Displaying 15 results from an estimated 15 matches for "pr27299".
2016 May 24
5
The state of IRPGO (3 remaining work items)
...g (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/D15828 did 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're not
opposed to having more passes, which might be beneficial for non-game
workloads (w...
2016 May 25
3
The state of IRPGO (3 remaining work items)
...reshold 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/D15828 did 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're not
>> opposed to having more passes, which m...
2016 May 25
0
The state of IRPGO (3 remaining work items)
...e 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/D15828 did 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're not
> opposed to having more passes, which might be beneficial f...
2016 May 25
0
The state of IRPGO (3 remaining work items)
...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/D15828 did 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're
>>> not opposed to having...
2016 May 24
0
The state of IRPGO (3 remaining work items)
...e 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/D15828 did 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're not
> opposed to having more passes, which might be beneficial f...
2016 May 25
0
The state of IRPGO (3 remaining work items)
...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're
> not opposed to having more passes, which might be be...
2016 Jun 03
2
The state of IRPGO (3 remaining work items)
...l 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: for our use cases, inlining does all the work, but we're not opposed to having more passes, which mi...
2016 May 24
6
The state of IRPGO (3 remaining work items)
...s 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're not opposed to having more passes, which might be beneficial for non-game wo...
2016 May 25
2
The state of IRPGO (3 remaining work items)
...00
>> 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're
>> not opposed to having more passe...
2016 Jun 01
4
The state of IRPGO (3 remaining work items)
...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: for our use cases, inlining does all the work, but we're not opposed to having more passes, which might be beneficial for no...
2016 Jun 02
2
The state of IRPGO (3 remaining work items)
...eared 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: for our use cases, inlining does all the work, but we're not opposed to having more passes, which might be beneficia...
2016 Jun 02
4
The state of IRPGO (3 remaining work items)
...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: for our use cases, inlining does all the work, but we're not opposed to having more passes, which might be b...
2016 Jun 03
5
The state of IRPGO (3 remaining work items)
...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
>...
2016 Jun 13
2
The state of IRPGO (3 remaining work items)
...>>>> 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, inl...
2016 Jun 23
0
The state of IRPGO (3 remaining work items)
...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 t...