All of lore.kernel.org
 help / color / mirror / Atom feed
From: Martin Peres <martin.peres-GANU6spQydw@public.gmane.org>
To: nouveau-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org
Subject: Re: idea of extracting microcode
Date: Mon, 24 Nov 2014 14:50:36 +0100	[thread overview]
Message-ID: <547337AC.70706@free.fr> (raw)
In-Reply-To: <CAPM=9txGnz5eSdoVzbN2tb7GJUs2-wR+c_-10z6qZ_EfAJ3xMQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>

On 24/11/14 11:19, Dave Airlie wrote:
> Hey,
>
> Probably mostly for Ben, but I had this idea that it might be possible
> to write a userspace wrapper for the binary driver that run as root
> could be used to do the card init for the current or even a fake GPU
> then produce traces to pull the firmwares out.
>
> Not sure how insane it is, vs having to mmiotrace, but I just thought
> I'd see if anyone had considered it.
>
>
> Dave.

Hello Dave,

For the moment, we have an even-better solution developed by Ilia. This 
script extracts firmwares from the blob's installer without having to 
run it further than decompressing it.

Here it is: 
https://raw.githubusercontent.com/imirkin/re-vp2/master/extract_firmware.py
And there is a package for archlinux here: 
https://aur.archlinux.org/packages/nouveau-fw/

This solution is not perfect but, at XDC2014, we discussed it with Andy 
Ritger from NVIDIA and we are looking into a way for them to release 
their firmwares separately from the blob for our usage (along with the 
rights to redistribute them).

As for your proposal of faking the GPU, we would need to change the 
mmiotracing tool to fake any MMIO request and couple that with a minimal 
hw model of NVIDIA GPUs. That being done, we would still need a 
mmiotrace with the real card to produce the golden context (the initial 
values for the context-switched registers).

In the end, i am not sure it is worth the effort as we will need to have 
most chipsets anyway for testing purposes...

Martin
_______________________________________________
Nouveau mailing list
Nouveau@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/nouveau

      parent reply	other threads:[~2014-11-24 13:50 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-11-24 10:19 idea of extracting microcode Dave Airlie
     [not found] ` <CAPM=9txGnz5eSdoVzbN2tb7GJUs2-wR+c_-10z6qZ_EfAJ3xMQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-11-24 13:50   ` Martin Peres [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=547337AC.70706@free.fr \
    --to=martin.peres-ganu6spqydw@public.gmane.org \
    --cc=nouveau-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.