On 2012-10-20 06:24, Aurelien Jarno wrote:
> Agreed. On the other hand, the few lines just above the one your quoted
> would be more complicated:
>
> +if (!(call_flags & TCG_CALL_NO_READ_GLOBALS)) {
> +/* globals should be synced to memory */
> +
On Wed, Oct 10, 2012 at 10:11:54AM -0700, Richard Henderson wrote:
> On 10/09/2012 12:56 PM, Aurelien Jarno wrote:
> > +if (!(call_flags & (TCG_CALL_NO_WRITE_GLOBALS |
> > +TCG_CALL_NO_READ_GLOBALS))) {
>
> Code like this would be shorter
On 10/10/2012 10:12 AM, Richard Henderson wrote:
> On 10/09/2012 01:24 PM, Aurelien Jarno wrote:
>> Maybe NO_RG_SE?
>
> Perhaps, yes. But I certainly can't think of anything better.
How about NO_RWG_SE. I like having the fact that R implies W
be included in the symbol.
r~
On 10/09/2012 01:24 PM, Aurelien Jarno wrote:
> Maybe NO_RG_SE?
Perhaps, yes. But I certainly can't think of anything better.
r~
On 10/09/2012 12:56 PM, Aurelien Jarno wrote:
> +if (!(call_flags & (TCG_CALL_NO_WRITE_GLOBALS |
> +TCG_CALL_NO_READ_GLOBALS))) {
Code like this would be shorter, and perhaps clearer, by
> +/* Helper does not read globals (either directl
The current helper flags, TCG_CALL_CONST and TCG_CALL_PURE might be
confusing and doesn't provide enough granularity for some helpers (FP
helpers for example).
This patch changes them into the following helpers flags:
- TCG_CALL_NO_READ_GLOBALS means that the helper does not read globals,
either
On Tue, Oct 09, 2012 at 09:04:24PM +0100, Peter Maydell wrote:
> On 9 October 2012 20:56, Aurelien Jarno wrote:
> > +#define TCG_CALL_NO_RGSE(TCG_CALL_NO_RG | TCG_CALL_NO_SE)
>
> When I first read this (in the target-arm patch) I assumed it had
> to be a typo for NO_REGS...
>
> (alas I c
On 9 October 2012 20:56, Aurelien Jarno wrote:
> +#define TCG_CALL_NO_RGSE(TCG_CALL_NO_RG | TCG_CALL_NO_SE)
When I first read this (in the target-arm patch) I assumed it had
to be a typo for NO_REGS...
(alas I can't think of a better name)
-- PMM
