Re: [PATCH] QCA9984 VHT160 support

[Sebastian Gottschall]

i've seen you added a new firmware for 9984. is there anything you can 
tell about already?


Am 18.10.2016 um 10:20 schrieb Sebastian Gottschall:

Am 14.10.2016 um 12:14 schrieb Valo, Kalle:

Sebastian Gottschall  writes:


This patch adds full VHT160 support for QCA9984 chipsets
Tested on Netgear R7800. 80+80 is possible, but disabled so far since
it seems to contain glitches like missing vht station flags (this may
be firmware or mac80211 related)

So how did you disable 80+80 exactly? I can't find it and I would like
to add a comment to the code so that it's easy to enable it later (once
it's working).



IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160MHZ  =vht160 only
IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ = vht160 and 80+80 
support


+if ((ar->vht_cap_info & IEEE80211_VHT_CAP_SHORT_GI_160) &&
+!(ar->vht_cap_info & 
IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ))

+vht_cap.cap |= IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160MHZ;
+

so if the firmware itself doesnt contain the correct vht flags but 
just short_gi_160 flag. i assume it supports vht160 and set the vht160 
only flag.
newer firmwares on codeaurora contain the correct flags. so this code 
here gets avoided. unfortunatly the newer i found is also not stable 
and crashes often.
you can also enable 80+80 support with the current firmware.but all 
station ht and vht flags seem to be empty then in vht160 mode too. so 
it might be a firmware issue still with 80+80 or mac80211.
i did not debug that much. but vht160 works. i also tested it now with 
a 88w6964 marvell client in vht160 which worked too






--
Mit freundlichen Grüssen / Regards

Sebastian Gottschall / CTO

NewMedia-NET GmbH - DD-WRT
Firmensitz:  Berliner Ring 101, 64625 Bensheim
Registergericht: Amtsgericht Darmstadt, HRB 25473
Geschäftsführer: Peter Steinhäuser, Christian Scheele
http://www.dd-wrt.com
email: s.gottsch...@dd-wrt.com
Tel.: +496251-582650 / Fax: +496251-5826565


___
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: [PATCH] QCA9984 VHT160 support

[Sebastian Gottschall]

Am 14.10.2016 um 12:14 schrieb Valo, Kalle:

Sebastian Gottschall  writes:


This patch adds full VHT160 support for QCA9984 chipsets
Tested on Netgear R7800. 80+80 is possible, but disabled so far since
it seems to contain glitches like missing vht station flags (this may
be firmware or mac80211 related)

So how did you disable 80+80 exactly? I can't find it and I would like
to add a comment to the code so that it's easy to enable it later (once
it's working).



IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160MHZ  =vht160 only
IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ = vht160 and 80+80 
support


+if ((ar->vht_cap_info & IEEE80211_VHT_CAP_SHORT_GI_160) &&
+!(ar->vht_cap_info & 
IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ))

+vht_cap.cap |= IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160MHZ;
+

so if the firmware itself doesnt contain the correct vht flags but just 
short_gi_160 flag. i assume it supports vht160 and set the vht160 only 
flag.
newer firmwares on codeaurora contain the correct flags. so this code 
here gets avoided. unfortunatly the newer i found is also not stable and 
crashes often.
you can also enable 80+80 support with the current firmware.but all 
station ht and vht flags seem to be empty then in vht160 mode too. so it 
might be a firmware issue still with 80+80 or mac80211.
i did not debug that much. but vht160 works. i also tested it now with a 
88w6964 marvell client in vht160 which worked too



--
Mit freundlichen Grüssen / Regards

Sebastian Gottschall / CTO

NewMedia-NET GmbH - DD-WRT
Firmensitz:  Berliner Ring 101, 64625 Bensheim
Registergericht: Amtsgericht Darmstadt, HRB 25473
Geschäftsführer: Peter Steinhäuser, Christian Scheele
http://www.dd-wrt.com
email: s.gottsch...@dd-wrt.com
Tel.: +496251-582650 / Fax: +496251-5826565


___
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: [PATCH] QCA9984 VHT160 support

[Valo, Kalle]

Sebastian Gottschall  writes:

> This patch adds full VHT160 support for QCA9984 chipsets
> Tested on Netgear R7800. 80+80 is possible, but disabled so far since
> it seems to contain glitches like missing vht station flags (this may
> be firmware or mac80211 related)

So how did you disable 80+80 exactly? I can't find it and I would like
to add a comment to the code so that it's easy to enable it later (once
it's working).

-- 
Kalle Valo
___
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: [PATCH] QCA9984 VHT160 support

[Valo, Kalle]

Sebastian Gottschall  writes:

> Am 10.10.2016 um 18:06 schrieb Valo, Kalle:
>> Kalle Valo  writes:
>>
>>> Sebastian Gottschall  writes:
>>>
 This patch adds full VHT160 support for QCA9984 chipsets
 Tested on Netgear R7800. 80+80 is possible, but disabled so far since
 it seems to contain glitches like missing vht station flags (this may
 be firmware or mac80211 related)
>>> There are some compilation and checkpatch warnings. Also you forgot to
>>> CC linux-wireless and sent the patch as an attachment. I'll send v2 to
>>> fix those.
>
> gen_dbglog_cfg? i did not touch anything on this struct in my patch.
> this warning doesnt seem to belong to my patch

According to patchwork you did, see below. Most likely just merge damage
from rebasing, I fixed it in v2.

--- a/drivers/net/wireless/ath/ath10k/wmi-tlv.c
+++ b/drivers/net/wireless/ath/ath10k/wmi-tlv.c
@@ -2468,7 +2468,7 @@  ath10k_wmi_tlv_op_gen_force_fw_hang(struct ath10k *ar,
 }
 
 static struct sk_buff *
-ath10k_wmi_tlv_op_gen_dbglog_cfg(struct ath10k *ar, u64 module_enable,
+ath10k_wmi_tlv_op_gen_dbglog_cfg(struct ath10k *ar, u32 module_enable,
 u32 log_level) {
struct wmi_tlv_dbglog_cmd *cmd;
struct wmi_tlv *tlv;

-- 
Kalle Valo
___
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: [PATCH] QCA9984 VHT160 support

[Sebastian Gottschall]

Am 10.10.2016 um 18:06 schrieb Valo, Kalle:

Kalle Valo  writes:


Sebastian Gottschall  writes:


This patch adds full VHT160 support for QCA9984 chipsets
Tested on Netgear R7800. 80+80 is possible, but disabled so far since
it seems to contain glitches like missing vht station flags (this may
be firmware or mac80211 related)

There are some compilation and checkpatch warnings. Also you forgot to
CC linux-wireless and sent the patch as an attachment. I'll send v2 to
fix those.
gen_dbglog_cfg? i did not touch anything on this struct in my patch. 
this warning doesnt seem to belong to my patch

Forgot the actual warnings:

drivers/net/wireless/ath/ath10k/wmi-tlv.c:3520:2: warning: initialization from 
incompatible pointer type [enabled by default]
drivers/net/wireless/ath/ath10k/wmi-tlv.c:3520:2: warning: (near initialization 
for 'wmi_tlv_ops.gen_dbglog_cfg') [enabled by default]
drivers/net/wireless/ath/ath10k/wmi-tlv.c:3520:27: warning: incorrect type in 
initializer (incompatible argument 2 (different type sizes))
drivers/net/wireless/ath/ath10k/wmi-tlv.c:3520:27:expected struct sk_buff 
*( *gen_dbglog_cfg )( ... )
drivers/net/wireless/ath/ath10k/wmi-tlv.c:3520:27:got struct sk_buff *( static 
[toplevel] * )( ... )
drivers/net/wireless/ath/ath10k/wmi-tlv.h:546: code indent should use tabs 
where possible
drivers/net/wireless/ath/ath10k/wmi-tlv.h:546: please, no spaces at the start 
of a line
drivers/net/wireless/ath/ath10k/mac.c:2556: braces {} should be used on all 
arms of this statement
drivers/net/wireless/ath/ath10k/mac.c:2560: line over 90 characters
drivers/net/wireless/ath/ath10k/mac.c:4301: line over 90 characters
drivers/net/wireless/ath/ath10k/mac.c:4301: braces {} are not necessary for 
single statement blocks




--
Mit freundlichen Grüssen / Regards

Sebastian Gottschall / CTO

NewMedia-NET GmbH - DD-WRT
Firmensitz:  Berliner Ring 101, 64625 Bensheim
Registergericht: Amtsgericht Darmstadt, HRB 25473
Geschäftsführer: Peter Steinhäuser, Christian Scheele
http://www.dd-wrt.com
email: s.gottsch...@dd-wrt.com
Tel.: +496251-582650 / Fax: +496251-5826565


___
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: [PATCH] QCA9984 VHT160 support

[Valo, Kalle]

Kalle Valo  writes:

> Sebastian Gottschall  writes:
>
>> This patch adds full VHT160 support for QCA9984 chipsets
>> Tested on Netgear R7800. 80+80 is possible, but disabled so far since
>> it seems to contain glitches like missing vht station flags (this may
>> be firmware or mac80211 related)
>
> There are some compilation and checkpatch warnings. Also you forgot to
> CC linux-wireless and sent the patch as an attachment. I'll send v2 to
> fix those.

Forgot the actual warnings:

drivers/net/wireless/ath/ath10k/wmi-tlv.c:3520:2: warning: initialization from 
incompatible pointer type [enabled by default]
drivers/net/wireless/ath/ath10k/wmi-tlv.c:3520:2: warning: (near initialization 
for 'wmi_tlv_ops.gen_dbglog_cfg') [enabled by default]
drivers/net/wireless/ath/ath10k/wmi-tlv.c:3520:27: warning: incorrect type in 
initializer (incompatible argument 2 (different type sizes))
drivers/net/wireless/ath/ath10k/wmi-tlv.c:3520:27:expected struct sk_buff 
*( *gen_dbglog_cfg )( ... )
drivers/net/wireless/ath/ath10k/wmi-tlv.c:3520:27:got struct sk_buff *( 
static [toplevel] * )( ... )
drivers/net/wireless/ath/ath10k/wmi-tlv.h:546: code indent should use tabs 
where possible
drivers/net/wireless/ath/ath10k/wmi-tlv.h:546: please, no spaces at the start 
of a line
drivers/net/wireless/ath/ath10k/mac.c:2556: braces {} should be used on all 
arms of this statement
drivers/net/wireless/ath/ath10k/mac.c:2560: line over 90 characters
drivers/net/wireless/ath/ath10k/mac.c:4301: line over 90 characters
drivers/net/wireless/ath/ath10k/mac.c:4301: braces {} are not necessary for 
single statement blocks

-- 
Kalle Valo
___
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k