Implementation notes: amd64, waldorf, crypto_dh/prjfp128bk

Computer: waldorf
Architecture: amd64
CPU ID: GenuineIntel-000106e5-bfebfbff
SUPERCOP version: 20160715
Operation: crypto_dh
Primitive: prjfp128bk
TimeImplementationCompilerBenchmark dateSUPERCOP version
836120v01/w8s16glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
838188v01/w8s32glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
846864v01/w8s16glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
849376v01/w8s32glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
857484v01/w8s16glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
858364v01/w8s32glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
872684v01/w8s08glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
872768v01/w8s02glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
873348v01/w8s08glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
878060v01/w8s32glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
886184v01/w8s04glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
887888v01/w8s04glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
888588v01/w8s04glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
897268v01/w8s04glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
908232v01/w8s02glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
908588v01/w8s08glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
920644v01/w8s02glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
933256v01/w8s02glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
976320v01/w4s32glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
982736v01/w4s32glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
985360v01/w4s16glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
986436v01/w8s01glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
986440v01/w8s16glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
988752v01/w8s08glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
989084v01/w4s32glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
997268v01/w8s01glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
999612v01/w4s16glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
1004568v01/w8s01glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
1026768v01/w4s04glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
1029764v01/w4s08glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
1043328v01/w4s08glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
1044432v01/w4s08glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
1060060v01/w4s04glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
1091900v01/w4s02glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
1116988v01/w4s08glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
1117236v01/w4s02glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
1123120v01/w4s32glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
1127720v01/w8s01glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
1149860v01/w4s16glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
1174524v01/w4s16glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
1208452v01/w4s04glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
1218932v01/w4s04glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
1248444v01/w4s01glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
1260912v01/w4s01glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
1276764v01/w4s02glv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
1305984v01/w4s02glv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
1324800v01/varglv4gcc -march=native -mtune=native -O2 -fomit-frame-pointer -fwrapv2016071820160715
1343176v01/varglv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
1347680v01/varglv4gcc -march=native -mtune=native -Os -fomit-frame-pointer -fwrapv2016071820160715
1365352v01/varglv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715
1443520v01/w4s01glv4gcc -march=native -mtune=native -O3 -fomit-frame-pointer -fwrapv2016071820160715
1446104v01/w4s01glv4gcc -march=native -mtune=native -O -fomit-frame-pointer -fwrapv2016071820160715

Compiler output

Implementation: crypto_dh/prjfp128bk/v01/w4s01glv4
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments
divisor.c: In file included from divisor.c:13:
divisor.c: ./finite128.h:363:13: warning: inline function 'hec_fp_add_2e128mc_g2i_x8664_asm' is not defined [-Wundefined-inline]
divisor.c: inline void hec_fp_add_2e128mc_g2i_x8664_asm(TEMP_hec_fp_2e128mc t, DIV_hec_fp_2e128mc D2);
divisor.c: ^
divisor.c: divisor.c:41:3: note: used here
divisor.c: hec_fp_add_2e128mc_g2i_x8664_asm(t, D2);
divisor.c: ^
divisor.c: In file included from divisor.c:13:
divisor.c: ./finite128.h:364:13: warning: inline function 'hec_fp_aadd_2e128mc_g2i_x8664_asm' is not defined [-Wundefined-inline]
divisor.c: inline void hec_fp_aadd_2e128mc_g2i_x8664_asm(TEMP_hec_fp_2e128mc t, DIV_hec_fp_2e128mc D2);
divisor.c: ^
divisor.c: divisor.c:180:3: note: used here
divisor.c: hec_fp_aadd_2e128mc_g2i_x8664_asm(t, D2);
divisor.c: ^
divisor.c: In file included from divisor.c:13:
divisor.c: ./finite128.h:365:13: warning: inline function 'hec_fp_madd_2e128mc_g2i_x8664_asm' is not defined [-Wundefined-inline]
divisor.c: inline void hec_fp_madd_2e128mc_g2i_x8664_asm(TEMP_hec_fp_2e128mc t, DIV_hec_fp_2e128mc D2);
divisor.c: ^
divisor.c: divisor.c:263:3: note: used here
divisor.c: hec_fp_madd_2e128mc_g2i_x8664_asm(t, D2);
divisor.c: ^
divisor.c: In file included from divisor.c:13:
divisor.c: ./finite128.h:366:13: warning: inline function 'hec_fp_dbl_2e128mc_g2i_x8664_asm' is not defined [-Wundefined-inline]
divisor.c: inline void hec_fp_dbl_2e128mc_g2i_x8664_asm(TEMP_hec_fp_2e128mc t);
divisor.c: ^
divisor.c: ...
fp_2e128mc_x8664.c: In file included from fp_2e128mc_x8664.c:7:
fp_2e128mc_x8664.c: In file included from ./finite128.h:5:
fp_2e128mc_x8664.c: ./fpx.h:23:13: warning: inline function 'fp_cnt_bits' is not defined [-Wundefined-inline]
fp_2e128mc_x8664.c: inline void fp_cnt_bits(int *bc, const uni an, const uni_t al);
fp_2e128mc_x8664.c: ^
fp_2e128mc_x8664.c: fp_2e128mc_x8664.c:88:2: note: used here
fp_2e128mc_x8664.c: fp_cnt_bits(&i, n01, 1);
fp_2e128mc_x8664.c: ^
fp_2e128mc_x8664.c: 1 warning generated.
io.c: io.c:352:14: warning: equality comparison with extraneous parentheses [-Wparentheses-equality]
io.c: if((buf[i] == -1)){
io.c: ~~~~~~~^~~~~
io.c: io.c:352:14: note: remove extraneous parentheses around the comparison to silence this warning
io.c: if((buf[i] == -1)){
io.c: ~ ^ ~
io.c: io.c:352:14: note: use '=' to turn this equality comparison into an assignment
io.c: if((buf[i] == -1)){
io.c: ^~
io.c: =
io.c: io.c:379:14: warning: equality comparison with extraneous parentheses [-Wparentheses-equality]
io.c: if((buf[i] == -1)){
io.c: ~~~~~~~^~~~~
io.c: io.c:379:14: note: remove extraneous parentheses around the comparison to silence this warning
io.c: if((buf[i] == -1)){
io.c: ~ ^ ~
io.c: io.c:379:14: note: use '=' to turn this equality comparison into an assignment
io.c: if((buf[i] == -1)){
io.c: ^~
io.c: =
io.c: 2 warnings generated.
smul.c: In file included from smul.c:13:
smul.c: In file included from ./finite128.h:5:
smul.c: ./fpx.h:23:13: warning: inline function 'fp_cnt_bits' is not defined [-Wundefined-inline]
smul.c: inline void fp_cnt_bits(int *bc, const uni an, const uni_t al);
smul.c: ^
smul.c: smul.c:29:2: note: used here
smul.c: fp_cnt_bits(&i, kn, FP_LEN*2); /*printf("%d ", i);*/
smul.c: ^
smul.c: In file included from smul.c:13:
smul.c: In file included from ./finite128.h:5:
smul.c: ./fpx.h:24:13: warning: inline function 'find_nextwindow' is not defined [-Wundefined-inline]
smul.c: inline void find_nextwindow(int *v, int *k, uni_t kl, int *wd, const uni en, const int i, uni_t WINDOW_SIZE_LTR, uni_t TABLE_SIZE_LTR);
smul.c: ^
smul.c: smul.c:70:3: note: used here
smul.c: find_nextwindow(&ni, &k, 2*FP_LEN, &wd, kn, i, WINDOW_SIZE_LTR, TABLE_SIZE_LTR);
smul.c: ^
smul.c: 2 warnings generated.
smul_a2a3is0.c: In file included from smul_a2a3is0.c:13:
smul_a2a3is0.c: In file included from ./finite128.h:5:
smul_a2a3is0.c: ./fpx.h:23:13: warning: inline function 'fp_cnt_bits' is not defined [-Wundefined-inline]
smul_a2a3is0.c: inline void fp_cnt_bits(int *bc, const uni an, const uni_t al);
smul_a2a3is0.c: ^
smul_a2a3is0.c: smul_a2a3is0.c:29:2: note: used here
smul_a2a3is0.c: fp_cnt_bits(&i, kn, FP_LEN*2); /*printf("%d ", i);*/
smul_a2a3is0.c: ^
smul_a2a3is0.c: In file included from smul_a2a3is0.c:13:
smul_a2a3is0.c: In file included from ./finite128.h:5:
smul_a2a3is0.c: ./fpx.h:24:13: warning: inline function 'find_nextwindow' is not defined [-Wundefined-inline]
smul_a2a3is0.c: inline void find_nextwindow(int *v, int *k, uni_t kl, int *wd, const uni en, const int i, uni_t WINDOW_SIZE_LTR, uni_t TABLE_SIZE_LTR);
smul_a2a3is0.c: ^
smul_a2a3is0.c: smul_a2a3is0.c:70:3: note: used here
smul_a2a3is0.c: find_nextwindow(&ni, &k, 2*FP_LEN, &wd, kn, i, WINDOW_SIZE_LTR, TABLE_SIZE_LTR);
smul_a2a3is0.c: ^
smul_a2a3is0.c: 2 warnings generated.
smul_glv4.c: In file included from smul_glv4.c:11:
smul_glv4.c: In file included from ./finite128.h:5:
smul_glv4.c: ./fpx.h:23:13: warning: inline function 'fp_cnt_bits' is not defined [-Wundefined-inline]
smul_glv4.c: inline void fp_cnt_bits(int *bc, const uni an, const uni_t al);
smul_glv4.c: ^
smul_glv4.c: smul_glv4.c:223:2: note: used here
smul_glv4.c: fp_cnt_bits(&bt, k0->gt;v->gt;n, k0->gt;v->gt;l);
smul_glv4.c: ^
smul_glv4.c: 1 warning generated.
try.c: crypto_dh_prjfp128bk.a(fp_2e128mc_x8664.o): In function `fp_inv_2e128mc_x8664':
try.c: fp_2e128mc_x8664.c:(.text+0x...): undefined reference to `fp_cnt_bits'
try.c: crypto_dh_prjfp128bk.a(smul_a2a3is0.o): In function `hec_fp_smul_2e128mc_g2i_a2a3is0':
try.c: smul_a2a3is0.c:(.text+0x...): undefined reference to `fp_cnt_bits'
try.c: smul_a2a3is0.c:(.text+0x...): undefined reference to `find_nextwindow'
try.c: smul_a2a3is0.c:(.text+0x...): undefined reference to `find_nextwindow'
try.c: crypto_dh_prjfp128bk.a(smul_glv4.o): In function `hec_fp_smul_2e128mc_bk_glv4':
try.c: smul_glv4.c:(.text+0x...): undefined reference to `fp_cnt_bits'
try.c: smul_glv4.c:(.text+0x...): undefined reference to `fp_cnt_bits'
try.c: smul_glv4.c:(.text+0x...): undefined reference to `fp_cnt_bits'
try.c: smul_glv4.c:(.text+0x...): undefined reference to `fp_cnt_bits'
try.c: clang: error: linker command failed with exit code 1 (use -v to see invocation)

Number of similar (compiler,implementation) pairs: 12, namely:
CompilerImplementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments v01/w4s01glv4 v01/w4s02glv4 v01/w4s04glv4 v01/w4s08glv4 v01/w4s16glv4 v01/w4s32glv4 v01/w8s01glv4 v01/w8s02glv4 v01/w8s04glv4 v01/w8s08glv4 v01/w8s16glv4 v01/w8s32glv4

Compiler output

Implementation: crypto_dh/prjfp128bk/v01/varglv4
Compiler: clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments
divisor.c: In file included from divisor.c:13:
divisor.c: ./finite128.h:363:13: warning: inline function 'hec_fp_add_2e128mc_g2i_x8664_asm' is not defined [-Wundefined-inline]
divisor.c: inline void hec_fp_add_2e128mc_g2i_x8664_asm(TEMP_hec_fp_2e128mc t, DIV_hec_fp_2e128mc D2);
divisor.c: ^
divisor.c: divisor.c:41:3: note: used here
divisor.c: hec_fp_add_2e128mc_g2i_x8664_asm(t, D2);
divisor.c: ^
divisor.c: In file included from divisor.c:13:
divisor.c: ./finite128.h:364:13: warning: inline function 'hec_fp_aadd_2e128mc_g2i_x8664_asm' is not defined [-Wundefined-inline]
divisor.c: inline void hec_fp_aadd_2e128mc_g2i_x8664_asm(TEMP_hec_fp_2e128mc t, DIV_hec_fp_2e128mc D2);
divisor.c: ^
divisor.c: divisor.c:180:3: note: used here
divisor.c: hec_fp_aadd_2e128mc_g2i_x8664_asm(t, D2);
divisor.c: ^
divisor.c: In file included from divisor.c:13:
divisor.c: ./finite128.h:365:13: warning: inline function 'hec_fp_madd_2e128mc_g2i_x8664_asm' is not defined [-Wundefined-inline]
divisor.c: inline void hec_fp_madd_2e128mc_g2i_x8664_asm(TEMP_hec_fp_2e128mc t, DIV_hec_fp_2e128mc D2);
divisor.c: ^
divisor.c: divisor.c:263:3: note: used here
divisor.c: hec_fp_madd_2e128mc_g2i_x8664_asm(t, D2);
divisor.c: ^
divisor.c: In file included from divisor.c:13:
divisor.c: ./finite128.h:366:13: warning: inline function 'hec_fp_dbl_2e128mc_g2i_x8664_asm' is not defined [-Wundefined-inline]
divisor.c: inline void hec_fp_dbl_2e128mc_g2i_x8664_asm(TEMP_hec_fp_2e128mc t);
divisor.c: ^
divisor.c: ...
fp_2e128mc_x8664.c: In file included from fp_2e128mc_x8664.c:7:
fp_2e128mc_x8664.c: In file included from ./finite128.h:5:
fp_2e128mc_x8664.c: ./fpx.h:23:13: warning: inline function 'fp_cnt_bits' is not defined [-Wundefined-inline]
fp_2e128mc_x8664.c: inline void fp_cnt_bits(int *bc, const uni an, const uni_t al);
fp_2e128mc_x8664.c: ^
fp_2e128mc_x8664.c: fp_2e128mc_x8664.c:88:2: note: used here
fp_2e128mc_x8664.c: fp_cnt_bits(&i, n01, 1);
fp_2e128mc_x8664.c: ^
fp_2e128mc_x8664.c: 1 warning generated.
io.c: io.c:352:14: warning: equality comparison with extraneous parentheses [-Wparentheses-equality]
io.c: if((buf[i] == -1)){
io.c: ~~~~~~~^~~~~
io.c: io.c:352:14: note: remove extraneous parentheses around the comparison to silence this warning
io.c: if((buf[i] == -1)){
io.c: ~ ^ ~
io.c: io.c:352:14: note: use '=' to turn this equality comparison into an assignment
io.c: if((buf[i] == -1)){
io.c: ^~
io.c: =
io.c: io.c:379:14: warning: equality comparison with extraneous parentheses [-Wparentheses-equality]
io.c: if((buf[i] == -1)){
io.c: ~~~~~~~^~~~~
io.c: io.c:379:14: note: remove extraneous parentheses around the comparison to silence this warning
io.c: if((buf[i] == -1)){
io.c: ~ ^ ~
io.c: io.c:379:14: note: use '=' to turn this equality comparison into an assignment
io.c: if((buf[i] == -1)){
io.c: ^~
io.c: =
io.c: 2 warnings generated.
smul.c: In file included from smul.c:13:
smul.c: In file included from ./finite128.h:5:
smul.c: ./fpx.h:23:13: warning: inline function 'fp_cnt_bits' is not defined [-Wundefined-inline]
smul.c: inline void fp_cnt_bits(int *bc, const uni an, const uni_t al);
smul.c: ^
smul.c: smul.c:29:2: note: used here
smul.c: fp_cnt_bits(&i, kn, FP_LEN*2); /*printf("%d ", i);*/
smul.c: ^
smul.c: In file included from smul.c:13:
smul.c: In file included from ./finite128.h:5:
smul.c: ./fpx.h:24:13: warning: inline function 'find_nextwindow' is not defined [-Wundefined-inline]
smul.c: inline void find_nextwindow(int *v, int *k, uni_t kl, int *wd, const uni en, const int i, uni_t WINDOW_SIZE_LTR, uni_t TABLE_SIZE_LTR);
smul.c: ^
smul.c: smul.c:70:3: note: used here
smul.c: find_nextwindow(&ni, &k, 2*FP_LEN, &wd, kn, i, WINDOW_SIZE_LTR, TABLE_SIZE_LTR);
smul.c: ^
smul.c: 2 warnings generated.
smul_a2a3is0.c: In file included from smul_a2a3is0.c:13:
smul_a2a3is0.c: In file included from ./finite128.h:5:
smul_a2a3is0.c: ./fpx.h:23:13: warning: inline function 'fp_cnt_bits' is not defined [-Wundefined-inline]
smul_a2a3is0.c: inline void fp_cnt_bits(int *bc, const uni an, const uni_t al);
smul_a2a3is0.c: ^
smul_a2a3is0.c: smul_a2a3is0.c:29:2: note: used here
smul_a2a3is0.c: fp_cnt_bits(&i, kn, FP_LEN*2); /*printf("%d ", i);*/
smul_a2a3is0.c: ^
smul_a2a3is0.c: In file included from smul_a2a3is0.c:13:
smul_a2a3is0.c: In file included from ./finite128.h:5:
smul_a2a3is0.c: ./fpx.h:24:13: warning: inline function 'find_nextwindow' is not defined [-Wundefined-inline]
smul_a2a3is0.c: inline void find_nextwindow(int *v, int *k, uni_t kl, int *wd, const uni en, const int i, uni_t WINDOW_SIZE_LTR, uni_t TABLE_SIZE_LTR);
smul_a2a3is0.c: ^
smul_a2a3is0.c: smul_a2a3is0.c:70:3: note: used here
smul_a2a3is0.c: find_nextwindow(&ni, &k, 2*FP_LEN, &wd, kn, i, WINDOW_SIZE_LTR, TABLE_SIZE_LTR);
smul_a2a3is0.c: ^
smul_a2a3is0.c: 2 warnings generated.
smul_glv4.c: In file included from smul_glv4.c:11:
smul_glv4.c: In file included from ./finite128.h:5:
smul_glv4.c: ./fpx.h:23:13: warning: inline function 'fp_cnt_bits' is not defined [-Wundefined-inline]
smul_glv4.c: inline void fp_cnt_bits(int *bc, const uni an, const uni_t al);
smul_glv4.c: ^
smul_glv4.c: smul_glv4.c:223:2: note: used here
smul_glv4.c: fp_cnt_bits(&bt, k0->gt;v->gt;n, k0->gt;v->gt;l);
smul_glv4.c: ^
smul_glv4.c: 1 warning generated.
try.c: crypto_dh_prjfp128bk.a(fp_2e128mc_x8664.o): In function `fp_inv_2e128mc_x8664':
try.c: fp_2e128mc_x8664.c:(.text+0x...): undefined reference to `fp_cnt_bits'
try.c: crypto_dh_prjfp128bk.a(smul_glv4.o): In function `hec_fp_smul_2e128mc_bk_glv4':
try.c: smul_glv4.c:(.text+0x...): undefined reference to `fp_cnt_bits'
try.c: smul_glv4.c:(.text+0x...): undefined reference to `fp_cnt_bits'
try.c: smul_glv4.c:(.text+0x...): undefined reference to `fp_cnt_bits'
try.c: smul_glv4.c:(.text+0x...): undefined reference to `fp_cnt_bits'
try.c: clang: error: linker command failed with exit code 1 (use -v to see invocation)

Number of similar (compiler,implementation) pairs: 1, namely:
CompilerImplementations
clang -march=native -O3 -fomit-frame-pointer -fwrapv -Qunused-arguments v01/varglv4