Skip to content

T
tic
Switch branch/tag
  1. 03 Oct, 2019 1 commit
  3. 02 Oct, 2019 3 commits
  5. 01 Oct, 2019 4 commits
  7. 30 Sep, 2019 5 commits
  9. 29 Sep, 2019 3 commits
  11. 28 Sep, 2019 4 commits
  13. 27 Sep, 2019 6 commits
  15. 26 Sep, 2019 3 commits
  17. 25 Sep, 2019 11 commits
    • [RELAY]impose a max op limit to the op fusion pass (#4002) · d21f0ad5 
      * impose a max op limit to op fusion

* use cross platform data type
      Yida Wang committed
    • [TOPI] Move conv2d spatial pack schedule to dedicated file (#3972) · f1d2d46b 
      More schedules are making the conv2d.py file too large, so
we'd like to move the spatial pack schedule to dedicated file
before introducing NHWC schedule. No logic change in this patch.
      黎明灰烬 committed
    • Revert "Added tesnorizeation for avx2 based gemm. (#3982)" (#4007) · 4a3abb94 
      This reverts commit 23727eb4.
      Tianqi Chen committed
    • remove FLOP computation for 3rd party lib call (#4005) · 5f19e5a8
      Cody Hao Yu committed
    • [ARITH] Refactor to use explicit div/mod functions instead of operators. (#4000) · f0079a57 
      * [ARITH] Use explicit div/mod functions instead of operators.

* fix pooling case
      Tianqi Chen committed
    • Expose llvm.nearbyint intrinsic. This is a faster alternate to rounding. (#4001) · 17c2c0a1 
      * Expose llvm.nearbyint intrinsic. This is a faster alternate to rounding.

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Added python binding. Added test.

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:
      Kimish Patel committed
    • Change Vivado install instructions to version 2018.3 (#4003) · 9756b067
      Philipp Krones committed
    • Added tesnorizeation for avx2 based gemm. (#3982) · 23727eb4 
      * Added tesnorizeation for avx2 based gemm.

Summary:
Tensorized the same region as avx512. Names produce 16x1 int32 results.
Does by doing two sets of AVX2 instructions to do reduction on 8x4 int8
kernel with 1x4 data.

Test Plan:
on avx2 machine:
python tests/python/contrib/test_gemm_avx2_acc32.py

Reviewers:

Subscribers:

Tasks:

Tags:

* Fix lint errors. Removed commented out code.

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:
      Kimish Patel committed
    • [COMMUNITY] @yongwww-> reviewer (#3997) · 9baff009
      Tianqi Chen committed
    • add parser support for GREATER tflite operator (#3963) · 21353e5f 
      add test for GREATER
      Ina Dobreva committed
    • Changes to make tensorize work. These changes also fix the previously broken test. (#3981) · b410df8c 
      * Changes to make tensorize work. These changes also fix the previously
broken test.

Summary:
Tensorize was breaking  for a few reasons.
1)
Assert at: src/op/tensorize.cc:234 CHECK(is_one(e.region[j]->extent))
In some cases this cannot be proven, e.g.:
expected shape=[16, 4], given region=[range(min=((ax1.outer*16)/16), ext=(((((ax1.outer*16) + 15)/16) + 1) - ax1.outer)), range(min=((k.outer*4)/4), ext=(((((k.outer*4) + 3)/4) + 1) - k.outer)), range(min=0, ext=16), range(min=0, ext=4)]
The unprovable one is: ext=(((((ax1.outer*16) + 15)/16) + 1) - ax1.outer)).
This can be simplified but it is not because to simplify divide, it must
prove ax1.outer > 0 and since it is var it cannot. The fix for this to
just find all the vars in expr in relace them with some const value.

2) Equivalence between tensorized expr and one being asked to tensorize. For example,
the error would be.
TVMError: Check failed: Equal(lhs, rhs):
Failed to match the compute with TensorIntrin tensor_intrin's declaration
provided= reduce(combiner=comm_reducer(result=[(x + y)], lhs=[x], rhs=[y], identity_element=[(int16)0]), source=[(int16(data(k))*int16(kernel(((((((((k.outer.outer*64) + (k.outer.inner*2)) + k)/2)*128) + i) - (k.outer.inner*128)) - (k.outer.outer*4096)), ((((k.outer.outer*64) + (k.outer.inner*2)) + k) % 2))))], axis=[iter_var(k, range(min=0, ext=2))], where=(bool)1, value_index=0),
intrin=  reduce(combiner=comm_reducer(result=[(x + y)], lhs=[x], rhs=[y], identity_element=[(int16)0]), source=[(int16(data(k))*int16(kernel(i, k)))], axis=[iter_var(k, range(min=0, ext=2))], where=(bool)1, value_index=0)
Difference is mainly in the source part:
source=[(int16(data(k))*int16(kernel(((((((((k.outer.outer*64) + (k.outer.inner*2)) + k)/2)*128) + i) - (k.outer.inner*128)) - (k.outer.outer*4096)), ((((k.outer.outer*64) + (k.outer.inner*2)) + k) % 2))))]
source=[(int16(data(k))*int16(kernel(i, k)))], axis=[iter_var(k, range(min=0, ext=2))]
This was not being simpifiled due to compute_intrin_iter_space (map for
iter var to range) not containing leaf iter vars.

3) Here it fails with:
Check failed: is_one(Simplify(value->shape[i])): Argument b_buffer shape mismatch[16, 4] vs [(((((ax1.outer*16) + 15)/16) + 1) - ax1.outer), (((((k.outer*4) + 3)/4) + 1) - k.outer), 16, 4]
This is in buffer binding where it thinks expected and buffer bound
shape is different. Although if we could simplify expr, this would not
be the case.

Test Plan:
On skylake avx512 machine:
python tests/python/contrib/test_gemm_acc16.py

Reviewers:

Subscribers:

Tasks:

Tags:

* Implemented bounded analyzer which traverses tree and for reduce/for
statements binds the bound of the analyzer. Later this is used to
simplify expressions. Inspired from ir_mutator_with_analyzer

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Addressed comments.

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Added ASF header + define macro for the header file: TVM_ARITHMETIC_IR_VISITOR_WITH_ANALYZER_H_
Some lint fixes as well.

* Relax the assumption that dom_map must always contain all leaf itervars.

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* Disable copy constructor and move to raw ptr.

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:
      Kimish Patel committed