linux-kernel - Re: [PATCH V2 1/4] drivers/fpga/amd: Add new driver amd versal-pci

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Z6Vtz/Bb8wsIH0pG@yilunxu-OptiPlex-7050>
Date: Fri, 7 Feb 2025 10:19:59 +0800
From: Xu Yilun <yilun.xu@...ux.intel.com>
To: Yidong Zhang <yidong.zhang@....com>
Cc: linux-kernel@...r.kernel.org, linux-fpga@...r.kernel.org,
	mdf@...nel.org, hao.wu@...el.com, yilun.xu@...el.com,
	lizhi.hou@....com, DMG Karthik <Karthik.DMG@....com>,
	Nishad Saraf <nishads@....com>,
	Prapul Krishnamurthy <prapulk@....com>,
	Hayden Laccabue <hayden.laccabue@....com>
Subject: Re: [PATCH V2 1/4] drivers/fpga/amd: Add new driver amd versal-pci

> > No need to detach a specific driver, remove all devices in the
> > fpga-region. I imagine this could also be a generic flow for all PCI/e
> > based FPGA cards.
> 
> I see your point. Is there an existing example in current fpga driver for
> PCI/e based cards?

No. The fpga-region re-enumeration arch is still WIP, so no existing
implementation.

> 
> We will need to talk to our management team and re-design our driver.
> I think we have 2 approaches:
> 1) Align with linux fpga, having one driver for both userPF and mgmtPF; or

Don't get you. Linux FPGA doesn't require one driver for both PFs.

> 2) find a different location (maybe driver/misc) for the version-pci driver,
> because it is an utility driver, not need to be tied with fpga framework.

I'm not the misc maintainer, but I assume in-tree utility driver +
out-of-tree client driver is not generally welcomed.

Thanks,
Yilun

> 
> Please let me know you thoughts. Which way is acceptable by you.
> 
> Thanks,
> David
> > 
> > Thanks,
> > Yilun
> > 
> > > driver and allow the userPF driver re-enumerate all to match the new
> > > hardware.
> > > 
> > > I think my understanding is correct, it is doable.
> > > 
> > > As long as we can keep our userPF driver as separate driver, the code change
> > > won't be too big.
> > > 
> > > > 
> > > > Thanks,
> > > > Yilun