lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAFBinCB6Q67nG3Z33i=nDLcYGENCSHFKNq=ViJks0j9aXbcYpA@mail.gmail.com>
Date:   Mon, 3 Jul 2023 23:25:22 +0200
From:   Martin Blumenstingl <martin.blumenstingl@...glemail.com>
To:     Ping-Ke Shih <pkshih@...ltek.com>
Cc:     "linux-wireless@...r.kernel.org" <linux-wireless@...r.kernel.org>,
        "Lukas F. Hartmann" <lukas@...re.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "tony0620emma@...il.com" <tony0620emma@...il.com>,
        "jernej.skrabec@...il.com" <jernej.skrabec@...il.com>
Subject: Re: wifi: rtw88: question about SDIO RX aggregation limiting

Hello Ping-Ke,

sorry again for the long waiting time. I'll be quicker next time.

On Tue, Jun 20, 2023 at 7:26 AM Ping-Ke Shih <pkshih@...ltek.com> wrote:
[...]
> > > The unit of BIT_RXDMA_AGG_PG_TH is 1k bytes, so I think you can
> > > set mmc_host->max_req_size/1024.
> > I tried this but I got a result that I don't understand.
> > I've been testing with three BIT_RXDMA_AGG_PG_TH values on a SoC that
> > can handle 255 * 1024 bytes. Each time I connected to the same AP and
> > downloaded a bigger file over http(s).
> > BIT_RXDMA_AGG_PG_TH: biggest observed rx_len in rtw_sdio_rxfifo_recv()
> > 255: 20968
> > 6: 5122
> > 1: 1602
>
> Please also print out number of packets you receive, and then we can see how
> many packets aggregate.
sure - here are the results:
BIT_RXDMA_AGG_PG_TH: biggest observed rx_len in rtw_sdio_rxfifo_recv()
/ number of (aggregated) packets
255: 20824 / 12
6: 5128 / 4
1: 3132 / 1 (these were a few exceptions and I'm not able to reliably
reproduce it, 1602 / 1 is what I can easily reproduce)

> > The biggest rx_len I have observed for BIT_RXDMA_AGG_PG_TH 1 looks suspicious:
> > My understanding is that I shouldn't be seeing rx_len larger than
> > BIT_RXDMA_AGG_PG_TH * 1024.
> > BIT_RXDMA_AGG_PG_TH = 6 is within this limit but BIT_RXDMA_AGG_PG_TH =
> > 1 isn't (I'm seeing 578 extra bytes in addition to the 1024 bytes that
> > I was expecting).
>
> Assume threshold is 1k, and single one packet is larger than 1k. Hardware
> will not split it into two. Also, please make sure 0x280[29] BIT_EN_PRE_CALC
> is 1. Otherwise, it will possibly aggregate additional one packet to over
> the threshold.
>From the numbers above it seems most likely that we're hitting the
"one packet is larger than 1k" case.

Also I'm seeing:
  wlan0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue
state UP group default qlen 1000
My interface's MTU is 1500 bytes. Seeing 1602 bytes rx_len with one
packet is already odd (that would mean 102 bytes for overhead like RX
descriptor and other headers/metadata). But 3132 bytes rx_len is very
odd.

BIT_EN_PRE_CALC is set, see also the attached diff (it's not meant to
be applied anywhere - it's just so you understand what I've been
testing with).

> 0x280[15:8] is timeout time in unit of 1us for SDIO interface. When set
> threshold to 255, you can enlarge this to see if it can aggregate more as
> expected.
I did not experiment with this yet as I'd like to understand the above
findings first.


Best regards,
Martin

View attachment "rtw88-rx-agg-test.diff" of type "text/x-patch" (2834 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ