aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
| * Make sure FP16 tests are enabled only if the target actually supports them.Christophe Lyon2016-08-2417-125/+125
| | | | | | | | | | This implies checking __ARM_FP or __ARM_NEON_FP16_INTRINSICS to account for different compilers.
| * __set_neon_cumulative_sat() modifies the contents on the QC flag, andChristophe Lyon2015-01-2025-85/+90
| | | | | | | | | | | | | | | | | | | | | | | | | | some intrinsics do so too: this patch adds the explicit dependency on the asm statement, to avoid code reordering or removal. When writing QC, the asm statement now has a fake input dependency, which is the output of the intrinsic being tested. Modifying the __set_neon_cumulative_sat macro is necessary, to be able to accept all the possible input types. Update the generic code in ref_v_binary_sat_op.c and ref_v_unary_sat_op.c accordingly, as well as all the tests involving QC.
| * Fix comment formatting.Christophe Lyon2015-01-202-2/+2
| |
| * Fix typo in comment.Christophe Lyon2015-01-161-1/+1
| |
| * Rename helper macros:Christophe Lyon2014-12-16100-1688/+1687
| | | | | | | | | | TEST_VLOAD into VLOAD TEST_VDUP into VDUP
| * Add CPU variable in Makefiles.Christophe Lyon2014-12-162-6/+9
| |
| * Change vldX_dup test input buffer.Christophe Lyon2014-11-281-3/+1
| |
| * Fix detection of FP16 support: test __ARM_FP16_FORMAT_IEEE instead of ↵Christophe Lyon2014-09-0216-128/+126
| | | | | | | | __ARM_NEON_FP16_INTRINSICS.
| * Remove 2 incorrect entries in VCVT.Christophe Lyon2014-09-021-2/+0
| |
| * Make sure poly* types are handled as unsigned when they get printed.Christophe Lyon2014-07-113-38/+63
| |
| * Remove floating-point outputs to avoid miscomparison of results when using ↵Christophe Lyon2014-07-115-1742/+1742
| | | | | | | | different libC, some of which can output "-nan", the others always "nan".
| * Dump fp16 values in expected_input4gcc.txt too.Christophe Lyon2014-07-114-11/+7619
| | | | | | | | Add expected_input4gcc-nofp16.txt.
| * Fix vcvt input initialization when FP16 is supported.Christophe Lyon2014-07-111-1/+1
| |
| * Add missing seperator messages in expected_input4gcc.txt to help readability.Christophe Lyon2014-07-1113-0/+62
| |
| * Support Neon Cumulative Saturation flag for aarch64 too.Christophe Lyon2014-07-101-14/+30
| |
| * Enable build for aarch64.Christophe Lyon2014-07-10100-100/+100
| |
| * Add missing GCCTESTS_FILE definition.Christophe Lyon2014-06-131-1/+1
| |
| * Add a new expected results file, useful for the conversion of theChristophe Lyon2014-06-131-0/+7063
| | | | | | | | tests into GCC's dejagnu testsuite.
| * Additional Makefile to make it easier to check a particular GCC version.Christophe Lyon2014-06-131-0/+122
| |
| * Fix comment.Christophe Lyon2014-06-051-2/+1
| |
| * Renamed Neon overflow to Neon cumulative saturation, since it's the real name.Christophe Lyon2014-06-0333-2637/+2669
| |
| * Add new output file, to make conversion to GCC dejagnu testsuite easier.Christophe Lyon2014-05-1627-50/+141
| |
| * Add license description for files derived from ARM examples.Christophe Lyon2014-05-122-0/+466
| |
| * Added support for half-precision (fp16) variants.Christophe Lyon2013-04-1123-5329/+14472
| |
| * Fix typo in poly16_t printf format.Christophe Lyon2013-04-091-1/+1
| |
| * Typo in comments.Christophe Lyon2013-04-091-1/+1
| |
| * Remove useless code.Christophe Lyon2013-03-294-2/+0
| |
| * Update copyright years.Christophe Lyon2013-03-2922-22/+22
| |
| * Add support for polynomial variants (*_p8, *_p16).Christophe Lyon2013-03-2939-7917/+10437
| |
| * Change output format of vget_lane, for better readability.Christophe Lyon2013-03-286-4/+40
| |
| * Be compliant with Intel preprocessor.Christophe Lyon2013-02-221-1/+1
| |
| * Use union instead of cast to avoid GCC warning.Christophe Lyon2013-02-054-1/+7
| |
| * Reorganize input data buffer to try to detect some out of bounds accesses.Christophe Lyon2013-01-1713-92/+557
| | | | | | | | Add dedicated input for vdup, vld1_dup, vldX, vldX_dup, vldX_lane, vstX_lane.
| * Enable compilation of non-Neon tests with a non-ARM compiler.Christophe Lyon2013-01-141-2/+2
| |
| * Add test for vclz with 0 as input.Christophe Lyon2013-01-143-0/+50
| |
| * Fix typo.Christophe Lyon2013-01-143-2/+2
| | | | | | | | Reported by Victoria Zhislina.
| * Fix input data for vld3 and vld4 tests.Christophe Lyon2013-01-143-12/+15
| | | | | | | | Reported by Victoria Zhislina.
| * Fix output of 2nd chunk of data for vuzp, vzip and vtrn.Christophe Lyon2013-01-143-43/+43
| | | | | | | | Reported by Victoria Zhislina.
| * Improve support for endianness of the target processor. (eg when compiling ↵Christophe Lyon2012-08-311-1/+15
| | | | | | | | with llvm).
| * Update binary produced by GCC for ARM. This one produces accurate results.Christophe Lyon2012-06-211-0/+0
| |
| * Update binaries after building with newer binutils.Christophe Lyon2012-06-042-0/+0
| |
| * * Enable compilation & link with GCCChristophe Lyon2012-05-148-85/+6388
| | | | | | | | | | * Enable compilation with GCC and link with armlink * Separate integer/dsp tests as GCC does not support them currently
| * Make sure references to file descriptors do not pull internal CChristophe Lyon2012-05-111-1/+2
| | | | | | | | library dependencies.
| * Update memory ranges to permit linking GCC-generated code with armlink.Christophe Lyon2012-05-111-1/+1
| |
| * Add accessor functions for Neon_Overflow, such that its value can beChristophe Lyon2012-05-1126-34/+48
| | | | | | | | read/changed when compiling with GCC/ARM too.
| * Update expected results with new tests (and update test program as well)Christophe Lyon2012-03-092-32/+46
| |
| * Fix definition of NAN, INF under Win32.Christophe Lyon2011-10-171-4/+7
| | | | | | | | Fix build with GCC/ARM.
| * Fix/add tests with -NaN, -infChristophe Lyon2011-10-172-5/+25
| |
| * Define NAN and HUGE_VALF to be compatible with MSVC.Christophe Lyon2011-10-031-0/+3
| |
| * Update expected resultsChristophe Lyon2011-09-271-1/+252
| |
generated by cgit v1.2.3 (git 2.25.1) at 2024-05-28 16:40:37 +0000