Lovell Fuller 0cac224431 Replace use of "Duff's device" with for-loop to allow compiler to optimise. ...

Reduces CPU instruction count of conv_gen function by ~30%.
Improves performance of convolution operation by 20-25%.

2016-01-25 21:17:41 +00:00

..

compass.c

use g_assert_not_reached()

2016-01-24 13:21:53 +00:00

conv.c

use g_assert_not_reached()

2016-01-24 13:21:53 +00:00

convolution.c

more pyvips8 docs

2014-11-07 14:49:18 +00:00

convsep.c

rename VIPS_ANGLE_0 as VIPS_ANGLE_D0

2014-09-29 13:56:55 +01:00

correlation.c

add copy-to-memory

2015-04-19 09:45:20 +01:00

correlation.h

redo correlation funcs as classes

2013-11-08 14:09:42 +00:00

fastcor.c

use g_assert_not_reached()

2016-01-24 13:21:53 +00:00

gaussblur.c

mask builders take prec, not integer

2014-12-16 20:48:11 +00:00

im_aconv.c

use g_assert_not_reached()

2016-01-24 13:21:53 +00:00

im_aconvsep.c

use g_assert_not_reached()

2016-01-24 13:21:53 +00:00

im_conv_f.c

Replace use of "Duff's device" with for-loop to allow compiler to optimise.

2016-01-25 21:17:41 +00:00

im_conv.c

Replace use of "Duff's device" with for-loop to allow compiler to optimise.

2016-01-25 21:17:41 +00:00

Makefile.am

deprecate im_zerox(), not very useful

2014-01-17 18:03:46 +00:00

pconvolution.h

start vips_conv()

2013-10-01 14:26:13 +01:00

sharpen.c

mask builders take prec, not integer

2014-12-16 20:48:11 +00:00

spcor.c

use g_assert_not_reached()

2016-01-24 13:21:53 +00:00