From 4da6db51880289f0bf41b39e05cf9bb1c4769c47 Mon Sep 17 00:00:00 2001 From: Joseph Myers Date: Mon, 23 Jun 2014 20:12:33 +0000 Subject: Fix pow overflow in non-default rounding modes (bug 16315). This patch fixes bug 16315, bad pow handling of overflow/underflow in non-default rounding modes. Tests of pow are duly converted to ALL_RM_TEST to run all tests in all rounding modes. There are two main issues here. First, various implementations compute a negative result by negating a positive result, but this yields inappropriate overflow / underflow values for directed rounding, so either overflow / underflow results need recomputing in the correct sign, or the relevant overflowing / underflowing operation needs to be made to have a result of the correct sign. Second, the dbl-64 implementation sets FE_TONEAREST internally; in the overflow / underflow case, the result needs recomputing in the original rounding mode. Tested x86_64 and x86 and ulps updated accordingly. [BZ #16315] * sysdeps/i386/fpu/e_pow.S (__ieee754_pow): Ensure possibly overflowing or underflowing operations take place with sign of result. * sysdeps/i386/fpu/e_powf.S (__ieee754_powf): Likewise. * sysdeps/i386/fpu/e_powl.S (__ieee754_powl): Likewise. * sysdeps/ieee754/dbl-64/e_pow.c: Include . (__ieee754_pow): Recompute overflowing and underflowing results in original rounding mode. * sysdeps/x86/fpu/powl_helper.c: Include . (__powl_helper): Allow negative argument X and scale negated value as needed. Avoid passing value outside [-1, 1] to f2xm1. * sysdeps/x86_64/fpu/e_powl.S (__ieee754_powl): Ensure possibly overflowing or underflowing operations take place with sign of result. * sysdeps/x86_64/fpu/multiarch/e_pow.c [HAVE_FMA4_SUPPORT]: Include . * math/auto-libm-test-in: Add more tests of pow. * math/auto-libm-test-out: Regenerated. * math/libm-test.inc (pow_test): Use ALL_RM_TEST. (pow_tonearest_test_data): Remove. (pow_test_tonearest): Likewise. (pow_towardzero_test_data): Likewise. (pow_test_towardzero): Likewise. (pow_downward_test_data): Likewise. (pow_test_downward): Likewise. (pow_upward_test_data): Likewise. (pow_test_upward): Likewise. (main): Don't call removed functions. * sysdeps/i386/fpu/libm-test-ulps: Update. * sysdeps/x86_64/fpu/libm-test-ulps: Likewise. --- math/auto-libm-test-in | 65 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 65 insertions(+) (limited to 'math/auto-libm-test-in') diff --git a/math/auto-libm-test-in b/math/auto-libm-test-in index 6edad5ad7e..0053340e18 100644 --- a/math/auto-libm-test-in +++ b/math/auto-libm-test-in @@ -1613,6 +1613,71 @@ pow -max 0x1.ffffffffffffffffffffffffffffp+112 pow -max 0x1.ffffffffffffffffffffffffffffp+113 pow -max max +pow -0x1p65 2 +pow -0x1p65 3 +pow -0x1p65 4 +pow -0x1p65 5 +pow -0x1p43 3 +pow -0x1p43 4 +pow -0x1p43 5 +pow -0x1p33 4 +pow -0x1p33 5 +pow -0x1p26 5 +pow -0x1p-65 -2 +pow -0x1p-65 -3 +pow -0x1p-65 -4 +pow -0x1p-65 -5 +pow -0x1p-43 -3 +pow -0x1p-43 -4 +pow -0x1p-43 -5 +pow -0x1p-33 -4 +pow -0x1p-33 -5 +pow -0x1p-26 -5 + +pow -0x1p513 2 +pow -0x1p513 3 +pow -0x1p513 4 +pow -0x1p513 5 +pow -0x1p342 3 +pow -0x1p342 4 +pow -0x1p342 5 +pow -0x1p257 4 +pow -0x1p257 5 +pow -0x1p205 5 +pow -0x1p-513 -2 +pow -0x1p-513 -3 +pow -0x1p-513 -4 +pow -0x1p-513 -5 +pow -0x1p-342 -3 +pow -0x1p-342 -4 +pow -0x1p-342 -5 +pow -0x1p-257 -4 +pow -0x1p-257 -5 +pow -0x1p-205 -5 + +pow -0x1p8192 2 +pow -0x1p8192 3 +pow -0x1p8192 4 +pow -0x1p8192 5 +pow -0x1p5462 3 +pow -0x1p5462 4 +pow -0x1p5462 5 +pow -0x1p4097 4 +pow -0x1p4097 5 +pow -0x1p3277 5 +pow -0x1p64 257 +pow -0x1p-8192 -2 +pow -0x1p-8192 -3 +pow -0x1p-8192 -4 +pow -0x1p-8192 -5 +pow -0x1p-5462 -3 +pow -0x1p-5462 -4 +pow -0x1p-5462 -5 +pow -0x1p-4097 -4 +pow -0x1p-4097 -5 +pow -0x1p-3277 -5 +pow -0x1p-64 -257 + pow -0.5 126 pow -0.5 127 pow -0.5 -126 -- cgit 1.4.1