Comments (10)
lindstro
commented on June 18, 2024
@cgohlke Thanks for reporting this issue. Although I don't expect any difference, do earlier versions of zfp pass with 14.38.33130?
It's possible that the data model is not detected properly, even though both compilers report LLP64. Can you please build with
-DZFP_INT64="long long" -DZFP_INT64_SUFFIX=ll -DZFP_UINT64="unsigned long long" -DZFP_UINT64_SUFFIX=ull
and see if that fixes the issue?
from zfp.
cgohlke
commented on June 18, 2024
do earlier versions of zfp pass with 14.38.33130
No, but I only tried with 1.0.0 and 1.0.1 git tags so far.
see if that fixes the issue?
No, unfortunately not.
from zfp.
cgohlke
commented on June 18, 2024
zfp-0.5.5 passes with 14.38.33130
zfp version 0.5.5 (May 5, 2019)
library version 85
CODEC version 5
data model LLP64
testing 1D array of floats
compress: rate= 2 OK
decompress: rate= 2 1.626e+01 <= 1.627e+01 OK
compress: rate= 8 OK
decompress: rate= 8 8.276e-02 <= 8.277e-02 OK
compress: rate=32 OK
decompress: rate=32 0.000e+00 <= 0.000e+00 OK
compress: precision= 4 ratio= 7.474 OK
decompress: precision= 4 OK
compress: precision= 8 ratio= 4.995 OK
decompress: precision= 8 OK
compress: precision=16 ratio= 2.589 OK
decompress: precision=16 OK
compress: tolerance=9.766e-04 ratio= 2.589 OK
decompress: tolerance=9.766e-04 3.662e-04 <= 9.766e-04 OK
compress: tolerance=2.384e-07 ratio= 1.372 OK
decompress: tolerance=2.384e-07 0.000e+00 <= 2.384e-07 OK
compress: tolerance=0.000e+00 ratio= 1.194 OK
decompress: tolerance=0.000e+00 0.000e+00 <= 0.000e+00 OK
compress: reversible ratio= 2.253 OK
decompress: reversible OK
construct: 4.578e-05 <= 4.578e-05 OK
update: 2.155e-02 ~ 2.155e-02 OK
testing 1D array of doubles
compress: rate= 3 OK
decompress: rate= 3 1.626e+01 <= 1.627e+01 OK
compress: rate= 4 OK
decompress: rate= 4 1.600e+01 <= 1.601e+01 OK
compress: rate=16 OK
decompress: rate=16 1.831e-04 <= 1.832e-04 OK
compress: rate=64 OK
decompress: rate=64 0.000e+00 <= 0.000e+00 OK
compress: precision= 8 ratio= 8.943 OK
decompress: precision= 8 OK
compress: precision=16 ratio= 4.882 OK
decompress: precision=16 OK
compress: precision=32 ratio= 2.323 OK
decompress: precision=32 OK
compress: tolerance=9.766e-04 ratio= 4.882 OK
decompress: tolerance=9.766e-04 3.662e-04 <= 9.766e-04 OK
compress: tolerance=4.441e-16 ratio= 1.266 OK
decompress: tolerance=4.441e-16 0.000e+00 <= 4.441e-16 OK
compress: tolerance=0.000e+00 ratio= 1.127 OK
decompress: tolerance=0.000e+00 0.000e+00 <= 0.000e+00 OK
compress: reversible ratio= 4.210 OK
decompress: reversible OK
construct: 1.831e-04 <= 1.832e-04 OK
update: 2.155e-02 ~ 2.155e-02 OK
testing 2D array of floats
compress: rate= 2 OK
decompress: rate= 2 1.500e+00 <= 1.500e+00 OK
compress: rate= 8 OK
decompress: rate= 8 3.662e-03 <= 3.663e-03 OK
compress: rate=32 OK
decompress: rate=32 0.000e+00 <= 0.000e+00 OK
compress: precision= 4 ratio= 27.676 OK
decompress: precision= 4 OK
compress: precision= 8 ratio= 12.337 OK
decompress: precision= 8 OK
compress: precision=16 ratio= 3.737 OK
decompress: precision=16 OK
compress: tolerance=9.766e-04 ratio= 3.319 OK
decompress: tolerance=9.766e-04 3.395e-04 <= 9.766e-04 OK
compress: tolerance=2.384e-07 ratio= 1.481 OK
decompress: tolerance=2.384e-07 0.000e+00 <= 2.384e-07 OK
compress: tolerance=0.000e+00 ratio= 1.309 OK
decompress: tolerance=0.000e+00 0.000e+00 <= 0.000e+00 OK
compress: reversible ratio= 3.210 OK
decompress: reversible OK
construct: 7.629e-06 <= 7.630e-06 OK
update: 3.755e-01 ~ 3.755e-01 OK
testing 2D array of doubles
compress: rate= 1 OK
decompress: rate= 1 2.376e+01 <= 2.376e+01 OK
compress: rate= 4 OK
decompress: rate= 4 1.797e-01 <= 1.797e-01 OK
compress: rate=16 OK
decompress: rate=16 8.583e-06 <= 8.584e-06 OK
compress: rate=64 OK
decompress: rate=64 0.000e+00 <= 0.000e+00 OK
compress: precision= 8 ratio= 23.011 OK
decompress: precision= 8 OK
compress: precision=16 ratio= 7.314 OK
decompress: precision=16 OK
compress: precision=32 ratio= 2.597 OK
decompress: precision=32 OK
compress: tolerance=9.766e-04 ratio= 6.512 OK
decompress: tolerance=9.766e-04 3.395e-04 <= 9.766e-04 OK
compress: tolerance=4.441e-16 ratio= 1.260 OK
decompress: tolerance=4.441e-16 0.000e+00 <= 4.441e-16 OK
compress: tolerance=0.000e+00 ratio= 1.131 OK
decompress: tolerance=0.000e+00 0.000e+00 <= 0.000e+00 OK
compress: reversible ratio= 6.263 OK
decompress: reversible OK
construct: 8.583e-06 <= 8.584e-06 OK
update: 3.756e-01 ~ 3.755e-01 OK
testing 3D array of floats
compress: rate= 2 OK
decompress: rate= 2 1.500e+00 <= 1.500e+00 OK
compress: rate= 8 OK
decompress: rate= 8 9.583e-03 <= 9.583e-03 OK
compress: rate=32 OK
decompress: rate=32 0.000e+00 <= 0.000e+00 OK
compress: precision= 4 ratio=107.789 OK
decompress: precision= 4 OK
compress: precision= 8 ratio= 15.754 OK
decompress: precision= 8 OK
compress: precision=16 ratio= 3.562 OK
decompress: precision=16 OK
compress: tolerance=9.766e-04 ratio= 2.684 OK
decompress: tolerance=9.766e-04 2.203e-04 <= 9.766e-04 OK
compress: tolerance=2.384e-07 ratio= 1.394 OK
decompress: tolerance=2.384e-07 0.000e+00 <= 2.384e-07 OK
compress: tolerance=0.000e+00 ratio= 1.282 OK
decompress: tolerance=0.000e+00 0.000e+00 <= 0.000e+00 OK
compress: reversible ratio= 2.688 OK
decompress: reversible OK
construct: 3.147e-05 <= 3.148e-05 OK
update: 1.846e+00 ~ 1.846e+00 OK
testing 3D array of doubles
compress: rate= 1 OK
decompress: rate= 1 5.210e+00 <= 5.210e+00 OK
compress: rate= 4 OK
decompress: rate= 4 2.002e-01 <= 2.002e-01 OK
compress: rate=16 OK
decompress: rate=16 3.338e-05 <= 3.338e-05 OK
compress: rate=64 OK
decompress: rate=64 0.000e+00 <= 0.000e+00 OK
compress: precision= 8 ratio= 30.797 OK
decompress: precision= 8 OK
compress: precision=16 ratio= 7.087 OK
decompress: precision=16 OK
compress: precision=32 ratio= 2.558 OK
decompress: precision=32 OK
compress: tolerance=9.766e-04 ratio= 5.347 OK
decompress: tolerance=9.766e-04 2.201e-04 <= 9.766e-04 OK
compress: tolerance=4.441e-16 ratio= 1.208 OK
decompress: tolerance=4.441e-16 0.000e+00 <= 4.441e-16 OK
compress: tolerance=0.000e+00 ratio= 1.122 OK
decompress: tolerance=0.000e+00 0.000e+00 <= 0.000e+00 OK
compress: reversible ratio= 5.347 OK
decompress: reversible OK
construct: 3.338e-05 <= 3.338e-05 OK
update: 1.846e+00 ~ 1.846e+00 OK
testing 4D array of floats
compress: rate= 2 OK
decompress: rate= 2 1.372e+01 <= 1.373e+01 OK
compress: rate= 8 OK
decompress: rate= 8 6.632e-01 <= 6.633e-01 OK
compress: rate=32 OK
decompress: rate=32 0.000e+00 <= 0.000e+00 OK
compress: precision= 4 ratio=256.000 OK
decompress: precision= 4 OK
compress: precision= 8 ratio= 9.309 OK
decompress: precision= 8 OK
compress: precision=16 ratio= 2.798 OK
decompress: precision=16 OK
compress: tolerance=9.766e-04 ratio= 1.736 OK
decompress: tolerance=9.766e-04 1.359e-04 <= 9.766e-04 OK
compress: tolerance=2.384e-07 ratio= 1.166 OK
decompress: tolerance=2.384e-07 0.000e+00 <= 2.384e-07 OK
compress: tolerance=0.000e+00 ratio= 1.166 OK
decompress: tolerance=0.000e+00 0.000e+00 <= 0.000e+00 OK
compress: reversible ratio= 2.387 OK
decompress: reversible OK
testing 4D array of doubles
compress: rate= 1 OK
decompress: rate= 1 1.016e+01 <= 1.016e+01 OK
compress: rate= 4 OK
decompress: rate= 4 8.984e+00 <= 8.985e+00 OK
compress: rate=16 OK
decompress: rate=16 3.312e-03 <= 3.312e-03 OK
compress: rate=64 OK
decompress: rate=64 0.000e+00 <= 0.000e+00 OK
compress: precision= 8 ratio= 18.534 OK
decompress: precision= 8 OK
compress: precision=16 ratio= 5.588 OK
decompress: precision=16 OK
compress: precision=32 ratio= 2.331 OK
decompress: precision=32 OK
compress: tolerance=9.766e-04 ratio= 3.468 OK
decompress: tolerance=9.766e-04 1.360e-04 <= 9.766e-04 OK
compress: tolerance=4.441e-16 ratio= 1.076 OK
decompress: tolerance=4.441e-16 0.000e+00 <= 4.441e-16 OK
compress: tolerance=0.000e+00 ratio= 1.076 OK
decompress: tolerance=0.000e+00 0.000e+00 <= 0.000e+00 OK
compress: reversible ratio= 4.768 OK
decompress: reversible OK
all tests passed
from zfp.
cgohlke
commented on June 18, 2024
As a workaround, compile with the /Ob1 instead of the default /Ob2 compiler switch, for example using -DCMAKE_C_FLAGS_RELEASE:STRING=/MD /O2 /Ob1 /DNDEBUG for release builds.
/Ob1 "allows expansion only of functions marked inline, __inline, or __forceinline, or in a C++ member function defined in a class declaration", while /Ob2 "allows the compiler to expand any function not explicitly marked for no inlining" (https://learn.microsoft.com/en-us/cpp/build/reference/ob-inline-function-expansion?view=msvc-170).
from zfp.
lindstro
commented on June 18, 2024
Thanks for the suggestion. I doubt that we have access to the same compiler/OS/hardware setup, so debugging this from our end will be challenging. AppVeyor supports Visual Studio Community 2022 version 17.5.0. We can take a look and see if the issue can be reproduced with that compiler, though it seems unlikely as the issue seems confined to one particular compiler version and system.
By the way, zfp 1.0.x is not ABI compatible with 0.5.y. I assume you've rebuilt any binaries that depend on zfp.
from zfp.
cgohlke
commented on June 18, 2024
Yes, it could be a compiler bug introduced by the latest VS update. This issue is meant as a "heads-up" in case someone else runs into it. The workaround, to build with /Ob1, is good enough for now. I think the issue can be closed (?)
By the way, zfp 1.0.x is not ABI compatible with 0.5.y. I assume you've rebuilt any binaries that depend on zfp.
Sure, I was just rebuilding 0.5.5 temporarily for testing this issue.
from zfp.
Related Issues (20)
- 16-bit integer support HOT 3
- CUDA backend ignores ZFP_WITH_TIGHT_ERROR
- Is safe to read values in parallel from a const_array with cache_size=0 ? HOT 3
- race condition when using private_const_view from inside oneapi::tbb:parallel_for lambda HOT 13
- The predicted maximum compressed capacity is much larger than the actual HOT 2
- oldest-supported-numpy Erroneously Required in Apple Silicon Python 3.10 Wheel HOT 30
- Python binding installation fails HOT 18
- ppc32: index.hpp: error: integer constant is too large for ‘unsigned long’ type HOT 3
- How to run zfp as a c code with gcc HOT 6
- Is psnr formula error in ZFP? HOT 2
- Missing `void` in functions that take no parameters
- Tiny but above subnormal numbers not handled correctly
- Numpy 2.0 HOT 6
- Cuda mode ZFP examples HOT 4
- Compilation issues with zfp HOT 2
- Running issues with zfp and cuda HOT 3
- Errors reported with scikit-build when running CMake with BUILD_ZFPY enabled HOT 1
- 2D Point data curiosly compresses way better at 1D HOT 4
- Deserializing const_array HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from zfp.