On 20/01/15 13:26, Maxim Kuvyrkov wrote:
On Jan 20, 2015, at 1:24 PM, Richard Earnshaw wrote:
...
In general, how should someone tuning the compiler for this parameter
select a value that isn't one of (-1, m_i_q_d+1)?
From my experiments it seems there are 4 reasonable values for the paramet
On 20/01/15 13:26, Maxim Kuvyrkov wrote:
> On Jan 20, 2015, at 1:24 PM, Richard Earnshaw wrote:
> ...
In general, how should someone tuning the compiler for this parameter
select a value that isn't one of (-1, m_i_q_d+1)?
>>>
>>> From my experiments it seems there are 4 reasonable values
On Jan 20, 2015, at 1:24 PM, Richard Earnshaw wrote:
...
>>> In general, how should someone tuning the compiler for this parameter
>>> select a value that isn't one of (-1, m_i_q_d+1)?
>>
>> From my experiments it seems there are 4 reasonable values for the
>> parameter: (-1) autopref turned off
On 19/01/15 18:14, Maxim Kuvyrkov wrote:
> On Jan 19, 2015, at 6:05 PM, Richard Earnshaw wrote:
>
>> On 16/01/15 15:06, Maxim Kuvyrkov wrote:
>>> @@ -1874,7 +1889,8 @@ const struct tune_params arm_cortex_a15_tune =
>>> true, true, /* Prefer 32-bit encodings.
>
On 19/01/15 18:14, Maxim Kuvyrkov wrote:
On Jan 19, 2015, at 6:05 PM, Richard Earnshaw wrote:
On 16/01/15 15:06, Maxim Kuvyrkov wrote:
@@ -1874,7 +1889,8 @@ const struct tune_params arm_cortex_a15_tune =
true, true, /* Prefer 32-bit encodings. */
tru
On Jan 19, 2015, at 6:05 PM, Richard Earnshaw wrote:
> On 16/01/15 15:06, Maxim Kuvyrkov wrote:
>> @@ -1874,7 +1889,8 @@ const struct tune_params arm_cortex_a15_tune =
>> true, true, /* Prefer 32-bit encodings.
>> */
>> true,
On 16/01/15 15:06, Maxim Kuvyrkov wrote:
> @@ -1874,7 +1889,8 @@ const struct tune_params arm_cortex_a15_tune =
>true, true, /* Prefer 32-bit encodings.
> */
>true, /* Prefer Neon for
> stringops. */
>8,
On Fri, Jan 16, 2015 at 3:06 PM, Maxim Kuvyrkov
wrote:
> On Nov 19, 2014, at 12:27 PM, Ramana Radhakrishnan
> wrote:
>
>>
>
> Hi Ramana,
> Hi Vladimir,
>
> I still don't have SPEC2000/SPEC2006 benchmark numbers for this patch. Since
> stage3 is about to finish, I'm going to commit the target i
On Nov 19, 2014, at 12:27 PM, Ramana Radhakrishnan
wrote:
>
Hi Ramana,
Hi Vladimir,
I still don't have SPEC2000/SPEC2006 benchmark numbers for this patch. Since
stage3 is about to finish, I'm going to commit the target independent part of
the patch now (it was approved by Vladimir a while
On Nov 19, 2014, at 12:27 PM, Ramana Radhakrishnan
wrote:
>
>
> On 14/11/14 15:12, Maxim Kuvyrkov wrote:
>> On Nov 14, 2014, at 8:38 AM, Jeff Law wrote:
>>
>>> On 10/20/14 22:06, Maxim Kuvyrkov wrote:
Hi,
Ramana, this change requires benchmarking, which I can't easily do
at
>>
On 14/11/14 15:12, Maxim Kuvyrkov wrote:
On Nov 14, 2014, at 8:38 AM, Jeff Law wrote:
On 10/20/14 22:06, Maxim Kuvyrkov wrote:
Hi,
Ramana, this change requires benchmarking, which I can't easily do
at
the moment. I would appreciate any benchmarking results that you can
share. In particular
On 11/14/14 08:12, Maxim Kuvyrkov wrote:
...
Can this be built on top of Bin's work for insn fusion? There's a
lot of commonality in the structure of the insns you care about.
He's already got a nice little priority function that I think you
could utilize to to ensure the insns with smaller off
On Nov 14, 2014, at 8:38 AM, Jeff Law wrote:
> On 10/20/14 22:06, Maxim Kuvyrkov wrote:
>> Hi,
>> Ramana, this change requires benchmarking, which I can't easily do
>> at
> the moment. I would appreciate any benchmarking results that you can
> share. In particular, the value of PARAM_SCHED_AUTOPR
On Nov 14, 2014, at 4:57 AM, Vladimir Makarov wrote:
> On 2014-10-21 12:06 AM, Maxim Kuvyrkov wrote:
>
...
> I'd prefer symbolic constants for dont_delay. Also the address can contains
> other parts, e.g. index for some targets. It is not necessary to change the
> code but a comment would be
On 10/20/14 22:06, Maxim Kuvyrkov wrote:
Hi,
Ramana, this change requires benchmarking, which I can't easily do
at
the moment. I would appreciate any benchmarking results that you can
share. In particular, the value of PARAM_SCHED_AUTOPREF_QUEUE_DEPTH
needs to be tuned/confirmed for Cortex-A15.
On 2014-10-21 12:06 AM, Maxim Kuvyrkov wrote:
Hi,
This patch adds auto-prefetcher modeling to GCC scheduler. The auto-prefetcher
model is currently enabled only for ARM Cortex-A15, since this is the only CPU
that I know of to have the hardware auto-prefetcher unit.
The documentation on the a
On Oct 21, 2014, at 8:06 AM, Maxim Kuvyrkov wrote:
> Hi,
>
> This patch adds auto-prefetcher modeling to GCC scheduler. The
> auto-prefetcher model is currently enabled only for ARM Cortex-A15, since
> this is the only CPU that I know of to have the hardware auto-prefetcher unit.
>
> The doc
On Mon, Oct 20, 2014 at 9:06 PM, Maxim Kuvyrkov
wrote:
> Hi,
>
> This patch adds auto-prefetcher modeling to GCC scheduler. The
> auto-prefetcher model is currently enabled only for ARM Cortex-A15, since
> this is the only CPU that I know of to have the hardware auto-prefetcher unit.
That migh
18 matches
Mail list logo
