qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Sven Schnelle <svens@stackframe.org>
To: Gerd Hoffmann <kraxel@redhat.com>
Cc: "Helge Deller" <deller@gmx.de>,
	"Philippe Mathieu-Daudé" <f4bug@amsat.org>,
	qemu-devel@nongnu.org
Subject: Re: [PATCH 3/3] hw/display/artist: rewrite vram access mode handling
Date: Tue, 25 Jan 2022 17:29:53 +0100	[thread overview]
Message-ID: <87tudr6any.fsf@x1.stackframe.org> (raw)
In-Reply-To: <20220125082552.uupa7yyqdnpxgopw@sirius.home.kraxel.org> (Gerd Hoffmann's message of "Tue, 25 Jan 2022 09:25:52 +0100")

Gerd Hoffmann <kraxel@redhat.com> writes:

>   Hi,
>
>> +static void artist_vram_write4(ARTISTState *s, struct vram_buffer *buf,
>> +                               uint32_t offset, uint32_t data)
>
>> +static int get_vram_offset(ARTISTState *s, struct vram_buffer *buf,
>> +                           int pos, int posy)
>
>> +    case 0x13a0:
>> +        artist_vram_write4(s, buf, get_vram_offset(s, buf, pos >> 2, posy),
>> +                           data);
>
> That is asking for trouble.
>
> You should pass around offsets not pointers.  An offset can trivially be
> checked whenever it is within the valid range (i.e. smaller than vram
> size), or it can be masked to strip off high bits when accessing virtual
> vram.  You need that for robustness and security reasons (i.e. make sure
> the guest can't write to host memory by tricking your get_vram_offset
> calculations).

I'm not sure i understand the problem. get_vram_offset() returns an
offset, which is passed to artist_vram_write4() which itself doesn't
do anything on the buffer. artist_rop8() in the end accesses the buffer,
and that function checks whether it's < buf->size. Can you elaborate
a bit more? Maybe it's just so obvious that i don't see it.

Thanks,
Sven


  reply	other threads:[~2022-01-25 17:03 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-21 22:16 [PATCH 0/3] hw/display/artist: cursor & buffer mode fixes Sven Schnelle
2022-01-21 22:16 ` [PATCH 1/3] hw/display/artist: fix cursor position Sven Schnelle
2022-01-21 22:16 ` [PATCH 2/3] hw/display/artist: allow to disable/enable cursor Sven Schnelle
2022-01-25  8:11   ` Gerd Hoffmann
2022-01-21 22:16 ` [PATCH 3/3] hw/display/artist: rewrite vram access mode handling Sven Schnelle
2022-01-25  8:25   ` Gerd Hoffmann
2022-01-25 16:29     ` Sven Schnelle [this message]
2022-01-26  9:08       ` Gerd Hoffmann

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=87tudr6any.fsf@x1.stackframe.org \
    --to=svens@stackframe.org \
    --cc=deller@gmx.de \
    --cc=f4bug@amsat.org \
    --cc=kraxel@redhat.com \
    --cc=qemu-devel@nongnu.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).