platform/upstream/orc.git
4 years agoexecutor: x2 union members not endian safe
Doug Nazar [Wed, 28 Aug 2019 13:55:26 +0000 (09:55 -0400)]
executor: x2 union members not endian safe

4 years agopowerpc: Pre-load long constants & invariants
Doug Nazar [Sun, 25 Aug 2019 19:50:15 +0000 (15:50 -0400)]
powerpc: Pre-load long constants & invariants

PowerPC doesn't have an easy way to load a long constant into a
vector register so we emit the value into the code and then load it.

This moves those values to after the epilogue and then loads them
before starting the outer loop.

4 years agoorc: Consolidate orc_program_has_float()
Doug Nazar [Sun, 25 Aug 2019 19:41:59 +0000 (15:41 -0400)]
orc: Consolidate orc_program_has_float()

4 years agoorc: Make orc_*_emit_invariants() generic
Doug Nazar [Sun, 25 Aug 2019 18:50:11 +0000 (14:50 -0400)]
orc: Make orc_*_emit_invariants() generic

Move mmx/sse emit invariant function to orcompiler so it can
be used on other architectures.

4 years agoorccodemem: Mark exec mapping writable if debugging
Doug Nazar [Sun, 25 Aug 2019 18:27:23 +0000 (14:27 -0400)]
orccodemem: Mark exec mapping writable if debugging

gdb needs the mapping to be writable to set breakpoints.

4 years agoorctest: Fix ULP compare and make compatible with clang
Doug Nazar [Sat, 31 Aug 2019 15:08:58 +0000 (11:08 -0400)]
orctest: Fix ULP compare and make compatible with clang

4 years agobuild: fix werror build with clang
Matthew Waters [Tue, 27 Aug 2019 01:36:49 +0000 (11:36 +1000)]
build: fix werror build with clang

../subprojects/orc/orc-test/orcarray.c:230:47: error: taking the absolute value of unsigned type 'unsigned int' has no effect [-Werror,-Wabsolute-value]
          if ((a[i] < 0.0) == (b[i] < 0.0) && abs(*(orc_uint32 *)&a[i] - *(orc_uint32 *)&b[i]) <= 2) continue;
                                              ^
../subprojects/orc/orc-test/orcarray.c:230:47: note: remove the call to 'abs' since unsigned values cannot be negative
          if ((a[i] < 0.0) == (b[i] < 0.0) && abs(*(orc_uint32 *)&a[i] - *(orc_uint32 *)&b[i]) <= 2) continue;
                                              ^~~
../subprojects/orc/orc-test/orcarray.c:247:47: error: taking the absolute value of unsigned type 'unsigned long' has no effect [-Werror,-Wabsolute-value]
          if ((a[i] < 0.0) == (b[i] < 0.0) && abs(*(orc_uint64 *)&a[i] - *(orc_uint64 *)&b[i]) <= 2) continue;
                                              ^
../subprojects/orc/orc-test/orcarray.c:247:47: note: remove the call to 'abs' since unsigned values cannot be negative
          if ((a[i] < 0.0) == (b[i] < 0.0) && abs(*(orc_uint64 *)&a[i] - *(orc_uint64 *)&b[i]) <= 2) continue;
                                              ^~~
../subprojects/orc/orc-test/orctest.c:525:63: error: taking the absolute value of unsigned type 'unsigned int' has no effect [-Werror,-Wabsolute-value]
      if ((*(float *)ptr1 < 0.0) == (*(float *)ptr2 < 0.0) && abs(*(orc_uint32 *)ptr1 - *(orc_uint32 *)ptr2) <= 2) return TRUE;
                                                              ^
../subprojects/orc/orc-test/orctest.c:525:63: note: remove the call to 'abs' since unsigned values cannot be negative
      if ((*(float *)ptr1 < 0.0) == (*(float *)ptr2 < 0.0) && abs(*(orc_uint32 *)ptr1 - *(orc_uint32 *)ptr2) <= 2) return TRUE;
                                                              ^~~
../subprojects/orc/orc-test/orctest.c:530:65: error: taking the absolute value of unsigned type 'unsigned long' has no effect [-Werror,-Wabsolute-value]
      if ((*(double *)ptr1 < 0.0) == (*(double *)ptr2 < 0.0) && abs(*(orc_uint64 *)ptr1 - *(orc_uint64 *)ptr2) <= 2) return TRUE;
                                                                ^
../subprojects/orc/orc-test/orctest.c:530:65: note: remove the call to 'abs' since unsigned values cannot be negative
      if ((*(double *)ptr1 < 0.0) == (*(double *)ptr2 < 0.0) && abs(*(orc_uint64 *)ptr1 - *(orc_uint64 *)ptr2) <= 2) return TRUE;
                                                                ^~~

4 years agopowerpc: Add little endian support
Doug Nazar [Tue, 20 Aug 2019 08:50:56 +0000 (04:50 -0400)]
powerpc: Add little endian support

4 years agopowerpc: Fix whitespace
Doug Nazar [Tue, 20 Aug 2019 08:38:39 +0000 (04:38 -0400)]
powerpc: Fix whitespace

4 years agoorc: Add ORC_LIKELY/ORC_UNLIKELY macros
Doug Nazar [Sun, 18 Aug 2019 15:50:44 +0000 (11:50 -0400)]
orc: Add ORC_LIKELY/ORC_UNLIKELY macros

4 years agomeson: Fix ppc/ppc64 cpu family detection
Doug Nazar [Sun, 18 Aug 2019 15:48:10 +0000 (11:48 -0400)]
meson: Fix ppc/ppc64 cpu family detection

4 years agoorctest: print_array_val_float return value not used
Doug Nazar [Sun, 18 Aug 2019 15:47:06 +0000 (11:47 -0400)]
orctest: print_array_val_float return value not used

4 years agoadd Hygon Dhyana processor support
fanjinke [Tue, 7 May 2019 15:48:37 +0000 (23:48 +0800)]
add Hygon Dhyana processor support

Signed-off-by: fanjinke <fanjinke@hygon.cn>
4 years agotests: Use float constants and parameters when testing float opcodes.
Doug Nazar [Tue, 13 Aug 2019 19:38:55 +0000 (15:38 -0400)]
tests: Use float constants and parameters when testing float opcodes.

4 years agoorcarm: Fix flushing of icache when using dual map
Doug Nazar [Wed, 7 Aug 2019 01:44:02 +0000 (21:44 -0400)]
orcarm: Fix flushing of icache when using dual map

On a codemem_dual_map setup, we need to also flush the PROT_EXEC alias.

4 years agoorctest: Switch nearby float comparison to ULP method
Doug Nazar [Wed, 7 Aug 2019 04:11:35 +0000 (00:11 -0400)]
orctest: Switch nearby float comparison to ULP method

4 years agodoc: fix gtkdoc when orc is used as a subproject
Mathieu Duponchelle [Fri, 2 Aug 2019 16:41:52 +0000 (18:41 +0200)]
doc: fix gtkdoc when orc is used as a subproject

meson.source_root() will point to the root of the top level
project, and should not be used to compute paths in the current
project.

5 years agox86: add endbr32 and endbr64 instructions
Wim Taymans [Fri, 19 Jul 2019 14:18:04 +0000 (16:18 +0200)]
x86: add endbr32 and endbr64 instructions

Issue endbr instructions at the start of generated functions to
make IBT (indirect branch tracking) work.

Resolves #17

5 years agoci: run tests on windows
Tim-Philipp Müller [Tue, 23 Jul 2019 23:26:32 +0000 (00:26 +0100)]
ci: run tests on windows

5 years agoAdd Windows CI for msvc 2017 and msys2
Jordan Petridis [Thu, 11 Jul 2019 04:48:08 +0000 (07:48 +0300)]
Add Windows CI for msvc 2017 and msys2

Currently the container is based upon a windows 1607 image
and requires a runner with a server/hyper-v 1607 host
to be able to execute.

There are plan on moving the runner to a 1807 later.

5 years agouwp: workaround a false positive in certification of orc
Nirbheek Chauhan [Tue, 7 May 2019 14:48:34 +0000 (20:18 +0530)]
uwp: workaround a false positive in certification of orc

It seems that the Windows App Certification Kit searches all files and
binaries for the regex '\<reg\>' (or something like it) and throws
errors if it exists. Supposedly this is for preventing apps from
running REG.EXE

https://blogs.msdn.microsoft.com/appconsult/2017/08/16/how-to-validate-if-your-application-is-compliant-with-the-windows-store-polices-windows-10-and-windows-10-s/

5 years agoMeson: Do not always generate static library
Xavier Claessens [Fri, 12 Apr 2019 15:34:33 +0000 (11:34 -0400)]
Meson: Do not always generate static library

Since Meson 0.46.0 'default_library' can be set to 'both' when user
wants both shared and static libraries.

5 years agoBack to development
Tim-Philipp Müller [Mon, 15 Apr 2019 08:27:24 +0000 (09:27 +0100)]
Back to development

5 years agoRelease 0.4.29
Tim-Philipp Müller [Sun, 14 Apr 2019 22:47:03 +0000 (23:47 +0100)]
Release 0.4.29

5 years agoorc: fix C++ style comment
Tim-Philipp Müller [Fri, 22 Mar 2019 19:36:24 +0000 (19:36 +0000)]
orc: fix C++ style comment

5 years agoconfigure.ac: no more bugzilla
Tim-Philipp Müller [Tue, 5 Mar 2019 14:21:04 +0000 (14:21 +0000)]
configure.ac: no more bugzilla

5 years agoUpdate news file for upcoming release
Tim-Philipp Müller [Tue, 5 Mar 2019 13:52:54 +0000 (13:52 +0000)]
Update news file for upcoming release

5 years agoAdd pkg-config file for orc-test library
Tim-Philipp Müller [Tue, 5 Mar 2019 11:05:43 +0000 (11:05 +0000)]
Add pkg-config file for orc-test library

Fixes #16

5 years agoFix warnings with Cerbero's mingw (gcc 4.7)
Nirbheek Chauhan [Tue, 5 Feb 2019 13:29:11 +0000 (18:59 +0530)]
Fix warnings with Cerbero's mingw (gcc 4.7)

orcutils.h:59:38: error: "__STDC_VERSION__" is not defined [-Werror=undef]

5 years agomeson: Make orcc detectable from other subprojects
Seungha Yang [Mon, 28 Jan 2019 13:12:46 +0000 (22:12 +0900)]
meson: Make orcc detectable from other subprojects

If we are a subproject (e.g., gst-build), this can make it possible
to use orcc by other subprojects.

5 years agoPowerPC: Support ELFv2 ABI
A. Wilcox [Tue, 10 Jul 2018 02:18:07 +0000 (21:18 -0500)]
PowerPC: Support ELFv2 ABI

The ELFv2 ABI does not use function descriptors.  This change has been
tested on Power Mac G5, IBM POWER6, and Raptor Talos II, all using the
musl libc on Linux kernel 4.14.

5 years agoAdd support for ppc64le
Michel Normand [Mon, 2 Mar 2015 11:14:03 +0000 (12:14 +0100)]
Add support for ppc64le

Fixes https://gitlab.freedesktop.org/gstreamer/orc/issues/6

5 years agoci: add minimal ci
Tim-Philipp Müller [Wed, 2 Jan 2019 00:47:42 +0000 (00:47 +0000)]
ci: add minimal ci

5 years agoorc: Remove dead assignments
Edward Hervey [Mon, 17 Dec 2018 10:26:21 +0000 (11:26 +0100)]
orc: Remove dead assignments

All those temporary registry were not used and instead were being
assigned constant values.

5 years agomeson: Maintain macOS ABI compatibility
Nirbheek Chauhan [Wed, 14 Nov 2018 09:49:42 +0000 (15:19 +0530)]
meson: Maintain macOS ABI compatibility

Need to use the correct dylib versioning.

5 years agomeson: add option to disable orc-test library
Tim-Philipp Müller [Sat, 6 Oct 2018 12:04:01 +0000 (13:04 +0100)]
meson: add option to disable orc-test library

https://bugzilla.gnome.org/show_bug.cgi?id=736785

5 years agomeson: rename "enable-backend" option to "orc-backend"
Tim-Philipp Müller [Wed, 3 Oct 2018 09:31:47 +0000 (10:31 +0100)]
meson: rename "enable-backend" option to "orc-backend"

More consistent with the Meson option naming conventions.

5 years agomeson: add options to disable benchmarks, examples, tools
Tim-Philipp Müller [Wed, 3 Oct 2018 08:28:53 +0000 (09:28 +0100)]
meson: add options to disable benchmarks, examples, tools

Like we have in other modules.

5 years agomeson: make "tests" option an auto feature option
Tim-Philipp Müller [Wed, 3 Oct 2018 08:13:31 +0000 (09:13 +0100)]
meson: make "tests" option an auto feature option

Consistent with the rest of GStreamer and gst-build.

5 years agomeson: rename 'disable_gtkdoc' option to 'gtk_doc' and make it an auto/feature option
Tim-Philipp Müller [Mon, 30 Jul 2018 09:06:04 +0000 (10:06 +0100)]
meson: rename 'disable_gtkdoc' option to 'gtk_doc' and make it an auto/feature option

5 years agomeson: rename option disable_tests=false -> tests=true
Tim-Philipp Müller [Mon, 30 Jul 2018 08:56:14 +0000 (09:56 +0100)]
meson: rename option disable_tests=false -> tests=true

Bring it in line with convention.

6 years agomeson: Bump required version to 0.46
Nicolas Dufresne [Sun, 22 Jul 2018 13:53:39 +0000 (09:53 -0400)]
meson: Bump required version to 0.46

6 years agomeson: pass the right header when testing for posix_memalign
Christoph Reiter [Mon, 21 May 2018 15:06:12 +0000 (17:06 +0200)]
meson: pass the right header when testing for posix_memalign

has_function() needs to know the header to correctly detect the function in
all cases.

This fixes the build under mingw-w64 where posix_memalign was wrongfully being
detected as available.

See this glib bug for a similar fix:
    https://bugzilla.gnome.org/show_bug.cgi?id=795876

https://bugzilla.gnome.org/show_bug.cgi?id=796309

6 years agoFix MSVC compiler warning
Tim-Philipp Müller [Fri, 18 May 2018 08:49:01 +0000 (09:49 +0100)]
Fix MSVC compiler warning

The isnan() macro was redefined to _isnan() on MSVC, but it's
been in math.h since vs2013, so it should be safe to rely on
it unconditionally nowadays, and we have no intention of
supporting ancient vs versions anyway.

6 years agoorc: fix static build on windows
Tim-Philipp Müller [Fri, 18 May 2018 08:10:17 +0000 (09:10 +0100)]
orc: fix static build on windows

Orc users mustn't use __declspec(dllimport) when
linking against a static liborc.

6 years agoorctest: Allow enough room for sprintf()
Edward Hervey [Fri, 4 May 2018 07:21:16 +0000 (09:21 +0200)]
orctest: Allow enough room for sprintf()

We were writing more than 300 characters in some cases. Thanks gcc
8 for pointing that out :)

6 years agomeson: use add_project_arguments() instead of add_global_arguments()
Tim-Philipp Müller [Sat, 28 Apr 2018 22:48:39 +0000 (23:48 +0100)]
meson: use add_project_arguments() instead of add_global_arguments()

The latter won't play nice when being used as a subproject.

6 years agoorc: update some generated files
Tim-Philipp Müller [Sat, 28 Apr 2018 12:06:34 +0000 (13:06 +0100)]
orc: update some generated files

6 years agoorcfunctions.h: include orcutils.h for ORC_API define
Tim-Philipp Müller [Sat, 28 Apr 2018 12:04:11 +0000 (13:04 +0100)]
orcfunctions.h: include orcutils.h for ORC_API define

Fixes build for users that just include orcfunctions.h
directly instead of orc/orc.h, like libs in gst-plugins-base.

6 years agoorc: remove empty and unused orcrules-x86.c
Tim-Philipp Müller [Sat, 28 Apr 2018 00:10:59 +0000 (01:10 +0100)]
orc: remove empty and unused orcrules-x86.c

6 years agoorc: move unusable orc_code_chunk_free() declaration to internal header
Tim-Philipp Müller [Fri, 27 Apr 2018 23:37:35 +0000 (00:37 +0100)]
orc: move unusable orc_code_chunk_free() declaration to internal header

Nothing in any public header ever returned or used an OrcCodeChunk.

6 years agoautotools: control symbol visibility via -fvisibility
Tim-Philipp Müller [Fri, 27 Apr 2018 17:42:29 +0000 (18:42 +0100)]
autotools: control symbol visibility via -fvisibility

All symbols are hidden by default unless exported
explicitly via our public API decorators.

6 years agomeson: fix symbol export with MSVC and use -fvisibility elsewhere
Tim-Philipp Müller [Fri, 27 Apr 2018 16:45:23 +0000 (17:45 +0100)]
meson: fix symbol export with MSVC and use -fvisibility elsewhere

Use newly-added ORC_API decorators to export symbols explicitly.

Get rid of orc.map file and the ORC_EXPORT stuff that never did
anything anyway (define was unused).

6 years agoorc: decorate public symbols with ORC_API
Tim-Philipp Müller [Fri, 27 Apr 2018 15:56:11 +0000 (16:56 +0100)]
orc: decorate public symbols with ORC_API

We'll use that later to control symbol visibility and
exported symbols.

6 years agoorc-test: decorate public symbols with ORC_TEST_API
Tim-Philipp Müller [Fri, 27 Apr 2018 13:00:54 +0000 (14:00 +0100)]
orc-test: decorate public symbols with ORC_TEST_API

We'll use that later to control symbol visibility and
exported symbols.

6 years agoorcc: add --decorator command line argument
Tim-Philipp Müller [Fri, 27 Apr 2018 16:02:18 +0000 (17:02 +0100)]
orcc: add --decorator command line argument

So we can add things like FOO_API if the functions are
supposed to be public, for example.

6 years agomeson: use -Bsymbolic-functions when linking if available
Tim-Philipp Müller [Fri, 27 Apr 2018 11:47:42 +0000 (12:47 +0100)]
meson: use -Bsymbolic-functions when linking if available

cc.has_linker_argument() requires Meson 0.46, so guard
against older Meson versions for now.

6 years agoorc: make more functions static
Tim-Philipp Müller [Fri, 27 Apr 2018 23:06:03 +0000 (00:06 +0100)]
orc: make more functions static

All of these are only used internally and are
not in any header files, but were still exported
by accident because they start with orc_*.

6 years agogtk-doc: get rid of templates
Tim-Philipp Müller [Fri, 20 Apr 2018 11:34:17 +0000 (12:34 +0100)]
gtk-doc: get rid of templates

gtkdoc-mktmpl was removed in recent gtk-doc versions.

Update gtk-doc.make to 1.27, but modify to generate
xml/version.entities (i.e. add copy in right dir).

All a bit cargo-culted, but seems to work.

6 years agoorcmips: only enable if the DSPr2 ASE is present
James Cowgill [Fri, 6 Apr 2018 14:26:43 +0000 (15:26 +0100)]
orcmips: only enable if the DSPr2 ASE is present

The mips target uses various instructions that a part of the MIPS DSPr2
ASE which is not present on all processors. The easiest solution is to
gate the target on the presense of DSPr2 at runtime.

Unfortunately there isn't a "nice" way to detect if DSPr2 is present, so
resort to parsing /proc/cpuinfo to get the information.

https://bugzilla.gnome.org/show_bug.cgi?id=760834

6 years agoorcutils: refactor /proc/cpuinfo reading code from orccpu-arm.c
James Cowgill [Fri, 6 Apr 2018 14:24:47 +0000 (15:24 +0100)]
orcutils: refactor /proc/cpuinfo reading code from orccpu-arm.c

This code will later be used by mips target, so it makes sense
to use a common function instead of duplicating the code.

https://bugzilla.gnome.org/show_bug.cgi?id=760834

6 years agoconfigure: don't check for functions that aren't used anywhere
Tim-Philipp Müller [Mon, 9 Apr 2018 18:14:05 +0000 (19:14 +0100)]
configure: don't check for functions that aren't used anywhere

And remove some unneeded includes.

6 years agomeson: fix mips cpu_family test
James Cowgill [Fri, 6 Apr 2018 14:38:54 +0000 (15:38 +0100)]
meson: fix mips cpu_family test

The cpu family "mipsel" does not exist in meson so the current test to
enable the HAVE_MIPSEL macro is always false. The real cpu family is
"mips" but this also matches big endian platforms so also check that the
endian is little.

https://bugzilla.gnome.org/attachment.cgi?id=370595

6 years agoorccpu-arm: Allow 'cpuinfo' fallback on non-android
Edward Hervey [Wed, 29 Nov 2017 08:19:20 +0000 (09:19 +0100)]
orccpu-arm: Allow 'cpuinfo' fallback on non-android

Reading the cpu flags from cpuinfo is also useful on non-android
ARM platforms which might have mechanisms that prevent /proc/*/auxv
from being read.

6 years agomeson: Always build both shared and static libraries
Nirbheek Chauhan [Tue, 21 Nov 2017 20:12:23 +0000 (01:42 +0530)]
meson: Always build both shared and static libraries

This reuses objects from the shared library for the static library, so
it should not affect the build time at all.

orc_dep and orc_test_dep will still contain the library selected by
--default-library

6 years agoBuild liborc-test also as a properly versioned library with meson
Sebastian Dröge [Mon, 20 Nov 2017 17:16:55 +0000 (19:16 +0200)]
Build liborc-test also as a properly versioned library with meson

6 years agoAdd meson configuration for disabling tests and gtk-doc docs
Sebastian Dröge [Mon, 20 Nov 2017 16:41:01 +0000 (18:41 +0200)]
Add meson configuration for disabling tests and gtk-doc docs

https://bugzilla.gnome.org/show_bug.cgi?id=790626

6 years agoAllow building shared/static library based on configuration
Sebastian Dröge [Mon, 20 Nov 2017 16:36:20 +0000 (18:36 +0200)]
Allow building shared/static library based on configuration

Before only a shared library was possible

6 years agoBack to development
Edward Hervey [Mon, 20 Nov 2017 15:47:00 +0000 (16:47 +0100)]
Back to development

6 years agoRelease 0.4.28 sandbox/mzx/upstream-git
Edward Hervey [Mon, 20 Nov 2017 15:15:58 +0000 (16:15 +0100)]
Release 0.4.28

6 years agoorcutils: Use compiler offsetof() when available
Edward Hervey [Mon, 20 Nov 2017 14:08:57 +0000 (15:08 +0100)]
orcutils: Use compiler offsetof() when available

The way we were calculating the offset of a structure entry was
undefined behaviour.

Instead, use 'offsetof()' (defined in C89) when available

6 years agoorccpu-x86: Fix undefined behaviour compilation issue
Edward Hervey [Sun, 19 Nov 2017 17:21:34 +0000 (18:21 +0100)]
orccpu-x86: Fix undefined behaviour compilation issue

runtime error: left shift of 1 by 31 places cannot be represented in type 'int'

6 years agoconfigure.ac: Print configuration after configuration
Edward Hervey [Sun, 19 Nov 2017 17:19:22 +0000 (18:19 +0100)]
configure.ac: Print configuration after configuration

Makes it easier to spot/review detected flags

6 years agoorc: Add a configure option to disable tests
Edward Hervey [Sun, 19 Nov 2017 17:18:19 +0000 (18:18 +0100)]
orc: Add a configure option to disable tests

For targets where it doesn't make sense.

Note that this will disable the build of orc-bugreport since it
depends on the the orc_test helper library

6 years agotesuite: Fix some leaks
Edward Hervey [Sun, 19 Nov 2017 17:15:31 +0000 (18:15 +0100)]
tesuite: Fix some leaks

6 years agoorcprogram-c.c: Don't cast constant
Edward Hervey [Sat, 21 Oct 2017 07:26:39 +0000 (09:26 +0200)]
orcprogram-c.c: Don't cast constant

Clang compilers would complain with constant-conversion error (trying
to fit a 'int' into another type)

This was introduced to fix a warning in gcc ... which doesn't happen

6 years agoorc: Cast maximum constants
Edward Hervey [Sat, 21 Oct 2017 07:25:39 +0000 (09:25 +0200)]
orc: Cast maximum constants

Avoids constant-conversion warnings as such:

/src/orc/orc/orcemulateopcodes.c:158:13: error: implicit conversion from 'int' to 'orc_int8' (aka 'signed char') changes value from 255 to -1 [-Werror,-Wconstant-conversion]
    var34 = ORC_CLAMP_UB((orc_uint8)var32 + (orc_uint8)var33);
          ~ ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/src/orc/orc/orcemulateopcodes.c:29:48: note: expanded from macro 'ORC_CLAMP_UB'
                                               ^~~~~~~~~~
/src/orc/orc/orcemulateopcodes.c:18:20: note: expanded from macro 'ORC_UB_MAX'
                   ^~~
/src/orc/orc/orcemulateopcodes.c:12:55: note: expanded from macro 'ORC_CLAMP'
                                                      ^

7 years agoDist meson files with autotools make dist
Tim-Philipp Müller [Tue, 18 Jul 2017 11:33:15 +0000 (12:33 +0100)]
Dist meson files with autotools make dist

7 years agoBack to development
Tim-Philipp Müller [Mon, 17 Jul 2017 15:42:02 +0000 (16:42 +0100)]
Back to development

7 years agoRelease 0.4.27
Tim-Philipp Müller [Mon, 17 Jul 2017 15:16:35 +0000 (16:16 +0100)]
Release 0.4.27

7 years agoorcc: make coverity happy
Tim-Philipp Müller [Tue, 27 Jun 2017 08:43:07 +0000 (09:43 +0100)]
orcc: make coverity happy

Use typedefed enum so static analysis tools like coverity can know
we covered all possibilities in switch/case statements. CID 1413336.

7 years agoorc: preserve non volatile sse registers
Matej Knopp [Mon, 19 Jun 2017 02:00:31 +0000 (04:00 +0200)]
orc: preserve non volatile sse registers

required for MSVC
https://msdn.microsoft.com/en-us/library/9z1stfyw.aspx

orc already marks these register in compiler->save_regs,
but the actual saving and restoring was not done

https://bugzilla.gnome.org/show_bug.cgi?id=783931

7 years agomeson: Only use --version-script if not using MSVC
Nirbheek Chauhan [Fri, 23 Sep 2016 11:01:11 +0000 (16:31 +0530)]
meson: Only use --version-script if not using MSVC

7 years agomeson: use find_library() on compiler object
Tim-Philipp Müller [Sun, 28 May 2017 18:13:04 +0000 (19:13 +0100)]
meson: use find_library() on compiler object

The old stand-alone find_library() was deprecated.

7 years agomeson: add meson build
Tim-Philipp Müller [Thu, 15 Oct 2015 19:29:15 +0000 (20:29 +0100)]
meson: add meson build

Includes contributions from Nirbheek Chauhan <nirbheek@centricular.com>

7 years agoorcx86: Don't hard-code register size to zero in orc_x86_emit_*() functions
Igor Rondarev [Tue, 28 Feb 2017 09:23:55 +0000 (12:23 +0300)]
orcx86: Don't hard-code register size to zero in orc_x86_emit_*() functions

Instead use the size passed as argument. Fixes segmentation fault on QNX.

https://bugzilla.gnome.org/show_bug.cgi?id=779319

7 years agoorcx86: use HAVE_AMD64 and HAVE_I386 defines everywhere
Jan Schmidt [Fri, 28 Oct 2016 12:37:51 +0000 (23:37 +1100)]
orcx86: use HAVE_AMD64 and HAVE_I386 defines everywhere

Some places were checking for __amd64__ and __i386__ explicitly,
but MSVC doesn't define __amd64__ in 64-bit mode, so use the
define from config.h instead. Fixes incorrect asm generation
on 64-bit Windows

7 years agoBack to development
Sebastian Dröge [Wed, 31 Aug 2016 10:13:29 +0000 (13:13 +0300)]
Back to development

7 years agoRelease 0.4.26
Sebastian Dröge [Wed, 31 Aug 2016 10:13:04 +0000 (13:13 +0300)]
Release 0.4.26

7 years agopkgconfig: add orc-test path to the -uninstalled .pc file
Josep Torra [Thu, 25 Aug 2016 13:45:57 +0000 (15:45 +0200)]
pkgconfig: add orc-test path to the -uninstalled .pc file

Expose orc-test library in the uninstalled pkgconfig file.

7 years agouse 64bit arithmetic to increment the stride
Wim Taymans [Wed, 24 Aug 2016 08:10:09 +0000 (10:10 +0200)]
use 64bit arithmetic to increment the stride

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=770295

7 years agoorcx86: Fix generation of ModR/M / SIB bytes for the EBP, R12, R13 registers
Sebastian Dröge [Wed, 24 Aug 2016 16:11:18 +0000 (19:11 +0300)]
orcx86: Fix generation of ModR/M / SIB bytes for the EBP, R12, R13 registers

See http://wiki.osdev.org/X86-64_Instruction_Encoding#ModR.2FM

ESP and R12 are to be handled exactly the same, that is the addressing is
done with SIB instead of directly.

EBP and R13 have a special case for Mod 00, where they're expecting RIP/EIP
plus 32 bit displacement. If we handle them like all other registers, we would
instead of addressing offset 0, address something relative to the instruction
pointer and make part of the following instruction the 32 bit displacement.

Instead of using Mod 00 for offset 0 like for all other registers, we use Mod
01 and a 8 bit displacement of 0.

On Windows this rather low-level bug caused problems at a rather
high-level in the ORC video conversion code, where a movdqu was eating the
following paddw instruction and tried to load data from a rather random memory
location.

For whatever reason this didn't cause any other big problems in other ORC
generated code, and it also only seemed to have caused problems on Windows.

https://bugzilla.gnome.org/show_bug.cgi?id=765713

8 years agotests: fix a warning spotted when building in osx
Josep Torra [Sun, 24 Jul 2016 16:09:50 +0000 (18:09 +0200)]
tests: fix a warning spotted when building in osx

orctest.c:730:19: error: logical not is only applied to the left hand side of this comparison
       if (!float_compare (dest_emul[l-ORC_VAR_D1], dest_exec[l-ORC_VAR_D1], i, j) != 0) {

It seems some sort of typo. Fixing by removing the potentially leftover condition.

8 years agoFix test_parse with no executable backend
Pascal Terjan [Wed, 4 May 2016 09:36:21 +0000 (09:36 +0000)]
Fix test_parse with no executable backend

https://bugzilla.gnome.org/show_bug.cgi?id=765953

8 years agoBack to development
Tim-Philipp Müller [Sat, 5 Mar 2016 00:32:19 +0000 (00:32 +0000)]
Back to development

8 years agoRelease 0.4.25
Tim-Philipp Müller [Sat, 5 Mar 2016 00:01:25 +0000 (00:01 +0000)]
Release 0.4.25

8 years agoconfigure: fix bug report url
Tim-Philipp Müller [Fri, 4 Mar 2016 23:58:09 +0000 (23:58 +0000)]
configure: fix bug report url

8 years agoorcc: check version of liborc to make sure the right lib is being used
Tim-Philipp Müller [Wed, 2 Mar 2016 19:04:35 +0000 (19:04 +0000)]
orcc: check version of liborc to make sure the right lib is being used

Interesting compilation failures might otherwise ensue.

8 years agoAdd orc_version_string()
Tim-Philipp Müller [Wed, 2 Mar 2016 19:04:24 +0000 (19:04 +0000)]
Add orc_version_string()

8 years agouninstalled.pc: add support for non libtool build systems
Julien Isorce [Thu, 18 Feb 2016 14:18:32 +0000 (14:18 +0000)]
uninstalled.pc: add support for non libtool build systems

Currently the .la path is provided which requires to use libtool as
mentioned in the GStreamer manual section-helloworld-compilerun.html.
It is fine as long as the application is built using libtool.

So currently it is not possible to compile a GStreamer application
within gst-uninstalled with CMake or other build system different
than autotools.

This patch allows to do the following in gst-uninstalled env:
gcc test.c -o test $(pkg-config --cflags --libs orc-uninstalled)
Previously it required to prepend libtool --mode=link

https://bugzilla.gnome.org/show_bug.cgi?id=720778