Renamed a few "h5tri_t extended" variables to "h5tri_t was_extended".
Apparently, "extended" is a typedef name someplace in the headers used
on Solaris, which causes potentially confusing warnings to be emitted.
Tested on: h5committest
Solaris w/ Solaris Studio (emu)
Description:
XL compilers in ostrich (PowerPC64 linux) fail in test/dt_arith
because of the removal of the LLONG_TO_LDOUBLE_CORRECT (removed in
r26625) and LDOUBLE_TO_LLONG_ACCURATE (removed in r26623).
Solution:
Reverse revisions
r26623: bring back LDOUBLE_TO_LLONG_ACCURATE configure macro
r26625: bring back LLONG_TO_LDOUBLE_CORRECT configure macro
r26627: bring back WANT_DATA_ACCURACY configure macro which is
used together with the above two macros. This also brings
back the enable-dconv-accuracy configure option.
Tested:
h5committested.
Also tested in ostrich using the XL compilers.
a different ID in the same type. Added a new skiplist routine,
H5SL_try_free_safe, which iterates over items, freeing some of them, and which
intercepts and defers attempts to remove from the list outside of the main
iteration. Changed H5I_clear_type to use this function.
Tested: jam, koala, ostrich (h5committest); ummon
- Updated the preprocessor logic that defines the H5_ATTR_* macros. The Solaris
Studio compiler understands __attribute__ but not the specific attributes
that we use. The H5_ATTR_* macros are now defined as blank with that
compiler which significantly cleans up the warnings, making debugging
easier.
- Replaced a few lingering __attribute__ macros with H5_ATTR_* markup.
Tested on: h5committest
Solaris w/ Solaris Studio (emu)
Correct error with braces around macro (in production builds), also clean
up a few warnings.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial, parallel & production
Linux/32 2.6 (jam) w/serial & parallel
Clean up parameters to H5C__flush_single_entry: accept cache entry instead
of address, and make new flag for deleting entry from skiplist on destroy
instead of a separate parameter to the routine.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
Linux/32 2.6.x (jam) w/serial & parallel
Split parallel metadata cache code into separate source code modules.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
(too minor for h5committest)
Bring 'fail if unknown always' object header message flag from the
avoid_truncate branch, strengthen error checking for unknown messages, and try
to catch object header code configurations at compile time (instead of runtime).
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
(h5committest forthcoming)
Large code cleanups and optimizations, in preparation for the upcoming
v3 metadata cache merge.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
Linux/32 2.6.x (jam) w/serial & parallel
Normalize trunk against the metadata_cache_merge branch, in preparation
for merging the branch into the trunk.
Tested on:
MacOSX/64 10.10.3 (amazon) w/serial & parallel
Linux/64 2.6.x (koala) w/serial
Linux/32 2.6.x (jam) w/serial & parallel
The H5F_ACC_DEBUG symbol remains but has been defined to zero and
has been listed as deprecated.
Fixes: HDFFV-1074
Tested on:
h5committest
32-bit Linux w/ C++ and Fortran and multi VFD
32-bit Linux w/ C++ and Fortran and multi VFD (no deprec symbols)
When the selection is set to all, H5Sextent_copy did not update the number of
elements in the selection in the destination space. Fixed H5Sextent_copy to do
this. Added tests for this functionality.
Tested: jam, koala, ostrich (h5committest)
Bring r26543 from autotools_rework branch to trunk:
Remove the LLONG_TO_LDOUBLE_CORRECT macro/define, it's working around
bugs in very old SGI/FreeBSD/Windows compilers.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(Daily tested on branch for 2+ days)
Bring r26513 from autotools_rework branch to trunk:
Remove the LDOUBLE_TO_LLONG_ACCURATE macro/define, it's working around
bugs in older SGI, HP/UX, MacOSX and Windows .NET 2003 compilers.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(Daily tested on branch for 2+ days)
Bring r26511 from autotools_rework branch to trunk:
Remove the FP_TO_INTEGER_OVERFLOW_WORKS macro/define, which is for working
around bugs in the Cray X1 compiler and is no longer supported.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(Daily tested on branch for 2+ days)
Bring r26508 from autotools_rework branch to trunk:
Remove HW_FP_TO_LLONG_NOT_WORKS macro/define, it was only addressing
Windows .NET 2003 compiler issues.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(Daily tested on branch for 2+ days)
Bring r26503 & r26528 from autotools_rework branch back to the trunk:
Remove old platform configure files: craynv, dec-flags, hpux11.23,
ia64-linux-gnu, nec-superux14.1, sv1-cray, x86_64-redstorm-linux-gnu
Also remove CONVERT_DENORMAL_FLOAT, since this was only set in the
configure files being removed.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(Daily tested on branch for 2+ days)
Bring r26502 from autotools_rework branch to trunk:
Remove the H5_SW_ULONG_TO_FP_BOTTOM_BIT_WORKS and
H5_FP_TO_ULLONG_BOTTOM_BIT_WORKS macros/defines, as they are no longer used by
supported platforms.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(Daily tested on branch for 2+ days)
Bring r26501 from the autotools_rework branch to the trunk:
Remove ULLONG_TO_LDOUBLE_PRECISION macro/define, as it's targeting bugs
in the FreeBSD and Cygwin compilers.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(Daily tested on branch for 2+ days)
Bring r26500 from autotools_rework branch to trunk:
Remove the LLONG_TO_FP_CAST_WORKS macro/define, as it targets problems with
the Visual Studio 6 compilers.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(Daily tested on branch for 2+ days)
Bring r26499 from autotools_rework branch to trunk:
Remove ULLONG_TO_FP_CAST_WORKS macro/define, as it only applies to older
platforms we aren't supporting any longer.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(Daily tested on branch for 2+ days)
Bring r26494 from autotools_rework branch back to trunk:
Remove the LDOUBLE_TO_UINT_ACCURATE macro/define, it was addressing
problems with older Intel compilers on Linux that are no longer supported.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(Daily tested on branch for several days)
Bring r26492 from autotools_rework branch back to trunk:
Remove the FP_TO_ULLONG_ACCURATE and FP_TO_ULLONG_RIGHT_MAXIMUM
macros/defines, which were added to address problems with older PGI
compilers and HP-UX systems and are no longer supported.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(daily tested on branch for >1 week)
Bring r26485 from the autotools_rework branch to the trunk:
Remove the ULONG_TO_FLOAT_ACCURATE macro/define, we no longer support the
Sandia system where it was necessary.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(daily tested for >1 week)
Bring r26483 from autotools_rework branch back to trunk:
Remove INTEGER_TO_LDOUBLE_ACCURATE macro/define - we no longer support
SGI systems.
Tested on:
Linux/32 2.6.18 (jam) w/serial & parallel
(daily tested on branch for >1 week)
H5Pset_istore_k() will validate the "ik" value to not exceed the max v1 btree entries (2 bytes)
The same check for H5Pset_sym_k() "ik" value.
h5committested.
Bring r26258 from the autotools_rework branch to the trunk - remove
the LDOUBLE_TO_INTEGER_WORKS macro/define, which was addressing issues with
SGI systems that are no longer supported.
Tested on:
Linux 2.6.x (jam) w/serial & parallel
Daily tested on branch for >2 weeks
Bring r26257 from autotools_rework branch back to the trunk - removes
the LDOUBLE_TO_INTEGER_ACCURATE macro/define, which was addressing problems
with SGI systems and is no longer supported.
Tested on:
Linux 2.6.x/32 (jam) w/serial & parallel
Daily tested on branch for >2 weeks.
