Displaying 12 results from an estimated 12 matches for "d15828did".
2016 Jun 03
2
The state of IRPGO (3 remaining work items)
...nts (and we are glad to provide more info) the main benefit was inlining (also as expected). A simple pass of inlining at threshold 100 appeared to give all the benefits. Even inlining at threshold 0 gave almost all the benefits. For example, the passes initially proposed in http://reviews.llvm.org/D15828did <http://reviews.llvm.org/D15828did> not improve over just inlining with threshold 100.
>>>> >
>>>> > (due to PR27299 we also need to add simplifycfg after inlining to clean up, but this doesn't affect the instrumentation overhead in our measurements)
>>...
2016 Jun 01
4
The state of IRPGO (3 remaining work items)
...nts (and we are glad to provide more info) the main benefit was inlining (also as expected). A simple pass of inlining at threshold 100 appeared to give all the benefits. Even inlining at threshold 0 gave almost all the benefits. For example, the passes initially proposed in http://reviews.llvm.org/D15828did <http://reviews.llvm.org/D15828did> not improve over just inlining with threshold 100.
> >
> > (due to PR27299 we also need to add simplifycfg after inlining to clean up, but this doesn't affect the instrumentation overhead in our measurements)
> >
> > Bottom line:...
2016 May 25
0
The state of IRPGO (3 remaining work items)
...to provide more info) the main benefit was
> inlining (also as expected). A simple pass of inlining at threshold 100
> appeared to give all the benefits. Even inlining at threshold 0 gave almost
> all the benefits. For example, the passes initially proposed in
> http://reviews.llvm.org/D15828did not improve over just inlining with
> threshold 100.
> >
> > (due to PR27299 we also need to add simplifycfg after inlining to clean
> up, but this doesn't affect the instrumentation overhead in our
> measurements)
> >
> > Bottom line: for our use cases, inlinin...
2016 May 24
6
The state of IRPGO (3 remaining work items)
...nts (and we are glad to provide more info) the main benefit was inlining (also as expected). A simple pass of inlining at threshold 100 appeared to give all the benefits. Even inlining at threshold 0 gave almost all the benefits. For example, the passes initially proposed in http://reviews.llvm.org/D15828did not improve over just inlining with threshold 100.
>
> (due to PR27299 we also need to add simplifycfg after inlining to clean up, but this doesn't affect the instrumentation overhead in our measurements)
>
> Bottom line: for our use cases, inlining does all the work, but we'r...
2016 May 25
2
The state of IRPGO (3 remaining work items)
...info) the main benefit was
>> inlining (also as expected). A simple pass of inlining at threshold 100
>> appeared to give all the benefits. Even inlining at threshold 0 gave almost
>> all the benefits. For example, the passes initially proposed in
>> http://reviews.llvm.org/D15828did not improve over just inlining with
>> threshold 100.
>> >
>> > (due to PR27299 we also need to add simplifycfg after inlining to clean
>> up, but this doesn't affect the instrumentation overhead in our
>> measurements)
>> >
>> > Bottom line...
2016 Jun 02
2
The state of IRPGO (3 remaining work items)
...nts (and we are glad to provide more info) the main benefit was inlining (also as expected). A simple pass of inlining at threshold 100 appeared to give all the benefits. Even inlining at threshold 0 gave almost all the benefits. For example, the passes initially proposed in http://reviews.llvm.org/D15828did <http://reviews.llvm.org/D15828did> not improve over just inlining with threshold 100.
>> >
>> > (due to PR27299 we also need to add simplifycfg after inlining to clean up, but this doesn't affect the instrumentation overhead in our measurements)
>> >
>> &...
2016 Jun 02
4
The state of IRPGO (3 remaining work items)
...nts (and we are glad to provide more info) the main benefit was inlining (also as expected). A simple pass of inlining at threshold 100 appeared to give all the benefits. Even inlining at threshold 0 gave almost all the benefits. For example, the passes initially proposed in http://reviews.llvm.org/D15828did <http://reviews.llvm.org/D15828did> not improve over just inlining with threshold 100.
>>> >
>>> > (due to PR27299 we also need to add simplifycfg after inlining to clean up, but this doesn't affect the instrumentation overhead in our measurements)
>>> >...
2016 Jun 03
5
The state of IRPGO (3 remaining work items)
...inlining (also as expected). A simple pass of inlining at threshold 100
>>>>> appeared to give all the benefits. Even inlining at threshold 0 gave almost
>>>>> all the benefits. For example, the passes initially proposed in
>>>>> http://reviews.llvm.org/D15828did not improve over just inlining with
>>>>> threshold 100.
>>>>> >
>>>>> > (due to PR27299 we also need to add simplifycfg after inlining to
>>>>> clean up, but this doesn't affect the instrumentation overhead in our
>>>&g...
2016 May 24
0
The state of IRPGO (3 remaining work items)
On Mon, May 23, 2016 at 8:23 PM, Sean Silva <chisophugis at gmail.com> wrote:
> Jake and I have been integrating IRPGO on PS4, and we've identified 3
> remaining work items.
>
Sean, thanks for the write up. It matches very well with what we think as
well.
>
>
> - Driver changes
>
> We'd like to make IRPGO the default on PS4. We also think that it would be
2016 Jun 13
2
The state of IRPGO (3 remaining work items)
...ted). A simple pass of inlining at threshold 100
>>>>>>> appeared to give all the benefits. Even inlining at threshold 0 gave almost
>>>>>>> all the benefits. For example, the passes initially proposed in
>>>>>>> http://reviews.llvm.org/D15828did not improve over just inlining
>>>>>>> with threshold 100.
>>>>>>> >
>>>>>>> > (due to PR27299 we also need to add simplifycfg after inlining to
>>>>>>> clean up, but this doesn't affect the instrumentatio...
2016 Jun 23
0
The state of IRPGO (3 remaining work items)
...lso as expected). A simple pass of inlining at threshold 100
>>>>>> appeared to give all the benefits. Even inlining at threshold 0 gave almost
>>>>>> all the benefits. For example, the passes initially proposed in
>>>>>> http://reviews.llvm.org/D15828did not improve over just inlining
>>>>>> with threshold 100.
>>>>>> >
>>>>>> > (due to PR27299 we also need to add simplifycfg after inlining to
>>>>>> clean up, but this doesn't affect the instrumentation overhead in ou...
2016 May 24
5
The state of IRPGO (3 remaining work items)
Jake and I have been integrating IRPGO on PS4, and we've identified 3
remaining work items.
- Driver changes
We'd like to make IRPGO the default on PS4. We also think that it would be
beneficial to make IRPGO the default PGO on all platforms (coverage would
continue to use FE instr as it does currently, of course). In previous
conversations (e.g. http://reviews.llvm.org/D15829) it has