QUERY · ISSUE
esp32: float/*_special tests fail
port-esp32
The following two tests fail on the esp32: float/cmath_fun_special.py and float/math_domain_special.py:
> ./run-tests --device /dev/ttyUSB0 --target esp32 float/cmath_fun_special.py float/math_domain_special.py
FAIL float/cmath_fun_special.py
FAIL float/math_domain_special.py
2 tests performed (105 individual testcases)
0 tests passed
2 tests failed: cmath_fun_special math_domain_special
> ./run-tests --print-failures
FAILURE float_math_domain_special.py
--- float_math_domain_special.py.exp 2020-04-11 09:47:24.921644291 -0700
+++ float_math_domain_special.py.out 2020-04-11 09:47:24.921644291 -0700
@@ -45,7 +45,7 @@
gamma ValueError
1.0000
inf
-gamma ValueError
+inf
nan
lgamma ValueError
lgamma ValueError
FAILURE float_cmath_fun_special.py
--- float_cmath_fun_special.py.exp 2020-04-11 09:47:24.621649221 -0700
+++ float_cmath_fun_special.py.out 2020-04-11 09:47:24.621649221 -0700
@@ -7,10 +7,10 @@
complex(1, -0.68219)
complex(-0.30103, 0)
complex(-0.30103, 1.3644)
-complex(-0.15051, 0.34109)
-complex(-0.15051, 1.0233)
-complex(-0.15051, -0.34109)
-complex(-0.15051, -1.0233)
+complex(-0.15052, 0.34109)
+complex(-0.15052, 1.0233)
+complex(-0.15052, -0.34109)
+complex(-0.15052, -1.0233)
complex(0.12448, 0.51506)
complex(0.12448, 0.84932)
complex(0.12448, -0.51506)
>
CANDIDATE · PULL REQUEST
esp32: Workaround math.gamma(-inf) result.
port-esp32
Summary
Without this workaround, math.gamma(-float("inf")) returns inf instead of raising a math domain ValueError.
Found while testing #15523.
Upstream issue reported as https://github.com/espressif/esp-idf/issues/14273
Testing
tests/float/math_domain_special.py fails on esp32 without this fix.
This work was funded through GitHub Sponsors.