search for: tapir

...ini via llvm-dev < llvm-dev at lists.llvm.org> wrote: > > > On Jan 19, 2017, at 11:36 AM, Adve, Vikram Sadanand via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > > > > Hi Johannes, > > > >> I am especially curious where you get your data from. Tapir [0] (and to > >> some degree PIR [1]) have shown that, counterintuitively, only a few > changes > >> to LLVM passes are needed. Tapir was recently used in an MIT class with > a > >> lot of students and it seemed to work well with only minimal changes > >> t...

...t lists.llvm.org> wrote: > >> >> > On Jan 19, 2017, at 11:36 AM, Adve, Vikram Sadanand via llvm-dev < >> llvm-dev at lists.llvm.org> wrote: >> > >> > Hi Johannes, >> > >> >> I am especially curious where you get your data from. Tapir [0] (and to >> >> some degree PIR [1]) have shown that, counterintuitively, only a few >> changes >> >> to LLVM passes are needed. Tapir was recently used in an MIT class >> with a >> >> lot of students and it seemed to work well with only minimal cha...

Hi Johannes, > I am especially curious where you get your data from. Tapir [0] (and to > some degree PIR [1]) have shown that, counterintuitively, only a few changes > to LLVM passes are needed. Tapir was recently used in an MIT class with a > lot of students and it seemed to work well with only minimal changes > to analysis and especially transformation passe...

...lt;llvm-dev at lists.llvm.org >>> <mailto:llvm-dev at lists.llvm.org>> wrote: >>> > >>> > Hi Johannes, >>> > >>> >> I am especially curious where you get your data from. >>> Tapir [0] (and to >>> >> some degree PIR [1]) have shown that, counterintuitively, >>> only a few changes >>> >> to LLVM passes are needed. Tapir was recently used in an >>> MIT class with a >>> >> lot...

...t;>>> On Jan 19, 2017, at 11:36 AM, Adve, Vikram Sadanand via llvm-dev >>>>> <llvm-dev at lists.llvm.org> wrote: >>>>> >>>>> Hi Johannes, >>>>> >>>>>> I am especially curious where you get your data from. Tapir [0] (and >>>>>> to >>>>>> some degree PIR [1]) have shown that, counterintuitively, only a few >>>>>> changes >>>>>> to LLVM passes are needed. Tapir was recently used in an MIT class >>>>>> with a >>&gt...

...anand via llvm-dev >>>>>>>> <llvm-dev at lists.llvm.org> wrote: >>>>>>>> >>>>>>>> Hi Johannes, >>>>>>>> >>>>>>>>> I am especially curious where you get your data from. Tapir [0] >>>>>>>>> (and to some degree PIR [1]) have shown that, >>>>>>>>> counterintuitively, only a few changes to LLVM passes are >>>>>>>>> needed. Tapir was recently used in an MIT class with a lot of >>>&g...

...AM, Adve, Vikram Sadanand via llvm-dev >>>>>>> <llvm-dev at lists.llvm.org> wrote: >>>>>>> >>>>>>> Hi Johannes, >>>>>>> >>>>>>>> I am especially curious where you get your data from. Tapir [0] >>>>>>>> (and to some degree PIR [1]) have shown that, >>>>>>>> counterintuitively, only a few changes to LLVM passes are >>>>>>>> needed. Tapir was recently used in an MIT class with a lot of >>>>>>&g...

...anand via llvm-dev >>>>>>>> <llvm-dev at lists.llvm.org> wrote: >>>>>>>> >>>>>>>> Hi Johannes, >>>>>>>> >>>>>>>>> I am especially curious where you get your data from. Tapir [0] >>>>>>>>> (and to some degree PIR [1]) have shown that, >>>>>>>>> counterintuitively, only a few changes to LLVM passes are >>>>>>>>> needed. Tapir was recently used in an MIT class with a lot of >>>&g...

...ore effectively, we see no reason the parallel IR would conflict with any such extensions. (On the contrary, the parallel IR would seem to help compiler analyses of higher-level parallel constructs by exposing logical parallelism.) (4) Feasibility and Impact: The Tapir and PIR prototypes demonstrate the feasibility of this approach. The Tapir prototype [13] has recently proven its robustness as the standard compiler in the MIT class on parallel programming. It was implemented in ~ 5k LOC. However, >1k are explicit parallel optimization, 1k is used to add new i...

...>>>>>>>>> <llvm-dev at lists.llvm.org> wrote: >>>>>>>>> >>>>>>>>> Hi Johannes, >>>>>>>>> >>>>>>>>>> I am especially curious where you get your data from. Tapir [0] >>>>>>>>>> (and to some degree PIR [1]) have shown that, >>>>>>>>>> counterintuitively, only a few changes to LLVM passes are >>>>>>>>>> needed. Tapir was recently used in an MIT class with a lot of &g...

...ct with any such extensions. (On the > > contrary, the parallel IR would seem to help compiler analyses > > of higher-level parallel constructs by exposing logical > > parallelism.) > > > > > > (4) Feasibility and Impact: > > The Tapir and PIR prototypes demonstrate the feasibility of this > > approach. The Tapir prototype [13] has recently proven its robustness as > > the standard compiler in the MIT class on parallel programming. It was > > implemented in ~ 5k LOC. However, >1k are explicit parallel > &g...

...ue is reaching critical mass. A few of us — Hal, Jeff (Vetter), Johannes, TB Schardl, Xinmin, I, and a few others — have been discussing design ideas, building on lessons from some of these ongoing projects: * Intel’s new OpenMP compiler * ARES from Los Alamos and Oak Ridge * PIR from Saarland. * TAPIR from MIT * HPVM from Illinois Our goal is to put together a design for how to add representations for parallel programs that can leverage LLVM capabilities and features without requiring excessive changes to the existing infrastructure. We’ll submit design docs, patches for review, and one or mor...

...>>>>>>>>> <llvm-dev at lists.llvm.org> wrote: >>>>>>>>> >>>>>>>>> Hi Johannes, >>>>>>>>> >>>>>>>>>> I am especially curious where you get your data from. Tapir [0] >>>>>>>>>> (and to some degree PIR [1]) have shown that, >>>>>>>>>> counterintuitively, only a few changes to LLVM passes are >>>>>>>>>> needed. Tapir was recently used in an MIT class with a lot of &g...

...rary") ip <- ip[ip[,1]!="base" ,1] install.packages(ip) This time around that's produced some warning grumbles: Warning messages: 1: packages 'dynamicGraph', 'ggplot', 'grid', 'Qtoolbox', 'ref', 'splines', 'stats4', 'tapiR', 'tcltk', 'tools' are not available 2: dependencies 'Design', 'graph', 'RBGL' are not available Sure enough, those in warning 1 were all in the 2.4.1 library directory tree and those in warning 2 weren't. Now I'm trying to work out how these fa...

...>>>>> <llvm-dev at lists.llvm.org> wrote: >>>>>>>>>> >>>>>>>>>> Hi Johannes, >>>>>>>>>> >>>>>>>>>>> I am especially curious where you get your data from. Tapir [0] >>>>>>>>>>> (and to some degree PIR [1]) have shown that, >>>>>>>>>>> counterintuitively, only a few changes to LLVM passes are >>>>>>>>>>> needed. Tapir was recently used in an MIT class with...

...Sadanand via llvm-dev >>>>>>>> <llvm-dev at lists.llvm.org> wrote: >>>>>>>> >>>>>>>> Hi Johannes, >>>>>>>> >>>>>>>>> I am especially curious where you get your data from. Tapir [0] >>>>>>>>> (and to some degree PIR [1]) have shown that, >>>>>>>>> counterintuitively, only a few changes to LLVM passes are >>>>>>>>> needed. Tapir was recently used in an MIT class with a lot of >>>>&g...

...ensions. (On the >>> contrary, the parallel IR would seem to help compiler analyses >>> of higher-level parallel constructs by exposing logical >>> parallelism.) >>> >>> >>> (4) Feasibility and Impact: >>> The Tapir and PIR prototypes demonstrate the feasibility of this >>> approach. The Tapir prototype [13] has recently proven its >>> robustness as the standard compiler in the MIT class on parallel >>> programming. It was implemented in ~ 5k LOC. However, >1k are >>>...

...contrary, the parallel IR would seem to help compiler analyses > >>> of higher-level parallel constructs by exposing logical > >>> parallelism.) > >>> > >>> > >>> (4) Feasibility and Impact: > >>> The Tapir and PIR prototypes demonstrate the feasibility of this > >>> approach. The Tapir prototype [13] has recently proven its > >>> robustness as the standard compiler in the MIT class on parallel > >>> programming. It was implemented in ~ 5k LOC. However, >1k ar...

...m to help compiler analyses >>>>>> of higher-level parallel constructs by exposing logical >>>>>> parallelism.) >>>>>> >>>>>> >>>>>> (4) Feasibility and Impact: >>>>>> The Tapir and PIR prototypes demonstrate the feasibility of this >>>>>> approach. The Tapir prototype [13] has recently proven its >>>>>> robustness as the standard compiler in the MIT class on parallel >>>>>> programming. It was implemented in ~ 5k LOC....

...> <llvm-dev at lists.llvm.org> wrote: >>>>>>>>>>> >>>>>>>>>>> Hi Johannes, >>>>>>>>>>> >>>>>>>>>>>> I am especially curious where you get your data from. Tapir [0] >>>>>>>>>>>> (and to some degree PIR [1]) have shown that, >>>>>>>>>>>> counterintuitively, only a few changes to LLVM passes are >>>>>>>>>>>> needed. Tapir was recently used in an MIT...