summaryrefslogtreecommitdiff
path: root/celt
Commit message (Collapse)AuthorAgeFilesLines
* Whitespace fixesexp_mips_optJean-Marc Valin2014-06-193-3/+1
|
* MIPS optimizationsRhishikesh Agashe2014-06-1913-0/+1040
| | | | Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Adds SIG2WORD16() to fixed_debug.hJean-Marc Valin2014-06-181-0/+10
|
* Suppress some coverity false positives.Gregory Maxwell2014-04-162-1/+2
|
* Fix iOS builds with assembly.Timothy B. Terriberry2014-03-261-3/+3
| | | | | | | | | | | | The patch in 76e831d917ff got us most of the way there, but out-of-tree builds required a second Makefile.am rule, which was missing @ARM2GNU_PARAMS@. Also, the arm2gnu.pl was terminating argument processing on any argument beginning with --, rather than an argument that was just -- by itself (as is the normal convention in GNU programs). That meant it never saw the --apple flag even when it was passed. Thanks to Jonathan Lennox for the report and for testing.
* Make the arm2gnu.pl converter handle apple specific detailsMartin Storsjo2014-03-191-6/+17
| | | | | | | | | This allows building the arm assembly for iOS. This checks for the __APPLE__ preprocessor built-in define to determine whether this extra handling should be enabled. Signed-off-by: Timothy B. Terriberry <tterribe@xiph.org>
* Add separate labels for the start of public functionsMartin Storsjo2014-03-191-2/+4
| | | | | | | This avoids having to use the public symbol name when jumping here, on platforms where the public symbols have an underscore prefix. Signed-off-by: Timothy B. Terriberry <tterribe@xiph.org>
* arm: Use the UAL syntax for instructionsMartin Storsjo2014-02-242-9/+11
| | | | | | | This is required in order to build using the built-in assembler in clang. Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Fixes an aliasing bug in the MDCT when the frame size isn't a multiple of 4.Marcello Caramma (mcaramma)2014-02-241-2/+2
| | | | | | | In that case, the yp0 and yp1 ended up aliasing for the last element, despite being marked as restrict. Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Adds missing license for arm2gnu.plJean-Marc Valin2014-02-241-0/+24
|
* Skips comb filter overlap when the parameters didn't change.Jean-Marc Valin2014-01-311-0/+3
|
* Minor div optimization -- reducing denominatorsJean-Marc Valin2014-01-302-2/+2
|
* Optimizing divisions with a signed numeratorJean-Marc Valin2014-01-302-2/+15
|
* Speeding up ec_tell_frac()Jean-Marc Valin2014-01-291-0/+22
|
* Speeding up extract_collapse_mask() slightlyJean-Marc Valin2014-01-231-1/+3
|
* Using SSAT in SIG2WORD16() on ARMv6Jean-Marc Valin2014-01-224-11/+29
|
* Unrolled version of the comb filter for ARM (eliminates MOVs)Jean-Marc Valin2014-01-211-2/+32
|
* Making decode_pulses() compute the L2-norm on the flyJean-Marc Valin2014-01-213-19/+31
|
* Speed up the comb filter on ARM by using MAC16_32_Q16()Jean-Marc Valin2014-01-216-1/+56
|
* Save more integer divisions on ARM when we know the operands are positiveJean-Marc Valin2014-01-215-13/+18
|
* Fixes use of uninitialized values in dynalloc_analysis()Jean-Marc Valin2014-01-201-3/+3
|
* Using a table on ARM for unsigned division by small (<=256) integers.Jean-Marc Valin2014-01-204-2/+62
| | | | Saves 0.6% for 64 kb/s and 1.8% for 128 kb/s when decoding on arm7tdmi.
* Minor cleanup in dynalloc_analysis()Jean-Marc Valin2014-01-191-9/+11
|
* Adds a median filter to make dynalloc_analysis() more conservativeJean-Marc Valin2014-01-191-0/+76
| | | | | This should prevent extreme dynalloc behaviour in cases where some bands are heavily attenuated.
* Fix declaration after statement in fixed point.Gregory Maxwell2014-01-081-2/+2
|
* pseudostack instrumentation (off by default)Jean-Marc Valin2014-01-071-1/+7
|
* Fixes SMALL_FOOTPRINT for floatJean-Marc Valin2014-01-071-4/+7
|
* Don't allocate pulses on the stack when calling the SILK PLC.Jean-Marc Valin2014-01-071-1/+1
| | | | Also minor C89 fix for the previous commit
* Moves CELT PLC pitch search to a separate function to reduce peak stackJean-Marc Valin2014-01-071-9/+17
|
* Adds SMALL_FOOTPRINT hack to the CELT PLC tooJean-Marc Valin2014-01-071-2/+10
|
* Moves deemphasis() call out of celt_decode_lost() to reduce peak stackJean-Marc Valin2014-01-071-11/+7
|
* Cleaning up leftovers of "freq" in celt_decode_with_ec()Jean-Marc Valin2014-01-061-2/+0
|
* Reduces the decoder stack use by removing the pcm_silk buffer in fixed-pointJean-Marc Valin2014-01-063-19/+50
| | | | We only keep when concealing less than 10ms with SILK.
* Moves the remains of compute_inv_mdcts() to celt_synthesis()Jean-Marc Valin2014-01-061-37/+24
|
* Hack that makes the SMALL_FOOTPRINT CELT decoder use only 4.25 kB of stack.Jean-Marc Valin2014-01-061-8/+16
|
* Reduces decoder stack usage by only storing one channel of denormalized MDCTJean-Marc Valin2014-01-065-98/+125
|
* Some cleaning up of the synthesis code.Jean-Marc Valin2014-01-054-31/+24
|
* Making exp_rotation1() use MAC16_16(), which saves a few cycles on ARMJean-Marc Valin2014-01-041-4/+6
|
* Silences unused parameter warningJean-Marc Valin2014-01-041-0/+1
|
* Moving the radix-2 to expose trivial twiddle factorsJean-Marc Valin2014-01-033-46/+73
|
* Improving the accuracy of the fixed-point radix-3 and radix-5Jean-Marc Valin2013-12-311-3/+14
|
* Minor cleanup -- nothing to see hereJean-Marc Valin2013-12-293-10/+2
|
* Fixed-point: slight accuracy improvement in the comb filterJean-Marc Valin2013-12-291-6/+6
|
* Remove a SAVE_STACK that was pasted accidentally in the previous commitJean-Marc Valin2013-12-291-1/+0
|
* Unifying scaling of fixed-point and float FFTJean-Marc Valin2013-12-295-33/+32
|
* Fixes C89 issueJean-Marc Valin2013-12-291-2/+4
|
* Getting rid of some negationsJean-Marc Valin2013-12-282-10/+15
| | | | | Since we're doing two rotations, we can invert the sign on both. Also adding a few comments for optimizing the FFT.
* Slightly improving the accuracy of the fixed-point MDCT downscaleJean-Marc Valin2013-12-281-13/+9
| | | | Also simplifying the code
* Commit 99968ab was causing us to allocate too much stack in the MDCTJean-Marc Valin2013-12-271-1/+1
|
* Removes the separate 1/8N rotation in the (I)MDCT and unmerges the MDCT sizesJean-Marc Valin2013-12-274-260/+794
| | | | Undoes commits f7547a4e and 72513f3c