User cannot choose D.U.M.A. versions prior to 2.5.15 in config,
so remove useless patches.
Signed-off-by: Kirill K. Smirnov <kirill.k.smirnov@gmail.com>
Fixed compile error for DUMA 2.5.15 due to different function prototypes
for new operator in c++03 vs c++11
Signed-off-by: Yogesh Sharma <ysharm01@harris.com>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69959
...was observed while trying to build gcc-5.3.0 on latest (at the time
of this change) archlinux using gcc-6.1.1.
This patch fixes the issue.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
Currently, native GDB 7.11 fails to build with uClibc-ng due to
undefined reference to _obstack_free.
On IRC
[http://crosstool-ng.osuosl.org/download/ibot-logs/2016-02-28.html], it
has been suggested to disable obstack in uClibc configuration. I think
it is a workaround rather than a fix: if another library/app needs
obstack, this leaves no viable configuration. IMO, if uClibc seeks to
mimic the glibc API, it should also provide _obstack_free call (an
alias for which it already has, even though commented out).
Signed-off-by: Alexey Neyman <stilor@att.net>
There is invalid assembly in dmalloc for PowerPC. The issue is that
'stw' expects a memory operand, and =g constraint allows both registers
and memory. Newer GCC tends to choose register even at -O0, resulting in
invalid assembly. Instead, force a register constraint in 'mflr' and let
GCC decide if it wants to store it into memory at all.
Reported this upstream.
Signed-off-by: Alexey Neyman <stilor@att.net>
CFLAGS_FOR_TARGET and CXXFLAGS_FOR_TARGET are rewritten in gcc-4.8.x and
gcc-4.9.x, so libstdc++ does not get any flags passed to gcc configure.
Backport fixes for config/mt-gnu and config/mt-ospace that preserve these
flags. With these fixes libstdc++ gets built with flags specified in
CT_TARGET_CFLAGS.
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Despite the documentation and the comment in xtensa_move_literals, in
the presence of --text-section-literals and --auto-litpools literals are
moved from the separate literal sections into .init and .fini, because
the check in the xtensa_move_literals is incorrect.
This fixes build errors seen with projects that have .init/.fini and use
text-section-literals.
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Binutils change 1058c7532d0b "Use signed data type for R_XTENSA_DIFF*
relocation offsets." changed signedness of BFD_RELOC_XTENSA_DIFF*
relocations substituted for BFD_RELOC_*. This made it impossible to
encode arbitrary 8-, 16- and 32-bit values, which broke e.g. debug info
encoding by .loc directive. Revert this part and add test.
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
This updates patches for 2.26. Most notably, the removal of the xtensa
patches which seem to be upstream in binutils now.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
OSX SDK has a declaration for `wcsncasecmp` since 10.7, which conflicts with
the definition provided by the patch for binutils.
Signed-off-by: Kasumi Fukuda <kasumi@rollingapple.net>
Cygwin follows POSIX IEEE Std 1003.1, 2004 Edition spec
for the select() fuction as described at:
http://pubs.opengroup.org/onlinepubs/009695399/functions/pselect.html
.. so we must #include <sys/select.h> for the declaration.
Signed-off-by: Ray Donnelly <mingw.android@gmail.com>
Define PICFLAG, UNDEFINEDPREAMBLE and UNDEFINEDCODE to nothing since Windows
doesn't have -fPIC (GCC warns) and can't handle undefined variable references
(which are not needed anyway).
Signed-off-by: Ray Donnelly <mingw.android@gmail.com>
The buildroot patches have xtensa for the supported versions of binutils
we use, and a few others that might help us keep away from host
polution.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
Now that versions of gcc that required PPL are no longer supported
( >= gcc-4.5.x AND <= gcc-4.7.x )
...we no longer require PPL or CLooG/PPL.
This commit:
* Removes PPL
* Removes CLooG/PPL
* Updates the documentation
* Updates build script for CLooG and GCC
* Removes PPL and CLooG/PPL from scripts/addToolVersion.sh and
scripts/showSamples.sh
* Adds ISL to scripts/addToolVersion.sh and scripts/showSamples.sh
I know that sounds like a lot for one commit, but it was all kind of
inter-tangled.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
1. Need to include <string> not <cstring> in gold's binary.cc
2. wcsncasecmp isn't always defined in Mac SDK
3. PTHREAD_ONCE_INIT is {0x30B1BCBA, {0}}, so can't be trivially
assigned in an intializer list.
Signed-off-by: Ray Donnelly <mingw.android@gmail.com>
This commit removes ncurses-5.9 and adds 6.0.
I also provide the stable patch updates in patches/ncurses/6.0.
I have also added an experimental toggle for enabling the new ABI
support.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
I was going to start doing some autoconf work, and noticed that
configure.in was executable. Then I noticed Makefile.in was executable.
o.O
So, I ran ```find . -type f -executable``` and found a bunch of files
that shouldn't be set executable.
This commit makes them normal files again.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
This commit sync gcc patches with buildroot.
I found this useful for fixing a few uClibc related issues.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
.. they're needed for the RPC generation in glibc
on both Cygwin and MinGW-w64.
Neither are built on GNU/Linux and iconv is not
built on Darwin.
Two patches for gettext are needed, one so that
-O0 works and one so that static builds can be
made.
They can take a good while to build, so if not
needed for_host or for_build then they are not
built.
Signed-off-by: Ray Donnelly <mingw.android@gmail.com>
AM_PROG_LEX sets this for some weird reason; it should
look for a program only and not a library. Then later
it gets linked to ar, ranlib, dlltool, windres, windmc
and itbl-test despite nothing in the code #include'ing
FlexLexer.h
This isn't a big deal but it did cause a build failure
on Cygwin as it triggered a bug with their flex package
dependencies which I reported at:
https://www.cygwin.com/ml/cygwin/2015-10/msg00433.html
Arguably I should remove all traces of LIBLEX in each
Makefile.am instead?
Signed-off-by: Ray Donnelly <mingw.android@gmail.com>
pthread_mutextattr_settype -> pthread_mutexattr_settype
.. I'm not sure why this didn't fail everywhere, unless
no one has tried to build gold?
Signed-off-by: Ray Donnelly <mingw.android@gmail.com>
