Re: [Qemu-devel] [PATCH 6/9] Convert cpu_memory_rw_debug to use MMUAccessType

[David Gibson <david@gibson.dropbear.id.au>]

[-- Attachment #1: Type: text/plain, Size: 1266 bytes --]

On Sun, Jul 10, 2016 at 08:32:32PM +0100, Peter Maydell wrote:
> On 8 July 2016 at 04:42, David Gibson <david@gibson.dropbear.id.au> wrote:
> > My only concern here is that the constants are named
> > *MMU*_DATA_... whereas these are physical memory accesses not
> > involving the MMU.  I can't actually see any current users of
> > MMUAccessType which makes me a bit confused as to what it's intended
> > meaning was
> 
> If you grep for MMU_DATA_LOAD/MMU_DATA_STORE/MMU_INST_FETCH
> you'll see the uses. A lot of the softmmu code uses the
> convention of 0=read,1=write,2=insn (which developed I
> think historically from a bool "is_write", which you'll
> still see in some function argument names, that was
> augmented to handle insn-fetch separately). The enum
> gives us some symbolic names for the constant values.
> (There's a proposed patch somewhere to change the
> 'int is_write' arguments to actually use the enum type.)

Ah, yes, I see.  Still surprisingly few, actually.

My concern about the potentially misleading name still stands..

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]