search for: congruency

...l: - The machine level can have more control over the outlining process. We could have optimized parameterization, alignment handling, etc. ---- Adapting the algorithm to pre-RA IR ---- The analysis portion of the IR outliner is already IR agnostic for the most part. It works on indices into the congruency vector for instructions and their inputs/outputs. This would mean that a preRA outliner would only have to define the MIR specific portions: Congruency detection, cost analysis, parameter/output optimizations, and the outlining of beneficial candidates. -- Implementation -- https://github.com/Riv...

On Wed, Sep 27, 2017 at 6:07 PM, Matthias Braun <mbraun at apple.com> wrote: > > On Sep 27, 2017, at 3:23 PM, Davide Italiano via llvm-dev > <llvm-dev at lists.llvm.org> wrote: > > On Wed, Sep 27, 2017 at 9:28 AM, Jessica Paquette via llvm-dev > <llvm-dev at lists.llvm.org> wrote: > > > I think that, given previous discussion on the topic, we might want

> > --- Algorithmic differences with the Machine Outliner ---- > > There was a lot of confusion on how exactly the algorithm I am > proposing differs from what is available in the Machine Outliner. The > similarities of the two outliners lie in the usage of a string matching > algorithm and candidate pruning. The first step in the algorithm is to > basically do

> On Sep 27, 2017, at 3:23 PM, Davide Italiano via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > On Wed, Sep 27, 2017 at 9:28 AM, Jessica Paquette via llvm-dev > <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote: >> >> I think that, given previous discussion on the topic, we might want a split >> like this: >>

...ne level can have more control over the outlining process. We could have optimized parameterization, alignment handling, etc. > > ---- Adapting the algorithm to pre-RA IR ---- > The analysis portion of the IR outliner is already IR agnostic for the most part. It works on indices into the congruency vector for instructions and their inputs/outputs. This would mean that a preRA outliner would only have to define the MIR specific portions: Congruency detection, cost analysis, parameter/output optimizations, and the outlining of beneficial candidates. That should apply the other way around too: t...

...can have more control over the outlining process. We > could have optimized parameterization, alignment handling, etc. > > ---- Adapting the algorithm to pre-RA IR ---- > The analysis portion of the IR outliner is already IR agnostic for the > most part. It works on indices into the congruency vector for instructions > and their inputs/outputs. This would mean that a preRA outliner would only > have to define the MIR specific portions: Congruency detection, cost > analysis, parameter/output optimizations, and the outlining of beneficial > candidates. > > That should app...

...have more control over the outlining process. We could have optimized parameterization, alignment handling, etc. >> >> ---- Adapting the algorithm to pre-RA IR ---- >> The analysis portion of the IR outliner is already IR agnostic for the most part. It works on indices into the congruency vector for instructions and their inputs/outputs. This would mean that a preRA outliner would only have to define the MIR specific portions: Congruency detection, cost analysis, parameter/output optimizations, and the outlining of beneficial candidates. > That should apply the other way around t...

> On Jul 25, 2017, at 9:24 AM, Jessica Paquette <jpaquette at apple.com> wrote: > >> The two passes are pretty different in their approaches to congruency finding, so I don't think it helps to group them as though they were interchangeable "outliner technology". The two passes might be totally orthogonal. > > I think that based off how River described his algorithm, the actual underlying method should be pretty similar. If it is,...

Daniel, OK, I see your point. My understanding was that VNDAG is an internal representation, not meant to be used by VN's customers (that should really care for classes of congruency, nothing more). Also, I thought that VN's results are available for a single function at a time. It seems that NewGVN provides access to VNDAGs computed for the whole program -- which enables its usage in Jessica's pass. Still, some issues remain. I wonder what leaf nodes for reads of &quo...

...n on the placement of the > outliner in LLVM. > > > My fear with a new framework is that we are going to split the effort for > pushing the outliner technology forward and I’d like to avoid that if at > all possible. > The two passes are pretty different in their approaches to congruency finding, so I don't think it helps to group them as though they were interchangeable "outliner technology". The two passes might be totally orthogonal. I can imagine two extreme outcomes (reality is probably somewhere in between): 1. if you run the LLVM IR level code size outliner,...

1. ————— Dan, The reasoning given below for how GVN operates seems odd to me, Poison is an attribute of a value, just like nsw is an attribute of an operation, So when GVN sees a pair of equal values, one of which has an extra attribute, The proper choice for representative value is the one without the attribute, Just like when GVN sees a pair of add operations, one of which has an

Daniel, Thanks for taking the time to respond. Regarding GVN and newGVN, I recently finished a search through the llvm-dev archives for “nsw” in the subject line, and GVN was discussed in some of those threads [1]. In particular it was claimed that there was a right choice for GVN to make given two ADD instructions, one with the “nsw” attribute and one without, the one without ‘nsw’

...ontrol over the outlining process. We could have optimized parameterization, alignment handling, etc. >>> >>> ---- Adapting the algorithm to pre-RA IR ---- >>> The analysis portion of the IR outliner is already IR agnostic for the most part. It works on indices into the congruency vector for instructions and their inputs/outputs. This would mean that a preRA outliner would only have to define the MIR specific portions: Congruency detection, cost analysis, parameter/output optimizations, and the outlining of beneficial candidates. >> That should apply the other way arou...

> On Feb 27, 2017, at 9:39 AM, Daniel Berlin <dberlin at dberlin.org> wrote: > > > > On Mon, Feb 27, 2017 at 9:29 AM, Adam Nemet <anemet at apple.com <mailto:anemet at apple.com>> wrote: > >> On Feb 27, 2017, at 7:27 AM, Hal Finkel <hfinkel at anl.gov <mailto:hfinkel at anl.gov>> wrote: >> >> >> On 02/27/2017 06:29 AM,

Daniel, I wonder what the NewGVN would generate for the following C code: int a, b; int foo() { return a + b; } int bar() { return a + b; } ? Obviously, the expressions would be the same ("value1 + value2"), but a single operator is not worthy to be outlined. What classes of congruency would be assigned to operands? The same for both reads of "a" and "b"? If yes, it would be incorrect, as obviously, these two additions are not guaranteed to be semantically equivalent (a's and b's values can be changed in-between). Also note that one should be able to...

Hi Andrey, Questions and feedback are very much welcome. - The explanation as to why the improvements can vary between the IR and MIR outliner mainly boil down to the level of abstraction that each are working at. The MIR level has very accurate heuristics and is effectively the last post ISel target independent code gen pass. The IR outliner on the other hand has more estimation in the cost

Hello, I need to do a principal component analysis with EQUAMAX-rotation. Unfortunately the function principal() I use normally for PCA does not offer this rotation specification. I could find out that this might be possible somehow with the package GPArotation but until now I could not figure out how to use this in the principal component analysis. Maybe someone can give an example on how to do

...e benefits tend to outnumber the regressions). The performance section contains data for both on a variety of benchmarks. -- Why at the IR level -- The decision to place the outliner at the IR level comes from a couple of major factors: - Desire to be target independent - Most opportunities for congruency The downside to having this type of transformation be at the IR level is it means there will be less accuracy in the cost model - we can somewhat accurately model the cost per instruction but we can’t get information on how a window of instructions may lower. This can cause regressions depending...

...The performance > section contains data for both on a variety of benchmarks. > > -- Why at the IR level -- > > The decision to place the outliner at the IR level comes from a couple of > major factors: > > - Desire to be target independent > > - Most opportunities for congruency > > The downside to having this type of transformation be at the IR level is > it means there will be less accuracy in the cost model - we can somewhat > accurately model the cost per instruction but we can’t get information on > how a window of instructions may lower. This can caus...

There's another issue with re-running the vectorizer (which I support, btw - I'm just saying there are more problems to solve on the way :-) ) Historically, we haven't even tried to evaluate the cost of the "constant" (not per-iteration) vectorization overhead - things like alias checks. Instead, we have hard bounds - we won't perform alias checks that are "too