lp://staging/~tom-gall/linaro/libjpeg-turbo

Branch merges

Propose for merging

No branches dependent on this one.

Related source package recipes

1 recipe using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

11. By Tom Gall on 2011-03-07

* address small set of lintian warnings.
  fixup libjpeg-turbo62.shlibs
  add override for lintian warning about libjpeg
* add a test to validate libjpeg62 is installed before installing
  dpkg-divert

10. By Siarhei Siamashka <email address hidden> on 2010-11-10

ARM NEON optimizations for quantization code in 'forward_DCT'

Integer divisions replaced with floating point multiplications.
Code is still bitexact (calculations produce the same results
for all the possible range of input values).

This optimized function is now almost 3x faster than original
C variant and improves encoding performance somewhat.

9. By Siarhei Siamashka <email address hidden> on 2010-11-10

ARM assembly optimizations for 'encode_one_block'

Almost 2x faster than original C variant.

8. By Siarhei Siamashka <email address hidden> on 2010-11-10

ARM NEON optimized version of 'rgb_ycc_convert'

Is approximately 10x faster than original C variant.

7. By Siarhei Siamashka <email address hidden> on 2010-11-10

ARM NEON optimized version of 'ycc_rgb_convert'

Is approximately 6x faster than original C variant.

6. By Siarhei Siamashka <email address hidden> on 2010-11-10

A minor ARM NEON optimization for 'convsamp'

5. By Siarhei Siamashka <email address hidden> on 2010-11-10

ARM NEON optimized version of 'jpeg_fdct_ifast'

Is approximately 4x faster than original C variant.

4. By Siarhei Siamashka <email address hidden> on 2010-11-10

ARM NEON optimized version of 'jpeg_idct_ifast'

Is approximately 4x faster than original C variant.

3. By Siarhei Siamashka <email address hidden> on 2010-11-10

ARM NEON optimized version of 'jpeg_idct_4x4'

Is approximately 4x faster than original C variant.

2. By Siarhei Siamashka <email address hidden> on 2010-11-10

Added support for ARM optimizations to configure.ac

ARM NEON optimizations need the use of 16-bit data types and
benefit from 16-byte alignemnt, so we want to have macro
WITH_SIMD defined in the code.

Also automated tests are updated to work correctly on ARM.
And the required stubs for ARM optimizations added (just
copied from 'jsimd_none.c').