Featured content is also available via:

glibc

The GNU C library.

Yesterday's Activity

Merge mktime, timegm from upstream Gnulib
Paul Eggert: [BZ #23603][BZ #16346] This fixes some obscure problems with integer overflow.

Other Activity This Week

Add new pow implementation
Szabolcs Nagy: The algorithm is exp(y * log(x)), where log(x) is computed with about 1.3*2^-68 relative error (1.5*2^-68 without fma), returning the result in two doubles, and the exp part uses the same algorithm (and lookup tables) as exp, but takes the input as two doubles and a sign (to handle negative bases with odd integer exponent).

Recent Activity

Merge mktime, timegm from upstream Gnulib
Paul Eggert: [BZ #23603][BZ #16346] This fixes some obscure problems with integer overflow.

Add new pow implementation
Szabolcs Nagy: The algorithm is exp(y * log(x)), where log(x) is computed with about 1.3*2^-68 relative error (1.5*2^-68 without fma), returning the result in two doubles, and the exp part uses the same algorithm (and lookup tables) as exp, but takes the input as two doubles and a sign (to handle negative bases with odd integer exponent).

Fix MIPS n32 pr_sigpend, pr_sighold, pr_flag type (bug 23656).
Joseph Myers: As discussed at <

Add new log2 implementation
Szabolcs Nagy: Similar algorithm is used as in log: log2(2^k x) = k + log2(c) + log2(x/c) where the last term is approximated by a polynomial of x/c - 1, the first order coefficient is about 1/ln2 in this case. There is separate code path when fma instruction is not available for computing x/c - 1 precisely, for which the table size is doubled. The worst case error is 0.547 ULP (0.55 without fma), the read only global data size is 1168 bytes (2192 without fma) on aarch64.

Add build-many-glibcs.py support for building more GCC libraries
Joseph Myers: Every so often we get libsanitizer or libgo builds breaking with new glibc because of some change in the glibc headers. glibc's build-many-glibcs.py deliberately disables libsanitizer and GCC languages other than C and C++ because the point is to test glibc and find glibc problems (including problems shown up by new compiler warnings in new GCC), not to test libsanitizer or libgo; if the compiler build fails because of libsanitizer or libgo failing to build, that could hide the existence of new problems in glibc.

Add build-many-glibcs.py --enable-obsolete-* configs
Joseph Myers: We've had issues before with build failures (with new GCC) in code only built with --enable-obsolete-rpc or --enable-obsolete-nsl not being reported for a while because build-many-glibcs.py does not test those configure options.

Indian and similar locales: Set the correct date format (bug 17426).
Rafal Luzynski: This commit also fixes d_fmt in bn_BD which is identical to bn_IN, in ne_NP which is identical to ne_IN (not supported by Glibc but supported by CLDR), and in ta_LK which is identical to ta_IN. For those locales which are supported by CLDR data is imported from CLDR v33.

Update NEWS for sinf improvements
Wilco Dijkstra: The NEWS entry for sinf improvements is listed for 2.28, while it was committed in 2.29, so move it there and mention tanf.

Add test-in-container infrastructure
DJ Delorie: - Makefile (testroot.pristine): New rules to initialize the test-in-container "testroot". - Makerules (all-testsuite): Add tests-container. - Rules (tests-expected): Add tests-container.

Add NT_VMCOREDD, AT_MINSIGSTKSZ from Linux 4.18 to elf.h
Joseph Myers: This patch adds two new constants from Linux 4.18 to elf.h, NT_VMCOREDD and AT_MINSIGSTKSZ. Tested for x86_64. - elf/elf.c (NT_VMCOREDD): New macro.

glibc Release Tags

Open master branch for glibc 2.29 development
Carlos O'Donell

Update for 2.25 release
Siddhesh Poyarekar

Update version.h and include/features.h for 2.22 release
Carlos O'Donell

Restore _POSIX2_C_VERSION definition (bug 438).
Joseph Myers: My review of conformtest expectations for POSIX showed up that the _POSIX2_C_VERSION macro, required by POSIX and XPG standards before 2001, was missing in unistd.h, having been removed on 2003-04-03 despite those standards still being supported.

Bump required version of texinfo to 4.7
Will Newton: It seems we require texinfo 4.7 for the --plaintext option, so document that and check for the correct version in configure. ChangeLog: 2014-12-15 Will Newton - manual/install.texi: Bump required version of texinfo to 4.7 from 4.5. - INSTALL: Regenerated. - configure.ac: Check for makeinfo version 4.7 and above. - configure: Regenerated.

Rework compiler version check in configure
Roland McGrath

Update version.h and include/features.h for 2.19 release
Allan McRae

Accept make versions 4.0 and greater
Marc-Antoine Perennou

Update version.h and include/features.h for 2.18 release
David S. Miller

Make _LIB_VERSION a weak symbol
Siddhesh Poyarekar: That way it can live alongside _LIB_VERSION in libieee.a for statically compiled programs.

Most Popular In The Past Month

Add test-in-container infrastructure
DJ Delorie: - Makefile (testroot.pristine): New rules to initialize the test-in-container "testroot". - Makerules (all-testsuite): Add tests-container. - Rules (tests-expected): Add tests-container.

Update NEWS for sinf improvements
Wilco Dijkstra: The NEWS entry for sinf improvements is listed for 2.28, while it was committed in 2.29, so move it there and mention tanf.

Merge mktime, timegm from upstream Gnulib
Paul Eggert: [BZ #23603][BZ #16346] This fixes some obscure problems with integer overflow.

Add new log2 implementation
Szabolcs Nagy: Similar algorithm is used as in log: log2(2^k x) = k + log2(c) + log2(x/c) where the last term is approximated by a polynomial of x/c - 1, the first order coefficient is about 1/ln2 in this case. There is separate code path when fma instruction is not available for computing x/c - 1 precisely, for which the table size is doubled. The worst case error is 0.547 ULP (0.55 without fma), the read only global data size is 1168 bytes (2192 without fma) on aarch64.

Add new pow implementation
Szabolcs Nagy: The algorithm is exp(y * log(x)), where log(x) is computed with about 1.3*2^-68 relative error (1.5*2^-68 without fma), returning the result in two doubles, and the exp part uses the same algorithm (and lookup tables) as exp, but takes the input as two doubles and a sign (to handle negative bases with odd integer exponent).

Indian and similar locales: Set the correct date format (bug 17426).
Rafal Luzynski: This commit also fixes d_fmt in bn_BD which is identical to bn_IN, in ne_NP which is identical to ne_IN (not supported by Glibc but supported by CLDR), and in ta_LK which is identical to ta_IN. For those locales which are supported by CLDR data is imported from CLDR v33.

Add build-many-glibcs.py support for building more GCC libraries
Joseph Myers: Every so often we get libsanitizer or libgo builds breaking with new glibc because of some change in the glibc headers. glibc's build-many-glibcs.py deliberately disables libsanitizer and GCC languages other than C and C++ because the point is to test glibc and find glibc problems (including problems shown up by new compiler warnings in new GCC), not to test libsanitizer or libgo; if the compiler build fails because of libsanitizer or libgo failing to build, that could hide the existence of new problems in glibc.

Fix MIPS n32 pr_sigpend, pr_sighold, pr_flag type (bug 23656).
Joseph Myers: As discussed at <

Add build-many-glibcs.py --enable-obsolete-* configs
Joseph Myers: We've had issues before with build failures (with new GCC) in code only built with --enable-obsolete-rpc or --enable-obsolete-nsl not being reported for a while because build-many-glibcs.py does not test those configure options.

Legal Disclaimer & Disclosure | Contact Us
Copyright © 2010 - 2018 by Phoronix Media. All rights reserved.
All trademarks and copyrights used are properties of their respective owners.
Powered by the Phoronix Content Management System (PHXCMS-8).