Received-SPF: pass (google.com: best guess record for domain of linux-wireless-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <CAHmv74wK2NYuj2JuDUz-og5w+e=H1=ef9eCSMTbO6otaYQatKg@mail.gmail.com>
 <9172166a-eb77-a508-ba54-08fdf96912fc@lwfinger.net> <CAHmv74wNOM1QKUPp2Vax2Z+-JyoFC81WHVZ+XbYgiRMUGxCFUQ@mail.gmail.com>
 <9492be2d-abd1-0e1d-5d80-1a72b0250749@lwfinger.net>
In-Reply-To: <9492be2d-abd1-0e1d-5d80-1a72b0250749@lwfinger.net>
From:   Sergey Kharitonov <haritonovsb@gmail.com>
Date:   Sat, 31 Aug 2019 01:45:07 +0400
Message-ID: <CAHmv74x008FvhDztqOq-EC=d7fSXN7=kEOhQm3TPZn1PDZP=fQ@mail.gmail.com>
Subject: Re: [Linux Realtek Wi-Fi support] rtl8192ee related bug
To:     Larry Finger <Larry.Finger@lwfinger.net>
Cc:     linux-wireless@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Sender: linux-wireless-owner@vger.kernel.org
Precedence: bulk

Sounds good!
I made a little change to compile that old rtl8192ee_revised.tar.bz2
driver on Linux 5.2.X and I get <1ms overall ping to the router with
rare spikes to 1-2ms.
I'm excited to know the right driver can provide the same performance,
it's starting to look now like the hardware's achievable maximum.
Is there any chance an improved rtl8192ee would be included in Linux 5.3?

On Sat, Aug 31, 2019 at 12:36 AM Larry Finger <Larry.Finger@lwfinger.net> wrote:
>
> On 8/10/19 1:13 PM, Sergey Kharitonov wrote:
> > Dear Mr. Finger!
> > Thank you for paying attention to the problem. I've just posted an
> > updated information to that bug before receiving this message.
> >
> > I get these results
> > ~]$ ping 192.168.88.1
> > PING 192.168.88.1 (192.168.88.1) 56(84) bytes of data.
> > 64 bytes from 192.168.88.1: icmp_seq=1 ttl=64 time=0.746 ms
> > 64 bytes from 192.168.88.1: icmp_seq=2 ttl=64 time=0.868 ms
> > 64 bytes from 192.168.88.1: icmp_seq=3 ttl=64 time=0.853 ms
> > 64 bytes from 192.168.88.1: icmp_seq=4 ttl=64 time=0.868 ms
> > 64 bytes from 192.168.88.1: icmp_seq=5 ttl=64 time=0.818 ms
> > 64 bytes from 192.168.88.1: icmp_seq=6 ttl=64 time=0.844 ms
> > 64 bytes from 192.168.88.1: icmp_seq=7 ttl=64 time=0.841 ms
> > 64 bytes from 192.168.88.1: icmp_seq=8 ttl=64 time=0.913 ms
> > 64 bytes from 192.168.88.1: icmp_seq=9 ttl=64 time=0.844 ms
> > 64 bytes from 192.168.88.1: icmp_seq=10 ttl=64 time=0.886 ms
> > ^C
> > --- 192.168.88.1 ping statistics ---
> > 10 packets transmitted, 10 received, 0% packet loss, time 387ms
> > rtt min/avg/max/mdev = 0.746/0.848/0.913/0.044 ms
> >
> > with a driver from rtl8192ee_revised.tar.bz2 file in
> > https://github.com/lwfinger/rtlwifi_new/tree/extended repository built
> > against Linux 4.19.65-1-lts ArchLinux current LTS kernel.
>
> I have completed but not yet sent some updates for rtl8192ee. Using them and
> kernel 5.3-rc6, I am getting the following:
>
> --- 192.168.1.1 ping statistics ---
> 100 packets transmitted, 100 received, 0% packet loss, time 99134ms
> rtt min/avg/max/mdev = 0.881/1.487/8.198/0.844 ms
>
> This is using the "old" firmware with the following md5sum:
>
> c8d25646cbc9efdcb818fc95ad5836a9  /lib/firmware/rtlwifi/rtl8192eefw.bin
>
> The new driver patches convert a set of macros to read and write the descriptors
> into inline routines. The calling sequences are also converted from u8 * to
> __le32 *, but I would not expect much improvement with the latter changes.
>
> Using a local speed test web site, I am getting 100 Mbps download and 80 Mbps
> upload rates.
>
> Larry
>
>
>