CPU features

Compiler CPU flags

customize the TensorFlow source build to take advantage of the availability of some CPU features that contribute to a speedier execution of TensorFlow code

Available CPU flags on target system can be found with following command and Linux kernel source helps unravel the meaning for each flag

$ more /proc/cpuinfo | grep flags

optimization flags will be supplied when configuring the TensorFlow source build. The following command is used to populate the optimization flags:

$grep flags -m1 /proc/cpuinfo | cut -d ":" -f 2 | tr '[:upper:]' '[:lower:]' | { read FLAGS; OPT="-march=native"; for flag in $FLAGS; do case "$flag" in "sse4_1" | "sse4_2" | "ssse3" | "fma" | "cx16" | "popcnt" | "avx" | "avx2") OPT+=" -m$flag";; esac; done; MODOPT=${OPT//_/\.}; echo "$MODOPT"; }

and followings are an output example on AMD EPYC Process

-march=native -mssse3 -mfma -mcx16 -msse4.1 -msse4.2 -mpopcnt -mavx -mavx2

No	Flag	CPU Feature	Additional Info
1	ssse3	Supplemental Streaming SIMD Extensions 3 (SSSE-3) instruction set
2	sse4_1	Streaming SIMD Extensions 4.1 (SSE-4.1) instruction set
3	sse4_2	Streaming SIDM Extensions 4.2 (SSE-4.2) instruction set
4	fma	Fused multiply-add (FMA) instruction set
5	cx16	CMPXCHG16B instruction (double-width compare-and-swap)
6	popcnt	Population count instruction (count number of bits set to 1)
7	avx	Advanced Vector Extensions
8	avx2	Advanced Vector Extension 2