This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
PING! This patch introduces a new wrapper for scalbln functions in order to set errno to ERANGE on overflow and underflow.The existing aliases that uses __scalbln functions are replaced by aliases to the new wrapper. The corresponding tests in libm-test.inc are updated to check for errno via ERRNO_PLUS_OFLOW etc.
Please test and give ok for commit. This patch only covers the scalbln-part of bug 6803. (The attached file is identical to the posted one from 08.05.2014) Bye On 05/08/2014 04:31 PM, Stefan Liebler wrote:
For these the behaviour is the same like before. Errno is not set and the testcases will fail. Now the scalbln-aliases are removed in i386/m68 and the wrappers are used when calling the scalbln-functions. On ia64 only scalblnf has its own implementation. For scalbln and scalblnl the ieee754/dbl-64 and ieee754/ldbl-96 are used, thus the wrappers are needed, too. Bye --- 2014-05-08 Stefan Liebler <stli@linux.vnet.ibm.com> [BZ #6803] * math/libm-test.inc (scalbln_test_date): Add errno expectations. * math/w_scalblnf.c: New File. Add wrapper which checks for setting errno to ERANGE. Add weak_alias for corresponding scalbln function. * math/w_scalbln.c: Likewise. * math/w_scalblnl.c: Likewise. * math/Makefile (libm-calls): Add w_scalbln. * sysdeps/ieee754/flt-32/s_scalblnf.c: Remove weak_alias for corresponding scalbln function. * sysdeps/ieee754/dbl-64/s_scalbln.c: Likewise. * sysdeps/ieee754/dbl-64/wordsize-64/s_scalbln.c: Likewise. * sysdeps/ieee754/ldbl-96/s_scalblnl.c: Likewise. * sysdeps/ieee754/ldbl-128/s_scalblnl.c: Likewise. * sysdeps/sparc/sparc64/soft-fp/s_scalblnl.c: Likewise. * sysdeps/i386/fpu/s_scalbnf.S: Likewise. * sysdeps/i386/fpu/s_scalbn.S: Likewise. * sysdeps/i386/fpu/s_scalbnl.S: Likewise. * sysdeps/m68k/m680x0/fpu/s_scalbn.c: Likewise. * sysdeps/ieee754/ldbl-64-128/s_scalblnl.c: Remove long_double_symbol for scalblnl function in libm, libc. * sysdeps/ieee754/ldbl-64-128/w_scalblnl.c: New File. Add wrapper which checks for setting errno to ERANGE. Add long_double_symbol for scalblnl function in libm, libc. * sysdeps/ieee754/ldbl-128ibm/s_scalblnl.c: Remove long_double_symbol for scalblnl in libm. * sysdeps/ieee754/ldbl-128ibm/w_scalblnl.c: New File. Add wrapper which checks for setting errno to ERANGE. Add long_double_symbol for scalblnl function in libm. * sysdeps/ia64/fpu/w_scalblnf.c: New File. Do not use wrapper because of own implementation. --- On 05/08/2014 12:24 PM, Andreas Schwab wrote:Stefan Liebler <stli@linux.vnet.ibm.com> writes:In case of sysdeps/i386 and sysdeps/m68k/m680x0, the existing s_scalbln.c files are empty, because the scalbln-functions are aliases for the scalbn-functions, which are not covered by this patch. Thus the wrapper-function is not used.Does that mean you are introducing a regression? Andreas.
Attachment:
patchglibc_math_scalbln_20140508
Description: Text document
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |