Date: Thu, 25 Nov 2021 16:20:30 +1030 From: Daniel O'Connor via freebsd-stable <freebsd-stable@freebsd.org> To: freebsd-stable <freebsd-stable@freebsd.org> Subject: Re: Poor USB performance on ASUS 520 motherboard (no IRQ?) Message-ID: <2DB18886-CC19-4061-9D70-E56FEBCB1F1F@dons.net.au> In-Reply-To: <1279E19E-50E4-41E6-8090-466A0AC8CFD3@dons.net.au> References: <1E968920-8819-4C2B-9572-38CC1002DC89@dons.net.au> <1279E19E-50E4-41E6-8090-466A0AC8CFD3@dons.net.au>
next in thread | previous in thread | raw e-mail | index | archive | help
> On 25 Nov 2021, at 14:19, Daniel O'Connor <darius@dons.net.au> wrote:
>> One thing I did notice is this dmesg output:
>> xhci0: <XHCI (generic) USB 3.0 controller> mem 0xfcfa0000-0xfcfa7fff =
at device 0.0 on pci1
>> xhci0: 32 bytes context size, 64-bit DMAxhci1: <XHCI (generic) USB =
3.0 controller
>> ...
>> xhci1: <XHCI (generic) USB 3.0 controller> mem 0xfcb00000-0xfcbfffff =
at device 0.3 on pci7
>> xhci1: 64 bytes context size, 64-bit DMA
>> ..
>> xhci2: <XHCI (generic) USB 3.0 controller> mem 0xfca00000-0xfcafffff =
at device 0.4 on pci7
>> xhci2: 64 bytes context size, 64-bit DMA
>=20
> It seems this is a bit of a red herring as vmstat -i does show it =
getting IRQs, not sure why they don't show up in dmesg though. Some =
discussion on IRC suggests it is because the system has no legacy PCI =
IRQs.
>=20
> However on the system in question the IRQ rate tops out at 1kHz and on =
the other it is 8kHz which I think explains my problem.
>=20
> I'm going to have a look for BIOS settings today but getting a bit =
desperate for ideas..
I had a PCIe USB card lying around:
xhci1@pci0:3:0:0: class=3D0x0c0330 rev=3D0x01 hdr=3D0x00 =
vendor=3D0x1106 device=3D0x3483 subvendor=3D0x1106 subdevice=3D0x3483
vendor =3D 'VIA Technologies, Inc.'
device =3D 'VL805/806 xHCI USB 3.0 Controller'
class =3D serial bus
subclass =3D USB
bar [10] =3D type Memory, range 64, base 0xfce00000, size 4096, =
enabled
And connecting my device to that fixes the speed problem (and shows =
>8000 IRQ/sec).
The difference between the A320 and A520 motherboards seem strange, =
since I would expect their USB controllers to be identical.
--
Daniel O'Connor
"The nice thing about standards is that there
are so many of them to choose from."
-- Andrew Tanenbaum
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?2DB18886-CC19-4061-9D70-E56FEBCB1F1F>