files to /dev/null.
>>>
>>> Some addtional information that might help:
>>> Downloading from encrypted https links or sftp server triggers the problem much easier.
>>> Disabling SMP in kernel seems to fix the problem.
>>> Sometimes the emac just stops sending and receiving frames, sometimes it produces kernel panics like below.
>>>
>>> [ 2191.975729] BUG: Bad page state in process swapper/0 pfn:8d960
>>> [ 2191.982354] page:ef463180 count:-1 mapcount:0 mapping: (null) index:0x0
>>> [ 2191.989047] flags: 0x0()
>>> [ 2191.991594] page dumped because: nonzero _count
>>> [ 2191.996127] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.4.0-rc6 #114
>>> [ 2192.002475] Hardware name: Rockchip (Device Tree)
>>> [ 2192.007174] Backtrace:
>>> [ 2192.009658] [<c00134d4>] (dump_backtrace) from [<c0013680>] (show_stack+0x18/0x1c)
>>> [ 2192.017220] r7:c051c4f8 r6:ef463180 r5:c05b7000 r4:00000000
>>> [ 2192.022948] [<c0013668>] (show_stack) from [<c0219d90>] (dump_stack+0x90/0xa0)
>>> [ 2192.030176] [<c0219d00>] (dump_stack) from [<c00b2cd4>] (bad_page+0xdc/0x12c)
>>> [ 2192.037302] r5:c059a100 r4:c05f430c
>>> [ 2192.040913] [<c00b2bf8>] (bad_page) from [<c00b606c>] (get_page_from_freelist+0x388/0x95c)
>>> [ 2192.049166] r9:00000008 r8:ef463180 r7:c051c4d0 r6:00000000 r5:00000000 r4:c051c4e4
>>> [ 2192.056982] [<c00b5ce4>] (get_page_from_freelist) from [<c00b6880>] (__alloc_pages_nodemask+0xd8/0x8e8)
>>> [ 2192.066362] r10:c001b068 r9:00000000 r8:ee0b02b0 r7:60000113 r6:00000003 r5:02095220
>>> [ 2192.074254] r4:c05ca1c0
>>> [ 2192.076809] [<c00b67a8>] (__alloc_pages_nodemask) from [<c00b7140>] (__alloc_page_frag+0xb0/0x160)
>>> [ 2192.085757] r10:c001b068 r9:00000000 r8:ee0b02b0 r7:60000113 r6:02080020 r5:00000740
>>> [ 2192.093650] r4:eedbc884
>>> [ 2192.096207] [<c00b7090>] (__alloc_page_frag) from [<c03273b4>] (__netdev_alloc_skb+0xa0/0x104)
>>> [ 2192.104806] r7:60000113 r6:eedbc884 r5:ee0b0000 r4:00000740
>>> [ 2192.110525] [<c0327314>] (__netdev_alloc_skb) from [<c02aac00>] (arc_emac_poll+0x318/0x57c)
>>> [ 2192.118865] r9:00000000 r8:ee0b02b0 r7:0000019c r6:ee163780 r5:00000670 r4:ee0b0000
>>> [ 2192.126683] [<c02aa8e8>] (arc_emac_poll) from [<c0339ed8>] (net_rx_action+0x1f0/0x2ec)
>>> [ 2192.134590] r10:c0599df8 r9:c059a100 r8:00073760 r7:0000012c r6:00000028 r5:c02aa8e8
>>> [ 2192.142483] r4:ee0b04e0
>>> [ 2192.145040] [<c0339ce8>] (net_rx_action) from [<c0026f5c>] (__do_softirq+0x134/0x258)
>>> [ 2192.152860] r10:c059a080 r9:40000003 r8:00000003 r7:00000100 r6:c0598000 r5:c059a08c
>>> [ 2192.160751] r4:00000000
>>> [ 2192.163303] [<c0026e28>] (__do_softirq) from [<c0027344>] (irq_exit+0xb8/0x120)
>>> [ 2192.170604] r10:c04818a4 r9:c059a4a0 r8:ee808000 r7:00000000 r6:00000000 r5:0000001a
>>> [ 2192.178496] r4:c05943bc
>>> [ 2192.181051] [<c002728c>] (irq_exit) from [<c00656a4>] (__handle_domain_irq+0x68/0xbc)
>>> [ 2192.188871] r5:0000001a r4:c05943bc
>>> [ 2192.192479] [<c006563c>] (__handle_domain_irq) from [<c0009438>] (gic_handle_irq+0x40/0x78)
>>> [ 2192.200818] r9:c059a4a0 r8:f0803100 r7:f0802100 r6:c0599f00 r5:f080210c r4:c059a87c
>>> [ 2192.208630] [<c00093f8>] (gic_handle_irq) from [<c0014214>] (__irq_svc+0x54/0x70)
>>> [ 2192.216105] Exception stack(0xc0599f00 to 0xc0599f48)
>>> [ 2192.221161] 9f00: 00000001 00000000 00000000 c001e940 c0598000 c059a454 00000000 c05ca291
>>> [ 2192.229336] 9f20: 00000000 c059a4a0 c04818a4 c0599f5c c0599f60 c0599f50 c0010208 c001020c
>>> [ 2192.237505] 9f40: 60000013 ffffffff
>>> [ 2192.240990] r9:c059a4a0 r8:00000000 r7:c0599f34 r6:ffffffff r5:60000013 r4:c001020c
>>> [ 2192.248810] [<c00101cc>] (arch_cpu_idle) from [<c005bbec>] (default_idle_call+0x28/0x34)
>>> [ 2192.256899] [<c005bbc4>] (default_idle_call) from [<c005be60>] (cpu_startup_entry+0x214/0x270)
>>> [ 2192.265515] [<c005bc4c>] (cpu_startup_entry) from [<c047a6f4>] (rest_init+0x80/0x84)
>>> [ 2192.273248] r7:ffffffff
>>> [ 2192.275808] [<c047a674>] (rest_init) from [<c055fca0>] (start_kernel+0x348/0x354)
>>> [ 2192.283288] [<c055f958>] (start_kernel) from [<60008078>] (0x60008078)
>>> [ 2192.289798] Disabling lock debugging due to kernel taint
>>> [ 2192.300779] rockchip_emac 10204000.ethernet eth0: BUG! Tx Ring full when queue awake!
>>>
>>> I'm glad to provide more information if needed.
>>>
>>> Shuyu Wei
>>>
>>> _______________________________________________
>>> Linux-rockchip mailing list
>>> Linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org
>>> http://lists.infradead.org/mailman/listinfo/linux-rockchip
>>
>>
>> --
>> Thanks,
>> Caesar
>>
>
 		 	   		  

Re: rk3188 emac stuck at high load

[Caesar Wang]

Hi Shuyu,


在 2016年01月26日 11:33, Shuyu Wei 写道:
> Hi Caesar,
> I checked /sys/kernel/debug/clk/clk_summary, sclk_macref is under dpll.
> Besides, the problem exists even if I build the kernel with cpufreq disabled.
> Maybe my previous patch is right. Shall I resend it to get it merged?

Thanks your tests. Yep, up to you.
Can you resend with cc the linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org?


> ----------------------------------------
>> From: shuyu.w-1ViLX0X+lBJBDgjK7y7TUQ@public.gmane.org
>> To: caesar.upstream-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org
>> CC: linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org; heiko-4mtYJXux2i+zQB+pC5nmwQ@public.gmane.org
>> Subject: RE: rk3188 emac stuck at high load
>> Date: Mon, 25 Jan 2016 13:33:02 +0000
>>
>> Hi Caesar,
>>
>> Thanks for your reply, I thought I have fixed the issue by this patch:
>> http://lists.infradead.org/pipermail/linux-rockchip/2016-January/006412.html
>> Maybe I sent it to the wrong maintainer so it didn't get noticed.
>> I'll try your solution and report as soon as possible.
>> ----------------------------------------
>>> Subject: Re: rk3188 emac stuck at high load
>>> To: shuyu.w-1ViLX0X+lBJBDgjK7y7TUQ@public.gmane.org
>>> CC: linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org; heiko-4mtYJXux2i+zQB+pC5nmwQ@public.gmane.org
>>> From: caesar.upstream-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org
>>> Date: Mon, 25 Jan 2016 20:49:36 +0800
>>>
>>> Hi Shuyu,
>>>
>>> Thanks your report! Do you have fixed this issue?
>>> I guess that caused by the cpufreq.
>>>
>>> if you run 'cat sys/kernel/debug/clk/clk_summary' to check the emac
>>> parent clock, Is it the APLL?
>>> Maybe you can change the parent clock to fix this issue if it's.
>>>
>>> Fox example:
>>> uboot：
>>> https://github.com/rockchip-linux/u-boot/commit/d64ef6b272d84c92bd02a7925f500880633c8599
>>>
>>> kernel:
>>> https://patchwork.kernel.org/patch/8108231/
>>>
>>>
>>> I try to install many somethings with ubuntu os on rk3036. (that's also
>>> working with the emac..)
>>>
>>> .....
>>> Get:35 http://ports.ubuntu.com/ubuntu-ports/ wily/main
>>> libpython2.7-minimal armhf 2.7.10-4ubuntu1 [338 kB]
>>> Get:36 http://ports.ubuntu.com/ubuntu-ports/ wily/main
>>> libpython2.7-stdlib armhf 2.7.10-4ubuntu1 [1750 kB]
>>> Get:37 http://ports.ubuntu.com/ubuntu-ports/ wily/main libpython2.7
>>> armhf 2.7.10-4ubuntu1 [935 kB]
>>> Get:38 http://ports.ubuntu.com/ubuntu-ports/ wily/main vim-runtime all
>>> 2:7.4.712-2ubuntu4 [4984 kB]
>>> 85% [38 vim-runtime 4405 kB/4984 kB 88%] 195 kB/s 11s[ 314.685641] BUG:
>>> Bad page state in process swapper/0 pfn:7e0
>>> [ 314.691607] page:dffcee00 count:-1 mapcount:0 mapping: (null) index:0x0
>>> [ 314.698311] flags: 0x0()
>>> [ 314.700866] page dumped because: nonzero _count
>>> 88% [38 vim-runtime 4738 kB/4984 kB 95%] 195 kB/s 9s[ 316.096013]
>>> skbuff: skb_over_panic: text:c04ae19c len:2878 p>
>>> [ 316.108471] ------------[ cut here ]------------
>>> [ 316.113109] kernel BUG at
>>> /home/wxt/work/android/brillo/kernel/net/core/skbuff.c:102!
>>> [ 316.120938] Internal error: Oops - BUG: 0 [#1] SMP ARM
>>> [ 316.126079] Modules linked in:
>>> [ 316.129166] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G B
>>> 4.4.0-rc8+ #35
>>> [ 316.136732] Hardware name: Rockchip (Device Tree)
>>> [ 316.141444] task: c0f0c0b0 ti: c0f00000 task.ti: c0f00000
>>> [ 316.146862] PC is at skb_panic+0x60/0x6c
>>> [ 316.150799] LR is at _raw_spin_unlock_irqrestore+0x1c/0x24
>>> [ 316.156291] pc : [<c093f694>] lr : [<c0947fb0>] psr: 60000113
>>> [ 316.156291] sp : c0f01d88 ip : c0f01c08 fp : c0f01dbc
>>> [ 316.167761] r10: df718540 r9 : deabbcc0 r8 : 000003a0
>>> [ 316.172989] r7 : c0c3ba0a r6 : debb0c00 r5 : debb0c64 r4 : debb17a2
>>> [ 316.179517] r3 : c0e933b8 r2 : 00000000 r1 : 60000113 r0 : 0000007d
>>> ...
>>> [ 316.372994] [<c093f694>] (skb_panic) from [<c07353b8>]
>>> (skb_put+0x54/0x60)
>>> [ 316.379888] [<c07353b8>] (skb_put) from [<c04ae19c>]
>>> (arc_emac_poll+0x26c/0x4bc)
>>> [ 316.387299] [<c04ae19c>] (arc_emac_poll) from [<c07456b4>]
>>> (net_rx_action+0xf4/0x308)
>>> [ 316.395146] [<c07456b4>] (net_rx_action) from [<c012365c>]
>>> (__do_softirq+0x10c/0x2d8)
>>> [ 316.402991] [<c012365c>] (__do_softirq) from [<c0123adc>]
>>> (irq_exit+0x94/0x104)
>>> [ 316.410313] [<c0123adc>] (irq_exit) from [<c016a778>]
>>> (__handle_domain_irq+0x98/0xbc)
>>> [ 316.418156] [<c016a778>] (__handle_domain_irq) from [<c010143c>]
>>> (gic_handle_irq+0x58/0xa0)
>>> [ 316.426514] [<c010143c>] (gic_handle_irq) from [<c010be94>]
>>> (__irq_svc+0x54/
>>>
>>> 在 2015年12月24日 23:52, Shuyu Wei 写道:
>>>> Hi,
>>>> I'm testing Radxa Rock Pro board with mainline kernel 4.4.0-rc6.
>>>> The problem can be reproduced by simultaneously wget two large files at high speed to /dev/null.
>>>>
>>>> $wget https://192.168.1.1/file.bin -O /dev/null -q &
>>>> $wget https://192.168.1.1/file.bin -O /dev/null
>>>>
>>>> or open two sftp session to download two large files to /dev/null.
>>>>
>>>> Some addtional information that might help:
>>>> Downloading from encrypted https links or sftp server triggers the problem much easier.
>>>> Disabling SMP in kernel seems to fix the problem.
>>>> Sometimes the emac just stops sending and receiving frames, sometimes it produces kernel panics like below.
>>>>
>>>> [ 2191.975729] BUG: Bad page state in process swapper/0 pfn:8d960
>>>> [ 2191.982354] page:ef463180 count:-1 mapcount:0 mapping: (null) index:0x0
>>>> [ 2191.989047] flags: 0x0()
>>>> [ 2191.991594] page dumped because: nonzero _count
>>>> [ 2191.996127] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.4.0-rc6 #114
>>>> [ 2192.002475] Hardware name: Rockchip (Device Tree)
>>>> [ 2192.007174] Backtrace:
>>>> [ 2192.009658] [<c00134d4>] (dump_backtrace) from [<c0013680>] (show_stack+0x18/0x1c)
>>>> [ 2192.017220] r7:c051c4f8 r6:ef463180 r5:c05b7000 r4:00000000
>>>> [ 2192.022948] [<c0013668>] (show_stack) from [<c0219d90>] (dump_stack+0x90/0xa0)
>>>> [ 2192.030176] [<c0219d00>] (dump_stack) from [<c00b2cd4>] (bad_page+0xdc/0x12c)
>>>> [ 2192.037302] r5:c059a100 r4:c05f430c
>>>> [ 2192.040913] [<c00b2bf8>] (bad_page) from [<c00b606c>] (get_page_from_freelist+0x388/0x95c)
>>>> [ 2192.049166] r9:00000008 r8:ef463180 r7:c051c4d0 r6:00000000 r5:00000000 r4:c051c4e4
>>>> [ 2192.056982] [<c00b5ce4>] (get_page_from_freelist) from [<c00b6880>] (__alloc_pages_nodemask+0xd8/0x8e8)
>>>> [ 2192.066362] r10:c001b068 r9:00000000 r8:ee0b02b0 r7:60000113 r6:00000003 r5:02095220
>>>> [ 2192.074254] r4:c05ca1c0
>>>> [ 2192.076809] [<c00b67a8>] (__alloc_pages_nodemask) from [<c00b7140>] (__alloc_page_frag+0xb0/0x160)
>>>> [ 2192.085757] r10:c001b068 r9:00000000 r8:ee0b02b0 r7:60000113 r6:02080020 r5:00000740
>>>> [ 2192.093650] r4:eedbc884
>>>> [ 2192.096207] [<c00b7090>] (__alloc_page_frag) from [<c03273b4>] (__netdev_alloc_skb+0xa0/0x104)
>>>> [ 2192.104806] r7:60000113 r6:eedbc884 r5:ee0b0000 r4:00000740
>>>> [ 2192.110525] [<c0327314>] (__netdev_alloc_skb) from [<c02aac00>] (arc_emac_poll+0x318/0x57c)
>>>> [ 2192.118865] r9:00000000 r8:ee0b02b0 r7:0000019c r6:ee163780 r5:00000670 r4:ee0b0000
>>>> [ 2192.126683] [<c02aa8e8>] (arc_emac_poll) from [<c0339ed8>] (net_rx_action+0x1f0/0x2ec)
>>>> [ 2192.134590] r10:c0599df8 r9:c059a100 r8:00073760 r7:0000012c r6:00000028 r5:c02aa8e8
>>>> [ 2192.142483] r4:ee0b04e0
>>>> [ 2192.145040] [<c0339ce8>] (net_rx_action) from [<c0026f5c>] (__do_softirq+0x134/0x258)
>>>> [ 2192.152860] r10:c059a080 r9:40000003 r8:00000003 r7:00000100 r6:c0598000 r5:c059a08c
>>>> [ 2192.160751] r4:00000000
>>>> [ 2192.163303] [<c0026e28>] (__do_softirq) from [<c0027344>] (irq_exit+0xb8/0x120)
>>>> [ 2192.170604] r10:c04818a4 r9:c059a4a0 r8:ee808000 r7:00000000 r6:00000000 r5:0000001a
>>>> [ 2192.178496] r4:c05943bc
>>>> [ 2192.181051] [<c002728c>] (irq_exit) from [<c00656a4>] (__handle_domain_irq+0x68/0xbc)
>>>> [ 2192.188871] r5:0000001a r4:c05943bc
>>>> [ 2192.192479] [<c006563c>] (__handle_domain_irq) from [<c0009438>] (gic_handle_irq+0x40/0x78)
>>>> [ 2192.200818] r9:c059a4a0 r8:f0803100 r7:f0802100 r6:c0599f00 r5:f080210c r4:c059a87c
>>>> [ 2192.208630] [<c00093f8>] (gic_handle_irq) from [<c0014214>] (__irq_svc+0x54/0x70)
>>>> [ 2192.216105] Exception stack(0xc0599f00 to 0xc0599f48)
>>>> [ 2192.221161] 9f00: 00000001 00000000 00000000 c001e940 c0598000 c059a454 00000000 c05ca291
>>>> [ 2192.229336] 9f20: 00000000 c059a4a0 c04818a4 c0599f5c c0599f60 c0599f50 c0010208 c001020c
>>>> [ 2192.237505] 9f40: 60000013 ffffffff
>>>> [ 2192.240990] r9:c059a4a0 r8:00000000 r7:c0599f34 r6:ffffffff r5:60000013 r4:c001020c
>>>> [ 2192.248810] [<c00101cc>] (arch_cpu_idle) from [<c005bbec>] (default_idle_call+0x28/0x34)
>>>> [ 2192.256899] [<c005bbc4>] (default_idle_call) from [<c005be60>] (cpu_startup_entry+0x214/0x270)
>>>> [ 2192.265515] [<c005bc4c>] (cpu_startup_entry) from [<c047a6f4>] (rest_init+0x80/0x84)
>>>> [ 2192.273248] r7:ffffffff
>>>> [ 2192.275808] [<c047a674>] (rest_init) from [<c055fca0>] (start_kernel+0x348/0x354)
>>>> [ 2192.283288] [<c055f958>] (start_kernel) from [<60008078>] (0x60008078)
>>>> [ 2192.289798] Disabling lock debugging due to kernel taint
>>>> [ 2192.300779] rockchip_emac 10204000.ethernet eth0: BUG! Tx Ring full when queue awake!
>>>>
>>>> I'm glad to provide more information if needed.
>>>>
>>>> Shuyu Wei
>>>>
>>>> _______________________________________________
>>>> Linux-rockchip mailing list
>>>> Linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org
>>>> http://lists.infradead.org/mailman/listinfo/linux-rockchip
>>>
>>> --
>>> Thanks,
>>> Caesar
>>>
>  		 	   		  
> _______________________________________________
> Linux-rockchip mailing list
> Linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org
> http://lists.infradead.org/mailman/listinfo/linux-rockchip


-- 
caesar wang | software engineer | wxt-TNX95d0MmH73oGB3hsPCZA@public.gmane.org