Need to get msdu-chaining working.
I'm using raw-rx mode in my firmware, and it's throughput is extremely poor at higher speeds. I see large numbers of warnings about msdu_chaining, so I suspect this may be the reason.
Does anyone have any suggestions for what needs to be done to handle these sorts of frames properly?
/* FIXME: we do not support chaining yet.
* this needs investigation */
if (msdu_chaining) {
ath10k_warn("htt rx msdu_chaining is true\n");
ath10k_htt_rx_free_msdu_chain(msdu_head);
continue;
}Thanks,
Ben
Signature · 8 lines
Ben Greear <[email protected]> Candela Technologies Inc http://www.candelatech.com _______________________________________________ ath10k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 4 lines · Ben Greear
I'm using raw-rx mode in my firmware, and it's throughput is extremely poor at higher speeds. I see large numbers of warnings about
msdu_chaining, so I suspect this may be the reason.
I assume this requires your custom FW?
Quoted 10 lines
Does anyone have any suggestions for what needs to be done to handle these sorts of frames properly?
/* FIXME: we do not support chaining yet. * this needs investigation */ if (msdu_chaining) { ath10k_warn("htt rx msdu_chaining is true\n"); ath10k_htt_rx_free_msdu_chain(msdu_head); continue; }
From what I understand chained msdu is a msdu that hasn't fit into the rx buffer and is split across the popped amsdu list. I suspect only the first msdu in chain has the htt_rx_desc and all other have not (this is what the current code does, but you'll need to verify that).
I would try to concatenate all msdus into one (lots of memcpy :( ) or increase the HTT_RX_BUF_SIZE so that A-MSDU frames can fit into a single buffer (hopefully FW/HW is capable of doing that).
Michał
_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 7 lines · Michal Kazior
Quoted 4 lines · Ben Greear
I'm using raw-rx mode in my firmware, and it's throughput is extremely poor at higher speeds. I see large numbers of warnings about
msdu_chaining, so I suspect this may be the reason.I assume this requires your custom FW?
Yes. I am not sure you can use raw tx with standard firmware (yet)?
Or at least, you cannot have tx cooked and rx raw.
Quoted 11 lines
Quoted 10 lines
Does anyone have any suggestions for what needs to be done to handle these sorts of frames properly?
/* FIXME: we do not support chaining yet. * this needs investigation */ if (msdu_chaining) { ath10k_warn("htt rx msdu_chaining is true\n"); ath10k_htt_rx_free_msdu_chain(msdu_head); continue; }
Quoted 8 lines
Quoted 1 line
From what I understand chained msdu is a msdu that hasn't fit into the
rx buffer and is split across the popped amsdu list. I suspect only the first msdu in chain has the
htt_rx_descand all other have not (this is what the current code does, but you'll need to verify that).I would try to concatenate all msdus into one (lots of memcpy :( ) or increase the
HTT_RX_BUF_SIZEso that A-MSDU frames can fit into a single buffer (hopefully FW/HW is capable of doing that).
I think in raw mode the hardware/firmware may not try to re-assemble the skb, or also when no peer object can be found. This might be an issue in monitor mode as well, but I'm not certain about that.
I think there is a way to chain skb data chunks w/out doing the copying, but I am not certain how that would be done properly.
Thanks,
Ben
Signature · 8 lines
Ben Greear <[email protected]> Candela Technologies Inc http://www.candelatech.com _______________________________________________ ath10k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 8 lines · Michal Kazior
Quoted 1 line
From what I understand chained msdu is a msdu that hasn't fit into the
rx buffer and is split across the popped amsdu list. I suspect only the first msdu in chain has the
htt_rx_descand all other have not (this is what the current code does, but you'll need to verify that).I would try to concatenate all msdus into one (lots of memcpy :( ) or increase the
HTT_RX_BUF_SIZEso that A-MSDU frames can fit into a single buffer (hopefully FW/HW is capable of doing that).
I got this working, basically using memcpy approach.
Throughput is comparable what I was seeing on stock firmware, but I do notice an issue:
The reported rx speed is almost always 6Mbps when I drive at high speeds.
(At low speeds I see rx rate reported at 1.3Gbps most of the time.)
I am pretty sure this is related to the msdu-chaining somehow.
There are lots of FIXME's in the ath10k_htt_rx_amsdu_pop method.
Any idea where the rx rate problem might lie?
Thanks,
Ben
Signature · 8 lines
Ben Greear <[email protected]> Candela Technologies Inc http://www.candelatech.com _______________________________________________ ath10k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 10 lines · Ben Greear
Quoted 9 lines · Michal Kazior
Quoted 5 lines · Ben Greear
I'm using raw-rx mode in my firmware, and it's throughput is extremely poor at higher speeds. I see large numbers of warnings about
msdu_chaining, so I suspect this may be the reason.
Quoted 6 lines
Quoted 1 line
I assume this requires your custom FW?
Yes. I am not sure you can use raw tx with standard firmware (yet)?
Or at least, you cannot have tx cooked and rx raw.
I recall there were issues with different tx/rx modes. I think the 10.x branch suffers from this, but the FW 636 does not as it has tx_encap_type vdev parameter.
Quoted 25 lines
Quoted 16 lines
Quoted 9 lines
Does anyone have any suggestions for what needs to be done to handle these sorts of frames properly?
/* FIXME: we do not support chaining yet.
* this needs investigation */
if (msdu_chaining) {
ath10k_warn("htt rxmsdu_chainingis
true\n");Quoted 6 lines
ath10k_htt_rx_free_msdu_chain(msdu_head); continue; }From what I understand chained msdu is a msdu that hasn't fit into the
Quoted 7 lines
rx buffer and is split across the popped amsdu list. I suspect only the first msdu in chain has the
htt_rx_descand all other have not (this is what the current code does, but you'll need to verify that).I would try to concatenate all msdus into one (lots of memcpy :( ) or increase the
HTT_RX_BUF_SIZEso that A-MSDU frames can fit into a single buffer (hopefully FW/HW is capable of doing that).
Quoted 7 lines
I think in raw mode the hardware/firmware may not try to re-assemble the skb,
or also when no peer object can be found. This might be an issue in monitor mode as well, but I'm not certain about that.I think there is a way to chain skb data chunks w/out doing the copying, but I am not certain how that would be done properly.
Did you try increasing the HTT_RX_BUF_SIZE and see if you get complete frames instead of chained ones?
Michał
_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 7 lines · Ben Greear
Quoted 5 lines · Michal Kazior
Quoted 1 line
From what I understand chained msdu is a msdu that hasn't fit into the
rx buffer and is split across the popped amsdu list. I suspect only the first msdu in chain has the
htt_rx_descand all other have not (this is what the current code does, but you'll need to verify that).
Quoted 8 lines
Quoted 3 lines
I would try to concatenate all msdus into one (lots of memcpy :( ) or increase the
HTT_RX_BUF_SIZEso that A-MSDU frames can fit into a single buffer (hopefully FW/HW is capable of doing that).I got this working, basically using memcpy approach.
Throughput is comparable what I was seeing on stock firmware, but I do notice an issue:
Doing memcpy() is pretty bad, though. It'll hurt you, sooner or later, or hurt other with less powerful host systems.
The problem is chained msdus suggest A-MSDU rx. This means you re-assemble the aggregate, just to pass it to mac80211 which splits the frame into 802.3 frames (again, memcpy). This can easily flush you d-cache.
Quoted 9 lines
The reported rx speed is almost always 6Mbps when I drive at high speeds.
(At low speeds I see rx rate reported at 1.3Gbps most of the time.)
I am pretty sure this is related to the msdu-chaining somehow.
There are lots of FIXME's in the
ath10k_htt_rx_amsdu_popmethod.Any idea where the rx rate problem might lie?
Interesting. Rx rates are computed from the htt_rx_indication structure, not popped frames from rx ring. I'm guessing FW/HW doesn't fill in some stuff for raw rx at all. You probably can't do much about it if it's HW. You can probably hack your way around if it's just FW not copying some register values.
Michał
_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 17 lines · Michal Kazior
Quoted 7 lines · Ben Greear
Quoted 5 lines · Michal Kazior
Quoted 1 line
From what I understand chained msdu is a msdu that hasn't fit into the
rx buffer and is split across the popped amsdu list. I suspect only the first msdu in chain has the
htt_rx_descand all other have not (this is what the current code does, but you'll need to verify that).Quoted 8 lines
Quoted 3 lines
I would try to concatenate all msdus into one (lots of memcpy :( ) or increase the
HTT_RX_BUF_SIZEso that A-MSDU frames can fit into a single buffer (hopefully FW/HW is capable of doing that).I got this working, basically using memcpy approach.
Throughput is comparable what I was seeing on stock firmware, but I do notice an issue:
Quoted 7 lines
Doing
memcpy()is pretty bad, though. It'll hurt you, sooner or later, or hurt other with less powerful host systems.The problem is chained msdus suggest A-MSDU rx. This means you re-assemble the aggregate, just to pass it to mac80211 which splits the frame into 802.3 frames (again, memcpy). This can easily flush you d-cache.
That is interesting...maybe a better API would allow us to pass a-msdu chains directly to mac80211?
Quoted 8 lines
Quoted 8 lines
The reported rx speed is almost always 6Mbps when I drive at high speeds.
(At low speeds I see rx rate reported at 1.3Gbps most of the time.)
I am pretty sure this is related to the msdu-chaining somehow.
There are lots of FIXME's in the
ath10k_htt_rx_amsdu_popmethod.
Quoted 7 lines
Quoted 1 line
Any idea where the rx rate problem might lie?
Interesting. Rx rates are computed from the
htt_rx_indicationstructure, not popped frames from rx ring. I'm guessing FW/HW doesn't fill in some stuff for raw rx at all. You probably can't do much about it if it's HW. You can probably hack your way around if it's just FW not copying some register values.
Ok, I'll go look at this in more detail.
I do see expected rates at lower speeds (where a-msdu does not happen), so it's not *just* a problem with raw rx mode.
Thanks,
Ben
Signature · 8 lines
Ben Greear <[email protected]> Candela Technologies Inc http://www.candelatech.com _______________________________________________ ath10k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 7 lines · Michal Kazior
Quoted 7 lines
Quoted 7 lines
rx buffer and is split across the popped amsdu list. I suspect only the first msdu in chain has the
htt_rx_descand all other have not (this is what the current code does, but you'll need to verify that).I would try to concatenate all msdus into one (lots of memcpy :( ) or increase the
HTT_RX_BUF_SIZEso that A-MSDU frames can fit into a single buffer (hopefully FW/HW is capable of doing that).
Just FYI: At least on my firmware in raw rx mode, increasing the HTT_RX_BUF_SIZE (to 4 * 1920) and at least some chaining remains. Performance did not change noticeably. I'm using fairly powerful core i7 processor systems, so maybe the memcpy doesn't make enough difference to notice in my tests.
I did not put any effort into figuring out why.
I'm currently getting about 540Mbps upload TCP goodput, and only 420Mbps download TCP goodput. Not sure why the discrepancy, but perhaps the rx raw performance is worse for a variety of reasons. My firmware changes to support multiple stations to same AP may also be slowing things down, though these numbers are from a single station test...
Thanks,
Ben
Signature · 8 lines
Ben Greear <[email protected]> Candela Technologies Inc http://www.candelatech.com _______________________________________________ ath10k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 10 lines · Ben Greear
Quoted 7 lines · Michal Kazior
Quoted 7 lines
Quoted 7 lines
rx buffer and is split across the popped amsdu list. I suspect only the first msdu in chain has the
htt_rx_descand all other have not (this is what the current code does, but you'll need to verify that).I would try to concatenate all msdus into one (lots of memcpy :( ) or increase the
HTT_RX_BUF_SIZEso that A-MSDU frames can fit into a single buffer (hopefully FW/HW is capable of doing that).
Quoted 7 lines
Just FYI: At least on my firmware in raw rx mode, increasing the
HTT_RX_BUF_SIZE(to 4 * 1920) and at least some chaining remains. Performance did not change noticeably. I'm using fairly powerful core i7 processor systems, so maybe the memcpy doesn't make enough difference to notice in my tests.I did not put any effort into figuring out why.
Getting rid of memcpy() was a huge performance win for AP135 and its MIPS processor.
Quoted 7 lines
I'm currently getting about 540Mbps upload TCP goodput, and only 420Mbps download TCP goodput. Not sure why the discrepancy, but perhaps the rx raw performance is worse for a variety of reasons. My firmware changes to support multiple stations to same AP may also be slowing things down, though these numbers are from a single station test...
Hmm, I assume you test this without any bridging. It's probably going to be a little slower due to tx timings being directly visible to the TCP subsystem because both TCP and ath10k are locally on the same machine. You could try moving the actual TCP endpoints behind bridges.
Or you're actually seeing the memcpy() at work...
Did you try to test performance on vanilla driver/firmware?
Michał
_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 17 lines · Michal Kazior
Quoted 16 lines · Ben Greear
Quoted 7 lines · Michal Kazior
Quoted 7 lines
Quoted 7 lines
rx buffer and is split across the popped amsdu list. I suspect only the first msdu in chain has the
htt_rx_descand all other have not (this is what the current code does, but you'll need to verify that).I would try to concatenate all msdus into one (lots of memcpy :( ) or increase the
HTT_RX_BUF_SIZEso that A-MSDU frames can fit into a single buffer (hopefully FW/HW is capable of doing that).Just FYI: At least on my firmware in raw rx mode, increasing the
HTT_RX_BUF_SIZE(to 4 * 1920) and at least some chaining remains. Performance did not change noticeably. I'm using fairly powerful core i7 processor systems, so maybe the memcpy doesn't make enough difference to notice in my tests.
Quoted 4 lines
Quoted 1 line
I did not put any effort into figuring out why.
Getting rid of
memcpy()was a huge performance win for AP135 and its MIPS processor.
No doubt, but at this point, my problems appear to lie elsewhere.
Quoted 8 lines
Quoted 7 lines
I'm currently getting about 540Mbps upload TCP goodput, and only 420Mbps download TCP goodput. Not sure why the discrepancy, but perhaps the rx raw performance is worse for a variety of reasons. My firmware changes to support multiple stations to same AP may also be slowing things down, though these numbers are from a single station test...
Quoted 8 lines
Hmm, I assume you test this without any bridging. It's probably going to be a little slower due to tx timings being directly visible to the TCP subsystem because both TCP and ath10k are locally on the same machine. You could try moving the actual TCP endpoints behind bridges.
Or you're actually seeing the
memcpy()at work...Did you try to test performance on vanilla driver/firmware?
I have used vanilla firmware on AP for all tests, because my firmware will not do AP mode on WLE900VX for some reason. Using my slightly modified driver has no noticeable difference (and it now works virtually identical to upstream code when not using my modified firmware).
For station machine, vanilla firmware performs no better than my firmware, and I see the same issue where upload is 150Mbps or so faster than download.
I tried putting TCP/UDP endpoints on AP, and using AP as bridge, and both cases have similar throughput. Interestingly to me, UDP and TCP have similar thoughput, so it is unlikely that we are actually hitting limits on the spectrum (otherwise, UDP would do better because it has no ACK packets and generally runs much faster total throughput on wifi in my experience with /a/b/g/n NICs).
With vanilla firmware, there should be little to no amsdu packets (I assume), so it is unlikely to be related to memcpy. perf top shows no obvious hot spots in download test (running about 380Mbps in this case):
2.24% [kernel] [k] swiotlb_tbl_unmap_single 1.88% [kernel] [k] do_raw_spin_lock 1.88% [kernel] [k] ioread32 1.84% [kernel] [k] tcp_packet 1.53% [mac80211] [k] ieee80211_rx_handlers 1.31% [kernel] [k] copy_user_generic_string 1.19% [ath10k_core] [k] ath10k_htt_rx_amsdu.isra.29 1.14% btserver [.] do_big_while() 1.09% [kernel] [k] _raw_spin_lock_irqsave
What throughputs are you seeing, and what NICs are you using for AP and stations?
Thanks,
Ben
Signature · 8 lines
Ben Greear <[email protected]> Candela Technologies Inc http://www.candelatech.com _______________________________________________ ath10k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 58 lines · Ben Greear
Quoted 17 lines · Michal Kazior
Quoted 16 lines · Ben Greear
Quoted 7 lines · Michal Kazior
Quoted 7 lines
Quoted 7 lines
rx buffer and is split across the popped amsdu list. I suspect only the first msdu in chain has the
htt_rx_descand all other have not (this is what the current code does, but you'll need to verify that).I would try to concatenate all msdus into one (lots of memcpy :( ) or increase the
HTT_RX_BUF_SIZEso that A-MSDU frames can fit into a single buffer (hopefully FW/HW is capable of doing that).Just FYI: At least on my firmware in raw rx mode, increasing the
HTT_RX_BUF_SIZE(to 4 * 1920) and at least some chaining remains. Performance did not change noticeably. I'm using fairly powerful core i7 processor systems, so maybe the memcpy doesn't make enough difference to notice in my tests.Quoted 4 lines
Quoted 1 line
I did not put any effort into figuring out why.
Getting rid of
memcpy()was a huge performance win for AP135 and its MIPS processor.No doubt, but at this point, my problems appear to lie elsewhere.
Quoted 8 lines
Quoted 7 lines
I'm currently getting about 540Mbps upload TCP goodput, and only 420Mbps download TCP goodput. Not sure why the discrepancy, but perhaps the rx raw performance is worse for a variety of reasons. My firmware changes to support multiple stations to same AP may also be slowing things down, though these numbers are from a single station test...
Quoted 8 lines
Hmm, I assume you test this without any bridging. It's probably going to be a little slower due to tx timings being directly visible to the TCP subsystem because both TCP and ath10k are locally on the same machine. You could try moving the actual TCP endpoints behind bridges.
Or you're actually seeing the
memcpy()at work...Did you try to test performance on vanilla driver/firmware?
I have used vanilla firmware on AP for all tests, because my firmware will not do AP mode on WLE900VX for some reason. Using my slightly modified driver has no noticeable difference (and it now works virtually identical to upstream code when not using my modified firmware).
For station machine, vanilla firmware performs no better than my firmware, and I see the same issue where upload is 150Mbps or so faster than download.
I tried putting TCP/UDP endpoints on AP, and using AP as bridge, and both cases have similar throughput. Interestingly to me, UDP and TCP have similar thoughput, so it is unlikely that we are actually hitting limits on the spectrum (otherwise, UDP would do better because it has no ACK packets and generally runs much faster total throughput on wifi in my experience with /a/b/g/n NICs).
Quoted 2 lines
With vanilla firmware, there should be little to no amsdu packets (I assume), so it is unlikely to be related to memcpy. perf top
Incorrect. There's actually quite a lot of amsdu with vanilla firmware (keep in mind I refer to nwifi rx). In the early days ath10k was stiching msdus back just to be teared down again in mac80211 and this was hitting performance.
Quoted 13 lines
shows no obvious hot spots in download test (running about 380Mbps in this case):
2.24% [kernel] [k] swiotlb_tbl_unmap_single 1.88% [kernel] [k] do_raw_spin_lock 1.88% [kernel] [k] ioread32 1.84% [kernel] [k] tcp_packet 1.53% [mac80211] [k] ieee80211_rx_handlers 1.31% [kernel] [k] copy_user_generic_string 1.19% [ath10k_core] [k] ath10k_htt_rx_amsdu.isra.29 1.14% btserver [.] do_big_while() 1.09% [kernel] [k] _raw_spin_lock_irqsave
Quoted 2 lines
What throughputs are you seeing, and what NICs are you using for AP and stations?
With current master branch CUS223-CUS223 should get over 800mbps in udp tx/rx and 700mbps in tcp tx/rx with cabled setup on a poor AP135. Since AP135's CPU doesn't have any time to idle around I expect x86 to perform better and I don't think OTA should be terribly slower in a reasonably clean environment.
Michał
_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 4 lines · Michal Kazior
Incorrect. There's actually quite a lot of amsdu with vanilla firmware (keep in mind I refer to nwifi rx). In the early days ath10k was stiching msdus back just to be teared down again in mac80211 and this was hitting performance.
I guess with native wifi you do not see any chained amsdu frames at ath10k now? Out of curiosity, how do amsdu frames appear in ath10k? Does the firmware take care of everything before it ever gets to the host for native wifi?
Quoted 8 lines
Quoted 2 lines
What throughputs are you seeing, and what NICs are you using for AP and stations?
With current master branch CUS223-CUS223 should get over 800mbps in udp tx/rx and 700mbps in tcp tx/rx with cabled setup on a poor AP135. Since AP135's CPU doesn't have any time to idle around I expect x86 to perform better and I don't think OTA should be terribly slower in a reasonably clean environment.
Over-the-air, CUS223 wasn't any better than WLE900VX, so probably not an issue with the NICs. Arranging the antennas a bit gave me a bit of a boost, but still only around 600Mbps TCP upload, and download was still in the 400Mbps range. There are no APs in scan range in the lower end of 5Ghz where I am running, so it should be pretty clean.
I cabled WLE900VX NICs up with 30dB attenuation, and I do see about 800Mbps UDP upload, 700Mbps TCP upload, and 460Mbps TCP download. Download still sucks compared to upload for me.
I switched functions: AP logic moved to STA machine, vice-versa. The mis-match between upload follows the stations/APs, not the hardware.
(In this test, both machines are using stock 467.1-1 firmware.)
Maybe something with how I am configuring APs interfaces is the problem.
Can you verify that you see similar upload and download speeds for your test scenario?
Also, at 800Mbps UDP, perf top main offender is a 17% raw spin lock, but in general my system was not working too hard as far as I can tell.
Thanks,
Ben
Signature · 8 lines
Ben Greear <[email protected]> Candela Technologies Inc http://www.candelatech.com _______________________________________________ ath10k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath10k
Quoted 11 lines · Ben Greear
Quoted 4 lines · Michal Kazior
Incorrect. There's actually quite a lot of amsdu with vanilla firmware (keep in mind I refer to nwifi rx). In the early days ath10k was stiching msdus back just to be teared down again in mac80211 and this was hitting performance.
I guess with native wifi you do not see any chained amsdu frames at ath10k now? Out of curiosity, how do amsdu frames appear in ath10k? Does the firmware take care of everything before it ever gets to the host for native wifi?
For nwifi amsdu you get linked skbuffs. Each skbuff is a 802.11 non-qos data frame. ath10k fixes the header (using decapped header in rx_hdr_status) and submits each subframe separately as an independent 802.11 frame. This required a small change to be done in mac80211 (RX_FLAG_AMSDU_MORE) because all subframes share the same sequence number and it conflicted with retransmission/deduplication countermeasures.
Quoted 22 lines
Quoted 8 lines
Quoted 2 lines
What throughputs are you seeing, and what NICs are you using for AP and stations?
With current master branch CUS223-CUS223 should get over 800mbps in udp tx/rx and 700mbps in tcp tx/rx with cabled setup on a poor AP135. Since AP135's CPU doesn't have any time to idle around I expect x86 to perform better and I don't think OTA should be terribly slower in a reasonably clean environment.
Over-the-air, CUS223 wasn't any better than WLE900VX, so probably not an issue with the NICs. Arranging the antennas a bit gave me a bit of a boost, but still only around 600Mbps TCP upload, and download was still in the 400Mbps range. There are no APs in scan range in the lower end of 5Ghz where I am running, so it should be pretty clean.
I cabled WLE900VX NICs up with 30dB attenuation, and I do see about 800Mbps UDP upload, 700Mbps TCP upload, and 460Mbps TCP download. Download still sucks compared to upload for me.
I switched functions: AP logic moved to STA machine, vice-versa. The mis-match between upload follows the stations/APs, not the hardware.
Quoted 6 lines
(In this test, both machines are using stock 467.1-1 firmware.)
Maybe something with how I am configuring APs interfaces is the problem.
Can you verify that you see similar upload and download speeds for your test scenario?
I'm busy so I can't investigate this much, sorry.
Like I mentioned in my other mail - on AP135 UDP is 800mbps, TCP is 700mbps, both ways, and the bottleneck is AP135's CPU (MIPS 74Kc @ 720MHz).
You might try to change your qdiscs to pfifo and check if it makes any difference for you.
Quoted 2 lines
Also, at 800Mbps UDP, perf top main offender is a 17% raw spin lock, but in general my system was not working too hard as far as I can tell.
To be honest I haven't tested x86-x86 performance for some time now. I've been mainly focused on AP135. It bottlenecks elsewhere (it is non-preemptive so spinlocks work completely different).
Michał
_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k