Date: Sat, 4 Dec 2021 12:42:07 +1030 From: Daniel O'Connor via freebsd-usb <freebsd-usb@freebsd.org> To: Hans Petter Selasky <hps@selasky.org> Cc: freebsd-usb@freebsd.org Subject: Re: Poor USB performance on ASUS 520 motherboard (no IRQ?) Message-ID: <D9637948-8B50-45D3-93BD-ADB5FE7D964B@dons.net.au> In-Reply-To: <abf22f4f-e8fc-86c3-e463-d98d4353591a@selasky.org> References: <A4A84F57-67F2-4402-B81C-9742F2C2B218@dons.net.au> <abf22f4f-e8fc-86c3-e463-d98d4353591a@selasky.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> On 4 Dec 2021, at 05:11, Hans Petter Selasky <hps@selasky.org> wrote: > On 11/24/21 23:33, Daniel O'Connor wrote: >> I found that the Gigabyte GA-A320M-H one works fine but the ASUS = A520M-A/CSM one is significantly slower for tests where there are many = back and forth messages (streaming data seems fine). >=20 > There is a register in the XHCI PCI area, to set the IRQ latency. You = may want to check the PCI IDs towards Linux if these chips require some = kind of special quirks. Thanks, I had a look at the Linux code (searched via lxr.linux.no) but = couldn't find any mention of that product ID (0x43ec). There are several other AMD related quirks but I don't think any of them = impact this. I wonder if the BIOS sets them differently and FreeBSD never changes it = (I haven't tried Linux). How can I change it? -- 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?D9637948-8B50-45D3-93BD-ADB5FE7D964B>