<![CDATA[Armv6-M CPUs (e.g. Arm Cortex M0+) have 32-bit multiplications but only return the low 32 bits of the result; a software routine must be used to compute the high 32 bits.]]>

<![CDATA[Armv6-M CPUs (e.g. Arm Cortex M0+) have 32-bit multiplications but only return the low 32 bits of the result; a software routine must be used to compute the high 32 bits.]]>Armv6-M CPUs (e.g. Arm Cortex M0+) have 32-bit multiplications but only return the low 32 bits of the result; a software routine must be used to compute the high 32 bits. The one that comes with the C compiler is usually bad (e.g. the one from GCC has cost 61 cycles and is not constant-time). I made a better one (constant-time, 24 cycles for a full 64x64->64 multiplication, 20 cycles for the 32x32->64 variants).
https://github.com/pornin/armv6m-longmul

]]>https://board.circlewithadot.net/topic/a859884c-251a-4a06-bae3-dc3c50f7b808/armv6-m-cpus-e.g.-arm-cortex-m0-have-32-bit-multiplications-but-only-return-the-low-32-bits-of-the-result-a-software-routine-must-be-used-to-compute-the-high-32-bits.RSS for NodeMon, 25 May 2026 12:41:38 GMTWed, 06 May 2026 19:00:10 GMT60<![CDATA[Reply to Armv6-M CPUs (e.g. Arm Cortex M0+) have 32-bit multiplications but only return the low 32 bits of the result; a software routine must be used to compute the high 32 bits. on Wed, 06 May 2026 20:19:02 GMT]]>@pornin This looks incredibly cool, even to someone who will never use it directly.

]]>https://board.circlewithadot.net/post/https://infosec.exchange/users/paulehoffman/statuses/116529479204742278https://board.circlewithadot.net/post/https://infosec.exchange/users/paulehoffman/statuses/116529479204742278Wed, 06 May 2026 20:19:02 GMT