Re: [Qemu-devel] [PATCH 3/5] Add support for receiving via receive buffers. While the Intel documentation claims this is unsupported, the OS X drivers use it, causing an assertion failure since rx buffer size is 0.

[malc <av1474@comtv.ru>]

On Wed, 12 Aug 2009, Reimar D?ffinger wrote:

> On Wed, Aug 12, 2009 at 03:04:17AM +0400, malc wrote:
> > On Tue, 11 Aug 2009, Reimar D?ffinger wrote:
> > 

[..snip..]

> 
> Well, I had mostly the same thoughts while working on this, but I
> thought it preferable to keep with the current "style" even if it is
> bad/problematic.

I sort of thought that would be the case.

> Discussing the best way to do it and then fixing all the code at a later
> point seemed like a better idea (well, if the goal is to get this
> applied without having to fix all those little issues with the current
> code first).

Sure.

> Particularly in this case it is simple and still readable to just use
> lduw_phys I think if you prefer...
> I attached a proof-of concept patch that converts parts of it, but doing
> the same for the tx_t and statistics_t structs would get a bit messy
> without changing the overall code a bit.
> Also I expect it to be full of stuff that can be discussed for weeks,
> i.e. typical bikeshed material...
> And maybe this is the kind of thing Stefan Weil might want to take care
> of?
> 
> P.S.: And I'd appreciate it if you wouldn't CC me, I am subscribed to the
> list. Thanks.

Well, Anthony changed the list settings to CC everyone, not CCing someone
is actually more work (couple of keystrokes) than doing it, so all your
Cc credits do belong to him.

-- 
mailto:av1474@comtv.ru