On Tue, 11 Jun 2019, Kewen.Lin wrote:
> >> If my understanding on this question is correct, IMHO we should try to make
> >> IVOPTs conservative than optimistic, since once the predict is wrong from
> >> too optimistic decision, the costing on the doloop use is wrong, it's very
> >> possible to aff
>> If my understanding on this question is correct, IMHO we should try to make
>> IVOPTs conservative than optimistic, since once the predict is wrong from
>> too optimistic decision, the costing on the doloop use is wrong, it's very
>> possible to affect the global optimal set. It looks we don't
on 2019/5/21 下午6:20, Richard Biener wrote:
> On Tue, 21 May 2019, Kewen.Lin wrote:
>
>> on 2019/5/21 上午12:37, Segher Boessenkool wrote:
>>> On Mon, May 20, 2019 at 08:43:59AM -0600, Jeff Law wrote:
> I think we should have two hooks: one is called with the struct loop as
> parameter; and t
On 5/21/19 4:20 AM, Richard Biener wrote:
> On Tue, 21 May 2019, Kewen.Lin wrote:
>
>> on 2019/5/21 上午12:37, Segher Boessenkool wrote:
>>> On Mon, May 20, 2019 at 08:43:59AM -0600, Jeff Law wrote:
> I think we should have two hooks: one is called with the struct loop as
> parameter; and th
On Tue, May 21, 2019 at 12:20:50PM +0200, Richard Biener wrote:
> On Tue, 21 May 2019, Kewen.Lin wrote:
>
> > on 2019/5/21 上午12:37, Segher Boessenkool wrote:
> > > On Mon, May 20, 2019 at 08:43:59AM -0600, Jeff Law wrote:
> > >>> I think we should have two hooks: one is called with the struct loop
On Tue, May 21, 2019 at 02:03:04PM +0800, Kewen.Lin wrote:
> on 2019/5/21 上午12:37, Segher Boessenkool wrote:
> > On Mon, May 20, 2019 at 08:43:59AM -0600, Jeff Law wrote:
> >>> I think we should have two hooks: one is called with the struct loop as
> >>> parameter; and the other is called for every
On Tue, May 21, 2019 at 01:50:31PM +0800, Kewen.Lin wrote:
> on 2019/5/20 下午10:43, Jeff Law wrote:
> > On 5/20/19 4:24 AM, Segher Boessenkool wrote:
> >> Let me try to answer a bit here...
> >>
> >> On Mon, May 20, 2019 at 11:28:26AM +0200, Richard Biener wrote:
> >>> On Thu, 16 May 2019, li...@lin
On Tue, 21 May 2019, Kewen.Lin wrote:
> on 2019/5/21 上午12:37, Segher Boessenkool wrote:
> > On Mon, May 20, 2019 at 08:43:59AM -0600, Jeff Law wrote:
> >>> I think we should have two hooks: one is called with the struct loop as
> >>> parameter; and the other is called for every statement in the lo
On Mon, 20 May 2019, Segher Boessenkool wrote:
> On Mon, May 20, 2019 at 08:43:59AM -0600, Jeff Law wrote:
> > > I think we should have two hooks: one is called with the struct loop as
> > > parameter; and the other is called for every statement in the loop, if
> > > the hook isn't null anyway. O
On Tue, May 21, 2019 at 1:50 PM Kewen.Lin wrote:
>
> on 2019/5/20 下午10:43, Jeff Law wrote:
> > On 5/20/19 4:24 AM, Segher Boessenkool wrote:
> >> Let me try to answer a bit here...
> >>
> >> On Mon, May 20, 2019 at 11:28:26AM +0200, Richard Biener wrote:
> >>> On Thu, 16 May 2019, li...@linux.ibm.
on 2019/5/21 上午12:37, Segher Boessenkool wrote:
> On Mon, May 20, 2019 at 08:43:59AM -0600, Jeff Law wrote:
>>> I think we should have two hooks: one is called with the struct loop as
>>> parameter; and the other is called for every statement in the loop, if
>>> the hook isn't null anyway. Or perh
on 2019/5/20 下午10:43, Jeff Law wrote:
> On 5/20/19 4:24 AM, Segher Boessenkool wrote:
>> Let me try to answer a bit here...
>>
>> On Mon, May 20, 2019 at 11:28:26AM +0200, Richard Biener wrote:
>>> On Thu, 16 May 2019, li...@linux.ibm.com wrote:
>>
>>> So the better way would be to expose that via
on 2019/5/20 下午5:28, Richard Biener wrote:
> On Thu, 16 May 2019, li...@linux.ibm.com wrote:
>
>> From: Kewen Lin
>>
>> Hi,
>>
>> Previous version link:
>> https://gcc.gnu.org/ml/gcc-patches/2019-05/msg00654.html
>>
>> Comparing with the previous version, I moved the generic
>> parts of rs6000 ta
on 2019/5/20 下午5:28, Richard Biener wrote:
> On Thu, 16 May 2019, li...@linux.ibm.com wrote:
>
>> From: Kewen Lin
>>
>> Hi,
>>
>> Previous version link:
>> https://gcc.gnu.org/ml/gcc-patches/2019-05/msg00654.html
>>
>> Comparing with the previous version, I moved the generic
>> parts of rs6000 ta
On Mon, May 20, 2019 at 08:43:59AM -0600, Jeff Law wrote:
> > I think we should have two hooks: one is called with the struct loop as
> > parameter; and the other is called for every statement in the loop, if
> > the hook isn't null anyway. Or perhaps we do not need that second one.
> I'd wait to
On 5/20/19 4:24 AM, Segher Boessenkool wrote:
> Let me try to answer a bit here...
>
> On Mon, May 20, 2019 at 11:28:26AM +0200, Richard Biener wrote:
>> On Thu, 16 May 2019, li...@linux.ibm.com wrote:
>>> +/* Predict whether the given loop in gimple will be transformed in the RTL
>>> + doloop_o
Let me try to answer a bit here...
On Mon, May 20, 2019 at 11:28:26AM +0200, Richard Biener wrote:
> On Thu, 16 May 2019, li...@linux.ibm.com wrote:
> > +/* Predict whether the given loop in gimple will be transformed in the RTL
> > + doloop_optimize pass. This is for rs6000 target specific. *
On Thu, 16 May 2019, li...@linux.ibm.com wrote:
> From: Kewen Lin
>
> Hi,
>
> Previous version link:
> https://gcc.gnu.org/ml/gcc-patches/2019-05/msg00654.html
>
> Comparing with the previous version, I moved the generic
> parts of rs6000 target hook to IVOPTs. But I still kept
> the target h
Hi Segher,
on 2019/5/17 下午2:49, Segher Boessenkool wrote:
> Hi Kewen,
>
> On Thu, May 16, 2019 at 10:35:30PM -0500, li...@linux.ibm.com wrote:
>> 2) For the other part of target invalid stmt check, as the
>> hook invalid_within_doloop grep data shows, no all targets
>> need to check whether inval
On Fri, May 17, 2019 at 03:30:41PM +1000, Kugan Vivekanandarajah wrote:
> On Fri, 17 May 2019 at 13:37, wrote:
> > + if (GET_CODE (body) == SET)
> > + {
> > + rtx set_val = XEXP (body, 1);
> > + enum rtx_code code = GET_CODE (set_val);
> > + enum rtx_class cls =
Hi Kewen,
On Thu, May 16, 2019 at 10:35:30PM -0500, li...@linux.ibm.com wrote:
> 2) For the other part of target invalid stmt check, as the
> hook invalid_within_doloop grep data shows, no all targets
> need to check whether invalid instructions exist in doloop.
> If we scan all stmts as generic,
on 2019/5/17 下午1:30, Kugan Vivekanandarajah wrote:
> Hi,
>
> On Fri, 17 May 2019 at 13:37, wrote:
>>
>> From: Kewen Lin
>>
>> +/* Check whether number of iteration computation is too costly for doloop
>> + transformation. It expands the gimple sequence to equivalent RTL insn
>> + sequence,
Hi,
On Fri, 17 May 2019 at 13:37, wrote:
>
> From: Kewen Lin
>
> Hi,
>
> Previous version link:
> https://gcc.gnu.org/ml/gcc-patches/2019-05/msg00654.html
>
> Comparing with the previous version, I moved the generic
> parts of rs6000 target hook to IVOPTs. But I still kept
> the target hook as
23 matches
Mail list logo
