ath9k: Support RSN Mesh
PATCH 1/1 ath9k: Support RSN Mesh
From: Chun-Yeow Yeoh <[email protected]>
Date: Fri, 18 Nov 2011 10:04:38 +0800
Subject: [PATCH] ath9k: Support RSN Mesh
Enable support of HW crypt for Mesh to achieve high throughput.
--- drivers/net/wireless/ath/ath9k/main.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)
(quoted diff) +0 -0
b/drivers/net/wireless/ath/ath9k/main.c
index 93fbe6f..1536515 100644
drivers/net/wireless/ath/ath9k/main.c +2 -1
@@ -1912,7 +1912,8 @@ static int ath9k_set_key(struct ieee80211_hw *hw,if (ath9k_modparam_nohwcrypt)return -ENOSPC ; - if (vif->type == NL80211_IFTYPE_ADHOC && + if ((vif->type == NL80211_IFTYPE_ADHOC || + vif->type == NL80211_IFTYPE_MESH_POINT) && (key->cipher == WLAN_CIPHER_SUITE_TKIP || key->cipher == WLAN_CIPHER_SUITE_CCMP) &&
!(key->flags & IEEE80211_KEY_FLAG_PAIRWISE)) {--
1.7.0.4
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20111202/b81588d8/attachment.htm
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ath9k-rsn-mesh.patch
Type: text/x-patch
Size: 990 bytes
Desc: not available
Url : http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20111202/b81588d8/attachment.bin
PATCH 1/1 ath9k: Support RSN Mesh
Hi Chun-Yeow Yeoh,
Quoted 5 lines · Yeoh Chun-Yeow
From: Chun-Yeow Yeoh <[email protected]>
Date: Fri, 18 Nov 2011 10:04:38 +0800
Subject: [PATCH] ath9k: Support RSN MeshEnable support of HW crypt for Mesh to achieve high throughput.
i think you are doing the other way around in your patch.
Quoted 32 lines
Signed-off-by Chun-Yeow Yeoh <[email protected]>---
?drivers/net/wireless/ath/ath9k/main.c |??? 3 ++-
?1 files changed, 2 insertions(+), 1 deletions(-)drivers/net/wireless/ath/ath9k/main.c +2
-1@@ -1912,7 +1912,8 @@ static int ath9k_set_key(struct ieee80211_hw *hw, ??????? if (ath9k_modparam_nohwcrypt) ??????????????? return -ENOSPC; -?????? if (vif->type == NL80211_IFTYPE_ADHOC && +?????? if ((vif->type == NL80211_IFTYPE_ADHOC || +??????????? vif->type == NL80211_IFTYPE_MESH_POINT) &&??????????? (
key->cipher==WLAN_CIPHER_SUITE_TKIP|| ????????????key->cipher==WLAN_CIPHER_SUITE_CCMP) && ??????????? !(key->flags &IEEE80211_KEY_FLAG_PAIRWISE)) { --
1.7.0.4_______________________________________________
ath9k-devel mailing list
ath9k-devel at lists.ath9k.org
https://lists.ath9k.org/mailman/listinfo/ath9k-devel
Signature · 1 line
shafi
PATCH 1/1 ath9k: Support RSN Mesh
Hi, Shafi
Appreciate if you can comment on this. Previous, for mesh node setup, we need to load the ath9k.ko with nohwcrypt=1. After applying this patch, we can simply load the ath9k.ko without nohwcrypt=1 and able to ensure the mesh node communicate. Besides, we can achieve HT throughput for single hop mesh setup, but still problematic on multi-hop mesh setup.
Thanks
Regards,
Chun-Yeow
Quoted 39 lines · Mohammed Shafi
Hi Chun-Yeow Yeoh,
Quoted 5 lines · Yeoh Chun-Yeow
From: Chun-Yeow Yeoh <[email protected]>
Date: Fri, 18 Nov 2011 10:04:38 +0800
Subject: [PATCH] ath9k: Support RSN MeshEnable support of HW crypt for Mesh to achieve high throughput.
i think you are doing the other way around in your patch.
Quoted 27 lines
Signed-off-by Chun-Yeow Yeoh <[email protected]>--- drivers/net/wireless/ath/ath9k/main.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)drivers/net/wireless/ath/ath9k/main.c +2
-1@@ -1912,7 +1912,8 @@ static int ath9k_set_key(struct ieee80211_hw *hw,if (ath9k_modparam_nohwcrypt)return -ENOSPC ; - if (vif->type == NL80211_IFTYPE_ADHOC && + if ((vif->type == NL80211_IFTYPE_ADHOC || + vif->type == NL80211_IFTYPE_MESH_POINT) && (key->cipher == WLAN_CIPHER_SUITE_TKIP || key->cipher == WLAN_CIPHER_SUITE_CCMP) &&!(key->flags & IEEE80211_KEY_FLAG_PAIRWISE)) {--
1.7.0.4
Quoted 11 lines
Quoted 5 lines
_______________________________________________
ath9k-devel mailing list
ath9k-devel at lists.ath9k.org
https://lists.ath9k.org/mailman/listinfo/ath9k-devel--
shafi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20111202/b60bc5f5/attachment.htm
PATCH 1/1 ath9k: Support RSN Mesh
Hi Chun-Yeow Yeoh,
ath9k.ko with nohwcrypt=1 disables hardware encryption and your patch also does the same for mesh mode. i need to look into the code what it does, as i don't know about the details and also i have not tested mesh.
also what card you are using? please make sure you are using the latest ath9k build.
please send patches to linux wireless mailing list
http://linuxwireless.org/en/developers/Documentation/git-guide
thank you
Quoted 68 lines · Yeoh Chun-Yeow
Hi, Shafi
Appreciate if you can comment on this. Previous, for mesh node setup, we need to load the ath9k.ko with nohwcrypt=1. After applying this patch, we can simply load the ath9k.ko without nohwcrypt=1 and able to ensure the mesh node communicate. Besides, we can achieve HT throughput for single hop mesh setup, but still problematic on multi-hop mesh setup.
Thanks
Regards,
Chun-YeowQuoted 50 lines · Mohammed Shafi
Hi Chun-Yeow Yeoh,
Quoted 5 lines · Yeoh Chun-Yeow
From: Chun-Yeow Yeoh <[email protected]>
Date: Fri, 18 Nov 2011 10:04:38 +0800
Subject: [PATCH] ath9k: Support RSN MeshEnable support of HW crypt for Mesh to achieve high throughput.
i think you are doing the other way around in your patch.
Quoted 27 lines
Signed-off-by Chun-Yeow Yeoh <[email protected]>---
?drivers/net/wireless/ath/ath9k/main.c |??? 3 ++-
?1 files changed, 2 insertions(+), 1 deletions(-)drivers/net/wireless/ath/ath9k/main.c +2
-1@@ -1912,7 +1912,8 @@ static int ath9k_set_key(struct ieee80211_hw *hw, ??????? if (ath9k_modparam_nohwcrypt) ??????????????? return -ENOSPC; -?????? if (vif->type == NL80211_IFTYPE_ADHOC && +?????? if ((vif->type == NL80211_IFTYPE_ADHOC || +??????????? vif->type == NL80211_IFTYPE_MESH_POINT) &&??????????? (
key->cipher==WLAN_CIPHER_SUITE_TKIP|| ????????????key->cipher==WLAN_CIPHER_SUITE_CCMP) && ??????????? !(key->flags &IEEE80211_KEY_FLAG_PAIRWISE)) { --
1.7.0.4Quoted 5 lines
_______________________________________________
ath9k-devel mailing list
ath9k-devel at lists.ath9k.org
https://lists.ath9k.org/mailman/listinfo/ath9k-devel--
shafi
Signature · 1 line
shafi
PATCH 1/1 ath9k: Support RSN Mesh
Hi, Shafi
I am using AR9280 and AR9160. Let me explain the situation:
Actually, we need to use per-STA RX GTK for secured mesh. Thus, we use software crypto for our multicast/broadcast frame. Without this patch, we need to load the ath9k with nohwcrypt=1. Thus, we suffer the throughput performance of our mesh nodes on unicast frame (no software crypto as well). After applying this patch, we are able to get HT throughput for single hop communication in mesh setup (802.11s) by enabling HW crypto for unicast.
However, we are still having throughput performance on multihop communication. Do you have any idea?
Regards,
Chun-Yeow
Quoted 89 lines · Mohammed Shafi
Hi Chun-Yeow Yeoh,
ath9k.ko with nohwcrypt=1 disables hardware encryption and your patch also does the same for mesh mode. i need to look into the code what it does, as i don't know about the details and also i have not tested mesh.
also what card you are using? please make sure you are using the latest ath9k build.
please send patches to linux wireless mailing list
http://linuxwireless.org/en/developers/Documentation/git-guide
thank youQuoted 5 lines · Yeoh Chun-Yeow
Hi, Shafi
Appreciate if you can comment on this. Previous, for mesh node setup, we need to load the ath9k.ko with nohwcrypt=1. After applying this patch, we can simply load the ath9k.ko without nohwcrypt=1 and able to ensure the
mesh
Quoted 1 line
node communicate. Besides, we can achieve HT throughput for single hop
mesh
Quoted 30 lines
setup, but still problematic on multi-hop mesh setup.
Thanks
Regards,
Chun-YeowOn Fri, Dec 2, 2011 at 5:18 PM, Mohammed Shafi <[email protected]
Quoted 19 lines · wrote:
Hi Chun-Yeow Yeoh,
Quoted 5 lines · Yeoh Chun-Yeow
From: Chun-Yeow Yeoh <[email protected]>
Date: Fri, 18 Nov 2011 10:04:38 +0800
Subject: [PATCH] ath9k: Support RSN MeshEnable support of HW crypt for Mesh to achieve high throughput.
i think you are doing the other way around in your patch.
Quoted 6 lines
Signed-off-by Chun-Yeow Yeoh <[email protected]>--- drivers/net/wireless/ath/ath9k/main.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)Quoted 6 lines
Quoted 6 lines
drivers/net/wireless/ath/ath9k/main.c static int ath9k_set_key(struct ieee80211_hw
Quoted 6 lines
drivers/net/wireless/ath/ath9k/main.c +0
-0@@ -1912,7 +1912,8 @@ static int ath9k_set_key(struct ieee80211_hw*hw,
Quoted 23 lines
Quoted 23 lines
Quoted 20 lines
drivers/net/wireless/ath/ath9k/main.c +2
-2if (ath9k_modparam_nohwcrypt)return -ENOSPC ; -if (vif->type ==NL80211_IFTYPE_ADHOC && +if ((vif->type ==NL80211_IFTYPE_ADHOC || + vif->type ==NL80211_IFTYPE_MESH_POINT ) && (key->cipher ==WLAN_CIPHER_SUITE_TKIP || key->cipher ==WLAN_CIPHER_SUITE_CCMP ) && !(key->flags &IEEE80211_KEY_FLAG_PAIRWISE )) { --1.7.0.4
_______________________________________________
ath9k-devel mailing list
ath9k-devel at lists.ath9k.org
https://lists.ath9k.org/mailman/listinfo/ath9k-develQuoted 4 lines
Quoted 2 lines
--
shafi
Quoted 3 lines
--
shafi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20111202/1cb02ca3/attachment-0001.htm
PATCH 1/1 ath9k: Support RSN Mesh
Hi, Shafi
The patch is actually based on the following:
http://ns.spinics.net/lists/linux-wireless/msg66579.html
Regards,
Chun-Yeow
Quoted 12 lines · Yeoh Chun-Yeow
Hi, Shafi
I am using AR9280 and AR9160. Let me explain the situation:
Actually, we need to use per-STA RX GTK for secured mesh. Thus, we use software crypto for our multicast/broadcast frame. Without this patch, we need to load the ath9k with nohwcrypt=1. Thus, we suffer the throughput performance of our mesh nodes on unicast frame (no software crypto as well). After applying this patch, we are able to get HT throughput for single hop communication in mesh setup (802.11s) by enabling HW crypto for unicast.
Quoted 104 lines
However, we are still having throughput performance on multihop communication. Do you have any idea?
Regards,
Chun-YeowQuoted 94 lines · Mohammed Shafi
Hi Chun-Yeow Yeoh,
ath9k.ko with nohwcrypt=1 disables hardware encryption and your patch also does the same for mesh mode. i need to look into the code what it does, as i don't know about the details and also i have not tested mesh.
also what card you are using? please make sure you are using the latest ath9k build.
please send patches to linux wireless mailing list
http://linuxwireless.org/en/developers/Documentation/git-guide
thank youQuoted 4 lines · Yeoh Chun-Yeow
Hi, Shafi
Appreciate if you can comment on this. Previous, for mesh node setup, we need to load the ath9k.ko with nohwcrypt=1. After applying this patch,
we
Quoted 1 line
can simply load the ath9k.ko without nohwcrypt=1 and able to ensure the
mesh
Quoted 1 line
node communicate. Besides, we can achieve HT throughput for single hop
mesh
Quoted 9 lines
setup, but still problematic on multi-hop mesh setup.
Thanks
Regards,
Chun-YeowOn Fri, Dec 2, 2011 at 5:18 PM, Mohammed Shafi <
shafi.wireless at gmail.com>
Quoted 21 lines
Quoted 20 lines · wrote:
Hi Chun-Yeow Yeoh,
On Fri, Dec 2, 2011 at 2:21 PM, Yeoh Chun-Yeow <[email protected]
Quoted 5 lines · wrote:
From: Chun-Yeow Yeoh <[email protected]>
Date: Fri, 18 Nov 2011 10:04:38 +0800
Subject: [PATCH] ath9k: Support RSN MeshEnable support of HW crypt for Mesh to achieve high throughput.
i think you are doing the other way around in your patch.
Quoted 6 lines
Signed-off-by Chun-Yeow Yeoh <[email protected]>--- drivers/net/wireless/ath/ath9k/main.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)Quoted 6 lines
Quoted 6 lines
drivers/net/wireless/ath/ath9k/main.c static int ath9k_set_key(struct ieee80211_hw
Quoted 6 lines
drivers/net/wireless/ath/ath9k/main.c +0
-0@@ -1912,7 +1912,8 @@ static int ath9k_set_key(struct ieee80211_hw*hw,
Quoted 23 lines
Quoted 23 lines
Quoted 20 lines
drivers/net/wireless/ath/ath9k/main.c +2
-2if (ath9k_modparam_nohwcrypt)return -ENOSPC ; -if (vif->type ==NL80211_IFTYPE_ADHOC && +if ((vif->type ==NL80211_IFTYPE_ADHOC || + vif->type ==NL80211_IFTYPE_MESH_POINT ) && (key->cipher ==WLAN_CIPHER_SUITE_TKIP || key->cipher ==WLAN_CIPHER_SUITE_CCMP ) && !(key->flags &IEEE80211_KEY_FLAG_PAIRWISE )) { --1.7.0.4
_______________________________________________
ath9k-devel mailing list
ath9k-devel at lists.ath9k.org
https://lists.ath9k.org/mailman/listinfo/ath9k-develQuoted 4 lines
Quoted 2 lines
--
shafi--
shafi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20111202/9ef60d28/attachment.htm
PATCH 1/1 ath9k: Support RSN Mesh
In reply to Yeoh Chun-YeowHi Chun-Yeow ,
sorry i have no idea and have not tested mesh. please also send a mail to linux wireless mailing list, where you may get some help from mesh developers from cozy bit.
Quoted 114 lines · Yeoh Chun-Yeow
Hi, Shafi
I am using AR9280 and AR9160. Let me explain the situation:
Actually, we need to use per-STA RX GTK for secured mesh. Thus, we use software crypto for our multicast/broadcast frame. Without this patch, we need to load the ath9k with nohwcrypt=1. Thus, we suffer the throughput performance of our mesh nodes on unicast frame (no software crypto as well). After applying this patch, we are able to get HT throughput for single hop communication in mesh setup (802.11s) by enabling HW crypto for unicast.
However, we are still having throughput performance on multihop communication. Do you have any idea?
Regards,
Chun-YeowQuoted 93 lines · Mohammed Shafi
Hi Chun-Yeow Yeoh,
ath9k.ko with nohwcrypt=1 disables hardware encryption and your patch also does the same for mesh mode. i need to look into the code what it does, as i don't know about the details and also i have not tested mesh.
also what card you are using? please make sure you ?are using the latest ath9k build.
please send patches to linux wireless mailing list
http://linuxwireless.org/en/developers/Documentation/git-guide
thank youQuoted 13 lines · Yeoh Chun-Yeow
Hi, Shafi
Appreciate if you can comment on this. Previous, for mesh node setup, we need to load the ath9k.ko with nohwcrypt=1. After applying this patch, we
can simply load the ath9k.ko without nohwcrypt=1 and able to ensure the mesh
node communicate. Besides, we can achieve HT throughput for single hop mesh
setup, but still problematic on multi-hop mesh setup.Thanks
Quoted 60 lines
Regards,
Chun-YeowQuoted 51 lines · Mohammed Shafi
Hi Chun-Yeow Yeoh,
Quoted 5 lines · Yeoh Chun-Yeow
From: Chun-Yeow Yeoh <[email protected]>
Date: Fri, 18 Nov 2011 10:04:38 +0800
Subject: [PATCH] ath9k: Support RSN MeshEnable support of HW crypt for Mesh to achieve high throughput.
i think you are doing the other way around in your patch.
Quoted 28 lines
Signed-off-by Chun-Yeow Yeoh <[email protected]>---
?drivers/net/wireless/ath/ath9k/main.c |??? 3 ++-
?1 files changed, 2 insertions(+), 1 deletions(-)drivers/net/wireless/ath/ath9k/main.c +2
-1@@ -1912,7 +1912,8 @@ static int ath9k_set_key(struct ieee80211_hw *hw, ??????? if (ath9k_modparam_nohwcrypt) ??????????????? return -ENOSPC; -?????? if (vif->type == NL80211_IFTYPE_ADHOC && +?????? if ((vif->type == NL80211_IFTYPE_ADHOC || +??????????? vif->type == NL80211_IFTYPE_MESH_POINT) &&??????????? (
key->cipher==WLAN_CIPHER_SUITE_TKIP|| ????????????key->cipher==WLAN_CIPHER_SUITE_CCMP) && ??????????? !(key->flags &IEEE80211_KEY_FLAG_PAIRWISE)) { --
1.7.0.4Quoted 5 lines
_______________________________________________
ath9k-devel mailing list
ath9k-devel at lists.ath9k.org
https://lists.ath9k.org/mailman/listinfo/ath9k-devel--
shafi--
shafi
Signature · 1 line
shafi
PATCH 1/1 ath9k: Support RSN Mesh
Hi, Shafi
I am active in open80211s mailing list. Thomas from cozybit is helping me quite a lot. Just wonder any peoples here working on adhoc multihop on ath9k facing the same problem when doing throughput test for multihop environment.
Regards,
Chun-Yeow
Quoted 131 lines · Mohammed Shafi
Hi Chun-Yeow ,
sorry i have no idea and have not tested mesh. please also send a mail to linux wireless mailing list, where you may get some help from mesh developers from cozy bit.
Quoted 8 lines · Yeoh Chun-Yeow
Hi, Shafi
I am using AR9280 and AR9160. Let me explain the situation:
Actually, we need to use per-STA RX GTK for secured mesh. Thus, we use software crypto for our multicast/broadcast frame. Without this patch, we need to load the ath9k with nohwcrypt=1. Thus, we suffer the throughput performance of our mesh nodes on unicast frame (no software crypto as
well).
Quoted 1 line
After applying this patch, we are able to get HT throughput for single
hop
Quoted 5 lines
communication in mesh setup (802.11s) by enabling HW crypto for unicast.
However, we are still having throughput performance on multihop communication. Do you have any idea?
Quoted 24 lines
Regards,
Chun-YeowOn Fri, Dec 2, 2011 at 5:38 PM, Mohammed Shafi <[email protected]
Quoted 18 lines · wrote:
Hi Chun-Yeow Yeoh,
ath9k.ko with nohwcrypt=1 disables hardware encryption and your patch also does the same for mesh mode. i need to look into the code what it does, as i don't know about the details and also i have not tested mesh.
also what card you are using? please make sure you are using the latest ath9k build.
please send patches to linux wireless mailing list
http://linuxwireless.org/en/developers/Documentation/git-guide
thank youQuoted 3 lines · Yeoh Chun-Yeow
Hi, Shafi
Appreciate if you can comment on this. Previous, for mesh node setup,
we
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
need to load the ath9k.ko with nohwcrypt=1. After applying this patch, we
can simply load the ath9k.ko without nohwcrypt=1 and able to ensurethe
Quoted 11 lines
Quoted 11 lines
Quoted 11 lines
mesh
node communicate. Besides, we can achieve HT throughput for single hop mesh
setup, but still problematic on multi-hop mesh setup.Thanks
Regards,
Chun-YeowQuoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 4 lines · Mohammed Shafi
Hi Chun-Yeow Yeoh,
On Fri, Dec 2, 2011 at 2:21 PM, Yeoh Chun-Yeow <
yeohchunyeow at gmail.com>
Quoted 52 lines
Quoted 52 lines
Quoted 49 lines
Quoted 47 lines
Quoted 5 lines · wrote:
From: Chun-Yeow Yeoh <[email protected]>
Date: Fri, 18 Nov 2011 10:04:38 +0800
Subject: [PATCH] ath9k: Support RSN MeshEnable support of HW crypt for Mesh to achieve high throughput.
i think you are doing the other way around in your patch.
Quoted 28 lines
Signed-off-by Chun-Yeow Yeoh <[email protected]>--- drivers/net/wireless/ath/ath9k/main.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)drivers/net/wireless/ath/ath9k/main.c +2
-1@@ -1912,7 +1912,8 @@ static int ath9k_set_key(struct ieee80211_hw *hw, if (ath9k_modparam_nohwcrypt) return -ENOSPC; - if (vif->type == NL80211_IFTYPE_ADHOC && + if ((vif->type == NL80211_IFTYPE_ADHOC || + vif->type == NL80211_IFTYPE_MESH_POINT) && (key->cipher == WLAN_CIPHER_SUITE_TKIP ||key->cipher == WLAN_CIPHER_SUITE_CCMP) && !(key->flags & IEEE80211_KEY_FLAG_PAIRWISE)) {--
1.7.0.4Quoted 5 lines
_______________________________________________
ath9k-devel mailing list
ath9k-devel at lists.ath9k.org
https://lists.ath9k.org/mailman/listinfo/ath9k-devel--
shafiQuoted 4 lines
Quoted 2 lines
--
shafi
Quoted 3 lines
--
shafi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20111202/8230de8e/attachment-0001.htm
PATCH 1/1 ath9k: Support RSN Mesh
Hi Chun-Yeow,
oh great, he is the right person. then please send a patch to linux wireless mailing list.
also please Cc him and other ath9k maintainers. I had also Cced Luis who knows more about mesh testing in ath9k.
thank you.
Quoted 151 lines · Yeoh Chun-Yeow
Hi, Shafi
I am active in open80211s mailing list. Thomas from cozybit is helping me quite a lot. Just wonder any peoples here working on adhoc multihop on ath9k facing the same problem when doing throughput test for multihop environment.
Regards,
Chun-YeowQuoted 137 lines · Mohammed Shafi
Hi ?Chun-Yeow ,
sorry i have no idea and have not tested mesh. please also send a mail to linux wireless mailing list, where you may get some help from mesh developers from cozy bit.
Quoted 17 lines · Yeoh Chun-Yeow
Hi, Shafi
I am using AR9280 and AR9160. Let me explain the situation:
Actually, we need to use per-STA RX GTK for secured mesh. Thus, we use software crypto for our multicast/broadcast frame. Without this patch, we
need to load the ath9k with nohwcrypt=1. Thus, we suffer the throughput performance of our mesh nodes on unicast frame (no software crypto as well).
After applying this patch, we are able to get HT throughput for single hop
communication in mesh setup (802.11s) by enabling HW crypto for unicast.However, we are still having throughput performance on multihop communication. Do you have any idea?
Quoted 106 lines
Regards,
Chun-YeowQuoted 98 lines · Mohammed Shafi
Hi Chun-Yeow Yeoh,
ath9k.ko with nohwcrypt=1 disables hardware encryption and your patch also does the same for mesh mode. i need to look into the code what it does, as i don't know about the details and also i have not tested mesh.
also what card you are using? please make sure you ?are using the latest ath9k build.
please send patches to linux wireless mailing list
http://linuxwireless.org/en/developers/Documentation/git-guide
thank youQuoted 17 lines · Yeoh Chun-Yeow
Hi, Shafi
Appreciate if you can comment on this. Previous, for mesh node setup, we
need to load the ath9k.ko with nohwcrypt=1. After applying this patch,
we
can simply load the ath9k.ko without nohwcrypt=1 and able to ensure the
mesh
node communicate. Besides, we can achieve HT throughput for single hop
mesh
setup, but still problematic on multi-hop mesh setup.Thanks
Quoted 61 lines
Regards,
Chun-YeowQuoted 52 lines · Mohammed Shafi
Hi Chun-Yeow Yeoh,
Quoted 5 lines · Yeoh Chun-Yeow
From: Chun-Yeow Yeoh <[email protected]>
Date: Fri, 18 Nov 2011 10:04:38 +0800
Subject: [PATCH] ath9k: Support RSN MeshEnable support of HW crypt for Mesh to achieve high throughput.
i think you are doing the other way around in your patch.
Quoted 28 lines
Signed-off-by Chun-Yeow Yeoh <[email protected]>---
?drivers/net/wireless/ath/ath9k/main.c |??? 3 ++-
?1 files changed, 2 insertions(+), 1 deletions(-)drivers/net/wireless/ath/ath9k/main.c +2
-1@@ -1912,7 +1912,8 @@ static int ath9k_set_key(struct ieee80211_hw *hw, ??????? if (ath9k_modparam_nohwcrypt) ??????????????? return -ENOSPC; -?????? if (vif->type == NL80211_IFTYPE_ADHOC && +?????? if ((vif->type == NL80211_IFTYPE_ADHOC || +??????????? vif->type == NL80211_IFTYPE_MESH_POINT) &&??????????? (
key->cipher==WLAN_CIPHER_SUITE_TKIP|| ????????????key->cipher==WLAN_CIPHER_SUITE_CCMP) && ??????????? !(key->flags &IEEE80211_KEY_FLAG_PAIRWISE)) { --
1.7.0.4Quoted 5 lines
_______________________________________________
ath9k-devel mailing list
ath9k-devel at lists.ath9k.org
https://lists.ath9k.org/mailman/listinfo/ath9k-devel--
shafi--
shafi--
shafi
Signature · 1 line
shafi
PATCH 1/1 ath9k: Support RSN Mesh
In reply to Yeoh Chun-YeowHi Chun-Yeow
I don't know if this is usefull for you, but we are using ad hoc with batman-adv, and we had some problems when the nodes didn't choose good routes for high throughputs for relatively long distances. Asking to the batman-adv list they explained us that the algorithm to choose the routes is based on low rate broadcasts, so the nodes could choose routes that with high rates become bad. They suggested us to increase the multicast rates, so that the nodes could "realize" if a route were not good for high throughputs, and it worked better, the nodes chose routes with more hops for long distances getting a better throughput than using less hops with larger distances.
By the way, why with nohwcrypt=1 the throughput get worse?
Regards
Gabriel
El 02/12/2011 07:04 a.m., Yeoh Chun-Yeow escribi?:
Quoted 167 lines
Hi, Shafi
I am active in open80211s mailing list. Thomas from cozybit is helping me quite a lot. Just wonder any peoples here working on adhoc multihop on ath9k facing the same problem when doing throughput test for multihop environment.
Regards,
Chun-YeowOn Fri, Dec 2, 2011 at 6:01 PM, Mohammed Shafi <shafi.wireless at gmail.com <mailto:[email protected]>> wrote:
Hi Chun-Yeow , sorry i have no idea and have not tested mesh. please also send a mail to linux wireless mailing list, where you may get some help from mesh developers from cozy bit. On Fri, Dec 2, 2011 at 3:21 PM, Yeoh Chun-Yeow <yeohchunyeow at gmail.com <mailto:[email protected]>> wrote: > Hi, Shafi > > I am using AR9280 and AR9160. Let me explain the situation: > > Actually, we need to use per-STA RX GTK for secured mesh. Thus, we use > software crypto for our multicast/broadcast frame. Without this patch, we > need to load the ath9k with nohwcrypt=1. Thus, we suffer the throughput > performance of our mesh nodes on unicast frame (no software crypto as well). > After applying this patch, we are able to get HT throughput for single hop > communication in mesh setup (802.11s) by enabling HW crypto for unicast. > > However, we are still having throughput performance on multihop > communication. Do you have any idea? > > Regards, > Chun-Yeow > > On Fri, Dec 2, 2011 at 5:38 PM, Mohammed Shafi <shafi.wireless at gmail.com <mailto:[email protected]>> > wrote: >> >> Hi Chun-Yeow Yeoh, >> >> ath9k.ko with nohwcrypt=1 disables hardware encryption and your patch >> also does the same for mesh mode. i need to look into the code what it >> does, as i don't know about the details and also i have not tested >> mesh. >> also what card you are using? please make sure you are using the >> latest ath9k build. >> please send patches to linux wireless mailing list >> http://linuxwireless.org/en/developers/Documentation/git-guide >> thank you >> >> On Fri, Dec 2, 2011 at 2:53 PM, Yeoh Chun-Yeow <yeohchunyeow at gmail.com <mailto:[email protected]>> >> wrote: >> > Hi, Shafi >> > >> > Appreciate if you can comment on this. Previous, for mesh node setup, we >> > need to load the ath9k.ko with nohwcrypt=1. After applying this patch, >> > we >> > can simply load the ath9k.ko without nohwcrypt=1 and able to ensure the >> > mesh >> > node communicate. Besides, we can achieve HT throughput for single hop >> > mesh >> > setup, but still problematic on multi-hop mesh setup. >> > >> > Thanks >> > >> > Regards, >> > Chun-Yeow >> > >> > >> > On Fri, Dec 2, 2011 at 5:18 PM, Mohammed Shafi >> > <shafi.wireless at gmail.com <mailto:[email protected]>> >> > wrote: >> >> >> >> Hi Chun-Yeow Yeoh, >> >> >> >> On Fri, Dec 2, 2011 at 2:21 PM, Yeoh Chun-Yeow <yeohchunyeow at gmail.com <mailto:[email protected]>> >> >> wrote: >> >> > From: Chun-Yeow Yeoh <[email protected] <mailto:[email protected]>> >> >> > Date: Fri, 18 Nov 2011 10:04:38 +0800 >> >> > Subject: [PATCH] ath9k: Support RSN Mesh >> >> > >> >> > Enable support of HW crypt for Mesh to achieve high throughput. >> >> >> >> i think you are doing the other way around in your patch. >> >> >> >> > >> >> > Signed-off-by: Chun-Yeow Yeoh <[email protected] <mailto:[email protected]>> >> >> > --- >> >> > drivers/net/wireless/ath/ath9k/main.c | 3 ++- >> >> > 1 files changed, 2 insertions(+), 1 deletions(-) >> >> > >> >> > diff --git a/drivers/net/wireless/ath/ath9k/main.c >> >> > b/drivers/net/wireless/ath/ath9k/main.c >> >> > index 93fbe6f..1536515 100644 >> >> > --- a/drivers/net/wireless/ath/ath9k/main.c >> >> > +++ b/drivers/net/wireless/ath/ath9k/main.c >> >> > @@ -1912,7 +1912,8 @@ static int ath9k_set_key(struct ieee80211_hw >> >> > *hw, >> >> > >> >> > if (ath9k_modparam_nohwcrypt) >> >> > return -ENOSPC; >> >> > >> >> > - if (vif->type == NL80211_IFTYPE_ADHOC && >> >> > + if ((vif->type == NL80211_IFTYPE_ADHOC || >> >> > + vif->type == NL80211_IFTYPE_MESH_POINT) && >> >> > (key->cipher == WLAN_CIPHER_SUITE_TKIP || >> >> > key->cipher == WLAN_CIPHER_SUITE_CCMP) && >> >> > !(key->flags & IEEE80211_KEY_FLAG_PAIRWISE)) { >> >> > -- >> >> > 1.7.0.4 >> >> > >> >> > >> >> > >> >> > _______________________________________________ >> >> > ath9k-devel mailing list >> >> > ath9k-devel at lists.ath9k.org <mailto:[email protected]> >> >> > https://lists.ath9k.org/mailman/listinfo/ath9k-devel >> >> > >> >> >> >> >> >> >> >> -- >> >> shafi >> > >> > >> >> >> >> -- >> shafi > > -- shafi_______________________________________________
ath9k-devel mailing list
ath9k-devel at lists.ath9k.org
https://lists.ath9k.org/mailman/listinfo/ath9k-devel
PATCH 1/1 ath9k: Support RSN Mesh
Hi, Gabriel
nohwcrypt=1 will make the ath9k driver to use software crypto instead of hardware crypto. My Atheros chipset is able to support hardware encryption and decryption for CCMP and AES. Thus, when we setup the 802.11s mesh network with security (SAE), the throughput for single hop communication degrades significantly if we use software crypto. We observe the HT throughput when we applied the patch to ensure unicast frame is sent with hardware encryption. However, the throughput is dropped for multi-hop communication. Possible due to the ath9k driver since we have no such problem when security is disable.
Regarding the adhoc batman-adv, did you setup the adhoc mode using security such as WPA2? What is the performance that you obtained?
Regards,
Chun-Yeow
Quoted 13 lines · On Fri, Dec 2, 2011 at 9:26 PM, <[email protected]> wrote:
Hi Chun-Yeow
I don't know if this is usefull for you, but we are using ad hoc with batman-adv, and we had some problems when the nodes didn't choose good routes for high throughputs for relatively long distances. Asking to the batman-adv list they explained us that the algorithm to choose the routes is based on low rate broadcasts, so the nodes could choose routes that with high rates become bad. They suggested us to increase the multicast rates, so that the nodes could "realize" if a route were not good for high throughputs, and it worked better, the nodes chose routes with more hops for long distances getting a better throughput than using less hops with larger distances.
Quoted 7 lines
By the way, why with nohwcrypt=1 the throughput get worse?
Regards
Gabriel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20111202/726b8e47/attachment.htm
PATCH 1/1 ath9k: Support RSN Mesh
Hi Chun-Yeow.
El 02/12/2011 12:28 p.m., Yeoh Chun-Yeow escribi?:
Quoted 11 lines
Hi, Gabriel
nohwcrypt=1 will make the ath9k driver to use software crypto instead of hardware crypto. My Atheros chipset is able to support hardware encryption and decryption for CCMP and AES. Thus, when we setup the 802.11s mesh network with security (SAE), the throughput for single hop communication degrades significantly if we use software crypto. We observe the HT throughput when we applied the patch to ensure unicast frame is sent with hardware encryption. However, the throughput is dropped for multi-hop communication. Possible due to the ath9k driver since we have no such problem when security is disable.
Ok, i understand, so if it works well without security it has nothing to do with the problem i mentioned.
Quoted 6 lines
Regarding the adhoc batman-adv, did you setup the adhoc mode using security such as WPA2? What is the performance that you obtained?
Regards,
Chun-Yeow
We are not using security yet, because we are having problems with throughput tests without encryption, and it's difficult to find where the problem is.
Regards
Gabriel
Quoted 25 lines
On Fri, Dec 2, 2011 at 9:26 PM, <[email protected] <mailto:[email protected]>> wrote:
Hi Chun-Yeow
I don't know if this is usefull for you, but we are using ad hoc with batman-adv, and we had some problems when the nodes didn't choose good routes for high throughputs for relatively long distances. Asking to the batman-adv list they explained us that the algorithm to choose the routes is based on low rate broadcasts, so the nodes could choose routes that with high rates become bad. They suggested us to increase the multicast rates, so that the nodes could "realize" if a route were not good for high throughputs, and it worked better, the nodes chose routes with more hops for long distances getting a better throughput than using less hops with larger distances.
By the way, why with nohwcrypt=1 the throughput get worse?
Regards
Gabriel
PATCH 1/1 ath9k: Support RSN Mesh
Hi, all
The problem of multihop low throughput is solved. This is due to the ath9k rate control algorithm must be set to minstrel instead of ath9k rate control. Anyone can comment on this?
Regards,
Chun-Yeow
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20111207/f3ae52c4/attachment.htm