Message ID | 544A3C3D.5070308@arm.com |
---|---|
State | New |
Headers | show |
On Fri, Oct 24, 2014 at 12:47 PM, Jiong Wang <jiong.wang@arm.com> wrote: > we should not add explicit declaration there. > > arm_neon.h contains those prototype already. they will be available if the > compiler configuration is with related builtin predefine, for example > __ARM_FEATURE_CRYPTO. > > so, actually, if there is any warning when compile these test programs, they > are expected, > and we rely on these warnings to check whether certain features are > available. > > previously, I only verified on arm-none-linux-gnueabi cross check, so have > not exposed > these regressions. I had also missed the vaes and vfma turning on by default by this change. This is OK. ramana > > no verified on arm-none-linux-gnueabihf, regression gone away on arm > directory. > > make check RUNTESTFLAGS="aapcs.exp neon.exp acle.exp simd.exp arm.exp" > > ok for trunk? > > gcc/testsuite/ > > * lib/target-supports.exp > (check_effective_target_arm_crypto_ok_nocache): Remove declaration > for > vaeseq_u8. > (check_effective_target_arm_neon_fp16_ok_nocache): Remove > declaration for > vcvt_f16_f32. > (check_effective_target_arm_neonv2_ok_nocache): Remove declaration > for > vfma_f32.
diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp index 91460c2..4398345 100644 --- a/gcc/testsuite/lib/target-supports.exp +++ b/gcc/testsuite/lib/target-supports.exp @@ -2404,7 +2404,6 @@ proc check_effective_target_arm_crypto_ok_nocache { } { foreach flags {"" "-mfloat-abi=softfp" "-mfpu=crypto-neon-fp-armv8" "-mfpu=crypto-neon-fp-armv8 -mfloat-abi=softfp"} { if { [check_no_compiler_messages_nocache arm_crypto_ok object { #include "arm_neon.h" - extern uint8x16_t vaeseq_u8 (uint8x16_t, uint8x16_t); uint8x16_t foo (uint8x16_t a, uint8x16_t b) { @@ -2549,7 +2548,6 @@ proc check_effective_target_arm_neon_fp16_ok_nocache { } { "-mfpu=neon-fp16 -mfloat-abi=softfp"} { if { [check_no_compiler_messages_nocache arm_neon_fp_16_ok object { #include "arm_neon.h" - extern float16x4_t vcvt_f16_f32 (float32x4_t); float16x4_t foo (float32x4_t arg) { @@ -2625,7 +2623,6 @@ proc check_effective_target_arm_neonv2_ok_nocache { } { foreach flags {"" "-mfloat-abi=softfp" "-mfpu=neon-vfpv4" "-mfpu=neon-vfpv4 -mfloat-abi=softfp"} { if { [check_no_compiler_messages_nocache arm_neonv2_ok object { #include "arm_neon.h" - extern float32x2_t vfma_f32 (float32x2_t, float32x2_t, float32x2_t); float32x2_t foo (float32x2_t a, float32x2_t b, float32x2_t c) {