[SYCL][FPGA] Enable translation of ac_fixed functions #653
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Contributor
Author
|
This review is created as a continue of #614 |
Contributor
Author
|
@AlexeySotkin @AlexeySachkov @vmaksimo @MrSidims Please review. |
mlychkov
force-pushed
the
private/mlychkov/ac_fixed
branch
from
271b96c to
b6b30a8
Compare
mlychkov
force-pushed
the
private/mlychkov/ac_fixed
branch
from
b6b30a8 to
7cf617c
Compare
mlychkov
changed the title
Contributor
|
LGTM |
| ; RUN: llvm-spirv %t.bc --spirv-ext=+SPV_INTEL_arbitrary_precision_integers,+SPV_INTEL_arbitrary_precision_fixed_point -o %t.spv | ||
| ; RUN: llvm-spirv %t.spv -to-text -o - | FileCheck %s --check-prefix=CHECK-SPIRV | ||
|
|
||
| ; RUN: llvm-spirv %t.bc --spirv-ext=+SPV_INTEL_arbitrary_precision_integers -spirv-text -o - | FileCheck %s --check-prefix=CHECK-SPIRV-NEGATIVE |
Contributor
There was a problem hiding this comment.
Will we produce a valid SPIR-V in this case? If no, I think we should report an error.
Contributor
Author
There was a problem hiding this comment.
In case if we translate to SPIR-V without SPV_INTEL_arbitrary_precision_fixed_point then those built-ins are translated as user defined function calls. I suppose that it is technically valid but linkage will be failed.
I may add error generation, but the same mechanism should be added for hls_float built-ins too. Sinse that feature has already merged I would address this change in a separate PR for both features later.
Contributor
|
Please resolve the conflicts. |
mlychkov
force-pushed
the
private/mlychkov/ac_fixed
branch
2 times, most recently
from
ea9c637 to
d3ed6f5
Compare
This patch introduces SPV_INTEL_arbitrary_precision_fixed_point extension, which adds operations for arbitrary precision fixed point numbers. The ac_fixed datatype is represented as a pseudo type using OpTypeInt. Its parameters are: * W - total width of the datatype. It's encoded in the width of the OpTypeInt. * I - determines the position of the decimal point. * S - determines if this is a signed or an unsigned number. Full specification (intel/llvm#1934): https://github.com/intel/llvm/blob/2f6e965e686354fbb25f9c177a667a646de302eb/sycl/doc/extensions/SPIRV/SPV_INTEL_arbitrary_precision_fixed_point.asciidoc
mlychkov
force-pushed
the
private/mlychkov/ac_fixed
branch
from
d3ed6f5 to
a277abb
Compare
Contributor
Author
|
@AlexeySachkov Could you please review? |
AlexeySachkov
approved these changes
Aug 7, 2020
vmaksimo
added a commit
to vmaksimo/SPIRV-LLVM-Translator
that referenced
this pull request
Aug 25, 2020
… extensions This patch addresses KhronosGroup#653 (comment) If the appropriate extension for arbitrary precision instruction is not enabled, invalid IR will be produced. Report an error in this case.
AlexeySachkov
pushed a commit
that referenced
this pull request
Sep 9, 2020
… extensions (#714) This patch addresses [comment][1] If the appropriate extension for arbitrary precision instruction is not enabled, invalid IR will be produced. Report an error in this case. [1]: #653 (comment)
vladimirlaz
pushed a commit
to vladimirlaz/llvm
that referenced
this pull request
Sep 13, 2020
… extensions (intel#714) This patch addresses [comment][1] If the appropriate extension for arbitrary precision instruction is not enabled, invalid IR will be produced. Report an error in this case. [1]: KhronosGroup/SPIRV-LLVM-Translator#653 (comment)
svenvh
pushed a commit
to svenvh/SPIRV-LLVM-Translator
that referenced
this pull request
Nov 4, 2020
…up#653) This patch introduces SPV_INTEL_arbitrary_precision_fixed_point extension, which adds operations for arbitrary precision fixed point numbers. The ac_fixed datatype is represented as a pseudo type using OpTypeInt. Its parameters are: * W - total width of the datatype. It's encoded in the width of the OpTypeInt. * I - determines the position of the decimal point. * S - determines if this is a signed or an unsigned number. Full specification can be found in intel/llvm#1934 Co-authored-by: Viktoria Maksimova <[email protected]>
svenvh
pushed a commit
to svenvh/SPIRV-LLVM-Translator
that referenced
this pull request
Nov 4, 2020
… extensions (KhronosGroup#714) This patch addresses [comment][1] If the appropriate extension for arbitrary precision instruction is not enabled, invalid IR will be produced. Report an error in this case. [1]: KhronosGroup#653 (comment)
svenvh
pushed a commit
that referenced
this pull request
Nov 12, 2020
This patch introduces SPV_INTEL_arbitrary_precision_fixed_point extension, which adds operations for arbitrary precision fixed point numbers. The ac_fixed datatype is represented as a pseudo type using OpTypeInt. Its parameters are: * W - total width of the datatype. It's encoded in the width of the OpTypeInt. * I - determines the position of the decimal point. * S - determines if this is a signed or an unsigned number. Full specification can be found in intel/llvm#1934 Co-authored-by: Viktoria Maksimova <[email protected]>
svenvh
pushed a commit
that referenced
this pull request
Nov 12, 2020
… extensions (#714) This patch addresses [comment][1] If the appropriate extension for arbitrary precision instruction is not enabled, invalid IR will be produced. Report an error in this case. [1]: #653 (comment)
DmitryBushev
pushed a commit
to DmitryBushev/SPIRV-LLVM-Translator
that referenced
this pull request
Sep 1, 2021
…up#653) This patch introduces SPV_INTEL_arbitrary_precision_fixed_point extension, which adds operations for arbitrary precision fixed point numbers. The ac_fixed datatype is represented as a pseudo type using OpTypeInt. Its parameters are: * W - total width of the datatype. It's encoded in the width of the OpTypeInt. * I - determines the position of the decimal point. * S - determines if this is a signed or an unsigned number. Full specification can be found in intel/llvm#1934 Co-authored-by: Viktoria Maksimova <[email protected]>
vmaksimo
pushed a commit
to vmaksimo/SPIRV-LLVM-Translator
that referenced
this pull request
Sep 1, 2022
…recision_fixed_point (KhronosGroup#653) Change-Id: I4773436f38a116b47d7cd41411838f6b0dfd2f69 Reviewed-on: https://git-amr-2.devtools.intel.com/gerrit/265338 Tested-by: LiuChen <[email protected]> Reviewed-by: LiuChen <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This patch introduces SPV_INTEL_arbitrary_precision_fixed_point extension,
which adds operations for arbitrary precision fixed point numbers.
The ac_fixed datatype is represented as a pseudo type using OpTypeInt. Its
parameters are:
Full specification (intel/llvm#1934):
https://github.com/intel/llvm/blob/2f6e965e686354fbb25f9c177a667a646de302eb/sycl/doc/extensions/SPIRV/SPV_INTEL_arbitrary_precision_fixed_point.asciidoc