On Oct 23, 2014, at 8:20 PM, Richard Sandiford
wrote:
> Maxim Kuvyrkov writes:
>> @@ -6701,6 +6753,24 @@ alloc_global_sched_pressure_data (void)
>>saved_reg_live = BITMAP_ALLOC (NULL);
>>region_ref_regs = BITMAP_ALLOC (NULL);
>> }
>> +
>> + /* Calculate number of CALL_
Maxim Kuvyrkov writes:
> @@ -6701,6 +6753,24 @@ alloc_global_sched_pressure_data (void)
> saved_reg_live = BITMAP_ALLOC (NULL);
> region_ref_regs = BITMAP_ALLOC (NULL);
> }
> +
> + /* Calculate number of CALL_USED_REGS in register classes that
> + we calculate regis
On Oct 23, 2014, at 3:45 AM, Vladimir Makarov wrote:
...
> Richard proposed 2-loops solution instead of 3. So I'd definitely prefer his.
>
> The patch is ok with Richard's 2-loops proposal. Thanks, Maxim. As I wrote
> the heuristic is an interesting one.
This is the patch I'm going to commi
On 2014-10-22 2:17 AM, Maxim Kuvyrkov wrote:
On Oct 22, 2014, at 4:24 AM, Vladimir Makarov wrote:
On 10/20/2014 02:57 AM, Maxim Kuvyrkov wrote:
Hi,
This patch improves register pressure scheduling (both SCHED_PRESSURE_WEIGHTED
and SCHED_PRESSURE_MODEL) to better estimate number of available
Maxim Kuvyrkov writes:
> + for (i = 0; i < FIRST_PSEUDO_REGISTER; ++i)
> + if (call_used_regs[i])
> + for (c = 0; c < ira_pressure_classes_num; ++c)
> + {
> + int j;
> + enum reg_class cl = ira_pressure_classes[c];
> +
> + for (j = 0; j < ira_cl
On Oct 22, 2014, at 4:24 AM, Vladimir Makarov wrote:
> On 10/20/2014 02:57 AM, Maxim Kuvyrkov wrote:
>> Hi,
>>
>> This patch improves register pressure scheduling (both
>> SCHED_PRESSURE_WEIGHTED and SCHED_PRESSURE_MODEL) to better estimate number
>> of available registers.
>>
>> At the momen
On 10/20/2014 02:57 AM, Maxim Kuvyrkov wrote:
> Hi,
>
> This patch improves register pressure scheduling (both
> SCHED_PRESSURE_WEIGHTED and SCHED_PRESSURE_MODEL) to better estimate number
> of available registers.
>
> At the moment the scheduler does not account for spills in the prologues and
Ramana Radhakrishnan wrote:
> We already have sched-pressure --param=sched-pressure-algorithm=1 on
> by default in the AArch64 backend from September.
> https://gcc.gnu.org/ml/gcc-patches/2014-09/msg01663.html went in a few
> days back.
>
> So if this patch is on then we are looking at uplifts wi
On Oct 21, 2014, at 10:39 AM, Ramana Radhakrishnan
wrote:
> On Mon, Oct 20, 2014 at 10:17 PM, Richard Sandiford
> wrote:
>> Maxim Kuvyrkov writes:
>>> [Adding ARM maintainers to CC]
>>>
>>> On Oct 21, 2014, at 9:44 AM, Sebastian Pop wrote:
>>>
Hi Maxim,
Maxim Kuvyrkov wrote:
> [Adding ARM maintainers to CC]
>
> On Oct 21, 2014, at 9:44 AM, Sebastian Pop wrote:
>
> > Hi Maxim,
> >
> > Maxim Kuvyrkov wrote:
> >> Thanks, benchmarking results are welcome! AArch64 doesn't use
> >> reg_pressure scheduling by default. Use "-fsched-pressure
> >> --param=sched-pressure-alg
On Mon, Oct 20, 2014 at 10:17 PM, Richard Sandiford
wrote:
> Maxim Kuvyrkov writes:
>> [Adding ARM maintainers to CC]
>>
>> On Oct 21, 2014, at 9:44 AM, Sebastian Pop wrote:
>>
>>> Hi Maxim,
>>>
>>> Maxim Kuvyrkov wrote:
Thanks, benchmarking results are welcome! AArch64 doesn't use reg_pre
Maxim Kuvyrkov wrote:
> [Adding ARM maintainers to CC]
>
> On Oct 21, 2014, at 9:44 AM, Sebastian Pop wrote:
>
> > Hi Maxim,
> >
> > Maxim Kuvyrkov wrote:
> >> Thanks, benchmarking results are welcome! AArch64 doesn't use reg_pressure
> >> scheduling by default. Use "-fsched-pressure
> >> --p
Maxim Kuvyrkov writes:
> [Adding ARM maintainers to CC]
>
> On Oct 21, 2014, at 9:44 AM, Sebastian Pop wrote:
>
>> Hi Maxim,
>>
>> Maxim Kuvyrkov wrote:
>>> Thanks, benchmarking results are welcome! AArch64 doesn't use reg_pressure
>>> scheduling by default. Use "-fsched-pressure
>>> --param=s
[Adding ARM maintainers to CC]
On Oct 21, 2014, at 9:44 AM, Sebastian Pop wrote:
> Hi Maxim,
>
> Maxim Kuvyrkov wrote:
>> Thanks, benchmarking results are welcome! AArch64 doesn't use reg_pressure
>> scheduling by default. Use "-fsched-pressure
>> --param=sched-pressure-algorithm=2" to enable
Hi Maxim,
Maxim Kuvyrkov wrote:
> Thanks, benchmarking results are welcome! AArch64 doesn't use reg_pressure
> scheduling by default. Use "-fsched-pressure
> --param=sched-pressure-algorithm=2" to enable same thing as on ARM. I would
> imagine C++ and Fortran floating-point code to be most aff
On Oct 21, 2014, at 8:11 AM, Sebastian Pop wrote:
> Maxim Kuvyrkov wrote:
>> Hi,
>>
>> This patch improves register pressure scheduling (both
>> SCHED_PRESSURE_WEIGHTED and SCHED_PRESSURE_MODEL) to better estimate number
>> of available registers.
>>
>> At the moment the scheduler does not ac
Maxim Kuvyrkov wrote:
> Hi,
>
> This patch improves register pressure scheduling (both
> SCHED_PRESSURE_WEIGHTED and SCHED_PRESSURE_MODEL) to better estimate number
> of available registers.
>
> At the moment the scheduler does not account for spills in the prologues and
> restores in the epil
Hi,
This patch improves register pressure scheduling (both SCHED_PRESSURE_WEIGHTED
and SCHED_PRESSURE_MODEL) to better estimate number of available registers.
At the moment the scheduler does not account for spills in the prologues and
restores in the epilogue, which occur from use of call-used
18 matches
Mail list logo
