Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

gh-130317: fix PyFloat_Pack/Unpack[24] for NaN's with payload#130452

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to ourterms of service andprivacy statement. We’ll occasionally send you account related emails.

Already on GitHub?Sign in to your account

Merged
vstinner merged 26 commits intopython:mainfromskirpichev:fix-nan-packing/130317
Apr 28, 2025

Conversation

skirpichev
Copy link
Contributor

@skirpichevskirpichev commentedFeb 22, 2025•
edited
Loading

Previously (1) all NaN's payload in PyFloat_Pack/Unpack2() was ignored and (2) type conversions (float <-> double) in PyFloat_Pack/Unpack4() broke pack-unpack round-trip for sNaN's.

@skirpichevskirpichev marked this pull request as ready for reviewFebruary 23, 2025 03:53
vstinner

This comment was marked as outdated.

@skirpichev

This comment was marked as outdated.

@skirpichevskirpichev marked this pull request as draftFebruary 25, 2025 06:04
@skirpichevskirpichev changed the titlegh-130317: fix PyFloat_Pack2/Unpack2 for NaN's with payloadgh-130317: fix PyFloat_Pack/Unpack[24] for NaN's with payloadFeb 25, 2025
@skirpichevskirpichev marked this pull request as ready for reviewFebruary 25, 2025 09:12
@skirpichev
Copy link
ContributorAuthor

Ok, I've added new tests in test_capi.test_floats (in principle, test_struct's tests are redundant).

win32 behavior (you can't unset "quiet" bit for NaN) looks as a bug for me.

@vstinner
Copy link
Member

win32 behavior (you can't unset "quiet" bit for NaN) looks as a bug for me.

Any idea why there is a bug only on Windows?

@skirpichev
Copy link
ContributorAuthor

Any idea why there is a bug only on Windows?

Only on win32. Though, I suspect the situation could be worse. Maybe I should revert win32-workaround and test this with some buildbots?

C17 says: "This specification does not define the behavior of signaling NaNs." But I don't think this means you can't flip appropriate bit in float/double to make a signaling NaN.

@skirpichev
Copy link
ContributorAuthor

Ok, I think it's ready for review.

Test for Windows failed in 32-bit mode (signaling NaN type not preserved in roundtrip), that seems to be a known issue:https://developercommunity.visualstudio.com/t/155064 (sNaN returned from function becomes qNaN). See alsohttps://developercommunity.visualstudio.com/t/903305 andhttps://en.delphipraxis.net/topic/12198-delphi-win32-quiets-signaling-nan-on-function-return/.

In principle, it's possible to workaround this for the struct module. But not for C-API (PyFloat_Unpack*).

Copy link
Member

@vstinnervstinner left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

LGTM. I just have some minor comments.

def test_pack_unpack_roundtrip_for_nans(self):
pack = _testcapi.float_pack
unpack = _testcapi.float_unpack
for _ in range(1000):
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

100 tests should be enough to validate the implementation, no?

Suggested change
for_inrange(1000):
for_inrange(100):

1000 tests might be a little bit too slow, I don't think that it's worth it.

Copy link
ContributorAuthor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

Up to you, the test looks instantaneous on my system. 0.3sec vs 0.03. Where the threshold?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

I don't think speed is really an issue here, but having the potential for 6000 failed test reports seems like major overkill. I think 10 would actually be plenty for this loop.

Co-authored-by: Victor Stinner <vstinner@python.org>
@vstinnervstinner merged commit6157135 intopython:mainApr 28, 2025
42 checks passed
@vstinner
Copy link
Member

Merged, thank you.

@vstinner
Copy link
Member

Should we backport this change? It's unclear to me if it's a new feature or a bugfix. The issue is reported as a bug report.

@skirpichevskirpichev deleted the fix-nan-packing/130317 branchApril 28, 2025 13:29
@skirpichev
Copy link
ContributorAuthor

The issue is reported as a bug report.

Strictly speaking, it's a bug, though maybe a very minor one (that's why there was no label for 3.13 backport).

From IEEE 754 (2008), 6.2.3: "Conversion of a quiet NaN from a narrower format to a wider format in the same radix, and then back to the same narrower format, should not change the quiet NaN payload in any way except to make it canonical."

E.g. in C nan's payload is preserved in conversions (e.g. float->double) or partially preserved (e.g. double -> float).

vstinner reacted with thumbs up emoji

@vstinner
Copy link
Member

test_capi.test_float fails on x86 Debian Installed with X 3.x:https://buildbot.python.org/#/builders/1244/builds/5176

test.pythoninfo:

CC.version: gcc (Debian 12.2.0-14) 12.2.0builtins.float.double_format: IEEE, little-endianbuiltins.float.float_format: IEEE, little-endianplatform.architecture: 32bit ELFplatform.freedesktop_os_release[NAME]: Debian GNU/Linuxplatform.freedesktop_os_release[VERSION]: 12 (bookworm)platform.libc_ver: glibc 2.36platform.platform: Linux-6.1.0-33-686-pae-i686-with-glibc2.36sys.float_info: sys.float_info(max=1.7976931348623157e+308, max_exp=1024, max_10_exp=308, min=2.2250738585072014e-308, min_exp=-1021, min_10_exp=-307, dig=15, mant_dig=53, epsilon=2.220446049250313e-16, radix=2, rounds=1)sys.float_repr_style: shortsys.maxsize: 2147483647

There are failures on 16, 32 and 64 bit floats (size: 2, 4, 8). Examples:

FAIL: test_pack_unpack_roundtrip_for_nans (test.test_capi.test_float.CAPIFloatTest.test_pack_unpack_roundtrip_for_nans)(data=b'\x7f\x98Pr', size=4, endian=0)----------------------------------------------------------------------Traceback (most recent call last):  File "test/test_capi/test_float.py", line 213, in test_pack_unpack_roundtrip_for_nans    self.assertEqual(data1, data2)    ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^AssertionError: b'\x7f\x98Pr' != b'\x7f\xd8Pr'======================================================================FAIL: test_pack_unpack_roundtrip_for_nans (test.test_capi.test_float.CAPIFloatTest.test_pack_unpack_roundtrip_for_nans) (data=b'|\xd6', size=2, endian=0)----------------------------------------------------------------------Traceback (most recent call last):  File "/buildbot/buildarea/3.x.ware-debian-x86.installed/build/target/lib/python3.14/test/test_capi/test_float.py", line 213, in test_pack_unpack_roundtrip_for_nans    self.assertEqual(data1, data2)    ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^AssertionError: b'|\xd6' != b'~\xd6'======================================================================FAIL: test_pack_unpack_roundtrip_for_nans (test.test_capi.test_float.CAPIFloatTest.test_pack_unpack_roundtrip_for_nans) (data=b'\x7f\xf2h\xf8\x1bU*\xc1', size=8, endian=0)----------------------------------------------------------------------Traceback (most recent call last):  File "/buildbot/buildarea/3.x.ware-debian-x86.installed/build/target/lib/python3.14/test/test_capi/test_float.py", line 213, in test_pack_unpack_roundtrip_for_nans    self.assertEqual(data1, data2)    ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^AssertionError: b'\x7f\xf2h\xf8\x1bU*\xc1' != b'\x7f\xfah\xf8\x1bU*\xc1'

@vstinner
Copy link
Member

I can reproduce the issue on Fedora 42 by building Python usinggcc -m32:

./configure CFLAGS="-m32" LDFLAGS="-m32"sed -i -e 's!#define PY_HAVE_PERF_TRAMPOLINE 1!#undef PY_HAVE_PERF_TRAMPOLINE!g' pyconfig.hmake./python -m test -v test_capi.test_float

@skirpichev
Copy link
ContributorAuthor

Hmm, again in all cases sNaN seems transformed to qNaN.

I think underlying reason is same and we should apply workaround for 32-bit modes.

@vstinner
Copy link
Member

I think underlying reason is same and we should apply workaround for 32-bit modes.

I replacedif signaling and sys.platform == 'win32': withif signaling and True: in test_capi.test_float, but it didn't work around the issue :-( Maybe_testcapi.float_set_snan() doesn't work on x86.

@vstinner
Copy link
Member

Aha, just copying a double to another double clears the sNaN flag:

(gdb) n33    double d = ((PyFloatObject *)obj)->ob_fval;(gdb) n35    switch (size)(gdb) p ((PyFloatObject *)obj)->ob_fval$3 = nan(0x5e5454683683)(gdb) p d$4 = nan(0x85e5454683683)

d is not equal toob_fval.

@vstinner
Copy link
Member

The sNaN flag is easily lost. I added some debug of the uint64_t value in different functions. Between float_pack() and PyFloat_Pack4(), the sNaN flag is lost:

float_set_snan(): v=fffbce9e80000000 (before)float_set_snan(): v=fff3ce9e80000000 (sNaN)float_pack():     v=fff3ce9e80000000PyFloat_Pack4():  v=fffbce9e80000000

PyFloat_Pack4() is called with the wrong value :-(

@skirpichev
Copy link
ContributorAuthor

Maybe _testcapi.float_set_snan() doesn't work on x86.

Hmm, indeed. I should finally do memcpy() to created PyFloat's ob_fval,then return it.

Does it work:

diff --git a/Modules/_testcapi/float.c b/Modules/_testcapi/float.cindex 2feeb205d8..b382883fa8 100644--- a/Modules/_testcapi/float.c+++ b/Modules/_testcapi/float.c@@ -181,9 +181,12 @@ _testcapi_float_set_snan(PyObject *module, PyObject *obj)     }     uint64_t v;     memcpy(&v, &d, 8);-    v &= ~(1ULL << 51); /* make sNaN */-    memcpy(&d, &v, 8);-    return PyFloat_FromDouble(d);+    if (v & (1ULL << 51)) {+        v -= (1ULL << 51); /* make sNaN */+    }+    PyObject *ret = PyFloat_FromDouble(0.0);+    memcpy(&(((PyFloatObject *)ret)->ob_fval), &v, 8);+    return ret; }  static PyMethodDef test_methods[] = {

?

vstinner reacted with thumbs up emoji

@vstinner
Copy link
Member

I wrote#133150 to fix 3 bugs, but it's not enough to fix the test on x86.

@vstinner
Copy link
Member

Tests were fixed by#133204.

zware reacted with thumbs up emoji

Sign up for freeto join this conversation on GitHub. Already have an account?Sign in to comment
Reviewers

@zwarezwarezware left review comments

@vstinnervstinnervstinner approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

4 participants
@skirpichev@vstinner@bedevere-bot@zware
[8]ページ先頭
©2009-2025 Movatter.jp