More work on syntax description

  Update grammar in report to match implementation

new version

new version coming next

  Merging in documentation changes from trunk

  Added trace, evals, and evecs to report

  Updating documentation

   Fixed typos

  Creating a branch to explore a pure CFG IL

  Working on language description

  added missing hats

  Added example

  New note

  Upated URL in header to diderot-language.cs.uchicago.edu

  Updated grammar to tack changes in implementation

  Tweak defs

  Added conditional expression syntax

  Added missing package

  Update example

  Fix continuity of ctmr

  Added discussion of partial-derivative operators

  Update compiler picture

  Working on report

  editing

  Working on report

  Added function types

  Starting to document the Diderot Basis

  Edits

  Edits

  Working on compiler

  remove generated files from repository

  minor edits

some tweaks

oops didnt mean to include GLK-specific paths

oops meant to do rename from run.sh

new test script

new test image

for gradient code

new test image

not as general purpose as hoped

  Update code to track code directory

  Update README

  Working on debugging

  Adding Gordon's test harness.

fixing up 1D and 3D convolution in C

  Added kernel definition

  Added C code for (D F) probe

  fix bug

  Fixed bounds

  Fixed OpenCL code

  Minor fix

  Added OpenCL 3d probe code

  More revisions

  Revising probe document

  A Didirot whitepaper based on the December 2009 NSF proposal

Added output state variable syntax

  Added discussion of image and field types

  Fix typos

  revising probe document

  Fix typo

  fixed bugs in probe doc.

  Adding notes on code for probing a field.

  Add compiler diagram

  Working on report

  Minor grammar tweak and started intro

  Grammar fixes

  Created Didirot report document

  Working on Diderot typing note

  editing

  Starting to write up some notes on the typechecking problem for Diderot.

  Minor edits

  Change tensor type syntax

added different multiplications and some possibilities for their syntax

* addition of a "rawten" type, also parameterized by \rho, which represents the type of tensors as they are stored in an image on disk or in memory.  As it was, images were only of type \rho, with no indication of the tensor structure.  The only annoyance now is that this allows one to store on disk or in memory tensors with int coefficients, which is probably nonsense, but its harmless.

* renamed the mapping of image to field from "sample" to "convolve".  However, its not clear to me if the definition as I wrote it is sufficiently clear that the mapping is from an image of rawten<o,d,\rho> to an field of tensor<o,d>, where the "o"s and "d"s must match up.

* added negation of fields

* moved what was in "tensor arithmatic" to the "addition" section.

fixing typos

  Added tensor+/-field operations

  Fixes.

  Preliminary design document

move math into doc

  Minor edit

  Working on type and operator design-space summary

  new file

  Initial commits