Perform runtime check for SSE 4.2 instructions#795
Merged
ohler55 merged 1 commit intoohler55:developfrom Jul 26, 2022
Merged
Conversation
stanhu
commented
Jul 25, 2022
stanhu
force-pushed
the
sh-runtime-sse42
branch
2 times, most recently
from
4560ef5 to
d566782
Compare
Owner
|
Detecting earlier is better. Running the check on each parse is expensive. I'd rather the test was done when the Makefile is generated. There is a check in the current extconf.rb. Is that not working? If not I could use some help getting that to work on some of the older machines you might have access to. Is that possible? |
If SSE 4.2 is not available on the system, don't attempt to use SIMD instructions. Relates to ohler55#789
Contributor
Author
Doesn't this change only do the detection when the module is initialized?
We package a build using SSE 4.2 platform, but some users run the code on older hardware. We either need a way to disable the SSE instructions outright, or make the check dynamic. Tensorflow behaves the same way: they ship a Python wheel that can use SSE instructions when available. |
Owner
|
I see, looks like what you have should be good then. I'll merge once tests are done and cut a release. |
ohler55
approved these changes
Jul 26, 2022
Owner
|
Released |
stanhu
added a commit
to stanhu/oj
that referenced
this pull request
Aug 2, 2022
Compiling with `-msse4.2` is problematic because it may cause a number of other SSE instructions to be used. For example: ``` $ ./elfx86exts oj.so MODE64 (call) CMOV (cmovne) SSE2 (movq) SSE41 (pinsrq) SSE1 (movups) SSSE3 (pshufb) SSE3 (movddup) SSE42 (pcmpestri) CPU Generation: Unknown ``` The runtime check in ohler55#795 is therefore not sufficient to prevent SSE instructions from leaking elsewhere in the shared library. As a result, loading Oj on older machines with these libraries will result in a "illegal instruction" crash. Introduce a `--with-sse42` flag that can be used to configure this gem.
stanhu
added a commit
to stanhu/oj
that referenced
this pull request
Aug 2, 2022
Compiling with `-msse4.2` is problematic because it may cause a number of other SSE instructions to be used. For example: ``` $ ./elfx86exts oj.so MODE64 (call) CMOV (cmovne) SSE2 (movq) SSE41 (pinsrq) SSE1 (movups) SSSE3 (pshufb) SSE3 (movddup) SSE42 (pcmpestri) CPU Generation: Unknown ``` The runtime check in ohler55#795 is therefore not sufficient to prevent SSE instructions from leaking elsewhere in the shared library. As a result, loading Oj on older machines with these libraries will result in a "illegal instruction" crash. Introduce a `--with-sse42` flag that can be used to configure this gem.
stanhu
added a commit
to stanhu/oj
that referenced
this pull request
Aug 2, 2022
Compiling with `-msse4.2` is problematic because it may cause a number of other SSE instructions to be used. For example: ``` $ ./elfx86exts oj.so MODE64 (call) CMOV (cmovne) SSE2 (movq) SSE41 (pinsrq) SSE1 (movups) SSSE3 (pshufb) SSE3 (movddup) SSE42 (pcmpestri) CPU Generation: Unknown ``` The runtime check in ohler55#795 is therefore not sufficient to prevent SSE instructions from leaking elsewhere in the shared library. As a result, loading Oj on older machines with these libraries will result in a "illegal instruction" crash. Introduce a `--with-sse42` flag that can be used to configure this gem.
ohler55
pushed a commit
that referenced
this pull request
Aug 6, 2022
Compiling with `-msse4.2` is problematic because it may cause a number of other SSE instructions to be used. For example: ``` $ ./elfx86exts oj.so MODE64 (call) CMOV (cmovne) SSE2 (movq) SSE41 (pinsrq) SSE1 (movups) SSSE3 (pshufb) SSE3 (movddup) SSE42 (pcmpestri) CPU Generation: Unknown ``` The runtime check in #795 is therefore not sufficient to prevent SSE instructions from leaking elsewhere in the shared library. As a result, loading Oj on older machines with these libraries will result in a "illegal instruction" crash. Introduce a `--with-sse42` flag that can be used to configure this gem.
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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.
If SSE 4.2 is not available on the system, don't attempt to use SIMD instructions.
Relates to #789