The non-reentrant versions of prng_* functions are thread-safe only in
OpenMP-enabled builds.

F...

Handle SSSE3 just like MMX and SSE2.

After a4c79d695d52c94647b1aff78548e5892d616b70 the MMX and SSE2 code
has some declarations after...

Fixes

check-formats.obj : error LNK2019: unresolved external symbol
_strcasecmp referenced in f...

In d1434d112ca5cd325e4fb85fc60afd1b9e902786 the benchmarks have been
extended to include other p...

The generated fast paths that were moved into the 'fast'
implementation in ec0e38cbb746a673f8e98...

So the redundant variables, memory reads/writes and reshuffles
can be safely removed. For exampl...

The SIMD optimized inner loops in the VMX/Altivec code are trying
to emulate unaligned accesses ...

This test program allocates an array of 16 * 7 uint32_ts and spawns 16
threads that each use 7 o...

The test for pthread_setspecific() can be used as a general test for
whether pthreads are availa...

Use a temporary variable s containing the absolute value of the stride
as the upper bound in the...

Commit 4312f077365bf9f59423b1694136089c6da6216b claimed to have made
print_image() work with neg...

The generated fetchers for NEAREST, BILINEAR, and
SEPARABLE_CONVOLUTION filters are fast paths a...

This iterator is really a fast path, so it belongs in the fast path
implementation.

Instead of having logic to swap the lines around when one of them
doesn't match, store the two l...

Pixman supports negative strides, but up until now they haven't been
tested outside of stress-te...

The affine-test, blitters-test, and scaling-test all have the ability
to print out the bytes of ...

By using this function instead of compute_crc32() the alpha masking
code and the call to image_e...

Converting a double precision number to 16.16 fixed point should be
done by multiplying with 655...

The separable convolution filter supports a subsample_bits of 0 which
corresponds to no subsampl...

This commit adds a new, empty SSSE3 implementation and the associated
build system support.

con...

At the moment iter buffers are only guaranteed to be aligned to a 4
byte boundary. SIMD implemen...

This new iterator uses the SSSE3 instructions pmaddubsw and pabsw to
implement a fast iterator f...

The loops are already unrolled, so it was just a matter of packing
4 pixels into a single XMM re...

The calloc call from pixman_image_create_bits may still
rely on http://en.wikipedia.org/wiki/Cop...

The default has been 7-bit for a while now, and the quality
improvement with 8-bit precision is ...

Scanline fetchers haven't been used for images other than bits for a
long time, so by making the...

Later versions of gcc-4.7.x are capable of generating iwMMXt
instructions properly, but gcc-4.8 ...

No memory is allocated in the error case, so a finalizer is not
necessary, and will cause proble...

This new iterator works in a separable way; that is, for a destination
scaline, it scales the tw...

This new benchmark scales a 320 x 240 test a8r8g8b8 image by all
ratios from 0.1, 0.2, ... up to...

Iterators may sometimes need to allocate auxillary memory. In order to
be able to free this memo...

The MSVC compiler is very strict about variable declarations after
statements.

Move all the dec...

I'm got bug in my system:

lcc: "scale.c", line 374: warning: function "gtk_scale_add_mark" decl...

Reviewed-by: Matt Turner <[email protected]>
Signed-off-by: Matthieu Herrb <matthieu.herrb@laas...

In preparation for sharing all iterator initialization code from all
the implementations, move t...

The SSE2, MMX, and fast implementations all have a copy of the
function iter_init_bits_stride th...

Similar to the SSE2 commit, information about the iterators is stored
in a table of pixman_iter_...

Now that we are using the new _pixman_implementation_iter_init(), the
old _src/_dest_iter_init()...

Similar to the SSE2 and MMX patches, this commit replaces a table of
fetcher_info_t with a table...

A new field, 'iter_info', is added to the implementation struct, and
all the implementations sto...

Similar to the changes to noop, put all the iterators into a table of
pixman_iter_info_t and the...

Instead of having a nest of if statements, store the information about
iterators in a table of a...

We only support alpha maps for BITS images, so it's always to ignore
the alpha map for non-BITS ...

This will be useful for putting iterators into tables where they can
be looked up by iterator fl...

These indicate whether the iterator is for a source or a destination
image. Note iterator initia...

Similar to c2230fe2aff, simply check against SAMPLES_COVER_CLIP_NEAREST
instead of comparing all...

The Loongson code is compiled with -march=loongson2f to enable the MMI
instructions, but binutil...

I look at that function and can never remember what it does or how it
manages to do it.

Build fix for platforms without a generated config.h, for example Win32.

Essentially all of it is obsolete by now.

Performance numbers before/after on MIPS-74kc @ 1GHz:

lowlevel-blt-bench results

Referent (bef...

Add necessary support to lowlevel-blt benchmark for benchmarking pixbuf and
rpixbuf fast paths. ...

Rounding logic was not implemented right.
Instead of using rounding version of the 8-bit shift, ...

Performance numbers before/after on MIPS-74kc @ 1GHz:

lowlevel-blt-bench results

Referent (bef...

After introducing new PRNG (pseudorandom number generator) a bug in two DSPr2
routines was revea...

Performance numbers before/after on MIPS-74kc @ 1GHz:

lowlevel-blt-bench results

Referent (bef...

Performance numbers before/after on MIPS-74kc @ 1GHz:

lowlevel-blt-bench results

Referent (bef...

Performance numbers before/after on MIPS-74kc @ 1GHz:

lowlevel-blt-bench results

Referent (bef...

The old code was calculating horizontal weights for right pixels
in the following way (for simpl...

Current blitters-test program had difficulties detecting a bug in
over_n_8888_8888_ca implementa...

They were found by codespell.

Signed-off-by: Stefan Weil <[email protected]>

Signed-off-by: Peter Breitenlohner <[email protected]>

The computations in pixman-gradient-walker.c currently take place at
very limited 8 bit precisio...

Some upcoming changes to pixman-gradient-walker.c will need this
macro.

This benchmark renders one of the radial gradients used in the
swfdec-youtube cairo trace 500 ti...

This program displays a linear gradient from blue to yellow. Due to
limited precision in pixman-...

- src_0888_8888_rev
- src_0888_0565_rev

Performance numbers before/after on MIPS-74kc @ 1GHz:
...

- over_8888_0565
- over_n_8_8

Performance numbers before/after on MIPS-74kc @ 1GHz:

lowlevel-...

GdkPixbufs are not premultiplied, so when using them to display pixman
images, there is some une...

The source, mask and destination buffers are initialised to 0xCC just after
they are allocated. ...

Some recent code added new type casts from pointer to unsigned long.
These type casts result in ...

This function takes a pixel in the format corresponding to the pixel
checker, and converts to a ...

This function returns the a, r, g, and b masks corresponding to the
pixel checker's format.

While releasing 0.29.2 the distcheck run produced a number of error
messages that had to be fixe...

When compiling with GCC this macro expands to __builtin_expect((expr), 0).
On other compilers, i...

The check-formats programs reveals that the 8 bit pipeline cannot meet
the current 0.004 accepta...

The check-formats.c test depends on the exact format of the strings
returned from these function...

Given an operator and two formats, this program will composite and
check all pixels where the re...

If we fail to find a composite function, don't update the fast path
cache with the dummy composi...

So that it can be used in other tests.

This test program contains a table of individual operator/pixel
combinations. For each pixel com...

In c2cb303d33ec11390b93cabd90f0f9, return_if_fail()s were added to
prevent the trapezoid rasteri...

The old one belongs to the email address [email protected], which
doesn't work anyore.

Also ...

In particular this affects single-core ARMs (e.g. ARM11, Cortex-A8), which
are usually configure...

The clip_color() function has some checks to avoid division by zero,
but they are done by compar...

Improved by adding preloads, combining writes and using the SEL
instruction.

add_8_8

Befor...

There was no previous attempt at accelerating these specifically for
ARMv6.

src_x888_8888

...

These are usable either as various composite operations, or via the
top-level function pixman_bl...