https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231293
[email protected] changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |[email protected] --- Comment #3 from [email protected] --- (In reply to Jan Beich from comment #1) (__ARM_FP & 2) refers to half precision support [1] gcc6 and gcc7 doesn't define __ARM_FP16_FORMAT_IEEE gcc7 -dM -E - < /dev/null | grep __ARM_FP16_FORMAT_IEEE but clang does: cc -dM -E - < /dev/null | grep __ARM_FP16_FORMAT_IEEE #define __ARM_FP16_FORMAT_IEEE 1 from [2]: The __fp16 type denotes half-precision (16-bit) floating-point. The recommended way to test for this hardware support is to test bit 1 in __ARM_FP and If 16-bit floating-point is available, one of __ARM_FP16_FORMAT_IEEE and __ARM_FP16_FORMAT_ALTERNATIVE will be defined to indicate the format in use could it be a clang bug? [1] http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0774g/chr1383660321827.html [2] http://infocenter.arm.com/help/topic/com.arm.doc.ihi0053c/IHI0053C_acle_2_0.pdf -- You are receiving this mail because: You are the assignee for the bug. You are on the CC list for the bug.
