From: Ben Avison bavi...@riscosopen.org
---
pixman/pixman-arm-simd-asm.h | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/pixman/pixman-arm-simd-asm.h b/pixman/pixman-arm-simd-asm.h
index 6543606..74400c1 100644
--- a/pixman/pixman-arm-simd-asm.h
+++
from combining the and inside the .if
statement with an earlier tst. The and was normally needed, except
for in one special case, where bits 4-31 were all shifted off the top of
the register later on in preload_leading_step2, so we didn't care about
their values.
v4, Pekka Paalanen pekka.paala
From: Ben Avison bavi...@riscosopen.org
in_reverse__ is one of the more commonly used operations in the
cairo-perf-trace suite that hasn't been in lowlevel-blt-bench until now.
v4, Pekka Paalanen pekka.paala...@collabora.co.uk :
Split from Add extra test to lowlevel-blt-bench
to be statistically significant. This claim is
based on comparing two 30-iteration identical before runs using the
exact same binaries, and observing changes from -0.4% to +0.5% with
=99% confidence.
Confidence is based on Welch's t-test.
v4, Pekka Paalanen pekka.paala...@collabora.co.uk
From: Ben Avison bavi...@riscosopen.org
Previously, the variable ARGS_STACK_OFFSET was available to extract values
from function arguments during the init macro. Now this changes dynamically
around stack operations in the function as a whole so that arguments can be
accessed at any point. It is
less than 1%
can be accounted as measurement errors, even if statistically
significant.
v4, Pekka Paalanen pekka.paala...@collabora.co.uk :
Use pixman_asm_function instead of startfunc.
Rebased. Re-benchmarked on Raspberry Pi.
Commit message.
---
pixman/pixman-arm-simd
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Several files define identically the asm macro pixman_asm_function.
Merge all these definitions into a new asm header.
The original definition is taken from pixman-arm-simd-asm-scaled.S with
the copyright/licence/author blurb verbatim
%
-6.2%
IOW, the flag has adverse effects on lowlevel-blt-bench performance,
but improves one or two Cairo traces slightly.
v4, Pekka Paalanen pekka.paala...@collabora.co.uk :
Rebased, re-benchmarked on Raspberry Pi, commit message.
---
Should I re-spin this without the flag? Ben
On Fri, 4 Apr 2014 08:24:18 +0300
Siarhei Siamashka siarhei.siamas...@gmail.com wrote:
On Mon, 31 Mar 2014 15:03:45 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
From: Ben Avison bavi...@riscosopen.org
Benchmark results, before is the patch
- ARMv6: Add fast path
On Mon, 31 Mar 2014 15:54:24 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
From: Ben Avison bavi...@riscosopen.org
Benchmark results, before is the patch
- ARMv6: Add fast path for over_n___ca
and after contains the additional patches:
- ARMv6: Add fast path flag to force
On Mon, 7 Apr 2014 10:50:03 +0300
Siarhei Siamashka siarhei.siamas...@gmail.com wrote:
On Fri, 4 Apr 2014 10:28:05 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
Could someone point me to a document describing how one uses these
testing tools properly? Hopefully it would answer all my
Hi,
this thread started from
http://lists.freedesktop.org/archives/pixman/2013-February/002619.html
and continued in
http://lists.freedesktop.org/archives/pixman/2013-March/002677.html
I'd like to hear what the thoughts of it are now, more questions below.
On Wed, 13 Feb 2013 08:37:06 +0100
From: Pekka Paalanen pekka.paala...@collabora.co.uk
The two ARM headers contained open-coded copies of pixman_asm_function,
replace these.
Since it seems customary that ARM headers do not use CPP include guards,
rely on the .S files to #include pixman-arm-asm.h first. They all
already do
On Fri, 4 Apr 2014 08:24:18 +0300
Siarhei Siamashka siarhei.siamas...@gmail.com wrote:
And this is not the first time we miss a bug because of running
blitters-test just a little bit shorter than would be necessary
to detect it:
On Mon, 7 Apr 2014 14:57:10 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
From: Pekka Paalanen pekka.paala...@collabora.co.uk
The two ARM headers contained open-coded copies of pixman_asm_function,
replace these.
Since it seems customary that ARM headers do not use CPP include guards
From: Ben Avison bavi...@riscosopen.org
Previously, the variable ARGS_STACK_OFFSET was available to extract values
from function arguments during the init macro. Now this changes dynamically
around stack operations in the function as a whole so that arguments can be
accessed at any point. It is
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Hi,
this is a re-send of the two series that ended with
ARMv6: Add fast path for over_n___ca and
ARMv6: Add fast path for in_reverse__, i.e.
http://lists.freedesktop.org/archives/pixman/2014-March/003219.html
http
transparent or opaque pixels.
v4, Pekka Paalanen pekka.paala...@collabora.co.uk :
Rebased, re-benchmarked on Raspberry Pi, commit message.
v5, Pekka Paalanen pekka.paala...@collabora.co.uk :
Rebased, re-benchmarked on Raspberry Pi due to a fix to
ARMv6: Add fast path
From: Ben Avison bavi...@riscosopen.org
---
pixman/pixman-arm-simd-asm.h | 14 +-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/pixman/pixman-arm-simd-asm.h b/pixman/pixman-arm-simd-asm.h
index 1bb8b45..0baf87a 100644
--- a/pixman/pixman-arm-simd-asm.h
+++
From: Pekka Paalanen pekka.paala...@collabora.co.uk
The two ARM headers contained open-coded copies of pixman_asm_function,
replace these.
Since it seems customary that ARM headers do not use CPP include guards,
rely on the .S files to #include pixman-arm-asm.h first. They all
do now.
v2: Fix
On Mon, 7 Apr 2014 12:22:15 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
Hi,
this thread started from
http://lists.freedesktop.org/archives/pixman/2013-February/002619.html
and continued in
http://lists.freedesktop.org/archives/pixman/2013-March/002677.html
I'd like to hear what
On Tue, 22 Apr 2014 03:40:14 +0200
soren.sandm...@gmail.com (Søren Sandmann) wrote:
Hi,
I have pushed this one and the two optimizations series to master.
Thank you very much!
- pq
___
Pixman mailing list
Pixman@lists.freedesktop.org
in Pixman and running time is not exactly the same.
Furthermore, in the above test, the CPU idle function was sampled 9%
before, and 15% after.
v4, Pekka Paalanen pekka.paala...@collabora.co.uk :
Re-benchmarked on Raspberry Pi, commit message.
---
pixman/pixman-arm-simd-asm.S | 77
On Thu, 01 May 2014 21:13:00 +0200
soren.sandm...@gmail.com (Søren Sandmann) wrote:
Pushed to master.
Thank you!
That was the last patch being pushed as is from the original series by
Ben.
Cheers,
pq
Pekka Paalanen ppaala...@gmail.com writes:
From: Ben Avison bavi...@riscosopen.org
On Tue, 3 Mar 2015 15:24:19 +
Ben Avison bavi...@riscosopen.org wrote:
There are many types of composite operation that are useful to benchmark but
which are omitted from the table. Continually having to add extra entries to
the table is a nuisance and is prone to human error, so this
On Tue, 3 Mar 2015 15:24:16 +
Ben Avison bavi...@riscosopen.org wrote:
Missing lsl meant that cases with a 32-bit source and/or mask, and an
8-bit destination, the code would not assemble.
---
pixman/pixman-arm-simd-asm.h |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
On Mon, 16 Mar 2015 12:11:28 +0100
Thilo Cestonaro th...@cestona.ro wrote:
Hey!
I try to understand how weston's fbdev compositor trys to transform the
screen.
As it uses pixman to do this, I come here to hopefully get some answers
:).
Can someone briefly explain the following
On Tue, 3 Mar 2015 15:24:18 +
Ben Avison bavi...@riscosopen.org wrote:
This reduces code size and eliminates the possibility of cut-and-paste errors
when extending the enums. It also highlighted a mismatch between the set of
formats accepted by format_name() and format_from_string(). For
On Tue, 3 Mar 2015 15:24:20 +
Ben Avison bavi...@riscosopen.org wrote:
---
test/Makefile.sources |1 +
test/affine-bench.c | 330
+
2 files changed, 331 insertions(+), 0 deletions(-)
create mode 100644 test/affine-bench.c
On Mon, 30 Mar 2015 10:25:04 -0700
Matt Turner matts...@gmail.com wrote:
On Mon, Mar 30, 2015 at 3:41 AM, Pekka Paalanen ppaala...@gmail.com wrote:
On Mon, 16 Mar 2015 13:56:53 +0200
Pekka Paalanen ppaala...@gmail.com wrote:
On Tue, 3 Mar 2015 15:24:16 +
Ben Avison bavi
,
that's what Siarhei suggested below. It seems to work too, I just got
it implemented.
Thanks,
pq
On Tue, Mar 3, 2015 at 8:49 AM, Pekka Paalanen ppaala...@gmail.com wrote:
On Mon, 2 Mar 2015 17:33:16 +0200
Siarhei Siamashka siarhei.siamas...@gmail.com wrote:
In some cases, the source
Hi,
I've been trying to use source clipping in Weston's Pixman-renderer,
and I found something that seems strange.
The scenario is that I have:
- source clip in source image coordinates
- transformation (matrix)
- destination clip in destination coordinates
- potentially a mask (to apply global
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 to push them?
I'm not capable of reviewing these.
Since Søren isn't
On Wed, 08 Apr 2015 02:27:48 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Tue, 17 Mar 2015 11:12:53 -, Pekka Paalanen ppaala...@gmail.com
wrote:
If there is no transform, why not return the original extents?
These have been reduced by a half unit in all four directions.
It's
On Wed, 8 Apr 2015 14:20:09 +0100
Ben Avison bavi...@riscosopen.org wrote:
---
.gitignore|1 +
test/Makefile.sources |1 +
test/affine-bench.c | 394
+
3 files changed, 396 insertions(+), 0 deletions(-)
create mode
On Thu, 23 Apr 2015 17:12:59 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Thu, 23 Apr 2015 12:46:56 +0100,
Pekka Paalanen ppaala...@gmail.com wrote:
I'm guessing yes, because you do that in affine-bench, reversed... am I
on the right track here?
Affine-bench only uses integer
On Tue, 21 Apr 2015 01:14:09 +0100
Ben Avison bavi...@riscosopen.org wrote:
There are a group of combiner types - SRC, OVER_REVERSE, IN, OUT and ADD -
where the source alpha affects only the destination alpha component. This
means that any fast path with a8r8g8b8 source and destination can
On Thu, 23 Apr 2015 20:28:58 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Thu, 23 Apr 2015 13:10:10 +0100, Pekka Paalanen ppaala...@gmail.com
wrote:
Affine-bench differs from lowlevel-blt-bench in the following:
- does not test different sized operations fitting to specific caches
On Fri, 24 Apr 2015 13:00:19 -0700
Bill Spitzak spit...@gmail.com wrote:
On 04/24/2015 07:31 AM, Pekka Paalanen wrote:
Would it be good to set a rule of thumb, that when you are scaling
images, first compute the final size, and then compute the
transformation paramaters for Pixman
On Wed, 22 Apr 2015 15:12:57 +0100
Ben Avison bavi...@riscosopen.org wrote:
This places a heavier emphasis on solid images than the other fuzz testers,
and tests both single-pixel repeating bitmap images as well as those created
using pixman_image_create_solid_fill(). In the former case, it
On Thu, 07 May 2015 19:31:36 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Wed, 06 May 2015 13:25:43 +0100, Pekka Paalanen ppaala...@gmail.com
wrote:
this is written based on blitters-test.c, right? That would have been
interesting to mention, so new reviewers like me can compare
, tested with both my reference build and a x86_64 optimized
build, and both pass.
With that one fix added,
Reviewed-by: Pekka Paalanen pekka.paala...@collabora.co.uk
I look forward to pushing this, say, mid next week. I can add that fix
myself, too.
Thanks,
pq
On Mon, 27 Apr 2015 09:58:18 +0100
James Cowgill james...@cowgill.org.uk wrote:
On Mon, 2015-04-27 at 11:35 +0300, Pekka Paalanen wrote:
On Fri, 24 Apr 2015 17:09:52 +0100 James Cowgill james...@cowgill.org.uk
wrote:
Hi,
Pixman has been failing to build on mips32r2 since 0.32.6
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Define a new token to simplify checking whether fence_malloc() actually
can catch out-of-bounds access.
This will be used in the future to skip tests that rely on fence_malloc
checking functionality.
Signed-off-by: Pekka Paalanen pekka.paala
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Useful for detecting out-of-bounds accesses in composite operations.
This will be used by follow-up patches adding new tests.
Signed-off-by: Pekka Paalanen pekka.paala...@collabora.co.uk
---
test/utils.c | 100
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Hi,
this is in reply to
http://lists.freedesktop.org/archives/pixman/2014-October/003457.html
item 37 (the last one).
I think it would be worthwhile to have a test that uses fence_malloc
for the source buffer and the matrix
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Tests that fence_malloc and fence_image_create_bits actually work: that
out-of-bounds and out-of-row (unused stride area) accesses trigger
SIGSEGV.
If fence_malloc is a dummy (FENCE_MALLOC_ACTIVE not defined), this test
is skipped.
Signed-off
On Tue, 05 May 2015 11:29:32 -0700
Bill Spitzak spit...@gmail.com wrote:
On 05/04/2015 11:57 PM, Pekka Paalanen wrote:
On Wed, 29 Apr 2015 11:44:17 -0700
Bill Spitzak spit...@gmail.com wrote:
Uses a wildcard to handle the majority which end in -test.
---
.gitignore | 45
On Thu, 14 May 2015 00:27:36 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Fri, 08 May 2015 13:45:37 +0100, Pekka Paalanen ppaala...@gmail.com
wrote:
+ok = test_image_faults (PIXMAN_a8r8g8b8, 7, 5);
+ok = test_image_faults (PIXMAN_r8g8b8, 7, 5);
+ok = test_image_faults
On Thu, 14 May 2015 00:22:03 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Fri, 08 May 2015 13:45:35 +0100, Pekka Paalanen ppaala...@gmail.com
wrote:
diff --git a/test/utils.h b/test/utils.h
index fc10524..6b88bf2 100644
--- a/test/utils.h
+++ b/test/utils.h
@@ -86,6 +86,15
On Thu, 14 May 2015 00:25:45 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Fri, 08 May 2015 13:45:36 +0100, Pekka Paalanen ppaala...@gmail.com
wrote:
+pixman_image_t *
+fence_image_create_bits (pixman_format_code_t format,
+ int min_width, int height
On Fri, 8 May 2015 11:38:45 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
On Thu, 7 May 2015 19:32:46 +0100
Ben Avison bavi...@riscosopen.org wrote:
This places a heavier emphasis on solid images than the other fuzz testers,
and tests both single-pixel repeating bitmap images as well
On Thu, 14 May 2015 00:19:40 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Fri, 08 May 2015 13:45:34 +0100, Pekka Paalanen ppaala...@gmail.com
wrote:
Ben, I think you know better what the actual tests for More accurate
COVER_CLIP_NEAREST|BILINEAR should look like. I hope
On Wed, 15 Apr 2015 12:39:09 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
On Tue, 14 Apr 2015 17:15:10 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Tue, 14 Apr 2015 09:36:38 +0100,
Pekka Paalanen ppaala...@gmail.com wrote:
So, can I take it that you gave your Reviewed
From: Pekka Paalanen pekka.paala...@collabora.co.uk
When doing component alpha, use a mask format that has all the color
channels instead of just a8. As Ben Avison explains it:
Lowlevel-blt-bench initialises all its images using memset(0xCC) so an
a8 solid image would be converted
)
+ box_48_16_t *transformed)
{
pixman_fixed_48_16_t tx1, ty1, tx2, ty2;
pixman_fixed_t x1, y1, x2, y2;
Reviewed-by: Pekka Paalanen pekka.paala...@collabora.co.uk
I'll push this on Friday unless anyone objects. I heard I should have
push access now. This is a nice small
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Previously there was a flat list of operators (pixman_op_t), used to
iterate over all operators when looking up an operator from name or
listing them. This cannot support name aliases.
To support name aliases (multiple name strings mapping
From: Ben Avison bavi...@riscosopen.org
This permits format_from_string(), list_formats(), list_operators() and
operator_from_string() to be used from tests other than check-formats.
Reviewed-by: Pekka Paalanen pekka.paala...@collabora.co.uk
---
test/check-formats.c | 192
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Let lowlevel-blt-bench parse the test name string from the command line,
allowing to run almost infinitely more tests. One is no longer limited
to the tests listed in the big table.
While you can use the old short-hand names like src__
From: Pekka Paalanen pekka.paala...@collabora.co.uk
This patch is inspired by lowlevel-blt-bench: Parse test name strings in
general case by Ben Avison. From Ben's commit message:
There are many types of composite operation that are useful to benchmark
but which are omitted from the table
On Tue, 14 Apr 2015 15:16:01 -0400
Lennart Sorensen lsore...@csclub.uwaterloo.ca wrote:
On Tue, Apr 14, 2015 at 07:20:50PM +0100, Ben Avison wrote:
On Tue, 14 Apr 2015 11:28:51 +0100, Pekka Paalanen ppaala...@gmail.com
wrote:
you asked about reading the cache sizes; I have no idea about
On Tue, 14 Apr 2015 17:15:10 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Tue, 14 Apr 2015 09:36:38 +0100,
Pekka Paalanen ppaala...@gmail.com wrote:
On Mon, 13 Apr 2015 18:42:45 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Mon, 13 Apr 2015 12:31:35 +0100,
Pekka Paalanen
Hi,
most pixman performance benchmarks currently rely on gettime() from
test/util.[ch]:
- lowlevel-blt-bench
- prng-test
- radial-perf-test
- scaling-bench
Furthermore, affine-bench has its own gettimei() which is essentially
gettime() but with uin32_t instead of double.
double
gettime (void)
{
On Tue, 2 Jun 2015 15:03:01 -0700
Bill Spitzak spit...@gmail.com wrote:
I would have the first call return 0.0 and all the others return the
difference between current time and when that first call was done. Then
there is no worry about accuracy of floating point. I do not think any
callers
On Wed, 3 Jun 2015 08:38:25 +0300
Siarhei Siamashka siarhei.siamas...@gmail.com wrote:
On Tue, 2 Jun 2015 10:32:35 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
Hi,
most pixman performance benchmarks currently rely on gettime() from
test/util.[ch]:
- lowlevel-blt-bench
- prng
On Wed, 3 Jun 2015 10:51:25 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
On Wed, 3 Jun 2015 08:38:25 +0300
Siarhei Siamashka siarhei.siamas...@gmail.com wrote:
On Tue, 2 Jun 2015 10:32:35 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
I would propose the following
From: Pekka Paalanen pekka.paala...@collabora.co.uk
When given just a single test pattern instead of all, print the test
details. This can be used to verify the pattern parser agrees with the
user, just like scaling settings are printed.
Signed-off-by: Pekka Paalanen pekka.paala
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Move explanation printing to a new file. This will help with
implementing a machine-readable output option.
Signed-off-by: Pekka Paalanen pekka.paala...@collabora.co.uk
---
test/lowlevel-blt-bench.c | 60
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Move printing of usage into a new function and use argv[0] as the
program name. This will help printing usage from multiple places.
Signed-off-by: Pekka Paalanen pekka.paala...@collabora.co.uk
---
test/lowlevel-blt-bench.c | 12 +---
1
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Refactor the Mpixels/s computations into a function. Easier to read and
better documents what is being computed.
Signed-off-by: Pekka Paalanen pekka.paala...@collabora.co.uk
---
test/lowlevel-blt-bench.c | 23 ---
1 file
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Hi,
most of the patches are trivial cleanups. The meat are the last two:
CSV output mode and skipping the memory speed benchmark.
Both new features are designed for an external benchmarking harness,
that runs several different versions
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Move the printing of the memory speed and scaling mode into a new
function. This will help with implementing a machine-readable output
option.
Signed-off-by: Pekka Paalanen pekka.paala...@collabora.co.uk
---
test/lowlevel-blt-bench.c | 37
From: Pekka Paalanen pekka.paala...@collabora.co.uk
The bench_* functions, that did not already do it, are modified to
return the number of pixels processed during the benchmark. This moves
the computation to the site that actually determines the number, and
simplifies bench_composite() a bit
From: Pekka Paalanen pekka.paala...@collabora.co.uk
We assign string literals to it, so it better be const.
Signed-off-by: Pekka Paalanen pekka.paala...@collabora.co.uk
---
test/lowlevel-blt-bench.c | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/test/lowlevel
From: Pekka Paalanen pekka.paala...@collabora.co.uk
The memcpy speed measurement takes several seconds. When you are running
single tests in a harness that iterates dozens or hundreds of times, the
repeated measurements are redundant and take a lot of time. It is also
an open question whether
On Wed, 03 Jun 2015 17:47:47 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Wed, 03 Jun 2015 08:51:25 +0100, Pekka Paalanen ppaala...@gmail.com
wrote:
If we fixed gettime() for clock() wraparounds rather than ignoring them,
I suppose there would be no reason to have gettimei().
Ben
On Tue, 16 Jun 2015 18:27:59 +0300
Oded Gabbay oded.gab...@gmail.com wrote:
From: Fernando Seiti Furusato ferse...@linux.vnet.ibm.com
Replaced usage of vec_lvsl to direct unaligned assignment
operation (=). That is because, according to Power ABI Specification,
the usage of lvsl is
On Tue, 26 May 2015 23:58:30 +0100
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
On Tue, 26 May 2015 23:58:25 +0100
Ben Avison bavi...@riscosopen.org wrote:
This macro does exactly the same thing as the platform-neutral macro
SIMPLE_NEAREST_FAST_PATH.
---
This is one of a number of (independent) patches arising from a survey of
scaled fast paths which highlighted the
)
/*/
Reviewed-by: Pekka Paalanen pekka.paala...@collabora.co.uk
Will push in a few days, if no-one objects.
Thanks,
pq
___
Pixman mailing list
Pixman@lists.freedesktop.org
http
, is that ok?
If the order is ok, then:
Reviewed-by: Pekka Paalanen pekka.paala...@collabora.co.uk
Thanks,
pq
___
Pixman mailing list
Pixman@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pixman
)
-
//
#define PIXMAN_MIPS_BIND_SCALED_BILINEAR_SRC_DST(flags, name, op,
\
Reviewed-by: Pekka Paalanen pekka.paala...@collabora.co.uk
Will push in a few days, if no-one objects.
Thanks,
pq
is ok, then:
Reviewed-by: Pekka Paalanen pekka.paala...@collabora.co.uk
Thanks,
pq
___
Pixman mailing list
Pixman@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pixman
On Thu, 28 May 2015 16:08:55 -0300
Fernando Seiti Furusato ferse...@linux.vnet.ibm.com wrote:
Hi Ben.
On 05/28/2015 01:57 PM, Fernando Seiti Furusato wrote:
Hello Ben.
On 05/27/2015 09:40 AM, Ben Avison wrote:
On Tue, 26 May 2015 19:16:32 +0100, I wrote:
+color.alpha =
On Mon, 1 Jun 2015 13:40:52 +0300
Siarhei Siamashka siarhei.siamas...@gmail.com wrote:
On Thu, 28 May 2015 16:53:16 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
I have made a branch with my R-b's and amended commit messages at:
https://git.collabora.com/cgit/user/pq/pixman.git/log/?h
On Fri, 29 May 2015 16:20:43 +0100
Ben Avison bavi...@riscosopen.org wrote:
Tested-by: Fernando Seiti Furusato ferse...@linux.vnet.ibm.com (ppc64le,
ppc64, powerpc)
Tested-by: Ben Avison bavi...@riscosopen.org (armv6l, armv7l, i686)
---
test/solid-test.c |9 ++---
1 files changed,
On Thu, 25 Jun 2015 14:41:37 +0300
Oded Gabbay oded.gab...@gmail.com wrote:
On Thu, Jun 25, 2015 at 2:05 PM, Pekka Paalanen ppaala...@gmail.com wrote:
On Tue, 16 Jun 2015 18:27:59 +0300
Oded Gabbay oded.gab...@gmail.com wrote:
From: Fernando Seiti Furusato ferse...@linux.vnet.ibm.com
On Thu, 02 Jul 2015 17:45:13 +0100
Ben Avison bavi...@riscosopen.org wrote:
On Wed, 10 Jun 2015 14:32:49 +0100, Pekka Paalanen ppaala...@gmail.com
wrote:
most of the patches are trivial cleanups. The meat are the last two:
CSV output mode and skipping the memory speed benchmark.
Both
From: Pekka Paalanen pekka.paala...@collabora.co.uk
If a user gives multiple patterns or extra arguments, only the last one
was used as the pattern while the former were just ignored. This is a
user error silently converted to something possibly unexpected.
In presence of extra arguments
On Tue, 7 Jul 2015 11:35:04 +0300
Pekka Paalanen ppaala...@gmail.com wrote:
From: Pekka Paalanen pekka.paala...@collabora.co.uk
If a user gives multiple patterns or extra arguments, only the last one
was used as the pattern while the former were just ignored. This is a
user error silently
on adding support
for ppc64le to pixman (fixing vmx fast-paths and adding new
implementation). Some of the patches have been upstreamed (by Pekka
Paalanen) and some are in the process of review (by Siarhei and
others).
From reading the above email thread, and from talking to Soren, Pekka
and others
On Thu, 20 Aug 2015 13:07:48 +0100
Ben Avison bavi...@riscosopen.org wrote:
---
I've been doing a fair bit of work on scaled plots again recently, and I was
finding that scaling-test omitted to mention lots of useful information
(notably whether a mask was used, and what the mask parameers
message]
Signed-off-by: Pekka Paalanen pekka.paala...@collabora.co.uk
---
pixman/pixman-arm-simd-asm.S | 41 +
pixman/pixman-arm-simd.c | 7 +++
2 files changed, 48 insertions(+)
diff --git a/pixman/pixman-arm-simd-asm.S b/pixman/pixman-arm-simd-asm.S
From: Pekka Paalanen pekka.paala...@collabora.co.uk
Enable the fast path added in the previous patch by moving the lookup
table entries to their proper locations.
Lowlevel-blt-bench benchmark statistics with 30 iterations, showing the
effect of adding this one patch on top of
pixman-fast-path
of benchmarks.
Benchmark results are included in the patch enabling this fast path.
[Pekka: disabled the fast path, commit message]
Signed-off-by: Pekka Paalanen pekka.paala...@collabora.co.uk
---
pixman/pixman-fast-path.c | 36
1 file changed, 36 insertions(+)
diff
Hi all,
as I promised a long time ago, probably in IRC, Patchwork for Pixman is
now up:
http://patchwork.freedesktop.org/project/pixman/list/
It has been there for a while, but only recently it was cleaned up from
old patch submissions. A huge thank-you to Oded Gabbay for the
clean-up, and to
On Tue, 14 Jul 2015 15:48:54 +0200
Thomas Petazzoni thomas.petazz...@free-electrons.com wrote:
Hello,
On Tue, 14 Jul 2015 15:34:30 +0300, Pekka Paalanen wrote:
There is one more bit of automation: when a commit gets pushed into the
upstream git repository, a git hook will run
On Tue, 14 Jul 2015 11:41:25 +0300
Siarhei Siamashka siarhei.siamas...@gmail.com wrote:
On Thu, 2 Jul 2015 13:04:11 +0300
Oded Gabbay oded.gab...@gmail.com wrote:
POWER8, 8 cores, 3.4GHz, RHEL 7.1 ppc64le.
reference memcpy speed = 24764.8MB/s (6191.2MP/s for 32bpp fills)
*Very*
On Wed, 15 Jul 2015 15:05:21 +0300
Oded Gabbay oded.gab...@gmail.com wrote:
On Wed, Jul 15, 2015 at 2:59 PM, Pekka Paalanen ppaala...@gmail.com wrote:
On Tue, 14 Jul 2015 11:41:25 +0300
Siarhei Siamashka siarhei.siamas...@gmail.com wrote:
On Thu, 2 Jul 2015 13:04:11 +0300
Oded Gabbay
only accepting a SIGSEGV as
> notification of invalid access.
>
> Fixes fence-image-self-test
>
> Reviewed-by: Pekka Paalanen <pekka.paala...@collabora.co.uk>
> ---
> test/fence-image-self-test.c | 12 +++-
> test/utils.c | 6 +++---
> 2 files cha
1 - 100 of 180 matches
Mail list logo