Re: [PATCH] Remove doc on OpenRISC 1000

[Jeremy Bennett <jeremy dot bennett at embecosm dot com>]

On 21/03/13 12:01, Franck Jullien wrote:
> 2013/3/20 Pedro Alves <palves@redhat.com>:
> > On 03/20/2013 09:35 AM, Yao Qi wrote:
> > > [My mail sent yesterday didn't show in the mail archive, so send it again.]
> > >
> > > On 03/19/2013 09:35 PM, Jeremy Bennett wrote:
> > > > It is quite likely the GDB code for OpenRISC 1000 was never submitted.
> > > > The whole GNU tool chain was developed around 2000-2002, but only
> > > > binutils ever committed its code. The tool chain is still widely used
> > > > and maintained atwww.opencores.org.
> > >
> > > Hi Jeremy,
> > > The patch was submitted in the link I gave in my first mail, but only
> > > the doc bit was approved and committed.  I can't tell why the non-doc
> > > bits were not approved.
> > >
> > > > I've copied Franck Jullien, who is the most active developer of the
> > > > OpenRISC GDB port at present (I was in the past). The alternative to
> > > > deleting the documentation is to submit the port.
> > >
> > > If the port can be submitted soon (in 2~3 months maybe?), I am fine to
> > > keep the doc there, because it has been there for 10 years.  Otherwise,
> > > I prefer to remove them first, personally.
> >
> > I'd prefer removing them from our tree too (and not wait).
> > 10 years have passed, and lots of non-"target remote" targets have
> > been yanked from the tree meanwhile -- I'm not familiar with OpenRISC,
> > but it's arguable whether we want a new "target jtag" nowadays,
> > compared to using the remote target (against something that talks jtag),
> > and perhaps we have better mechanisms for "info or1k spr" today
> > too (I don't know what that actually does).
> >
> > I think starting from scratch with a clean submission, that
> > includes the corresponding docs would make a lot of sense.
> >
> > --
> > Pedro Alves
>
> Hi,
>
> Our OpenRISC port is not ready yet. So, I think you're right.
> You can remove the old openrisc documentation and we'll
> start from scratch with a clean submission.
In the light of Pedro and Franck's comments, I agree. We should start 
from scratch. The documentation doesn't even reflect reality - 
proprietary JTAG interface has long gone from OpenRISC GDB. It has used 
standard Remote Serial Protocol for a long time.


Jeremy

--
Tel:     +44 (1590) 610184
Cell:    +44 (7970) 676050
SkypeId: jeremybennett
Email:   jeremy.bennett@embecosm.com
Web:     www.embecosm.com
Twitter: @embecosm