RX laggy/choppy/stuck without TX on Intel Wireless 7260 with iwlwifi

RX laggy/choppy/stuck without TX on Intel Wireless 7260 with iwlwifi

[Stephen Kent]

(Resending as plain text)

I have a strange issue with my Intel Wireless 7260 adapter on my Dell
XPS 13 (Haswell). It seems that packets are received but aren't
processed by the kernel or something until more packets are
transmitted. The net effect is that the laptop's wireless seems
somewhat choppy. I can demonstrate this by running something in a
remote ssh session that generates periodic output, and that output
doesn't appear until I generate more outgoing traffic. It's hard to
describe it beyond that, but I took a screen capture which
demonstrates this very well:
http://www.youtube.com/watch?v=4RMURSMszmc

lspci shows the wireless card as: 02:00.0 Network controller: Intel
Corporation Wireless 7260 (rev 6b)
uname -a: Linux dalek 3.11.0-12-generic #19-Ubuntu SMP Wed Oct 9
16:20:46UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

I googled this without success. I can gather diagnostic information if
you send me instructions, I'm just not sure where to start or what to
look for. Is this a bug in the driver? Maybe in the card microcode
itself? Has anyone else reported this issue? It seems very strange.
Any help is appreciated.

Thanks,

Stephen

RE: [Ilw] RX laggy/choppy/stuck without TX on Intel Wireless 7260 with iwlwifi

[Grumbach, Emmanuel]
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Re: [Ilw] RX laggy/choppy/stuck without TX on Intel Wireless 7260 with iwlwifi

[Stephen Kent]

On Sat, Feb 22, 2014 at 10:18 PM, Grumbach, Emmanuel
<emmanuel.grumbach@intel.com> wrote:
>>
>> (Resending as plain text)
>>
>> I have a strange issue with my Intel Wireless 7260 adapter on my Dell XPS 13
>> (Haswell). It seems that packets are received but aren't processed by the
>> kernel or something until more packets are transmitted. The net effect is
>> that the laptop's wireless seems somewhat choppy. I can demonstrate this
>> by running something in a remote ssh session that generates periodic
>> output, and that output doesn't appear until I generate more outgoing
>> traffic. It's hard to describe it beyond that, but I took a screen capture which
>> demonstrates this very well:
>> http://www.youtube.com/watch?v=4RMURSMszmc
>>
>> lspci shows the wireless card as: 02:00.0 Network controller: Intel Corporation
>> Wireless 7260 (rev 6b) uname -a: Linux dalek 3.11.0-12-generic #19-Ubuntu
>> SMP Wed Oct 9 16:20:46UTC 2013 x86_64 x86_64 x86_64 GNU/Linux
>>
>> I googled this without success. I can gather diagnostic information if you send
>> me instructions, I'm just not sure where to start or what to look for. Is this a
>> bug in the driver? Maybe in the card microcode itself? Has anyone else
>> reported this issue? It seems very strange.
>> Any help is appreciated.
>
> Can you please send the output of dmesg?
> Tracing output can help too - but let's start with dmesg output.
>

My full dmesg output is here: http://smkent.net/stuff/dmesg-dalek-20140222

Here's some selected output. I'm not sure if it's useful.

smkent@dalek Desktop $ grep -ie iwl -e 80211 dmesg-dalek-20140222
[    3.968070] cfg80211: Calling CRDA to update world regulatory domain
[    3.976500] iwlwifi 0000:02:00.0: can't disable ASPM; OS doesn't
have ASPM control
[    3.976585] iwlwifi 0000:02:00.0: irq 59 for MSI/MSI-X
[    4.115345] iwlwifi 0000:02:00.0: loaded firmware version 22.1.7.0
op_mode iwlmvm
[    4.157435] iwlwifi 0000:02:00.0: Detected Intel(R) Dual Band
Wireless AC 7260, REV=0x144
[    4.157523] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
[    4.157691] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
[    4.269377] cfg80211: World regulatory domain updated:
[    4.269383] cfg80211:   (start_freq - end_freq @ bandwidth),
(max_antenna_gain, max_eirp)
[    4.269387] cfg80211:   (2402000 KHz - 2472000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[    4.269390] cfg80211:   (2457000 KHz - 2482000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[    4.269393] cfg80211:   (2474000 KHz - 2494000 KHz @ 20000 KHz),
(300 mBi, 2000 mBm)
[    4.269396] cfg80211:   (5170000 KHz - 5250000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[    4.269399] cfg80211:   (5735000 KHz - 5835000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[    4.379283] ieee80211 phy0: Selected rate control algorithm 'iwl-mvm-rs'
[    4.709497] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
[    4.709671] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
[ 9676.817175] cfg80211: Calling CRDA to update world regulatory domain
[ 9676.823809] cfg80211: World regulatory domain updated:
[ 9676.823820] cfg80211:   (start_freq - end_freq @ bandwidth),
(max_antenna_gain, max_eirp)
[ 9676.823827] cfg80211:   (2402000 KHz - 2472000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[ 9676.823834] cfg80211:   (2457000 KHz - 2482000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[ 9676.823840] cfg80211:   (2474000 KHz - 2494000 KHz @ 20000 KHz),
(300 mBi, 2000 mBm)
[ 9676.823846] cfg80211:   (5170000 KHz - 5250000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[ 9676.823851] cfg80211:   (5735000 KHz - 5835000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[ 9680.014722] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
[ 9680.014890] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S

I haven't taken a kernel trace before but I'm happy to do so if you
can point me toward some instructions.

Thanks,

Stephen

RE: [Ilw] RX laggy/choppy/stuck without TX on Intel Wireless 7260 with iwlwifi

[Grumbach, Emmanuel]

PiANCj4gT24gU2F0LCBGZWIgMjIsIDIwMTQgYXQgMTA6MTggUE0sIEdydW1iYWNoLCBFbW1hbnVl
bA0KPiA8ZW1tYW51ZWwuZ3J1bWJhY2hAaW50ZWwuY29tPiB3cm90ZToNCj4gPj4NCj4gPj4gKFJl
c2VuZGluZyBhcyBwbGFpbiB0ZXh0KQ0KPiA+Pg0KPiA+PiBJIGhhdmUgYSBzdHJhbmdlIGlzc3Vl
IHdpdGggbXkgSW50ZWwgV2lyZWxlc3MgNzI2MCBhZGFwdGVyIG9uIG15IERlbGwgWFBTDQo+IDEz
DQo+ID4+IChIYXN3ZWxsKS4gSXQgc2VlbXMgdGhhdCBwYWNrZXRzIGFyZSByZWNlaXZlZCBidXQg
YXJlbid0IHByb2Nlc3NlZCBieSB0aGUNCj4gPj4ga2VybmVsIG9yIHNvbWV0aGluZyB1bnRpbCBt
b3JlIHBhY2tldHMgYXJlIHRyYW5zbWl0dGVkLiBUaGUgbmV0IGVmZmVjdCBpcw0KPiA+PiB0aGF0
IHRoZSBsYXB0b3AncyB3aXJlbGVzcyBzZWVtcyBzb21ld2hhdCBjaG9wcHkuIEkgY2FuIGRlbW9u
c3RyYXRlDQo+IHRoaXMNCj4gPj4gYnkgcnVubmluZyBzb21ldGhpbmcgaW4gYSByZW1vdGUgc3No
IHNlc3Npb24gdGhhdCBnZW5lcmF0ZXMgcGVyaW9kaWMNCj4gPj4gb3V0cHV0LCBhbmQgdGhhdCBv
dXRwdXQgZG9lc24ndCBhcHBlYXIgdW50aWwgSSBnZW5lcmF0ZSBtb3JlIG91dGdvaW5nDQo+ID4+
IHRyYWZmaWMuIEl0J3MgaGFyZCB0byBkZXNjcmliZSBpdCBiZXlvbmQgdGhhdCwgYnV0IEkgdG9v
ayBhIHNjcmVlbiBjYXB0dXJlDQo+IHdoaWNoDQo+ID4+IGRlbW9uc3RyYXRlcyB0aGlzIHZlcnkg
d2VsbDoNCj4gPj4gaHR0cDovL3d3dy55b3V0dWJlLmNvbS93YXRjaD92PTRSTVVSU01zem1jDQo+
ID4+DQo+ID4+IGxzcGNpIHNob3dzIHRoZSB3aXJlbGVzcyBjYXJkIGFzOiAwMjowMC4wIE5ldHdv
cmsgY29udHJvbGxlcjogSW50ZWwNCj4gQ29ycG9yYXRpb24NCj4gPj4gV2lyZWxlc3MgNzI2MCAo
cmV2IDZiKSB1bmFtZSAtYTogTGludXggZGFsZWsgMy4xMS4wLTEyLWdlbmVyaWMgIzE5LQ0KPiBV
YnVudHUNCj4gPj4gU01QIFdlZCBPY3QgOSAxNjoyMDo0NlVUQyAyMDEzIHg4Nl82NCB4ODZfNjQg
eDg2XzY0IEdOVS9MaW51eA0KPiA+Pg0KPiA+PiBJIGdvb2dsZWQgdGhpcyB3aXRob3V0IHN1Y2Nl
c3MuIEkgY2FuIGdhdGhlciBkaWFnbm9zdGljIGluZm9ybWF0aW9uIGlmIHlvdQ0KPiBzZW5kDQo+
ID4+IG1lIGluc3RydWN0aW9ucywgSSdtIGp1c3Qgbm90IHN1cmUgd2hlcmUgdG8gc3RhcnQgb3Ig
d2hhdCB0byBsb29rIGZvci4gSXMgdGhpcw0KPiBhDQo+ID4+IGJ1ZyBpbiB0aGUgZHJpdmVyPyBN
YXliZSBpbiB0aGUgY2FyZCBtaWNyb2NvZGUgaXRzZWxmPyBIYXMgYW55b25lIGVsc2UNCj4gPj4g
cmVwb3J0ZWQgdGhpcyBpc3N1ZT8gSXQgc2VlbXMgdmVyeSBzdHJhbmdlLg0KPiA+PiBBbnkgaGVs
cCBpcyBhcHByZWNpYXRlZC4NCj4gPg0KPiA+IENhbiB5b3UgcGxlYXNlIHNlbmQgdGhlIG91dHB1
dCBvZiBkbWVzZz8NCj4gPiBUcmFjaW5nIG91dHB1dCBjYW4gaGVscCB0b28gLSBidXQgbGV0J3Mg
c3RhcnQgd2l0aCBkbWVzZyBvdXRwdXQuDQo+ID4NCj4gDQo+IE15IGZ1bGwgZG1lc2cgb3V0cHV0
IGlzIGhlcmU6IGh0dHA6Ly9zbWtlbnQubmV0L3N0dWZmL2RtZXNnLWRhbGVrLQ0KPiAyMDE0MDIy
Mg0KPiANCg0KWW91ciBkbWVzZyBsb29rcyBoZWFsdGh5DQoNCj4gSGVyZSdzIHNvbWUgc2VsZWN0
ZWQgb3V0cHV0LiBJJ20gbm90IHN1cmUgaWYgaXQncyB1c2VmdWwuDQo+IA0KPiBzbWtlbnRAZGFs
ZWsgRGVza3RvcCAkIGdyZXAgLWllIGl3bCAtZSA4MDIxMSBkbWVzZy1kYWxlay0yMDE0MDIyMg0K
PiBbICAgIDMuOTY4MDcwXSBjZmc4MDIxMTogQ2FsbGluZyBDUkRBIHRvIHVwZGF0ZSB3b3JsZCBy
ZWd1bGF0b3J5IGRvbWFpbg0KPiBbICAgIDMuOTc2NTAwXSBpd2x3aWZpIDAwMDA6MDI6MDAuMDog
Y2FuJ3QgZGlzYWJsZSBBU1BNOyBPUyBkb2Vzbid0DQo+IGhhdmUgQVNQTSBjb250cm9sDQo+IFsg
ICAgMy45NzY1ODVdIGl3bHdpZmkgMDAwMDowMjowMC4wOiBpcnEgNTkgZm9yIE1TSS9NU0ktWA0K
PiBbICAgIDQuMTE1MzQ1XSBpd2x3aWZpIDAwMDA6MDI6MDAuMDogbG9hZGVkIGZpcm13YXJlIHZl
cnNpb24gMjIuMS43LjANCj4gb3BfbW9kZSBpd2xtdm0NCj4gWyAgICA0LjE1NzQzNV0gaXdsd2lm
aSAwMDAwOjAyOjAwLjA6IERldGVjdGVkIEludGVsKFIpIER1YWwgQmFuZA0KPiBXaXJlbGVzcyBB
QyA3MjYwLCBSRVY9MHgxNDQNCj4gWyAgICA0LjE1NzUyM10gaXdsd2lmaSAwMDAwOjAyOjAwLjA6
IEwxIEVuYWJsZWQ7IERpc2FibGluZyBMMFMNCj4gWyAgICA0LjE1NzY5MV0gaXdsd2lmaSAwMDAw
OjAyOjAwLjA6IEwxIEVuYWJsZWQ7IERpc2FibGluZyBMMFMNCj4gWyAgICA0LjI2OTM3N10gY2Zn
ODAyMTE6IFdvcmxkIHJlZ3VsYXRvcnkgZG9tYWluIHVwZGF0ZWQ6DQo+IFsgICAgNC4yNjkzODNd
IGNmZzgwMjExOiAgIChzdGFydF9mcmVxIC0gZW5kX2ZyZXEgQCBiYW5kd2lkdGgpLA0KPiAobWF4
X2FudGVubmFfZ2FpbiwgbWF4X2VpcnApDQo+IFsgICAgNC4yNjkzODddIGNmZzgwMjExOiAgICgy
NDAyMDAwIEtIeiAtIDI0NzIwMDAgS0h6IEAgNDAwMDAgS0h6KSwNCj4gKDMwMCBtQmksIDIwMDAg
bUJtKQ0KPiBbICAgIDQuMjY5MzkwXSBjZmc4MDIxMTogICAoMjQ1NzAwMCBLSHogLSAyNDgyMDAw
IEtIeiBAIDQwMDAwIEtIeiksDQo+ICgzMDAgbUJpLCAyMDAwIG1CbSkNCj4gWyAgICA0LjI2OTM5
M10gY2ZnODAyMTE6ICAgKDI0NzQwMDAgS0h6IC0gMjQ5NDAwMCBLSHogQCAyMDAwMCBLSHopLA0K
PiAoMzAwIG1CaSwgMjAwMCBtQm0pDQo+IFsgICAgNC4yNjkzOTZdIGNmZzgwMjExOiAgICg1MTcw
MDAwIEtIeiAtIDUyNTAwMDAgS0h6IEAgNDAwMDAgS0h6KSwNCj4gKDMwMCBtQmksIDIwMDAgbUJt
KQ0KPiBbICAgIDQuMjY5Mzk5XSBjZmc4MDIxMTogICAoNTczNTAwMCBLSHogLSA1ODM1MDAwIEtI
eiBAIDQwMDAwIEtIeiksDQo+ICgzMDAgbUJpLCAyMDAwIG1CbSkNCj4gWyAgICA0LjM3OTI4M10g
aWVlZTgwMjExIHBoeTA6IFNlbGVjdGVkIHJhdGUgY29udHJvbCBhbGdvcml0aG0gJ2l3bC1tdm0t
cnMnDQo+IFsgICAgNC43MDk0OTddIGl3bHdpZmkgMDAwMDowMjowMC4wOiBMMSBFbmFibGVkOyBE
aXNhYmxpbmcgTDBTDQo+IFsgICAgNC43MDk2NzFdIGl3bHdpZmkgMDAwMDowMjowMC4wOiBMMSBF
bmFibGVkOyBEaXNhYmxpbmcgTDBTDQo+IFsgOTY3Ni44MTcxNzVdIGNmZzgwMjExOiBDYWxsaW5n
IENSREEgdG8gdXBkYXRlIHdvcmxkIHJlZ3VsYXRvcnkgZG9tYWluDQo+IFsgOTY3Ni44MjM4MDld
IGNmZzgwMjExOiBXb3JsZCByZWd1bGF0b3J5IGRvbWFpbiB1cGRhdGVkOg0KPiBbIDk2NzYuODIz
ODIwXSBjZmc4MDIxMTogICAoc3RhcnRfZnJlcSAtIGVuZF9mcmVxIEAgYmFuZHdpZHRoKSwNCj4g
KG1heF9hbnRlbm5hX2dhaW4sIG1heF9laXJwKQ0KPiBbIDk2NzYuODIzODI3XSBjZmc4MDIxMTog
ICAoMjQwMjAwMCBLSHogLSAyNDcyMDAwIEtIeiBAIDQwMDAwIEtIeiksDQo+ICgzMDAgbUJpLCAy
MDAwIG1CbSkNCj4gWyA5Njc2LjgyMzgzNF0gY2ZnODAyMTE6ICAgKDI0NTcwMDAgS0h6IC0gMjQ4
MjAwMCBLSHogQCA0MDAwMCBLSHopLA0KPiAoMzAwIG1CaSwgMjAwMCBtQm0pDQo+IFsgOTY3Ni44
MjM4NDBdIGNmZzgwMjExOiAgICgyNDc0MDAwIEtIeiAtIDI0OTQwMDAgS0h6IEAgMjAwMDAgS0h6
KSwNCj4gKDMwMCBtQmksIDIwMDAgbUJtKQ0KPiBbIDk2NzYuODIzODQ2XSBjZmc4MDIxMTogICAo
NTE3MDAwMCBLSHogLSA1MjUwMDAwIEtIeiBAIDQwMDAwIEtIeiksDQo+ICgzMDAgbUJpLCAyMDAw
IG1CbSkNCj4gWyA5Njc2LjgyMzg1MV0gY2ZnODAyMTE6ICAgKDU3MzUwMDAgS0h6IC0gNTgzNTAw
MCBLSHogQCA0MDAwMCBLSHopLA0KPiAoMzAwIG1CaSwgMjAwMCBtQm0pDQo+IFsgOTY4MC4wMTQ3
MjJdIGl3bHdpZmkgMDAwMDowMjowMC4wOiBMMSBFbmFibGVkOyBEaXNhYmxpbmcgTDBTDQo+IFsg
OTY4MC4wMTQ4OTBdIGl3bHdpZmkgMDAwMDowMjowMC4wOiBMMSBFbmFibGVkOyBEaXNhYmxpbmcg
TDBTDQo+IA0KPiBJIGhhdmVuJ3QgdGFrZW4gYSBrZXJuZWwgdHJhY2UgYmVmb3JlIGJ1dCBJJ20g
aGFwcHkgdG8gZG8gc28gaWYgeW91DQo+IGNhbiBwb2ludCBtZSB0b3dhcmQgc29tZSBpbnN0cnVj
dGlvbnMuDQo+IA0KDQphcyByb290Og0KdHJhY2UtY21kIHJlY29yZCAtZSBpd2x3aWZpIC1lIGl3
bHdpZmlfbXNnLg0KDQpSZXByb2R1Y2UgdGhlIGJ1ZyB3aGlsZSBpdCdzIHJ1bm5pbmcuIFN0b3Ag
d2l0aCBDdHJsK0Mgd2hlbiB0aGUgYnVnIHJlcHJvZHVjZWQuDQpUaGlzIHdpbGwgcHJvZHVjZSBh
IHRyYWNlLmRhdCBmaWxlIHdoaWNoIHlvdSBjYW4gY29tcHJlc3MgYW5kIHNlbmQgb3ZlciBlbWFp
bC4NCg0KVGhhbmtzIQ0KDQo+IFRoYW5rcywNCj4gDQo+IFN0ZXBoZW4NCg==

Re: [Ilw] RX laggy/choppy/stuck without TX on Intel Wireless 7260 with iwlwifi

[Emmanuel Grumbach]

On 02/23/2014 09:02 AM, Grumbach, Emmanuel wrote:
>>
>> On Sat, Feb 22, 2014 at 10:18 PM, Grumbach, Emmanuel
>> <emmanuel.grumbach@intel.com> wrote:
>>>>
>>>> (Resending as plain text)
>>>>
>>>> I have a strange issue with my Intel Wireless 7260 adapter on my Dell XPS
>> 13
>>>> (Haswell). It seems that packets are received but aren't processed by the
>>>> kernel or something until more packets are transmitted. The net effect is
>>>> that the laptop's wireless seems somewhat choppy. I can demonstrate
>> this
>>>> by running something in a remote ssh session that generates periodic
>>>> output, and that output doesn't appear until I generate more outgoing
>>>> traffic. It's hard to describe it beyond that, but I took a screen capture
>> which
>>>> demonstrates this very well:
>>>> http://www.youtube.com/watch?v=4RMURSMszmc
>>>>
>>>> lspci shows the wireless card as: 02:00.0 Network controller: Intel
>> Corporation
>>>> Wireless 7260 (rev 6b) uname -a: Linux dalek 3.11.0-12-generic #19-
>> Ubuntu
>>>> SMP Wed Oct 9 16:20:46UTC 2013 x86_64 x86_64 x86_64 GNU/Linux
>>>>
>>>> I googled this without success. I can gather diagnostic information if you
>> send
>>>> me instructions, I'm just not sure where to start or what to look for. Is this
>> a
>>>> bug in the driver? Maybe in the card microcode itself? Has anyone else
>>>> reported this issue? It seems very strange.
>>>> Any help is appreciated.
>>>
>>> Can you please send the output of dmesg?
>>> Tracing output can help too - but let's start with dmesg output.
>>>
>>
>> My full dmesg output is here: http://smkent.net/stuff/dmesg-dalek-
>> 20140222
>>
> 
> Your dmesg looks healthy
> 
>> Here's some selected output. I'm not sure if it's useful.
>>
>> smkent@dalek Desktop $ grep -ie iwl -e 80211 dmesg-dalek-20140222
>> [    3.968070] cfg80211: Calling CRDA to update world regulatory domain
>> [    3.976500] iwlwifi 0000:02:00.0: can't disable ASPM; OS doesn't
>> have ASPM control
>> [    3.976585] iwlwifi 0000:02:00.0: irq 59 for MSI/MSI-X
>> [    4.115345] iwlwifi 0000:02:00.0: loaded firmware version 22.1.7.0
>> op_mode iwlmvm
>> [    4.157435] iwlwifi 0000:02:00.0: Detected Intel(R) Dual Band
>> Wireless AC 7260, REV=0x144
>> [    4.157523] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
>> [    4.157691] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
>> [    4.269377] cfg80211: World regulatory domain updated:
>> [    4.269383] cfg80211:   (start_freq - end_freq @ bandwidth),
>> (max_antenna_gain, max_eirp)
>> [    4.269387] cfg80211:   (2402000 KHz - 2472000 KHz @ 40000 KHz),
>> (300 mBi, 2000 mBm)
>> [    4.269390] cfg80211:   (2457000 KHz - 2482000 KHz @ 40000 KHz),
>> (300 mBi, 2000 mBm)
>> [    4.269393] cfg80211:   (2474000 KHz - 2494000 KHz @ 20000 KHz),
>> (300 mBi, 2000 mBm)
>> [    4.269396] cfg80211:   (5170000 KHz - 5250000 KHz @ 40000 KHz),
>> (300 mBi, 2000 mBm)
>> [    4.269399] cfg80211:   (5735000 KHz - 5835000 KHz @ 40000 KHz),
>> (300 mBi, 2000 mBm)
>> [    4.379283] ieee80211 phy0: Selected rate control algorithm 'iwl-mvm-rs'
>> [    4.709497] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
>> [    4.709671] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
>> [ 9676.817175] cfg80211: Calling CRDA to update world regulatory domain
>> [ 9676.823809] cfg80211: World regulatory domain updated:
>> [ 9676.823820] cfg80211:   (start_freq - end_freq @ bandwidth),
>> (max_antenna_gain, max_eirp)
>> [ 9676.823827] cfg80211:   (2402000 KHz - 2472000 KHz @ 40000 KHz),
>> (300 mBi, 2000 mBm)
>> [ 9676.823834] cfg80211:   (2457000 KHz - 2482000 KHz @ 40000 KHz),
>> (300 mBi, 2000 mBm)
>> [ 9676.823840] cfg80211:   (2474000 KHz - 2494000 KHz @ 20000 KHz),
>> (300 mBi, 2000 mBm)
>> [ 9676.823846] cfg80211:   (5170000 KHz - 5250000 KHz @ 40000 KHz),
>> (300 mBi, 2000 mBm)
>> [ 9676.823851] cfg80211:   (5735000 KHz - 5835000 KHz @ 40000 KHz),
>> (300 mBi, 2000 mBm)
>> [ 9680.014722] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
>> [ 9680.014890] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
>>
>> I haven't taken a kernel trace before but I'm happy to do so if you
>> can point me toward some instructions.
>>
> 
> as root:
> trace-cmd record -e iwlwifi -e iwlwifi_msg.
> 
> Reproduce the bug while it's running. Stop with Ctrl+C when the bug reproduced.
> This will produce a trace.dat file which you can compress and send over email.
> 

Could you please try to reproduce with power save disabled?

options iwlmvm power_scheme=1

Thanks.

Re: [Ilw] RX laggy/choppy/stuck without TX on Intel Wireless 7260 with iwlwifi

[Stephen Kent]

On Sun, Feb 23, 2014 at 12:02 PM, Emmanuel Grumbach <egrumbach@gmail.com> wrote:
>
>
>
> On 02/23/2014 09:02 AM, Grumbach, Emmanuel wrote:
> >>
> >> On Sat, Feb 22, 2014 at 10:18 PM, Grumbach, Emmanuel
> >> <emmanuel.grumbach@intel.com> wrote:
> >>>>
> >>>> (Resending as plain text)
> >>>>
> >>>> I have a strange issue with my Intel Wireless 7260 adapter on my Dell XPS
> >> 13
> >>>> (Haswell). It seems that packets are received but aren't processed by the
> >>>> kernel or something until more packets are transmitted. The net effect is
> >>>> that the laptop's wireless seems somewhat choppy. I can demonstrate
> >> this
> >>>> by running something in a remote ssh session that generates periodic
> >>>> output, and that output doesn't appear until I generate more outgoing
> >>>> traffic. It's hard to describe it beyond that, but I took a screen capture
> >> which
> >>>> demonstrates this very well:
> >>>> http://www.youtube.com/watch?v=4RMURSMszmc
> >>>>
> >>>> lspci shows the wireless card as: 02:00.0 Network controller: Intel
> >> Corporation
> >>>> Wireless 7260 (rev 6b) uname -a: Linux dalek 3.11.0-12-generic #19-
> >> Ubuntu
> >>>> SMP Wed Oct 9 16:20:46UTC 2013 x86_64 x86_64 x86_64 GNU/Linux
> >>>>
> >>>> I googled this without success. I can gather diagnostic information if you
> >> send
> >>>> me instructions, I'm just not sure where to start or what to look for. Is this
> >> a
> >>>> bug in the driver? Maybe in the card microcode itself? Has anyone else
> >>>> reported this issue? It seems very strange.
> >>>> Any help is appreciated.
> >>>
> >>> Can you please send the output of dmesg?
> >>> Tracing output can help too - but let's start with dmesg output.
> >>>
> >>
> >> My full dmesg output is here: http://smkent.net/stuff/dmesg-dalek-
> >> 20140222
> >>
> >
> > Your dmesg looks healthy
> >
> >> Here's some selected output. I'm not sure if it's useful.
> >>
> >> smkent@dalek Desktop $ grep -ie iwl -e 80211 dmesg-dalek-20140222
> >> [    3.968070] cfg80211: Calling CRDA to update world regulatory domain
> >> [    3.976500] iwlwifi 0000:02:00.0: can't disable ASPM; OS doesn't
> >> have ASPM control
> >> [    3.976585] iwlwifi 0000:02:00.0: irq 59 for MSI/MSI-X
> >> [    4.115345] iwlwifi 0000:02:00.0: loaded firmware version 22.1.7.0
> >> op_mode iwlmvm
> >> [    4.157435] iwlwifi 0000:02:00.0: Detected Intel(R) Dual Band
> >> Wireless AC 7260, REV=0x144
> >> [    4.157523] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
> >> [    4.157691] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
> >> [    4.269377] cfg80211: World regulatory domain updated:
> >> [    4.269383] cfg80211:   (start_freq - end_freq @ bandwidth),
> >> (max_antenna_gain, max_eirp)
> >> [    4.269387] cfg80211:   (2402000 KHz - 2472000 KHz @ 40000 KHz),
> >> (300 mBi, 2000 mBm)
> >> [    4.269390] cfg80211:   (2457000 KHz - 2482000 KHz @ 40000 KHz),
> >> (300 mBi, 2000 mBm)
> >> [    4.269393] cfg80211:   (2474000 KHz - 2494000 KHz @ 20000 KHz),
> >> (300 mBi, 2000 mBm)
> >> [    4.269396] cfg80211:   (5170000 KHz - 5250000 KHz @ 40000 KHz),
> >> (300 mBi, 2000 mBm)
> >> [    4.269399] cfg80211:   (5735000 KHz - 5835000 KHz @ 40000 KHz),
> >> (300 mBi, 2000 mBm)
> >> [    4.379283] ieee80211 phy0: Selected rate control algorithm 'iwl-mvm-rs'
> >> [    4.709497] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
> >> [    4.709671] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
> >> [ 9676.817175] cfg80211: Calling CRDA to update world regulatory domain
> >> [ 9676.823809] cfg80211: World regulatory domain updated:
> >> [ 9676.823820] cfg80211:   (start_freq - end_freq @ bandwidth),
> >> (max_antenna_gain, max_eirp)
> >> [ 9676.823827] cfg80211:   (2402000 KHz - 2472000 KHz @ 40000 KHz),
> >> (300 mBi, 2000 mBm)
> >> [ 9676.823834] cfg80211:   (2457000 KHz - 2482000 KHz @ 40000 KHz),
> >> (300 mBi, 2000 mBm)
> >> [ 9676.823840] cfg80211:   (2474000 KHz - 2494000 KHz @ 20000 KHz),
> >> (300 mBi, 2000 mBm)
> >> [ 9676.823846] cfg80211:   (5170000 KHz - 5250000 KHz @ 40000 KHz),
> >> (300 mBi, 2000 mBm)
> >> [ 9676.823851] cfg80211:   (5735000 KHz - 5835000 KHz @ 40000 KHz),
> >> (300 mBi, 2000 mBm)
> >> [ 9680.014722] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
> >> [ 9680.014890] iwlwifi 0000:02:00.0: L1 Enabled; Disabling L0S
> >>
> >> I haven't taken a kernel trace before but I'm happy to do so if you
> >> can point me toward some instructions.
> >>
> >
> > as root:
> > trace-cmd record -e iwlwifi -e iwlwifi_msg.
> >
> > Reproduce the bug while it's running. Stop with Ctrl+C when the bug reproduced.
> > This will produce a trace.dat file which you can compress and send over email.
> >
>
> Could you please try to reproduce with power save disabled?
>
> options iwlmvm power_scheme=1
>
> Thanks.


Huh, interesting. When I set that option the problem seems to go away.
I flipped back and forth between power_scheme=1 and power_scheme=2 in
iwlmvm (modinfo says 2 is the default), and the problem didn't
reproduce when power_scheme=1. I assume my battery life will suffer if
I permanently set power_scheme=1? :-)

Should I take a new trace or gather any other information?

I won't be my computer all day but I'll be able to respond / collect
any needed info intermittently.

Thanks,

Stephen

Re: [Ilw] RX laggy/choppy/stuck without TX on Intel Wireless 7260 with iwlwifi

[Emmanuel Grumbach]

>>>
>>
>> Could you please try to reproduce with power save disabled?
>>
>> options iwlmvm power_scheme=1
>>
>> Thanks.
> 
> 
> Huh, interesting. When I set that option the problem seems to go away.
> I flipped back and forth between power_scheme=1 and power_scheme=2 in
> iwlmvm (modinfo says 2 is the default), and the problem didn't
> reproduce when power_scheme=1. I assume my battery life will suffer if
> I permanently set power_scheme=1? :-)
> 
> Should I take a new trace or gather any other information?

Not needed

> 
> I won't be my computer all day but I'll be able to respond / collect
> any needed info intermittently.
> 

I guess the best would be to upgrade to 3.13.5 with new firmware... Or to stay with power_scheme=1 in the meantime.


> Thanks,
> 
> Stephen
> 

Re: [Ilw] RX laggy/choppy/stuck without TX on Intel Wireless 7260 with iwlwifi

[Stephen Kent]

On Sun, Feb 23, 2014 at 8:04 PM, Emmanuel Grumbach <egrumbach@gmail.com> wrote:
>
>
> >>>
> >>
> >> Could you please try to reproduce with power save disabled?
> >>
> >> options iwlmvm power_scheme=1
> >>
> >> Thanks.
> >
> >
> > Huh, interesting. When I set that option the problem seems to go away.
> > I flipped back and forth between power_scheme=1 and power_scheme=2 in
> > iwlmvm (modinfo says 2 is the default), and the problem didn't
> > reproduce when power_scheme=1. I assume my battery life will suffer if
> > I permanently set power_scheme=1? :-)
> >
> > Should I take a new trace or gather any other information?
>
> Not needed
>
> >
> > I won't be my computer all day but I'll be able to respond / collect
> > any needed info intermittently.
> >
>
> I guess the best would be to upgrade to 3.13.5 with new firmware... Or to stay with power_scheme=1 in the meantime.
>
>
> > Thanks,
> >
> > Stephen
> >


OK. Was this a bug that was identified and fixed between 3.11 and
3.13.5? I looked briefly at the kernel git log messages at
https://github.com/torvalds/linux/commits/master/drivers/net/wireless/iwlwifi/mvm
but I am completely unfamiliar with this code.

Until I upgrade, I'll just set power_scheme=1.

Thanks again for your help!

Stephen

Re: [Ilw] RX laggy/choppy/stuck without TX on Intel Wireless 7260 with iwlwifi

[Emmanuel Grumbach]

On 02/24/2014 06:19 AM, Stephen Kent wrote:
> On Sun, Feb 23, 2014 at 8:04 PM, Emmanuel Grumbach <egrumbach@gmail.com> wrote:
>>
>>
>>>>>
>>>>
>>>> Could you please try to reproduce with power save disabled?
>>>>
>>>> options iwlmvm power_scheme=1
>>>>
>>>> Thanks.
>>>
>>>
>>> Huh, interesting. When I set that option the problem seems to go away.
>>> I flipped back and forth between power_scheme=1 and power_scheme=2 in
>>> iwlmvm (modinfo says 2 is the default), and the problem didn't
>>> reproduce when power_scheme=1. I assume my battery life will suffer if
>>> I permanently set power_scheme=1? :-)
>>>
>>> Should I take a new trace or gather any other information?
>>
>> Not needed
>>
>>>
>>> I won't be my computer all day but I'll be able to respond / collect
>>> any needed info intermittently.
>>>
>>
>> I guess the best would be to upgrade to 3.13.5 with new firmware... Or to stay with power_scheme=1 in the meantime.
>>
>>
>>> Thanks,
>>>
>>> Stephen
>>>
> 
> 
> OK. Was this a bug that was identified and fixed between 3.11 and
> 3.13.5? I looked briefly at the kernel git log messages at
> https://github.com/torvalds/linux/commits/master/drivers/net/wireless/iwlwifi/mvm
> but I am completely unfamiliar with this code.

No - I can't think about any patch that could fix your issue. But 3.13 supports a newer firmware in which the power save code has been rewritten.

> 
> Until I upgrade, I'll just set power_scheme=1.
> 
> Thanks again for your help!
> 
> Stephen
>