linux-kernel - Re: [PATCH v1 1/1] usb: dwc3: gadget: Revert "set gadgets parent to the right controller"

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] [day] [month] [year] [list]

Message-ID: <YVwxORtF1aQDsT08@kroah.com>
Date:   Tue, 5 Oct 2021 13:04:25 +0200
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     Michael Grzeschik <mgr@...gutronix.de>
Cc:     Ferry Toth <fntoth@...il.com>,
        Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
        linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org,
        Felipe Balbi <balbi@...nel.org>
Subject: Re: [PATCH v1 1/1] usb: dwc3: gadget: Revert "set gadgets parent to
 the right controller"

On Tue, Oct 05, 2021 at 10:51:00AM +0200, Michael Grzeschik wrote:
> On Mon, Oct 04, 2021 at 10:35:57PM +0200, Ferry Toth wrote:
> > Hi,
> > 
> > Op 04-10-2021 om 16:18 schreef Andy Shevchenko:
> > > The commit c6e23b89a95d ("usb: dwc3: gadget: set gadgets parent to the right
> > > controller") changed the device for the UDC and broke the user space scripts
> > > that instantiate the USB gadget(s) via ConfigFS.
> > 
> > I confirm this regression on Intel Edison since at least 5.15-rc2 while
> > in 5.14.0 it was working fine.
> > 
> > This patch resolves the issue as tested on 5.15-rc4.
> > 
> > Tested-by: Ferry Toth<fntoth@...il.com>
> 
> NACK! Why should we resolv an issue by reverting it to solve not working
> userspace. We already have this patch as a solution for solving a deeper
> Problem, regarding the allocator addressing the right device.
> 
> > > Revert it for now until the better solution will be proposed.
> 
> So, I think fixing the userspace would be the right fix, not changing
> the kernel. Otherwise we should find a proper solution.

We only really have one rule in Linux kernel development:

	If a kernel change breaks userspace, the kernel change needs to
	be reverted.

Go fix up the userspace tools first, ensure everyone has updated, and
then we can consider taking the change back into the kernel tree.

thanks,

greg k-h