Jeff Law writes:
> On 05/17/14 01:33, Richard Sandiford wrote:
>> I suppose we could put the onus on the users of the iterator to invoke
>> a "handle subrtxes of this code" routine once they know what the code is.
>> That could make things a bit ugly though. E.g.:
>>
>>FOR_EACH_SUBRTX (iter,
On 05/17/14 01:33, Richard Sandiford wrote:
I suppose we could put the onus on the users of the iterator to invoke
a "handle subrtxes of this code" routine once they know what the code is.
That could make things a bit ugly though. E.g.:
FOR_EACH_SUBRTX (iter, array, expr, NONCONST)
if (
Thanks for the comments.
Jeff Law writes:
>> Implementation-wise, the main observation is that most subrtxes are part
>> of a single contiguous sequence of "e" fields. E.g. when compiling an
>> oldish combine.ii on x86_64-linux-gnu with -O2, we iterate over the
>> subrtxes of 7,636,542 rtxes. O
On Fri, May 9, 2014 at 8:18 AM, Jeff Law wrote:
> On 05/07/14 14:52, Richard Sandiford wrote:
>>
>> I noticed for_each_rtx showing up in profiles and thought I'd have a go
>> at using worklist-based iterators instead. So far I have three:
>>
>>FOR_EACH_SUBRTX: iterates over const_rtx subrtxes
On 05/07/14 14:52, Richard Sandiford wrote:
I noticed for_each_rtx showing up in profiles and thought I'd have a go
at using worklist-based iterators instead. So far I have three:
FOR_EACH_SUBRTX: iterates over const_rtx subrtxes of a const_rtx
FOR_EACH_SUBRTX_VAR: iterates over rtx subrt
On Thu, May 08, 2014 at 07:25:50AM +0100, Richard Sandiford wrote:
> Trevor Saunders writes:
> > On Wed, May 07, 2014 at 09:52:49PM +0100, Richard Sandiford wrote:
> >> I noticed for_each_rtx showing up in profiles and thought I'd have a go
> >> at using worklist-based iterators instead. So far I
Richard Sandiford writes:
> Trevor Saunders writes:
>> On Wed, May 07, 2014 at 09:52:49PM +0100, Richard Sandiford wrote:
>>> I noticed for_each_rtx showing up in profiles and thought I'd have a go
>>> at using worklist-based iterators instead. So far I have three:
>>>
>>> FOR_EACH_SUBRTX: it
Trevor Saunders writes:
> On Wed, May 07, 2014 at 09:52:49PM +0100, Richard Sandiford wrote:
>> I noticed for_each_rtx showing up in profiles and thought I'd have a go
>> at using worklist-based iterators instead. So far I have three:
>>
>> FOR_EACH_SUBRTX: iterates over const_rtx subrtxes of
On Wed, May 07, 2014 at 09:52:49PM +0100, Richard Sandiford wrote:
> I noticed for_each_rtx showing up in profiles and thought I'd have a go
> at using worklist-based iterators instead. So far I have three:
>
> FOR_EACH_SUBRTX: iterates over const_rtx subrtxes of a const_rtx
> FOR_EACH_SUBRTX
On May 7, 2014, at 1:52 PM, Richard Sandiford
wrote:
>
> I've locally replaced all for_each_rtx calls in the generic code with
> these iterators and they make things reproducably faster. The speed-up
> on full --enable-checking=release ./cc1 and ./cc1plus times is only about 1%,
> but maybe tha
I noticed for_each_rtx showing up in profiles and thought I'd have a go
at using worklist-based iterators instead. So far I have three:
FOR_EACH_SUBRTX: iterates over const_rtx subrtxes of a const_rtx
FOR_EACH_SUBRTX_VAR: iterates over rtx subrtxes of an rtx
FOR_EACH_SUBRTX_PTR: iterates ov
11 matches
Mail list logo
