Re: [PATCH v8 00/13] Add Broadcom VK driver

[Greg Kroah-Hartman]

On Tue, Jan 19, 2021 at 12:21:32PM -0800, Scott Branden wrote:
> Hi Greg,
> 
> On 2021-01-18 3:06 p.m., Olof Johansson wrote:
> > On Sun, Jan 17, 2021 at 11:17 PM Greg Kroah-Hartman
> > <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> >> On Sun, Jan 17, 2021 at 10:47:59AM -0800, Olof Johansson wrote:
> >>> Hi,
> >>>
> >>> On Mon, Nov 30, 2020 at 10:42 AM Scott Branden
> >>> <scott.branden@xxxxxxxxxxxx> wrote:
> >>>> This patch series drops previous patches in [1]
> >>>> that were incorporated by Kees Cook into patch series
> >>>> "Introduce partial kernel_read_file() support" [2].
> >>>>
> >>>> Remaining patches are contained in this series to add Broadcom VK driver.
> >>>> (which depends on request_firmware_into_buf API addition which has
> >>>> now been accepted into the upstream kernel as of v5.10-rc1).
> >>>>
> >>>> [1] https://lore.kernel.org/lkml/20200706232309.12010-1-scott.branden@xxxxxxxxxxxx/
> >>>> [2] https://lore.kernel.org/lkml/20201002173828.2099543-1-keescook@xxxxxxxxxxxx/
> >>>
> >>> I've been following this series for some time, and I think the code is
> >>> ready to go in.
> >>>
> >>> Greg, mind queuing this up in the misc tree?
> >> I will need a new version, this is long gone from my queue.
> > I'll let Scott repost then (with acks applied etc)
> I can send another patch version with Olof's acks applied to each patch.
> Please let me know if that is what you are looking for?

Yes please.

> >> And hopefully the tty layer abuse is gone... :)
> > There's a simple tty driver as the final patch in the series, but it's
> > pretty straightforward.
> >
> > If you've still got concerns with it, the rest of the series should
> > stand on its own and should be mergeable without that piece.
> Yes, I placed the patch at the end of the series so it can be dropped
> if that is what is required to accept the rest of the patches.
> 
> There has been no viable solution suggested for replacing this functionality.
> We need a tty-like interface that works via access to the circular buffers
> in PCIe BAR space and interrupts.
> 
> The vk tty devices are a direct replacement to attaching serial cables.
> In real production environment it is not possible to attach such cables.
> 
> I can work on an alternative tty solution and send such patch later
> but I don't know what is going to prevent the tty "abuse".

I can't remember the specifics (I review hundreds of patches a week),
but please just resend and I'll deal with it then :)

thanks,

greg k-h