generic-library/bionic
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
92b9cb2c89
bionic/libm/Android.mk

335 lines
13 KiB
Makefile
Raw Normal View History

Add a flag control that PRIVATE libm can be chosen Change-Id: I080456a2c45ac870965c25552c5821875568fe9b Author: Xiaokang, Qin <xiaokang.qin@intel.com> Signed-off-by: Xiaokang, Qin <xiaokang.qin@intel.com> Signed-off-by: Shuo Gao <shuo.gao@intel.com> Signed-off-by: Bruce Beare <bruce.j.beare@intel.com> Signed-off-by: Jack Ren <jack.ren@intel.com> Author-tracking-BZ: 63396
2012-10-20 20:48:43 +02:00
ifneq ($(TARGET_USE_PRIVATE_LIBM),true)
Revert "Switch libm to building with clang." This reverts commit aadc4b2ff03b12142f2c6163266ebdb1df2a6e76. The patch broke x86. Change-Id: I0cda903732a685838bbd1877f94e1593a2a32acc
2014-07-11 03:24:15 +02:00
LOCAL_PATH:= $(call my-dir)
Initial Contribution
2008-10-21 16:00:00 +02:00
Add coverage flag for libm. Change-Id: I195d38c5e1f50d1cc6d4353dc8857f11b803b7a3
2014-10-01 01:56:43 +02:00
bionic_coverage := false
Reimplement isinf/isnan/fpclassify. Also move isinf and isnan into libc like everyone else. Also move fpclassify to libc like the BSDs (but unlike glibc). We need this to be able to upgrade our float/double/long double parsing to gdtoa. Also add some missing aliases. We now have all of: isnan, __isnan, isnanf, __isnanf, isnanl, __isnanl, isinf, __isinf, isinff, __isinff, isinfl, __isinfl, __fpclassify, __fpclassifyd, __fpclassifyf, __fpclassifyl. Bug: 13469877 Change-Id: I407ffbac06c765a6c5fffda8106c37d7db04f27d
2014-04-12 02:02:20 +02:00
# TODO: this comes from from upstream's libc, not libm, but it's an
# implementation detail that should have hidden visibility, so it needs
# to be in whatever library the math code is in.
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
libm_common_src_files := \
digittoint.c \
# TODO: this is not in the BSDs.
libm_common_src_files += \
Clean-up _fpmath and fake_long_doubles - promoted IEEEld2bits to fpmath since most of the where the same for diffrent archs - removed _fpmath - reinstated weak_references - moved isfinite and isnormal to libc - clean up fake_long_doubles - clean up some useless ifdefs - added missing nexttoward* tests Bug: 14134235 Change-Id: I95639c4885653fe47fd7dc0570ee5bb3389bbc6b
2014-04-17 19:17:32 +02:00
significandl.c \
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
sincos.c \
libm_common_src_files += \
upstream-freebsd/lib/msun/bsdsrc/b_exp.c \
upstream-freebsd/lib/msun/bsdsrc/b_log.c \
upstream-freebsd/lib/msun/bsdsrc/b_tgamma.c \
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
upstream-freebsd/lib/msun/src/catrig.c \
upstream-freebsd/lib/msun/src/catrigf.c \
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
upstream-freebsd/lib/msun/src/e_acos.c \
upstream-freebsd/lib/msun/src/e_acosf.c \
upstream-freebsd/lib/msun/src/e_acosh.c \
upstream-freebsd/lib/msun/src/e_acoshf.c \
upstream-freebsd/lib/msun/src/e_asin.c \
upstream-freebsd/lib/msun/src/e_asinf.c \
upstream-freebsd/lib/msun/src/e_atan2.c \
upstream-freebsd/lib/msun/src/e_atan2f.c \
upstream-freebsd/lib/msun/src/e_atanh.c \
upstream-freebsd/lib/msun/src/e_atanhf.c \
upstream-freebsd/lib/msun/src/e_cosh.c \
upstream-freebsd/lib/msun/src/e_coshf.c \
upstream-freebsd/lib/msun/src/e_exp.c \
upstream-freebsd/lib/msun/src/e_expf.c \
upstream-freebsd/lib/msun/src/e_fmod.c \
upstream-freebsd/lib/msun/src/e_fmodf.c \
upstream-freebsd/lib/msun/src/e_gamma.c \
upstream-freebsd/lib/msun/src/e_gammaf.c \
upstream-freebsd/lib/msun/src/e_gammaf_r.c \
upstream-freebsd/lib/msun/src/e_gamma_r.c \
upstream-freebsd/lib/msun/src/e_hypot.c \
upstream-freebsd/lib/msun/src/e_hypotf.c \
upstream-freebsd/lib/msun/src/e_j0.c \
upstream-freebsd/lib/msun/src/e_j0f.c \
upstream-freebsd/lib/msun/src/e_j1.c \
upstream-freebsd/lib/msun/src/e_j1f.c \
upstream-freebsd/lib/msun/src/e_jn.c \
upstream-freebsd/lib/msun/src/e_jnf.c \
upstream-freebsd/lib/msun/src/e_lgamma.c \
upstream-freebsd/lib/msun/src/e_lgammaf.c \
upstream-freebsd/lib/msun/src/e_lgammaf_r.c \
upstream-freebsd/lib/msun/src/e_lgamma_r.c \
upstream-freebsd/lib/msun/src/e_log10.c \
upstream-freebsd/lib/msun/src/e_log10f.c \
upstream-freebsd/lib/msun/src/e_log2.c \
upstream-freebsd/lib/msun/src/e_log2f.c \
upstream-freebsd/lib/msun/src/e_log.c \
upstream-freebsd/lib/msun/src/e_logf.c \
upstream-freebsd/lib/msun/src/e_pow.c \
upstream-freebsd/lib/msun/src/e_powf.c \
upstream-freebsd/lib/msun/src/e_remainder.c \
upstream-freebsd/lib/msun/src/e_remainderf.c \
upstream-freebsd/lib/msun/src/e_rem_pio2.c \
upstream-freebsd/lib/msun/src/e_rem_pio2f.c \
upstream-freebsd/lib/msun/src/e_scalb.c \
upstream-freebsd/lib/msun/src/e_scalbf.c \
upstream-freebsd/lib/msun/src/e_sinh.c \
upstream-freebsd/lib/msun/src/e_sinhf.c \
upstream-freebsd/lib/msun/src/e_sqrt.c \
upstream-freebsd/lib/msun/src/e_sqrtf.c \
Clean-up _fpmath and fake_long_doubles - promoted IEEEld2bits to fpmath since most of the where the same for diffrent archs - removed _fpmath - reinstated weak_references - moved isfinite and isnormal to libc - clean up fake_long_doubles - clean up some useless ifdefs - added missing nexttoward* tests Bug: 14134235 Change-Id: I95639c4885653fe47fd7dc0570ee5bb3389bbc6b
2014-04-17 19:17:32 +02:00
upstream-freebsd/lib/msun/src/imprecise.c \
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
upstream-freebsd/lib/msun/src/k_cos.c \
upstream-freebsd/lib/msun/src/k_cosf.c \
upstream-freebsd/lib/msun/src/k_exp.c \
upstream-freebsd/lib/msun/src/k_expf.c \
upstream-freebsd/lib/msun/src/k_rem_pio2.c \
upstream-freebsd/lib/msun/src/k_sin.c \
upstream-freebsd/lib/msun/src/k_sinf.c \
upstream-freebsd/lib/msun/src/k_tan.c \
upstream-freebsd/lib/msun/src/k_tanf.c \
upstream-freebsd/lib/msun/src/s_asinh.c \
upstream-freebsd/lib/msun/src/s_asinhf.c \
upstream-freebsd/lib/msun/src/s_atan.c \
upstream-freebsd/lib/msun/src/s_atanf.c \
upstream-freebsd/lib/msun/src/s_carg.c \
upstream-freebsd/lib/msun/src/s_cargf.c \
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
upstream-freebsd/lib/msun/src/s_cargl.c \
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
upstream-freebsd/lib/msun/src/s_cbrt.c \
upstream-freebsd/lib/msun/src/s_cbrtf.c \
upstream-freebsd/lib/msun/src/s_ccosh.c \
upstream-freebsd/lib/msun/src/s_ccoshf.c \
upstream-freebsd/lib/msun/src/s_ceil.c \
upstream-freebsd/lib/msun/src/s_ceilf.c \
upstream-freebsd/lib/msun/src/s_cexp.c \
upstream-freebsd/lib/msun/src/s_cexpf.c \
upstream-freebsd/lib/msun/src/s_cimag.c \
upstream-freebsd/lib/msun/src/s_cimagf.c \
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
upstream-freebsd/lib/msun/src/s_cimagl.c \
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
upstream-freebsd/lib/msun/src/s_conj.c \
upstream-freebsd/lib/msun/src/s_conjf.c \
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
upstream-freebsd/lib/msun/src/s_conjl.c \
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
upstream-freebsd/lib/msun/src/s_copysign.c \
upstream-freebsd/lib/msun/src/s_copysignf.c \
upstream-freebsd/lib/msun/src/s_cos.c \
upstream-freebsd/lib/msun/src/s_cosf.c \
upstream-freebsd/lib/msun/src/s_cproj.c \
upstream-freebsd/lib/msun/src/s_cprojf.c \
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
upstream-freebsd/lib/msun/src/s_cprojl.c \
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
upstream-freebsd/lib/msun/src/s_creal.c \
upstream-freebsd/lib/msun/src/s_crealf.c \
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
upstream-freebsd/lib/msun/src/s_creall.c \
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
upstream-freebsd/lib/msun/src/s_csinh.c \
upstream-freebsd/lib/msun/src/s_csinhf.c \
upstream-freebsd/lib/msun/src/s_csqrt.c \
upstream-freebsd/lib/msun/src/s_csqrtf.c \
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
upstream-freebsd/lib/msun/src/s_csqrtl.c \
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
upstream-freebsd/lib/msun/src/s_ctanh.c \
upstream-freebsd/lib/msun/src/s_ctanhf.c \
upstream-freebsd/lib/msun/src/s_erf.c \
upstream-freebsd/lib/msun/src/s_erff.c \
upstream-freebsd/lib/msun/src/s_exp2.c \
upstream-freebsd/lib/msun/src/s_exp2f.c \
upstream-freebsd/lib/msun/src/s_expm1.c \
upstream-freebsd/lib/msun/src/s_expm1f.c \
upstream-freebsd/lib/msun/src/s_fabs.c \
upstream-freebsd/lib/msun/src/s_fabsf.c \
upstream-freebsd/lib/msun/src/s_fdim.c \
upstream-freebsd/lib/msun/src/s_finite.c \
upstream-freebsd/lib/msun/src/s_finitef.c \
upstream-freebsd/lib/msun/src/s_floor.c \
upstream-freebsd/lib/msun/src/s_floorf.c \
upstream-freebsd/lib/msun/src/s_fma.c \
upstream-freebsd/lib/msun/src/s_fmaf.c \
upstream-freebsd/lib/msun/src/s_fmax.c \
upstream-freebsd/lib/msun/src/s_fmaxf.c \
upstream-freebsd/lib/msun/src/s_fmin.c \
upstream-freebsd/lib/msun/src/s_fminf.c \
upstream-freebsd/lib/msun/src/s_frexp.c \
upstream-freebsd/lib/msun/src/s_frexpf.c \
upstream-freebsd/lib/msun/src/s_ilogb.c \
upstream-freebsd/lib/msun/src/s_ilogbf.c \
upstream-freebsd/lib/msun/src/s_llrint.c \
upstream-freebsd/lib/msun/src/s_llrintf.c \
upstream-freebsd/lib/msun/src/s_llround.c \
upstream-freebsd/lib/msun/src/s_llroundf.c \
upstream-freebsd/lib/msun/src/s_log1p.c \
upstream-freebsd/lib/msun/src/s_log1pf.c \
upstream-freebsd/lib/msun/src/s_logb.c \
upstream-freebsd/lib/msun/src/s_logbf.c \
upstream-freebsd/lib/msun/src/s_lrint.c \
upstream-freebsd/lib/msun/src/s_lrintf.c \
upstream-freebsd/lib/msun/src/s_lround.c \
upstream-freebsd/lib/msun/src/s_lroundf.c \
upstream-freebsd/lib/msun/src/s_modf.c \
upstream-freebsd/lib/msun/src/s_modff.c \
upstream-freebsd/lib/msun/src/s_nan.c \
upstream-freebsd/lib/msun/src/s_nearbyint.c \
upstream-freebsd/lib/msun/src/s_nextafter.c \
upstream-freebsd/lib/msun/src/s_nextafterf.c \
upstream-freebsd/lib/msun/src/s_remquo.c \
upstream-freebsd/lib/msun/src/s_remquof.c \
upstream-freebsd/lib/msun/src/s_rint.c \
upstream-freebsd/lib/msun/src/s_rintf.c \
upstream-freebsd/lib/msun/src/s_round.c \
upstream-freebsd/lib/msun/src/s_roundf.c \
upstream-freebsd/lib/msun/src/s_scalbln.c \
upstream-freebsd/lib/msun/src/s_scalbn.c \
upstream-freebsd/lib/msun/src/s_scalbnf.c \
upstream-freebsd/lib/msun/src/s_signgam.c \
upstream-freebsd/lib/msun/src/s_significand.c \
upstream-freebsd/lib/msun/src/s_significandf.c \
upstream-freebsd/lib/msun/src/s_sin.c \
upstream-freebsd/lib/msun/src/s_sinf.c \
upstream-freebsd/lib/msun/src/s_tan.c \
upstream-freebsd/lib/msun/src/s_tanf.c \
upstream-freebsd/lib/msun/src/s_tanh.c \
upstream-freebsd/lib/msun/src/s_tanhf.c \
upstream-freebsd/lib/msun/src/s_tgammaf.c \
upstream-freebsd/lib/msun/src/s_trunc.c \
upstream-freebsd/lib/msun/src/s_truncf.c \
upstream-freebsd/lib/msun/src/w_cabs.c \
upstream-freebsd/lib/msun/src/w_cabsf.c \
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
upstream-freebsd/lib/msun/src/w_cabsl.c \
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
upstream-freebsd/lib/msun/src/w_drem.c \
upstream-freebsd/lib/msun/src/w_dremf.c \
We actually have modfl. Change-Id: I22bb4d14f99a418819f2941c2dfec49094807c5f
2013-12-19 20:30:51 +01:00
libm_common_src_files += \
fake_long_double.c \
Clean-up _fpmath and fake_long_doubles - promoted IEEEld2bits to fpmath since most of the where the same for diffrent archs - removed _fpmath - reinstated weak_references - moved isfinite and isnormal to libc - clean up fake_long_doubles - clean up some useless ifdefs - added missing nexttoward* tests Bug: 14134235 Change-Id: I95639c4885653fe47fd7dc0570ee5bb3389bbc6b
2014-04-17 19:17:32 +02:00
signbit.c \
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
libm_ld128_src_files = \
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
upstream-freebsd/lib/msun/src/e_acosl.c \
upstream-freebsd/lib/msun/src/e_acoshl.c \
upstream-freebsd/lib/msun/src/e_asinl.c \
upstream-freebsd/lib/msun/src/e_atan2l.c \
upstream-freebsd/lib/msun/src/e_atanhl.c \
upstream-freebsd/lib/msun/src/e_fmodl.c \
upstream-freebsd/lib/msun/src/e_hypotl.c \
Fix lgamma_r/lgammaf_r/lgammal_r for -0. Upstream has implemented lgammal/lgammal_r for ld128, and fixed the sign problem we reported with all the lgamma*_r functions and -0. Bug: 17471883 Change-Id: Ibb175d9cab67efae75f1010796fd44c9ba6ce4fc
2014-09-18 20:23:58 +02:00
upstream-freebsd/lib/msun/src/e_lgammal.c \
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
upstream-freebsd/lib/msun/src/e_remainderl.c \
upstream-freebsd/lib/msun/src/e_sqrtl.c \
upstream-freebsd/lib/msun/src/s_asinhl.c \
upstream-freebsd/lib/msun/src/s_atanl.c \
upstream-freebsd/lib/msun/src/s_cbrtl.c \
upstream-freebsd/lib/msun/src/s_ceill.c \
upstream-freebsd/lib/msun/src/s_copysignl.c \
Get new coshl, sinhl, and tanhl implementations from upstream. Change-Id: I92f9cd53d12efb2e0f4f8a9590b9fe42398233f6
2014-09-13 01:09:40 +02:00
upstream-freebsd/lib/msun/src/e_coshl.c \
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
upstream-freebsd/lib/msun/src/s_cosl.c \
upstream-freebsd/lib/msun/src/s_fabsl.c \
upstream-freebsd/lib/msun/src/s_floorl.c \
upstream-freebsd/lib/msun/src/s_fmal.c \
upstream-freebsd/lib/msun/src/s_fmaxl.c \
upstream-freebsd/lib/msun/src/s_fminl.c \
We actually have modfl. Change-Id: I22bb4d14f99a418819f2941c2dfec49094807c5f
2013-12-19 20:30:51 +01:00
upstream-freebsd/lib/msun/src/s_modfl.c \
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
upstream-freebsd/lib/msun/src/s_frexpl.c \
upstream-freebsd/lib/msun/src/s_ilogbl.c \
upstream-freebsd/lib/msun/src/s_llrintl.c \
upstream-freebsd/lib/msun/src/s_llroundl.c \
upstream-freebsd/lib/msun/src/s_logbl.c \
upstream-freebsd/lib/msun/src/s_lrintl.c \
upstream-freebsd/lib/msun/src/s_lroundl.c \
upstream-freebsd/lib/msun/src/s_nextafterl.c \
upstream-freebsd/lib/msun/src/s_nexttoward.c \
Clean-up _fpmath and fake_long_doubles - promoted IEEEld2bits to fpmath since most of the where the same for diffrent archs - removed _fpmath - reinstated weak_references - moved isfinite and isnormal to libc - clean up fake_long_doubles - clean up some useless ifdefs - added missing nexttoward* tests Bug: 14134235 Change-Id: I95639c4885653fe47fd7dc0570ee5bb3389bbc6b
2014-04-17 19:17:32 +02:00
upstream-freebsd/lib/msun/src/s_nexttowardf.c \
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
upstream-freebsd/lib/msun/src/s_remquol.c \
upstream-freebsd/lib/msun/src/s_rintl.c \
upstream-freebsd/lib/msun/src/s_roundl.c \
upstream-freebsd/lib/msun/src/s_scalbnl.c \
Get new coshl, sinhl, and tanhl implementations from upstream. Change-Id: I92f9cd53d12efb2e0f4f8a9590b9fe42398233f6
2014-09-13 01:09:40 +02:00
upstream-freebsd/lib/msun/src/e_sinhl.c \
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
upstream-freebsd/lib/msun/src/s_sinl.c \
Get new coshl, sinhl, and tanhl implementations from upstream. Change-Id: I92f9cd53d12efb2e0f4f8a9590b9fe42398233f6
2014-09-13 01:09:40 +02:00
upstream-freebsd/lib/msun/src/s_tanhl.c \
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
upstream-freebsd/lib/msun/src/s_tanl.c \
upstream-freebsd/lib/msun/src/s_truncl.c \
Revert "Pull the pthread_key_t functions out of pthread.c." This reverts commit ad59322ae432d11ff36dcf046016af8cfe45fbe4 somehow my unfinished libm/Android.mk change got into here. Change-Id: I46be626c5269d60fb1ced9862f2ebaa380b4e0af
2013-02-11 21:18:16 +01:00
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
libm_ld128_src_files += \
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
upstream-freebsd/lib/msun/ld128/invtrig.c \
Fix lgamma_r/lgammaf_r/lgammal_r for -0. Upstream has implemented lgammal/lgammal_r for ld128, and fixed the sign problem we reported with all the lgamma*_r functions and -0. Bug: 17471883 Change-Id: Ibb175d9cab67efae75f1010796fd44c9ba6ce4fc
2014-09-18 20:23:58 +02:00
upstream-freebsd/lib/msun/ld128/e_lgammal_r.c \
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
upstream-freebsd/lib/msun/ld128/k_cosl.c \
upstream-freebsd/lib/msun/ld128/k_sinl.c \
upstream-freebsd/lib/msun/ld128/k_tanl.c \
Sync libm with upstream. Change-Id: I8ac8ee52122ee19a2e423c3211092023cb4896eb
2014-09-12 23:00:02 +02:00
upstream-freebsd/lib/msun/ld128/s_erfl.c \
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
upstream-freebsd/lib/msun/ld128/s_exp2l.c \
upstream-freebsd/lib/msun/ld128/s_expl.c \
upstream-freebsd/lib/msun/ld128/s_logl.c \
upstream-freebsd/lib/msun/ld128/s_nanl.c \
Revert "Pull the pthread_key_t functions out of pthread.c." This reverts commit ad59322ae432d11ff36dcf046016af8cfe45fbe4 somehow my unfinished libm/Android.mk change got into here. Change-Id: I46be626c5269d60fb1ced9862f2ebaa380b4e0af
2013-02-11 21:18:16 +01:00
# TODO: re-enable i387/e_sqrtf.S for x86, and maybe others.
Pull the pthread_key_t functions out of pthread.c. This was originally motivated by noticing that we were setting the wrong bits for the well-known tls entries. That was a harmless bug because none of the well-known tls entries has a destructor, but it's best not to leave land mines lying around. Also add some missing POSIX constants, a new test, and fix pthread_key_create's return value when we hit the limit. Change-Id: Ife26ea2f4b40865308e8410ec803b20bcc3e0ed1
2013-02-09 00:46:37 +01:00
Switch to upstream sleep(3) and usleep(3). Also fix the signature of usleep, and the definition of useconds_t which should be unsigned, as the 'u' in its name implies. This patch also cleans up the existing FreeBSD hacks by moving the libm stuff from <sys/cdefs.h> to a libm-private header, and adding comments about the hacks we use to build FreeBSD source. Change-Id: Ibe5067a380502df94a0a3a7901969b35411085b6
2013-11-21 01:09:06 +01:00
libm_common_cflags := \
-DFLT_EVAL_METHOD=0 \
Revert "Switch libm to building with clang." This reverts commit aadc4b2ff03b12142f2c6163266ebdb1df2a6e76. The patch broke x86. Change-Id: I0cda903732a685838bbd1877f94e1593a2a32acc
2014-07-11 03:24:15 +02:00
-std=c99 \
Switch to upstream sleep(3) and usleep(3). Also fix the signature of usleep, and the definition of useconds_t which should be unsigned, as the 'u' in its name implies. This patch also cleans up the existing FreeBSD hacks by moving the libm stuff from <sys/cdefs.h> to a libm-private header, and adding comments about the hacks we use to build FreeBSD source. Change-Id: Ibe5067a380502df94a0a3a7901969b35411085b6
2013-11-21 01:09:06 +01:00
-include $(LOCAL_PATH)/freebsd-compat.h \
Fix <math.h> to quieten most of our warnings. I've reported the wcsftime bug upstream, but we really just want to use -D to ensure the buggy code isn't built. (I've also brought our strftime a bit closer to upstream now we have the right define.) I don't think upstream is likely to fix all their sign-compare and uninitialized warnings, so let's just silence them. As for libm, again upstream isn't likely to fix all their warnings, and silencing those made the ones that were our fault stand out. I've fixed our <math.h> to fix the warnings caused by our lack of definitions for the non-imprecise long-double functions. I checked the C99 standard, and all these functions are there. Change-Id: Iee8e1182c1db375058fb2c451eceb212bab47a37
2014-05-06 05:28:28 +02:00
-Wno-missing-braces \
-Wno-parentheses \
-Wno-sign-compare \
-Wno-uninitialized \
Revert "Switch libm to building with clang." This reverts commit aadc4b2ff03b12142f2c6163266ebdb1df2a6e76. The patch broke x86. Change-Id: I0cda903732a685838bbd1877f94e1593a2a32acc
2014-07-11 03:24:15 +02:00
-Wno-unknown-pragmas \
Use -fvisibility=hidden to build libm. Bug: 11156955 Change-Id: I77b7772a44313a1eb3970f5d2c8e2af63690efa4
2014-05-20 01:58:52 +02:00
-fvisibility=hidden \
Switch to upstream sleep(3) and usleep(3). Also fix the signature of usleep, and the definition of useconds_t which should be unsigned, as the 'u' in its name implies. This patch also cleans up the existing FreeBSD hacks by moving the libm stuff from <sys/cdefs.h> to a libm-private header, and adding comments about the hacks we use to build FreeBSD source. Change-Id: Ibe5067a380502df94a0a3a7901969b35411085b6
2013-11-21 01:09:06 +01:00
Workaround the GCC "(long)fn -> lfn" optimization bug. Pass "-fno-builtin-rint -fno-builtin-rintl -fno-builtin-rintf" to avoid generating self recursions in lrint, lrintl, and lrintf. BUG: 14225968 Change-Id: Iffc4d8f0692580d636a529f4c8aa9a7a10d7acd8
2014-05-30 01:02:09 +02:00
# Workaround the GCC "(long)fn -> lfn" optimization bug which will result in
# self recursions for lrint, lrintf, and lrintl.
# BUG: 14225968
libm_common_cflags += -fno-builtin-rint -fno-builtin-rintf -fno-builtin-rintl
Revert "Pull the pthread_key_t functions out of pthread.c." This reverts commit ad59322ae432d11ff36dcf046016af8cfe45fbe4 somehow my unfinished libm/Android.mk change got into here. Change-Id: I46be626c5269d60fb1ced9862f2ebaa380b4e0af
2013-02-11 21:18:16 +01:00
libm_common_includes := $(LOCAL_PATH)/upstream-freebsd/lib/msun/src/
Reimplement isinf/isnan/fpclassify. Also move isinf and isnan into libc like everyone else. Also move fpclassify to libc like the BSDs (but unlike glibc). We need this to be able to upgrade our float/double/long double parsing to gdtoa. Also add some missing aliases. We now have all of: isnan, __isnan, isnanf, __isnanf, isnanl, __isnanl, isinf, __isinf, isinff, __isinff, isinfl, __isinfl, __fpclassify, __fpclassifyd, __fpclassifyf, __fpclassifyl. Bug: 13469877 Change-Id: I407ffbac06c765a6c5fffda8106c37d7db04f27d
2014-04-12 02:02:20 +02:00
Fix LP64 libm for 128-bit long doubles * reworked amd64/_fpmath.h and arm64/_fpmath.h to support 128-bit long doubles. * improved tests to cover long double cases * made modfl an alias for LP32 Tests pass on x86, x86_64, arm, arm64 and mips. Bug: 12921273 Change-Id: Ibe39acde57972447a8950fa45b1e702acc68ebeb
2014-03-14 18:56:46 +01:00
libm_ld_includes := $(LOCAL_PATH)/upstream-freebsd/lib/msun/ld128/
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
#
# libm.a for target.
#
Initial Contribution
2008-10-21 16:00:00 +02:00
include $(CLEAR_VARS)
Disable Clang, due to x86 LDBL_MANT_DIG errors. Clang -m32 has incorrect sizeof(long double) = 12. With -m64, LDBL_MANT_DIG is incorrect 64. BUG: 17163651 Change-Id: I4b157aa1a8572c3a10aece6070e119a292ab8e83
2014-09-04 21:35:19 +02:00
ifneq (,$(filter $(TARGET_ARCH),x86 x86_64))
# Clang has wrong long double sizes for x86.
LOCAL_CLANG := false
endif
Revert "Switch libm to building with clang." This reverts commit aadc4b2ff03b12142f2c6163266ebdb1df2a6e76. The patch broke x86. Change-Id: I0cda903732a685838bbd1877f94e1593a2a32acc
2014-07-11 03:24:15 +02:00
LOCAL_MODULE:= libm
Add dependencies on all the makefiles. This should help prevent broken builds next time I'm messing with assembler/compiler/linker flags... Change-Id: I30f15a3ce3c3f3c60cad7bc59aaba9f42d792224
2012-08-13 23:06:05 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(LOCAL_PATH)/Android.mk
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
LOCAL_ARM_MODE := arm
Use arch-specific LOCAL_ variables to build libm. Use arch-specific LOCAL_ variables to build libm for both TARGET_ARCH and TARGET_2ND_ARCH. Bug: 11654773 Change-Id: I6da794ba722bb68e7484d8869c6eb0425b7d17cb
2014-01-27 23:59:26 +01:00
LOCAL_CFLAGS := $(libm_common_cflags)
LOCAL_C_INCLUDES += $(libm_common_includes)
LOCAL_SRC_FILES := $(libm_common_src_files)
Initial Contribution
2008-10-21 16:00:00 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES := libc
Use arch-specific LOCAL_ variables to build libm. Use arch-specific LOCAL_ variables to build libm for both TARGET_ARCH and TARGET_2ND_ARCH. Bug: 11654773 Change-Id: I6da794ba722bb68e7484d8869c6eb0425b7d17cb
2014-01-27 23:59:26 +01:00
Add coverage flag for libm. Change-Id: I195d38c5e1f50d1cc6d4353dc8857f11b803b7a3
2014-10-01 01:56:43 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Use arch-specific LOCAL_ variables to build libm. Use arch-specific LOCAL_ variables to build libm for both TARGET_ARCH and TARGET_2ND_ARCH. Bug: 11654773 Change-Id: I6da794ba722bb68e7484d8869c6eb0425b7d17cb
2014-01-27 23:59:26 +01:00
# arch-specific settings
LOCAL_C_INCLUDES_arm := $(LOCAL_PATH)/arm
LOCAL_SRC_FILES_arm := arm/fenv.c
Clean-up _fpmath and fake_long_doubles - promoted IEEEld2bits to fpmath since most of the where the same for diffrent archs - removed _fpmath - reinstated weak_references - moved isfinite and isnormal to libc - clean up fake_long_doubles - clean up some useless ifdefs - added missing nexttoward* tests Bug: 14134235 Change-Id: I95639c4885653fe47fd7dc0570ee5bb3389bbc6b
2014-04-17 19:17:32 +02:00
LOCAL_C_INCLUDES_arm64 := $(libm_ld_includes)
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
LOCAL_SRC_FILES_arm64 := arm64/fenv.c $(libm_ld128_src_files)
Use arch-specific LOCAL_ variables to build libm. Use arch-specific LOCAL_ variables to build libm for both TARGET_ARCH and TARGET_2ND_ARCH. Bug: 11654773 Change-Id: I6da794ba722bb68e7484d8869c6eb0425b7d17cb
2014-01-27 23:59:26 +01:00
Clean-up _fpmath and fake_long_doubles - promoted IEEEld2bits to fpmath since most of the where the same for diffrent archs - removed _fpmath - reinstated weak_references - moved isfinite and isnormal to libc - clean up fake_long_doubles - clean up some useless ifdefs - added missing nexttoward* tests Bug: 14134235 Change-Id: I95639c4885653fe47fd7dc0570ee5bb3389bbc6b
2014-04-17 19:17:32 +02:00
LOCAL_C_INCLUDES_x86 := $(LOCAL_PATH)/i387
Use arch-specific LOCAL_ variables to build libm. Use arch-specific LOCAL_ variables to build libm for both TARGET_ARCH and TARGET_2ND_ARCH. Bug: 11654773 Change-Id: I6da794ba722bb68e7484d8869c6eb0425b7d17cb
2014-01-27 23:59:26 +01:00
LOCAL_SRC_FILES_x86 := i387/fenv.c
Clean-up _fpmath and fake_long_doubles - promoted IEEEld2bits to fpmath since most of the where the same for diffrent archs - removed _fpmath - reinstated weak_references - moved isfinite and isnormal to libc - clean up fake_long_doubles - clean up some useless ifdefs - added missing nexttoward* tests Bug: 14134235 Change-Id: I95639c4885653fe47fd7dc0570ee5bb3389bbc6b
2014-04-17 19:17:32 +02:00
LOCAL_C_INCLUDES_x86_64 := $(libm_ld_includes)
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
LOCAL_SRC_FILES_x86_64 := amd64/fenv.c $(libm_ld128_src_files)
Use arch-specific LOCAL_ variables to build libm. Use arch-specific LOCAL_ variables to build libm for both TARGET_ARCH and TARGET_2ND_ARCH. Bug: 11654773 Change-Id: I6da794ba722bb68e7484d8869c6eb0425b7d17cb
2014-01-27 23:59:26 +01:00
LOCAL_SRC_FILES_mips := mips/fenv.c
Clean-up _fpmath and fake_long_doubles - promoted IEEEld2bits to fpmath since most of the where the same for diffrent archs - removed _fpmath - reinstated weak_references - moved isfinite and isnormal to libc - clean up fake_long_doubles - clean up some useless ifdefs - added missing nexttoward* tests Bug: 14134235 Change-Id: I95639c4885653fe47fd7dc0570ee5bb3389bbc6b
2014-04-17 19:17:32 +02:00
LOCAL_C_INCLUDES_mips64 := $(libm_ld_includes)
Fix our <complex.h> support. We build libm with -fvisibility=hidden, so we weren't exporting any of the <complex.h> functions. We also weren't building many of the functions anyway. We were also missing the complex inverse trigonometric functions. And because we didn't even have perfunctory "call each function once" tests, we didn't notice that we weren't exporting any symbols, so this patch adds at least that level of testing. Change-Id: Ibcf2843f507126c51d134cc5fc8d67747e033a0d
2014-11-06 20:16:55 +01:00
LOCAL_SRC_FILES_mips64 := mips/fenv.c $(libm_ld128_src_files)
[MIPS64] libc/libm support libc/libm support for MIPS64 targets Change-Id: I8271941d418612a286be55495f0e95822f90004f Signed-off-by: Chris Dearman <chris.dearman@imgtec.com> Signed-off-by: Raghu Gandham <raghu.gandham@imgtec.com>
2014-02-06 03:51:43 +01:00
Don't use an STL for libm. The default is none right now, but this will be needed when the default becomes libc++. Change-Id: Ib022adac2b9dc08fbfae830908128043cfffb91e
2014-10-16 16:44:47 +02:00
LOCAL_CXX_STL := none
Initial Contribution
2008-10-21 16:00:00 +02:00
include $(BUILD_STATIC_LIBRARY)
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
#
# libm.so for target.
#
Initial Contribution
2008-10-21 16:00:00 +02:00
include $(CLEAR_VARS)
Disable Clang, due to x86 LDBL_MANT_DIG errors. Clang -m32 has incorrect sizeof(long double) = 12. With -m64, LDBL_MANT_DIG is incorrect 64. BUG: 17163651 Change-Id: I4b157aa1a8572c3a10aece6070e119a292ab8e83
2014-09-04 21:35:19 +02:00
ifneq (,$(filter $(TARGET_ARCH),x86 x86_64))
# Clang has wrong long double sizes for x86.
LOCAL_CLANG := false
endif
Revert "Switch libm to building with clang." This reverts commit aadc4b2ff03b12142f2c6163266ebdb1df2a6e76. The patch broke x86. Change-Id: I0cda903732a685838bbd1877f94e1593a2a32acc
2014-07-11 03:24:15 +02:00
LOCAL_MODULE:= libm
Add dependencies on all the makefiles. This should help prevent broken builds next time I'm messing with assembler/compiler/linker flags... Change-Id: I30f15a3ce3c3f3c60cad7bc59aaba9f42d792224
2012-08-13 23:06:05 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(LOCAL_PATH)/Android.mk
Initial Contribution
2008-10-21 16:00:00 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES := libc
Upgrade libm. This brings us up to date with FreeBSD HEAD, fixes various bugs, unifies the set of functions we support on ARM, MIPS, and x86, fixes "long double", adds ISO C99 support, and adds basic unit tests. It turns out that our "long double" functions have always been broken for non-normal numbers. This patch fixes that by not using the upstream implementations and just forwarding to the regular "double" implementation instead (since "long double" on Android is just "double" anyway, which is what BSD doesn't support). All the tests pass on ARM, MIPS, and x86, plus glibc on x86-64. Bug: 3169850 Bug: 8012787 Bug: https://code.google.com/p/android/issues/detail?id=6697 Change-Id: If0c343030959c24bfc50d4d21c9530052c581837
2013-01-31 04:06:37 +01:00
LOCAL_WHOLE_STATIC_LIBRARIES := libm
Revert "Revert "Prevent symbols from libgcc from being reexported."" This reverts commit 4e4e1a5d78ff6dd9394b9b116c012046da409e99. Change-Id: I8e78c3a59598e052e02965dd37b3fba8f567eed9
2014-06-24 06:31:08 +02:00
Add coverage flag for libm. Change-Id: I195d38c5e1f50d1cc6d4353dc8857f11b803b7a3
2014-10-01 01:56:43 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Don't use an STL for libm. The default is none right now, but this will be needed when the default becomes libc++. Change-Id: Ib022adac2b9dc08fbfae830908128043cfffb91e
2014-10-16 16:44:47 +02:00
LOCAL_CXX_STL := none
Revert "Revert "Prevent symbols from libgcc from being reexported."" This reverts commit 4e4e1a5d78ff6dd9394b9b116c012046da409e99. Change-Id: I8e78c3a59598e052e02965dd37b3fba8f567eed9
2014-06-24 06:31:08 +02:00
# We'd really like to do this for all architectures, but since this wasn't done
# before, these symbols must continue to be exported on LP32 for binary
# compatibility.
LOCAL_LDFLAGS_arm64 := -Wl,--exclude-libs,libgcc.a
LOCAL_LDFLAGS_mips64 := -Wl,--exclude-libs,libgcc.a
LOCAL_LDFLAGS_x86_64 := -Wl,--exclude-libs,libgcc.a
Initial Contribution
2008-10-21 16:00:00 +02:00
include $(BUILD_SHARED_LIBRARY)
Add a flag control that PRIVATE libm can be chosen Change-Id: I080456a2c45ac870965c25552c5821875568fe9b Author: Xiaokang, Qin <xiaokang.qin@intel.com> Signed-off-by: Xiaokang, Qin <xiaokang.qin@intel.com> Signed-off-by: Shuo Gao <shuo.gao@intel.com> Signed-off-by: Bruce Beare <bruce.j.beare@intel.com> Signed-off-by: Jack Ren <jack.ren@intel.com> Author-tracking-BZ: 63396
2012-10-20 20:48:43 +02:00
endif
Reference in New Issue Copy Permalink