[ANNOUNCE] pixman release 0.43.4 now available

on clang/arm32 Makoto Kato (1): pixman-arm: Fix build on clang/arm32 Matt Turner (5): pixman-x86: Use cpuid.h header pixman-x86: Move #include "cpuid.h" inside conditionals Revert "Allow to build

Re: [Pixman] [ANNOUNCE] pixman release 0.42.2 now available

On Wed, Nov 2, 2022 at 1:37 PM Matt Turner wrote: > > A new pixman release 0.42.2 is now available. This is a stable release > in the 0.42 series. > > This version contains a fix for a heap overflow. A CVE has been > requested, and I'll reply to this email with the number when

[Pixman] [ANNOUNCE] pixman release 0.42.2 now available

: 3476e2676e66756b1af61b1e532cd80c985c191fb7956eb01702b419726cce99e79163b7f287f74f66414680e7396d13c3fee525cd663f12b6ac4877070ff4e8 pixman-0.42.2.tar.xz GPG signature: https://cairographics.org/releases/pixman-0.42.2.tar.gz.sha512.asc (signed by [ultimate] Matt Turner [ultimate] Matt Turner

Re: [Pixman] Performance regression with pixman 0.40

Cc'ing the patch author, since I don't think he's subscribed. On Fri, Jun 4, 2021 at 12:15 AM wrote: > > Hi, > > We are developping a graphics framework called EGT dedicated to Microchip > parts: > https://github.com/linux4sam/egt > > We are using Cairo, and so Pixman, for the drawing part.

Re: [Pixman] [PATCH] Prevent empty top-level declaration

On Sun, Nov 17, 2019 at 4:48 PM Michael Forney wrote: > > The expansion of PIXMAN_DEFINE_THREAD_LOCAL(...) may end in a > function definition, so the following semicolon is considered an > empty top-level declaration, which is not allowed in ISO C. > --- > pixman/pixman-compiler.h | 6

[Pixman] [ANNOUNCE] pixman release 0.40.0 now available

://cairographics.org/releases/pixman-0.40.0.tar.gz.sha512.asc (signed by [ultimate] Matt Turner [ultimate] Matt Turner [ultimate] Matt Turner [ultimate] Matt Turner ) Git: https://gitlab.freedesktop.org/pixman/pixman.git tag: pixman-0.40.0 Log: Adam Jackson (17

Re: [Pixman] [PATCH 1/2] configure.ac: Use '-mloongson-mmi' for Loongson MMI.

On Thu, Mar 26, 2020 at 5:57 AM Shiyou Yin wrote: > > It's recommended to use '-mloongson-mmi' for MMI. > --- > configure.ac | 2 +- > meson.build | 2 +- > 2 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/configure.ac b/configure.ac > index 1ca3974..fd7df47 100644 > ---

Re: [Pixman] [PATCH v2] build: improve control logic for enabling MMI.

Thank you for the patch! On Fri, Mar 6, 2020 at 3:28 AM Shiyou Yin wrote: > > From: Yin Shiyou Should be yinshiyou-hf@loongson*.cn*? > > 1. Replace LS_CFLAGS with MMI_CFLAGS to express its intention more accurately. >LS_CFLAGS is still available, but it is not recommended. I'm not aware

Re: [Pixman] [PATCH v2 2/3] build: use '-mloongson-mmi' for Loongson MMI.

On Sat, Feb 22, 2020 at 6:34 AM YunQiang Su wrote: > > Shiyou Yin 于2020年2月22日周六 下午9:26写道： > > > > >-Original Message- > > >From: Adam Jackson [mailto:a...@redhat.com] > > >Sent: Friday, February 21, 2020 11:33 PM > > >To: Yin Shiyou; pixman@lists.freedesktop.org > > >Subject: Re:

Re: [Pixman] [PATCH v2 2/3] build: use '-mloongson-mmi' for Loongson MMI.

On Sat, Feb 22, 2020 at 5:26 AM Shiyou Yin wrote: > > >-Original Message- > >From: Adam Jackson [mailto:a...@redhat.com] > >Sent: Friday, February 21, 2020 11:33 PM > >To: Yin Shiyou; pixman@lists.freedesktop.org > >Subject: Re: [Pixman] [PATCH v2 2/3] build: use '-mloongson-mmi' for >

Re: [Pixman] [PATCH] pixman-combine: Fix wrong value of RB_MASK_PLUS_ONE.

On Thu, Feb 20, 2020 at 6:35 AM Shiyou Yin wrote: > Will this patch be merged? Yes, pushed. Thanks! ___ Pixman mailing list Pixman@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/pixman

Re: [Pixman] [PATCH] pixman-combine: Fix wrong value of RB_MASK_PLUS_ONE.

On Mon, Feb 3, 2020 at 1:56 AM Yin Shiyou wrote: > > --- > pixman/pixman-combine32.h | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/pixman/pixman-combine32.h b/pixman/pixman-combine32.h > index cdd56a6..59bb247 100644 > --- a/pixman/pixman-combine32.h > +++

Re: [Pixman] Optimize Graphic Routines for s390x in Pixman - Queries

On Sat, Jan 25, 2020 at 4:57 AM Naveen Naidu wrote: > > Hello Everyone, > > I am Naveen a Senior Year Computer Science Undergraduate from India. I am > planning to apply for Open Mainframe Project > Internship(https://github.com/openmainframeproject-internship/resources) > program, whose one

Re: [Pixman] [PATCH] [dither] Don't use GNU extension for binary numbers

Thanks. Pushed. ___ Pixman mailing list Pixman@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/pixman

Re: [Pixman] Dithering patches, v2

On Sat, May 11, 2019 at 7:42 AM Bryce Harrington wrote: > > On Tue, May 07, 2019 at 09:52:39AM -0700, Matt Turner wrote: > > On Sun, May 5, 2019 at 11:50 AM Bryce Harrington > > wrote: > > > > > > On Mon, Apr 22, 2019 at 09:26:48AM -0700, Matt Turner wrote:

Re: [Pixman] Dithering patches, v2

On Sun, May 5, 2019 at 11:50 AM Bryce Harrington wrote: > > On Mon, Apr 22, 2019 at 09:26:48AM -0700, Matt Turner wrote: > > On Fri, Apr 19, 2019 at 4:52 PM Bryce Harrington > > wrote: > > > Inkscape would love to see Basile's dithering patches included. Our > &g

Re: [Pixman] Dithering patches, v2

On Fri, Apr 19, 2019 at 4:52 PM Bryce Harrington wrote: > Inkscape would love to see Basile's dithering patches included. Our > testing shows that they make a huge quality difference for our users; > this solves a critical need. > > Mc and I have done some preliminary investigation into how to

Re: [Pixman] [PATCH 2/2] AVX2 implementation of OVER, ROVER, ADD, ROUT operators.

On Thu, Mar 28, 2019 at 10:41 PM Matt Turner wrote: > > On Wed, Mar 27, 2019 at 1:06 PM Matt Turner wrote: > > > > Thank you. I'll run some benchmarks on my KBL system to confirm and > > then commit them. > > > > I'm planning to do a 0.40 release soon with

[Pixman] [ANNOUNCE] pixman release 0.38.4 now available

: 87e1abc91ac4e5dfcc275f744f1d0ec3277ee7cd pixman-0.38.4.tar.bz2 GPG signature: https://cairographics.org/releases/pixman-0.38.4.tar.gz.sha1.asc (signed by [ultimate] Matt Turner [ultimate] Matt Turner [ultimate] Matt Turner [ultimate] Matt Turner ) Git: https://gitlab.freedesktop.org/pixman/pixman.git

[Pixman] [ANNOUNCE] pixman release 0.38.2 now available

pixman-0.38.2.tar.gz SHA1: ce40833fe4337aa6329ac5694d9ff342338219c1 pixman-0.38.2.tar.bz2 GPG signature: http://cairographics.org/releases/pixman-0.38.2.tar.gz.sha1.asc (signed by [ultimate] Matt Turner [ultimate] Matt Turner [ultimate] Matt Turner [ultimate] Matt Turner

Re: [Pixman] [PATCH 2/2] AVX2 implementation of OVER, ROVER, ADD, ROUT operators.

On Wed, Mar 27, 2019 at 1:06 PM Matt Turner wrote: > > Thank you. I'll run some benchmarks on my KBL system to confirm and > then commit them. > > I'm planning to do a 0.40 release soon with some Meson fixes and other > small things. Seems like these patches will be good

Re: [Pixman] [PATCH] void function should not return a value

Thanks. Merged. ___ Pixman mailing list Pixman@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/pixman

Re: [Pixman] [PATCH 1/2] Windows: Show compiler invocation

Thanks. Merged both. ___ Pixman mailing list Pixman@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/pixman

Re: [Pixman] [PATCH 2/2] AVX2 implementation of OVER, ROVER, ADD, ROUT operators.

Thank you. I'll run some benchmarks on my KBL system to confirm and then commit them. I'm planning to do a 0.40 release soon with some Meson fixes and other small things. Seems like these patches will be good to include to make the release have a new feature :)

[Pixman] [PATCH] avx2: Add fast path for over_reverse_n_8888

lowlevel-blt-bench, over_reverse_n_, 100 iterations: Before After Mean StdDev Mean StdDev Confidence Change L1 2372.6 2.50 4387.6 8.00100.00% +84.9% L2 2490.3 5.29 4326.5 20.79100.00% +73.7% M 2418.3 10.43 3718.0 38.55

Re: [Pixman] [PATCH 3/3] Rev2 of patch: AVX2 versions of OVER and ROVER operators.

On Wed, Jan 16, 2019 at 4:57 PM Raghuveer Devulapalli wrote: > > From: raghuveer devulapalli > > These were found to be upto 1.8 times faster (depending on the array > size) than the corresponding SSE2 version. The AVX2 and SSE2 were > benchmarked on a Intel(R) Core(TM) i5-6260U CPU @ 1.80GHz.

Re: [Pixman] [PATCH 2/3] Moving helper functions in pixman-sse2.c to pixman-sse2.h.

On Wed, Jan 16, 2019 at 4:57 PM Raghuveer Devulapalli wrote: > > From: raghuveer devulapalli > > These helper function will be reused in pixman-avx2.c implementations in > the future. > --- > pixman/pixman-sse2.c | 504 +-- > pixman/pixman-sse2.h | 502

Re: [Pixman] [PATCH] mmx: compile on MIPS for Loongson-3A MMI optimizations

On Tue, Sep 18, 2018 at 2:34 AM wrote: > > From: Xianju Diao > > make check: > when I enable the USE_OPENMP, the test of 'glyph-test' and > 'cover-test' will failed on Loongson-3A3000. > Neither of the two test examples passed without optimizing the > code.Maybe be multi-core

Re: [Pixman] [PATCH] Adding AVX2 implementation of the OVER and REVERSE-OVER operator

On Wed, Aug 29, 2018 at 12:09 PM Matt Turner wrote: > Trailing whitespace. There's a lot throughout this patch. I'm not > going to point them out individually. I just looked up how to configure git to alert you to bad whitespace: git config core.whitespace indent-with-non-tab,space-befo

Re: [Pixman] [PATCH] Adding AVX2 implementation of the OVER and REVERSE-OVER operator

On Wed, Aug 29, 2018 at 12:09 PM Matt Turner wrote: > > On Wed, Aug 22, 2018 at 10:03 AM raghuveer devulapalli > wrote: > > > > The AVX2 implementation of OVER and REVERSE OVER operator was > > found to be upto 2.2 times faster (depending on the array size) than

Re: [Pixman] [PATCH] Adding AVX2 implementation of the OVER and REVERSE-OVER operator

On Wed, Aug 22, 2018 at 10:03 AM raghuveer devulapalli wrote: > > The AVX2 implementation of OVER and REVERSE OVER operator was > found to be upto 2.2 times faster (depending on the array size) than > the corresponding SSE2 version. The AVX2 and SSE2 were benchmarked > on a Intel(R) Core(TM)

Re: [Pixman] [PATCH] Adding infrastructure to permit future AVX2 implementations

Thank you for the patches! Some comments inline. On Wed, Aug 22, 2018 at 10:03 AM raghuveer devulapalli wrote: > > --- > configure.ac| 44 > pixman/Makefile.am | 12 > pixman/pixman-avx2.c| 32

Re: [Pixman] [Patch 1/1] Clang compile failure due to use of __builtin_shuffle

On Tue, Aug 7, 2018 at 2:50 AM StormByte wrote: > > While playing with Clang and compiling a Gentoo system with it, I realized > that pixman is not compiling because of the use of __builtin_shuffle which > according to LLVM mailing list, should not be used directly [1]. > > As such, I

Re: [Pixman] Pushing unreviewed patches to the pixman git repository

On Tue, Jun 5, 2018 at 6:06 PM, Siarhei Siamashka wrote: > Hello, > > I noticed that some people with commit access started pushing patches > to the pixman git repository without giving the pixman mailing list > subscribers any reasonable chance to review them: > >

Re: [Pixman] [PATCH] test: Adjust for clang's removal of __builtin_shuffle

On Mon, Jun 4, 2018 at 10:37 AM, Adam Jackson wrote: > On Mon, 2018-06-04 at 10:04 -0700, Matt Turner wrote: > >> #ifdef HAVE_GCC_VECTOR_EXTENSIONS >> -const uint8x16 bswap_shufflemask = >> +# if __has_builtin(__builtin_shufflevector) >

[Pixman] [PATCH] test: Adjust for clang's removal of __builtin_shuffle

the same but have different API. Bugzilla: https://bugs.gentoo.org/646360 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104886 Tested-by: Philip Chimento Reviewed-by: Matt Turner --- I turned https://bugs.freedesktop.org/show_bug.cgi?id=104886#c2 into a Tested-by tag for Philip. I also

Re: [Pixman] [PATCH] vmx: Fix vector loads on ppc64le

Tested-by: Matt Turner <matts...@gmail.com> ___ Pixman mailing list Pixman@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/pixman

Re: [Pixman] Pixman not building on MacOS X 10.11

On Sun, Oct 11, 2015 at 10:34 AM, Andrea Canciani wrote: > On Sun, Oct 11, 2015 at 5:30 AM, Siarhei Siamashka > wrote: >> >> On Sun, 11 Oct 2015 04:53:08 +0300 >> Siarhei Siamashka wrote: >> >> > On Sat, 10 Oct 2015

Re: [Pixman] [PATCH] mmx: Use MMX2 intrinsics from xmmintrin.h directly.

On Sun, Oct 25, 2015 at 1:13 PM, Matt Turner <matts...@gmail.com> wrote: > On Sun, Oct 11, 2015 at 8:59 PM, Matt Turner <matts...@gmail.com> wrote: >> We had lots of hacks to handle the inability to include xmmintrin.h >> without compiling with -msse (lest SSE instru

Re: [Pixman] Pixman not building on MacOS X 10.11

On Wed, Nov 18, 2015 at 8:35 PM, Siarhei Siamashka <siarhei.siamas...@gmail.com> wrote: > On Wed, 18 Nov 2015 14:22:09 -0800 > Matt Turner <matts...@gmail.com> wrote: > >> On Sun, Oct 11, 2015 at 10:34 AM, Andrea Canciani <ranm...@gmail.com> wrote: >> >

Re: [Pixman] [PATCH] mmx: Use MMX2 intrinsics from xmmintrin.h directly.

On Sun, Oct 25, 2015 at 5:10 PM, Siarhei Siamashka <siarhei.siamas...@gmail.com> wrote: > On Sun, 25 Oct 2015 13:13:09 -0700 > Matt Turner <matts...@gmail.com> wrote: > >> On Sun, Oct 11, 2015 at 8:59 PM, Matt Turner <matts...@gmail.com> wrote: >> > We

Re: [Pixman] [PATCH] mmx: Use MMX2 intrinsics from xmmintrin.h directly.

On Sun, Oct 25, 2015 at 7:12 PM, Søren Sandmann wrote: > On Sun, Oct 25, 2015 at 8:10 PM, Siarhei Siamashka > wrote: > >> >> Or we could simply do nothing and finally retire MMX support on x86. >> If OLPC XO-1 users still do exist, they can

Re: [Pixman] [PATCH] mmx: Use MMX2 intrinsics from xmmintrin.h directly.

On Sun, Oct 11, 2015 at 8:59 PM, Matt Turner <matts...@gmail.com> wrote: > We had lots of hacks to handle the inability to include xmmintrin.h > without compiling with -msse (lest SSE instructions be used in > pixman-mmx.c). Some recent version of gcc relaxed this restrict

[Pixman] [PATCH] mmx: Use MMX2 intrinsics from xmmintrin.h directly.

We had lots of hacks to handle the inability to include xmmintrin.h without compiling with -msse (lest SSE instructions be used in pixman-mmx.c). Some recent version of gcc relaxed this restriction. Change configure.ac to test that xmmintrin.h can be included and that we can use some intrinsics

[Pixman] [PATCH] mmx: Use MMX2 intrinsics from xmmintrin.h directly.

after Signed-off-by: Matt Turner <matts...@gmail.com> --- Looks like _MM_SHUFFLE isn't defined by ARM's mmintrin.h. configure.ac| 15 - pixman/pixman-mmx.c | 64 - 2 files changed, 8 insertions(+), 71 deletions(-) diff

Re: [Pixman] [PATCH] mmx: Use MMX2 intrinsics from xmmintrin.h directly.

On Sun, Oct 11, 2015 at 8:41 PM, Siarhei Siamashka <siarhei.siamas...@gmail.com> wrote: > On Sun, 11 Oct 2015 14:55:28 -0700 > Matt Turner <matts...@gmail.com> wrote: > > Hello, > > Thanks. The patch looks good. In fact, it also allows the MMX code to > be com

Re: [Pixman] [PATCH 1/4] pixman-fast-path: Add over_n_8888 fast path (disabled)

On Thu, Aug 20, 2015 at 6:58 AM, Pekka Paalanen ppaala...@gmail.com wrote: From: Ben Avison bavi...@riscosopen.org This is a C fast path, useful for reference or for platforms that don't have their own fast path for this operation. This new fast path is initially disabled by putting the

Re: [Pixman] [PATCH] test: Add cover-test

On Tue, May 26, 2015 at 3:58 PM, Ben Avison bavi...@riscosopen.org wrote: This test aims to verify both numerical correctness and the honouring of array bounds for scaled plots (both nearest-neighbour and bilinear) at or close to the boundary conditions for applicability of cover type fast

Re: [Pixman] Is Pixman being maintained at all?

On Thu, Apr 2, 2015 at 2:26 AM, Pekka Paalanen ppaala...@gmail.com wrote: On Wed, 1 Apr 2015 18:46:10 -0700 Matt Turner matts...@gmail.com wrote: On Mon, Mar 30, 2015 at 10:58 AM, Bill Spitzak spit...@gmail.com wrote: On 03/30/2015 10:25 AM, Matt Turner wrote: Do you just need someone

Re: [Pixman] [PATCH 1/5] armv6: Fix typo in preload macro

Pushed. ___ Pixman mailing list Pixman@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pixman

Re: [Pixman] Is Pixman being maintained at all?

On Mon, Mar 30, 2015 at 10:58 AM, Bill Spitzak spit...@gmail.com wrote: On 03/30/2015 10:25 AM, Matt Turner wrote: Do you just need someone to push them? I'm not capable of reviewing these. Since Søren isn't really maintaining pixman anymore I'm not really sure how to proceed

Re: [Pixman] [PATCH 2/3] armv7: Faster fill operations

On Wed, Mar 4, 2015 at 5:56 PM, Ben Avison bavi...@riscosopen.org wrote: This eliminates a number of branches over blocks of code that are either empty or can be trivially combined with a separate code block at the start and end of each scanline. This has a surprisingly big effect, at least on

Re: [Pixman] Unable to build master on Raspberry PI

On Wed, Dec 3, 2014 at 9:18 AM, Andrea Giammarchi andrea.giammar...@gmail.com wrote: Thank you very much Siarhei, I am still building something huge and had no way to double check but at least I can confirm the gcc is 4.9.2. I will try to --disable-arm-iwmmxt when it shows arm6l as uname -m

Re: [Pixman] [PATCH 0/2] mmx nearest scaling paths

On Tue, Sep 23, 2014 at 12:24 PM, Søren Sandmann soren.sandm...@gmail.com wrote: IIRC, we have already discussed it before. Maybe we should just disable MMX support for x86 and use it only for MIPS Loongson and ARM IWMMXT? I don't really see the benefit. The bugs we've had have all been

[Pixman] [PATCH 1/2] mmx: Add nearest over_8888_n_8888

lowlevel-blt-bench -n, over__n_, 15 iterations on Loongson 2f: Before After Mean StdDev Mean StdDev Change L1 9.7 0.01 19.2 0.02 +98.2% L2 9.6 0.11 19.2 0.16 +99.5% M 7.3 0.02 12.5 0.01 +72.0%

[Pixman] [PATCH 0/2] mmx nearest scaling paths

Here are a couple of nearest scaling MMX paths I wrote a long time ago for Loongson and other things using the MMX code. I've got a few more patches for the MMX code that I'll send out as I benchmark them. I don't really expect any reviews, so barring objections I'll plan to commit them in a few

[Pixman] Test suite failures on 32-bit x86?

Building the 0.32.2 release and from git with CC=gcc -m32 ./autogen.sh make check PASS: prng-test PASS: a1-trap-test PASS: region-translate-test PASS: pdf-op-test PASS: region-test PASS: fetch-test ../test-driver: line 95: 3312 Segmentation fault $@ $log_file 21 FAIL: rotate-test PASS:

Re: [Pixman] Latest GIT source for 'pixman-sse2.c'

On Sun, Oct 6, 2013 at 1:50 AM, John Emmas john...@tiscali.co.uk wrote: On 05/10/2013 19:32, John Emmas wrote: On 5 Oct 2013, at 19:00, Siarhei Siamashka wrote: Andrea Canciani has already investigated the problem and submitted the fixes here:

Re: [Pixman] [PATCH 1/2] Add empty SSSE3 implementation

On Thu, Aug 29, 2013 at 10:02 AM, Søren Sandmann Pedersen sandm...@cs.au.dk wrote: This commit adds a new, empty SSSE3 implementation and the associated build system support. configure.ac: detect whether the compiler understands SSSE3 intrinsics and set up the required

[Pixman] [PATCH] mmx: Document implementation(s) of pix_multiply().

--- I look at that function and can never remember what it does or how it manages to do it. pixman/pixman-mmx.c | 23 +++ 1 file changed, 23 insertions(+) diff --git a/pixman/pixman-mmx.c b/pixman/pixman-mmx.c index 14790c0..746ecd6 100644 --- a/pixman/pixman-mmx.c +++

[Pixman] [PATCH] Use AC_LINK_IFELSE to check if the Loongson MMI code can link

From: Markos Chandras markos.chand...@imgtec.com The Loongson code is compiled with -march=loongson2f to enable the MMI instructions, but binutils refuses to link object code compiled with different -march settings, leading to link failures later in the compile. This avoids that problem by

Re: [Pixman] As per : Please report to pixman@lists.freedesktop.org

On Tue, Apr 9, 2013 at 2:39 PM, David Lisle da...@lisle.ca wrote: Thanks for responding, the problem remains a mystery bu the overall project now is operational. I appreciate that you took time. I really meant that there certainly must have been more error output that wasn't in your email. This

Re: [Pixman] As per : Please report to pixman@lists.freedesktop.org

On Mon, Apr 8, 2013 at 11:13 AM, David Lisle da...@lisle.ca wrote: === make[2]: *** [check-TESTS] Error 1 make[2]: Leaving directory `/usr/src/pixman-0.28.2/test' make[1]: *** [check-am] Error 2 make[1]: Leaving directory

Re: [Pixman] [PATCH 2/4] Added fast path for pad type repeats

On Tue, Feb 5, 2013 at 4:39 PM, Ben Avison bavi...@riscosopen.org wrote: diff --git a/test/Makefile.sources b/test/Makefile.sources index e323a8e..bcbca37 100644 --- a/test/Makefile.sources +++ b/test/Makefile.sources @@ -1,6 +1,7 @@ # Tests (sorted by expected completion time)

Re: [Pixman] 0.29.2

On Sun, Jan 27, 2013 at 11:43 AM, Siarhei Siamashka siarhei.siamas...@gmail.com wrote: Still, I'm not very happy about the code duplication. We already have similar iterators (fetch only, no writeback) in pixman-mmx.c:

Re: [Pixman] [PATCH] build: Support building Loongson code for 2e, 2f, 3a

Some preemptive explanations: On Sat, Jan 26, 2013 at 6:54 PM, Matt Turner matts...@gmail.com wrote: diff --git a/pixman/pixman-mips.c b/pixman/pixman-mips.c index 3048813..77bef5c 100644 --- a/pixman/pixman-mips.c +++ b/pixman/pixman-mips.c @@ -27,6 +27,10 @@ #if defined(USE_MIPS_DSPR2

Re: [Pixman] [PATCH] sse2: Implement simple bilinear scaling for x8r8g8b8 to a8r8g8b8

) static force_inline void +scaled_bilinear_scanline_sse2_0888__SRC (uint32_t * dst, Maybe some funny whitespace before dst? Or maybe just a spaces vs tabs issue. Anyway, Reviewed-by: Matt Turner matts...@gmail.com +const uint32_t * mask

[Pixman] [PATCH] Add new demos and tests to .gitignore

--- .gitignore | 3 +++ 1 file changed, 3 insertions(+) diff --git a/.gitignore b/.gitignore index a4d9f99..dcb3f8e 100644 --- a/.gitignore +++ b/.gitignore @@ -37,6 +37,7 @@ demos/quad2quad demos/radial-test demos/screen-test demos/srgb-test +demos/srgb-trap-test demos/trap-test

[Pixman] [PATCH] Convert INCLUDES to AM_CPPFLAGS

INCLUDES has been deprecated starting with automake 1.13. Convert all occurrences with the recommended AM_CPPFLAGS replacement. --- demos/Makefile.am | 2 +- pixman/Makefile.am | 2 +- test/Makefile.am | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/demos/Makefile.am

Re: [Pixman] 0.29.2

On Fri, Jan 18, 2013 at 4:15 PM, Søren Sandmann sandm...@cs.au.dk wrote: Hi, It's about time to get a 0.29.2 development snapshot out, but there are some outstanding patches I'd like to get my triple build loongson patch in, but haven't gotten any testers yet. I'll set up a chroot this

Re: [Pixman] [PATCH] build: Support building Loongson code for 2e, 2f, 3a

On Sun, Jan 6, 2013 at 7:46 PM, Cyril Brulebois k...@debian.org wrote: Hello Matt, Matt Turner matts...@gmail.com (06/01/2013): On Sat, Sep 15, 2012 at 11:59 PM, Matt Turner matts...@gmail.com wrote: pixman/Makefile.am contains a hack that allows pixman-mmx.c to be compiled with different

Re: [Pixman] [PATCH] sse2: Add fast paths for bilinear source with a solid mask

On Tue, Jan 8, 2013 at 12:55 PM, Chris Wilson ch...@chris-wilson.co.uk wrote: Based on the existing sse2__n_ nearest scaling routines. fishbowl on an i5-2500: 60.9s - 56.9s Signed-off-by: Chris Wilson ch...@chris-wilson.co.uk --- Looks good to me. Reviewed-by: Matt Turner matts

Re: [Pixman] [PATCH] build: Support building Loongson code for 2e, 2f, 3a

On Sat, Sep 15, 2012 at 11:59 PM, Matt Turner matts...@gmail.com wrote: pixman/Makefile.am contains a hack that allows pixman-mmx.c to be compiled with different overriding CFLAGS, since automake seriously doesn't have a way to do this. Seriously stupid. It works by defining a new rule

Re: [Pixman] [PATCH] Fix build with automake-1.13

On Wed, Jan 2, 2013 at 8:38 PM, Marko Lindqvist cazf...@gmail.com wrote: Automake-1.13 has removed long obsolete AM_CONFIG_HEADER macro ( http://lists.gnu.org/archive/html/automake/2012-12/msg00038.html ) and autoreconf errors out upon seeing it. Attached patch replaces obsolete

Re: [Pixman] [PATCH] sse2: Add a fast path for add_n_8888

On Wed, Jan 2, 2013 at 3:01 AM, Chris Wilson ch...@chris-wilson.co.uk wrote: This path is being exercised by inplace compositing of trapezoids, for instance as used in the firefox-asteroids cairo-trace. cairo-perf-trace numbers from firefox-asteroids would be cool to have in the commit message.

Re: [Pixman] [PATCH] sse2: Add a fast path for add_n_8888

On Wed, Jan 2, 2013 at 3:01 AM, Chris Wilson ch...@chris-wilson.co.uk wrote: This path is being exercised by inplace compositing of trapezoids, for instance as used in the firefox-asteroids cairo-trace. core2 @ 2.66GHz, reference memcpy speed = 4898.2MB/s (1224.6MP/s for 32bpp fills)

Re: [Pixman] [cairo] issue with blend modes in pixman

On Mon, Dec 31, 2012 at 1:05 PM, Rik Cabanier caban...@gmail.com wrote: Looking at the formulas, I can see what's wrong but I don't know who to contact. These mailing lists are perfect. ___ Pixman mailing list Pixman@lists.freedesktop.org

Re: [Pixman] [PATCH] Always use xmmintrin.h for 64 bit Windows

On Tue, Nov 13, 2012 at 10:44 AM, Stefan Weil s...@weilnetz.de wrote: MinGW-w64 uses the GNU compiler and does not define _MSC_VER. Nevertheless, it provides xmmintrin.h and must be handled here like the MS compiler. Otherwise compilation fails due to conflicting declarations. Signed-off-by:

Re: [Pixman] Questionable numbers from lowlevel-blt-bench

On Mon, Oct 1, 2012 at 1:17 AM, Jonathan Morton jonathan.mor...@movial.com wrote: On Sun, 30 Sep 2012 15:05:18 -0700, Matt Turner matts...@gmail.com wrote: In doing performance work, I've noticed some weird results from lowlevel-blt-bench. Often it has seemed that the RT results determined

[Pixman] Questionable numbers from lowlevel-blt-bench

Hi Jonathan, In doing performance work, I've noticed some weird results from lowlevel-blt-bench. Often it has seemed that the RT results determined the Kops/s almost entirely. I came across an instance of this today which was particularly striking: Before: add__ = L1: 47.01 L2: 36.84

[Pixman] [test PATCH] Use _mm_maddubs_epi16 in BILINEAR_INTERPOLATE_ONE_PIXEL

Siarhei, can you measure any performance improvement with this? I can't... :( --- pixman/pixman-sse2.c |8 +++- 1 files changed, 3 insertions(+), 5 deletions(-) diff --git a/pixman/pixman-sse2.c b/pixman/pixman-sse2.c index efed310..4fbc045 100644 --- a/pixman/pixman-sse2.c +++

Re: [Pixman] [PATCH 05/10] pixman-utils.c, pixman-private.h: Add floating point conversion routines

On Wed, Sep 26, 2012 at 1:43 PM, Søren Sandmann sandm...@cs.au.dk wrote: From: Søren Sandmann Pedersen s...@redhat.com A new struct argb_t containing a floating point pixel is added to pixman-private.h, and conversion routines are added to pixman-utils.c to convert normalized integers to and

[Pixman] [PATCH] sse2: mark pack_565_2x128_128 as static force_inline

--- pixman/pixman-sse2.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/pixman/pixman-sse2.c b/pixman/pixman-sse2.c index e273a95..cf21ef8 100644 --- a/pixman/pixman-sse2.c +++ b/pixman/pixman-sse2.c @@ -146,7 +146,7 @@ pack_565_2packedx128_128 (__m128i lo, __m128i hi)

[Pixman] [PATCH] build: Remove useless DEP_CFLAGS/DEP_LIBS variables

Reduces the size of the generated pixman/Makefile from 46k to 41k. --- configure.ac |2 -- pixman-1.pc.in |4 ++-- pixman/Makefile.am | 23 +-- 3 files changed, 7 insertions(+), 22 deletions(-) diff --git a/configure.ac b/configure.ac index

[Pixman] [PATCH] build: Support building Loongson code for 2e, 2f, 3a

pixman/Makefile.am contains a hack that allows pixman-mmx.c to be compiled with different overriding CFLAGS, since automake seriously doesn't have a way to do this. Seriously stupid. It works by defining a new rule and recursively calling make with modified CFLAGS set. Note the difference

Re: [Pixman] [PATCH] Make pixman-mmx.c compile on x86-32 without optimization

On Mon, Jul 9, 2012 at 10:19 PM, Matt Turner matts...@gmail.com wrote: Works for me. On second glance, did I just make a mistake in b87cd1f and write ifdef instead of ifndef? ___ Pixman mailing list Pixman@lists.freedesktop.org http

Re: [Pixman] [PATCH] Make pixman-mmx.c compile on x86-32 without optimization

On Mon, Jul 9, 2012 at 7:31 AM, Søren Sandmann sandm...@cs.au.dk wrote: From: Søren Sandmann Pedersen s...@redhat.com When not optimizing, write _mm_shuffle_pi16() as a statement expression with inline assembly. That way we avoid __builtin_ia32_pshufw(), which is only available when compiling

Re: [Pixman] [BUG pixman] f9c91ee2f27eaea68d8c3a130bf7d4bc0c860834 breaks compilation

On Mon, Jul 9, 2012 at 1:55 AM, Knut Petersen knut_peter...@t-online.de wrote: Søren, the bad commit was supposed to fix a gcc -O0 compile problem, but it breaks gcc -O0 compilation here. Reverting f9c91ee2 fixes the problem for me. Is this build automated? If it's an automated build that

Re: [Pixman] [PATCH 1/5] mmx: add scaled bilinear src_8888_8888

On Sun, Jul 1, 2012 at 12:56 PM, Søren Sandmann sandm...@cs.au.dk wrote: Matt Turner matts...@gmail.com writes: +SIMPLE_BILINEAR_FAST_PATH (SRC, a8r8g8b8, a8r8g8b8, mmx__ ), +SIMPLE_BILINEAR_FAST_PATH (SRC, a8r8g8b8, x8r8g8b8

Re: [Pixman] [PATCH] Use a compile-time constant for the K constraint in the MMX detection.

On Sun, Jul 1, 2012 at 5:03 PM, Søren Sandman sandm...@cs.au.dk wrote: From: Søren Sandmann Pedersen s...@redhat.com When compiling with -O0, gcc doesn't understand that in signed char x = 0; ... asm (..., : K (x)); x is constant. Fix this by using an immediate

Re: [Pixman] [ANNOUNCE] pixman release 0.26.2 now available

On Sat, Jun 30, 2012 at 3:04 AM, Andreas Radke a.ra...@arcor.de wrote: Somehow I get different checksums: [andyrtr@workstation64 trunk]$ md5sum pixman-0.26.2.tar.* 6b3e4c5300adb893a2baa9631c23efb2  pixman-0.26.2.tar.bz2 276242da5b3af1258d072cf205d18f0b  pixman-0.26.2.tar.gz Can you confirm

[Pixman] [ANNOUNCE] pixman release 0.26.2 now available

: ba71d029d174aa8b9d23b1072ab76e6b4ea3de59 pixman-0.26.2.tar.gz SHA1: c7cdb5803061ee6614acc66258b0339ad4e52314 pixman-0.26.2.tar.bz2 GPG signature: http://cairographics.org/releases/pixman-0.26.2.tar.gz.sha1.asc (signed by Matt Turner matts...@gmail.com) Git: git://git.freedesktop.org

[Pixman] [PATCH] mmx: Use expand_alpha instead of mask/shift

--- pixman/pixman-mmx.c |8 ++-- 1 files changed, 2 insertions(+), 6 deletions(-) diff --git a/pixman/pixman-mmx.c b/pixman/pixman-mmx.c index bff8585..071cdfd 100644 --- a/pixman/pixman-mmx.c +++ b/pixman/pixman-mmx.c @@ -1618,9 +1618,7 @@ mmx_composite_over__n_

Re: [Pixman] [PATCH 00/10] Cleanups to CPU detection

On Fri, Jun 29, 2012 at 5:20 PM, Alan Coopersmith alan.coopersm...@oracle.com wrote: On 06/29/12 01:44 PM, Søren Sandmann Pedersen wrote: I was looking at making use of some of the newer x86 SIMD instruction sets and realized that (a) we don't ever call cpuid on x86-64, we just assume that MMX

Re: [Pixman] [PATCH 3/5] mmx: add scaled bilinear over_8888_8_8888

On Wed, Jun 27, 2012 at 10:38 PM, Matt Turner matts...@gmail.com wrote: Reduces runtime of firefox-fishtank trace from 1510 to 1030 seconds on Loongson. ---  pixman/pixman-mmx.c |   84 +++  1 files changed, 84 insertions(+), 0 deletions

Re: [Pixman] [PATCH 5/5] mmx: optimize bilinear function when using 7-bit precision

On Wed, Jun 27, 2012 at 10:38 PM, Matt Turner matts...@gmail.com wrote: --- Reduces runtime of firefox-planet-gnome trace from 156 to 153 seconds on Loongson. Increases runtime of firefox-fishtank trace from 1030 to 1060 seconds. Why?  pixman/pixman-mmx.c |   45

[Pixman] Bilinear scaling patches for MMX

These five patches implement the same bilinear scaling compositing functions as provided by the SSE2 code. They pass the test suite on x86, Loongson, and iwMMXt, but I haven't done extensive benchmarking yet on iwMMXt. The fifth patch optimizes the functions for 7-bit bilinear interpolation, but

[Pixman] [PATCH 1/5] mmx: add scaled bilinear src_8888_8888

--- pixman/loongson-mmintrin.h | 73 ++ pixman/pixman-mmx.c| 93 2 files changed, 166 insertions(+), 0 deletions(-) diff --git a/pixman/loongson-mmintrin.h b/pixman/loongson-mmintrin.h index 1a114fe..f0931ac

[Pixman] [PATCH 3/5] mmx: add scaled bilinear over_8888_8_8888

Reduces runtime of firefox-fishtank trace from 1510 to 1030 seconds on Loongson. --- pixman/pixman-mmx.c | 84 +++ 1 files changed, 84 insertions(+), 0 deletions(-) diff --git a/pixman/pixman-mmx.c b/pixman/pixman-mmx.c index a504b60..ea732bb

[Pixman] [PATCH 5/5] mmx: optimize bilinear function when using 7-bit precision

--- Reduces runtime of firefox-planet-gnome trace from 156 to 153 seconds on Loongson. Increases runtime of firefox-fishtank trace from 1030 to 1060 seconds. Why? pixman/pixman-mmx.c | 45 - 1 files changed, 32 insertions(+), 13 deletions(-) diff

  1   2   3   >