You are not logged in.

Pages: 1

#1 2022-04-05 10:48:34

cribari
Member
From: Recife, Brazil
Registered: 2016-01-28
Posts: 84
Website

gcc optimization (for speed) flags not working

I run Arch on my laptop and Ubuntu on my desktop (Intel i7 on both). The corresponding gcc versions are 11.2.0 and 7.5.0. The optimization for speed flags (-O0, -O1, -O2, -O3, -Ofast) seem to work under Ubuntu but they make no difference under Arch. For instance (program: coin.c): 

/* coin.c */

#include <stdlib.h> 
#include <stdio.h>
#include <time.h>

#define COIN_MAX 10000000

void coin(unsigned int ntrials)
{
  unsigned int R2 = RAND_MAX/2; 
  unsigned int tail=0;
  usigned int temp; 
  double result; 

  srand(1750); 

  for (unsigned int s = 0; s < ntrials; s++){
      temp = rand(); 
      tail = temp > R2 ? tail : tail+1; 
  }
  
  result = 100.0*((double) tail)/ntrials;
  
  printf("Percentage of tails from %u tosses: %.2f\n", ntrials, result); 
}

int main(void)
{
  clock_t start_time, end_time;
 
  start_time = clock();
  
  for (unsigned int j = 10; j <= COIN_MAX; j *= 10){
     coin(j);
  }
  
  end_time = clock();
  
  printf("EXECUTION TIME: %.2f seconds\n",
              ((double)(end_time-start_time))/(CLOCKS_PER_SEC));
                          
  printf("EXECUTION TIME: %.2f minutes\n",
              ((double)(end_time-start_time))/(60*CLOCKS_PER_SEC));
                          
  return 0; 
}

The execution times I get under Arch are pretty much the same regardless of the optimization flag used (about 0.20 second). Under Ubuntu, however, they vary quite a bit (from approx. 0.08 second with -O3 to 0.14 second with -O0).

Suggestions are welcome. Thank you.

EDIT: Added missing #include<stdlib.h>

Last edited by cribari (2022-04-05 20:35:31)

Linux, street photography and Baroque music enthusiast.

Offline

#2 2022-04-05 11:04:38

Slithery
Administrator
From: Norfolk, UK
Registered: 2013-12-01
Posts: 5,776

Re: gcc optimization (for speed) flags not working

Exactly the same CPU in both systems?

No, it didn't "fix" anything. It just shifted the brokeness one space to the right. - jasonwryan
Closing -- for deletion; Banning -- for muppetry. - jasonwryan

aur - dotfiles

Offline

#3 2022-04-05 12:40:38

cribari
Member
From: Recife, Brazil
Registered: 2016-01-28
Posts: 84
Website

Re: gcc optimization (for speed) flags not working

Slithery wrote:

Exactly the same CPU in both systems?

No. My laptop (Arch): 

cribari@darwin7 ~ $ cat /proc/cpuinfo
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 142
model name      : Intel(R) Core(TM) i7-8550U CPU @ 1.80GHz
stepping        : 10
microcode       : 0xec
cpu MHz         : 2000.000
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 0
cpu cores       : 4
apicid          : 0
initial apicid  : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 22
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust sgx bmi1 avx2 smep bmi2 erms invpcid mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp md_clear flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer invvpid ept_x_only ept_ad ept_1gb flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest ple pml ept_mode_based_exec
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 3999.93
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 6
model           : 142
model name      : Intel(R) Core(TM) i7-8550U CPU @ 1.80GHz
stepping        : 10
microcode       : 0xec
cpu MHz         : 2000.000
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 1
cpu cores       : 4
apicid          : 2
initial apicid  : 2
fpu             : yes
fpu_exception   : yes
cpuid level     : 22
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust sgx bmi1 avx2 smep bmi2 erms invpcid mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp md_clear flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer invvpid ept_x_only ept_ad ept_1gb flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest ple pml ept_mode_based_exec
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 3999.93
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 2
vendor_id       : GenuineIntel
cpu family      : 6
model           : 142
model name      : Intel(R) Core(TM) i7-8550U CPU @ 1.80GHz
stepping        : 10
microcode       : 0xec
cpu MHz         : 2000.000
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 2
cpu cores       : 4
apicid          : 4
initial apicid  : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 22
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust sgx bmi1 avx2 smep bmi2 erms invpcid mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp md_clear flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer invvpid ept_x_only ept_ad ept_1gb flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest ple pml ept_mode_based_exec
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 3999.93
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 3
vendor_id       : GenuineIntel
cpu family      : 6
model           : 142
model name      : Intel(R) Core(TM) i7-8550U CPU @ 1.80GHz
stepping        : 10
microcode       : 0xec
cpu MHz         : 2000.000
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 3
cpu cores       : 4
apicid          : 6
initial apicid  : 6
fpu             : yes
fpu_exception   : yes
cpuid level     : 22
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust sgx bmi1 avx2 smep bmi2 erms invpcid mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp md_clear flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer invvpid ept_x_only ept_ad ept_1gb flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest ple pml ept_mode_based_exec
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 3999.93
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 4
vendor_id       : GenuineIntel
cpu family      : 6
model           : 142
model name      : Intel(R) Core(TM) i7-8550U CPU @ 1.80GHz
stepping        : 10
microcode       : 0xec
cpu MHz         : 1126.561
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 0
cpu cores       : 4
apicid          : 1
initial apicid  : 1
fpu             : yes
fpu_exception   : yes
cpuid level     : 22
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust sgx bmi1 avx2 smep bmi2 erms invpcid mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp md_clear flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer invvpid ept_x_only ept_ad ept_1gb flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest ple pml ept_mode_based_exec
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 3999.93
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 5
vendor_id       : GenuineIntel
cpu family      : 6
model           : 142
model name      : Intel(R) Core(TM) i7-8550U CPU @ 1.80GHz
stepping        : 10
microcode       : 0xec
cpu MHz         : 2000.000
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 1
cpu cores       : 4
apicid          : 3
initial apicid  : 3
fpu             : yes
fpu_exception   : yes
cpuid level     : 22
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust sgx bmi1 avx2 smep bmi2 erms invpcid mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp md_clear flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer invvpid ept_x_only ept_ad ept_1gb flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest ple pml ept_mode_based_exec
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 3999.93
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 6
vendor_id       : GenuineIntel
cpu family      : 6
model           : 142
model name      : Intel(R) Core(TM) i7-8550U CPU @ 1.80GHz
stepping        : 10
microcode       : 0xec
cpu MHz         : 2000.000
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 2
cpu cores       : 4
apicid          : 5
initial apicid  : 5
fpu             : yes
fpu_exception   : yes
cpuid level     : 22
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust sgx bmi1 avx2 smep bmi2 erms invpcid mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp md_clear flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer invvpid ept_x_only ept_ad ept_1gb flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest ple pml ept_mode_based_exec
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 3999.93
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 7
vendor_id       : GenuineIntel
cpu family      : 6
model           : 142
model name      : Intel(R) Core(TM) i7-8550U CPU @ 1.80GHz
stepping        : 10
microcode       : 0xec
cpu MHz         : 2000.000
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 3
cpu cores       : 4
apicid          : 7
initial apicid  : 7
fpu             : yes
fpu_exception   : yes
cpuid level     : 22
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust sgx bmi1 avx2 smep bmi2 erms invpcid mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp md_clear flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer invvpid ept_x_only ept_ad ept_1gb flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest ple pml ept_mode_based_exec
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 3999.93
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

My desktop (Ubuntu): 

cribari@edgeworth:~$ cat /proc/cpuinfo 
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 60
model name      : Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
stepping        : 3
microcode       : 0x28
cpu MHz         : 1297.715
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 0
cpu cores       : 4
apicid          : 0
initial apicid  : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts md_clear flush_l1d
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 5986.20
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 6
model           : 60
model name      : Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
stepping        : 3
microcode       : 0x28
cpu MHz         : 1297.790
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 1
cpu cores       : 4
apicid          : 2
initial apicid  : 2
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts md_clear flush_l1d
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 5986.20
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 2
vendor_id       : GenuineIntel
cpu family      : 6
model           : 60
model name      : Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
stepping        : 3
microcode       : 0x28
cpu MHz         : 1297.116
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 2
cpu cores       : 4
apicid          : 4
initial apicid  : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts md_clear flush_l1d
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 5986.20
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 3
vendor_id       : GenuineIntel
cpu family      : 6
model           : 60
model name      : Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
stepping        : 3
microcode       : 0x28
cpu MHz         : 1297.006
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 3
cpu cores       : 4
apicid          : 6
initial apicid  : 6
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts md_clear flush_l1d
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 5986.20
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 4
vendor_id       : GenuineIntel
cpu family      : 6
model           : 60
model name      : Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
stepping        : 3
microcode       : 0x28
cpu MHz         : 1295.750
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 0
cpu cores       : 4
apicid          : 1
initial apicid  : 1
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts md_clear flush_l1d
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 5986.20
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 5
vendor_id       : GenuineIntel
cpu family      : 6
model           : 60
model name      : Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
stepping        : 3
microcode       : 0x28
cpu MHz         : 1299.762
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 1
cpu cores       : 4
apicid          : 3
initial apicid  : 3
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts md_clear flush_l1d
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 5986.20
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 6
vendor_id       : GenuineIntel
cpu family      : 6
model           : 60
model name      : Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
stepping        : 3
microcode       : 0x28
cpu MHz         : 1297.678
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 2
cpu cores       : 4
apicid          : 5
initial apicid  : 5
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts md_clear flush_l1d
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 5986.20
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 7
vendor_id       : GenuineIntel
cpu family      : 6
model           : 60
model name      : Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz
stepping        : 3
microcode       : 0x28
cpu MHz         : 1297.029
cache size      : 8192 KB
physical id     : 0
siblings        : 8
core id         : 3
cpu cores       : 4
apicid          : 7
initial apicid  : 7
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts md_clear flush_l1d
bugs            : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips        : 5986.20
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

Linux, street photography and Baroque music enthusiast.

Offline

#4 2022-04-05 12:58:06

cribari
Member
From: Recife, Brazil
Registered: 2016-01-28
Posts: 84
Website

Re: gcc optimization (for speed) flags not working

Arch: 

 
cribari@darwin7 ~ $ gcc -O0 -o coin coin.c
cribari@darwin7 ~ $ ./coin
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.22 seconds
EXECUTION TIME: 0.00 minutes
cribari@darwin7 ~ $ gcc -O1 -o coin coin.c
cribari@darwin7 ~ $ ./coin
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.22 seconds
EXECUTION TIME: 0.00 minutes
cribari@darwin7 ~ $ gcc -O2 -o coin coin.c
cribari@darwin7 ~ $ ./coin
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.22 seconds
EXECUTION TIME: 0.00 minutes
cribari@darwin7 ~ $ gcc -O3 -o coin coin.c
cribari@darwin7 ~ $ ./coin
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.22 seconds
EXECUTION TIME: 0.00 minutes
cribari@darwin7 ~ $ gcc -Ofast -o coin coin.c
cribari@darwin7 ~ $ ./coin
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.22 seconds
EXECUTION TIME: 0.00 minutes

Ubuntu: 

bari@edgeworth:~$ ./coin
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.14 seconds
EXECUTION TIME: 0.00 minutes
cribari@edgeworth:~$ gcc -O1 -o coin coin.c
cribari@edgeworth:~$ ./coin
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.09 seconds
EXECUTION TIME: 0.00 minutes
cribari@edgeworth:~$ gcc -O2 -o coin coin.c
cribari@edgeworth:~$ ./coin
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.08 seconds
EXECUTION TIME: 0.00 minutes
cribari@edgeworth:~$ gcc -O3 -o coin coin.c
cribari@edgeworth:~$ ./coin
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.08 seconds
EXECUTION TIME: 0.00 minutes
cribari@edgeworth:~$ gcc -Ofast -o coin coin.c
cribari@edgeworth:~$ ./coin
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.08 seconds
EXECUTION TIME: 0.00 minutes
cribari@edgeworth:~$

Linux, street photography and Baroque music enthusiast.

Offline

#5 2022-04-05 13:18:19

Scimmia
Fellow
Registered: 2012-09-01
Posts: 13,729

Re: gcc optimization (for speed) flags not working

Try it with different output filenames for each build.

Online

#6 2022-04-05 15:04:32

cribari
Member
From: Recife, Brazil
Registered: 2016-01-28
Posts: 84
Website

Re: gcc optimization (for speed) flags not working

Scimmia wrote:

Try it with different output filenames for each build.

Arch:

cribari@darwin7 ~ $ gcc -O0 -o coin0 coin.c
cribari@darwin7 ~ $ ./coin0
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.19 seconds
EXECUTION TIME: 0.00 minutes
cribari@darwin7 ~ $ gcc -O1 -o coin1 coin.c
cribari@darwin7 ~ $ ./coin1
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.18 seconds
EXECUTION TIME: 0.00 minutes
cribari@darwin7 ~ $ gcc -O2 -o coin2 coin.c
cribari@darwin7 ~ $ ./coin2
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.18 seconds
EXECUTION TIME: 0.00 minutes
cribari@darwin7 ~ $ gcc -O3 -o coin3 coin.c
cribari@darwin7 ~ $ ./coin3
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.18 seconds
EXECUTION TIME: 0.00 minutes
cribari@darwin7 ~ $

Linux, street photography and Baroque music enthusiast.

Offline

#7 2022-04-05 15:16:15

loqs
Member
Registered: 2014-03-06
Posts: 18,970

Re: gcc optimization (for speed) flags not working

gcc -O3 coin.c 
coin.c: In function ‘coin’:
coin.c:10:21: error: ‘RAND_MAX’ undeclared (first use in this function)
   10 |   unsigned int R2 = RAND_MAX/2;
      |                     ^~~~~~~~
coin.c:10:21: note: each undeclared identifier is reported only once for each function it appears in
coin.c:15:3: warning: implicit declaration of function ‘srand’ [-Wimplicit-function-declaration]
   15 |   srand(1750);
      |   ^~~~~
coin.c:18:14: warning: implicit declaration of function ‘rand’ [-Wimplicit-function-declaration]
   18 |       temp = rand();

Adding in stdlib

gcc -O3 coin.c 
$ ./a.out 
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.05 seconds
EXECUTION TIME: 0.00 minutes
gcc -O0 coin.c 
$ ./a.out 
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.11 seconds
EXECUTION TIME: 0.00 minutes

Edit:
The output of the following might help

type -a gcc
gcc -O0 -v coin.c
gcc -O0 --S o coin0.S coin.c
gcc -O3 -S -o coin3.S coin.c
diff coin0.S coin3.S

Last edited by loqs (2022-04-05 15:26:44)

Offline

#8 2022-04-05 15:35:48

cribari
Member
From: Recife, Brazil
Registered: 2016-01-28
Posts: 84
Website

Re: gcc optimization (for speed) flags not working

The output of the following might help

type -a gcc
gcc -O0 -v coin.c
gcc -O0 --S o coin0.S coin.c
gcc -O3 -S -o coin3.S coin.c
diff coin0.S coin3.S

Here it is: 

cribari@darwin7 ~ $ type -a gcc
gcc is /usr/lib/ccache/bin/gcc
gcc is /usr/bin/gcc
cribari@darwin7 ~ $ gcc -O0 -v coin.c
Using built-in specs.
COLLECT_GCC=/usr/bin/gcc
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: /build/gcc/src/gcc/configure --enable-languages=c,c++,ada,fortran,go,lto,objc,obj-c++,d --enable-bootstrap --prefix=/usr --libdir=/usr/lib --libexecdir=/usr/lib --mandir=/usr/share/man --infodir=/usr/share/info --with-bugurl=https://bugs.archlinux.org/ --with-linker-hash-style=gnu --with-system-zlib --enable-__cxa_atexit --enable-cet=auto --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-default-ssp --enable-gnu-indirect-function --enable-gnu-unique-object --enable-linker-build-id --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --disable-libssp --disable-libstdcxx-pch --disable-werror --with-build-config=bootstrap-lto --enable-link-serialization=1 gdc_include_dir=/usr/include/dlang/gdc
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 11.2.0 (GCC) 
COLLECT_GCC_OPTIONS='-O0' '-v' '-mtune=generic' '-march=x86-64' '-dumpdir' 'a-'
 /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/cc1 -quiet -v coin.c -quiet -dumpdir a- -dumpbase coin.c -dumpbase-ext .c -mtune=generic -march=x86-64 -O0 -version -o /tmp/cc1GEVwG.s
GNU C17 (GCC) version 11.2.0 (x86_64-pc-linux-gnu)
        compiled by GNU C version 11.2.0, GMP version 6.2.1, MPFR version 4.1.0-p13, MPC version 1.2.1, isl version isl-0.24-GMP

GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072
ignoring nonexistent directory "/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/../../../../x86_64-pc-linux-gnu/include"
#include "..." search starts here:
#include <...> search starts here:
 /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/include
 /usr/local/include
 /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/include-fixed
 /usr/include
End of search list.
GNU C17 (GCC) version 11.2.0 (x86_64-pc-linux-gnu)
        compiled by GNU C version 11.2.0, GMP version 6.2.1, MPFR version 4.1.0-p13, MPC version 1.2.1, isl version isl-0.24-GMP

GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072
Compiler executable checksum: 65a5ec6b673b24273c0fdde359680a16
COLLECT_GCC_OPTIONS='-O0' '-v' '-mtune=generic' '-march=x86-64' '-dumpdir' 'a-'
 as -v --64 -o /tmp/ccbftIY7.o /tmp/cc1GEVwG.s
GNU assembler version 2.38 (x86_64-pc-linux-gnu) using BFD version (GNU Binutils) 2.38
COMPILER_PATH=/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/:/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/:/usr/lib/gcc/x86_64-pc-linux-gnu/:/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/:/usr/lib/gcc/x86_64-pc-linux-gnu/
LIBRARY_PATH=/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/:/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/../../../../lib/:/lib/../lib/:/usr/lib/../lib/:/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/../../../:/lib/:/usr/lib/
COLLECT_GCC_OPTIONS='-O0' '-v' '-mtune=generic' '-march=x86-64' '-dumpdir' 'a.'
 /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/collect2 -plugin /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/liblto_plugin.so -plugin-opt=/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/lto-wrapper -plugin-opt=-fresolution=/tmp/ccQ0xA7Z.res -plugin-opt=-pass-through=-lgcc -plugin-opt=-pass-through=-lgcc_s -plugin-opt=-pass-through=-lc -plugin-opt=-pass-through=-lgcc -plugin-opt=-pass-through=-lgcc_s --build-id --eh-frame-hdr --hash-style=gnu -m elf_x86_64 -dynamic-linker /lib64/ld-linux-x86-64.so.2 -pie /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/../../../../lib/Scrt1.o /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/../../../../lib/crti.o /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/crtbeginS.o -L/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0 -L/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/../../../../lib -L/lib/../lib -L/usr/lib/../lib -L/usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/../../.. /tmp/ccbftIY7.o -lgcc --push-state --as-needed -lgcc_s --pop-state -lc -lgcc --push-state --as-needed -lgcc_s --pop-state /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/crtendS.o /usr/lib/gcc/x86_64-pc-linux-gnu/11.2.0/../../../../lib/crtn.o
COLLECT_GCC_OPTIONS='-O0' '-v' '-mtune=generic' '-march=x86-64' '-dumpdir' 'a.'
cribari@darwin7 ~ $ gcc -O0 --S o coin0.S coin.c
gcc: error: unrecognized command-line option ‘--S’; did you mean ‘-S’?
cribari@darwin7 ~ $ gcc -O0 -S -o coin0.S coin.c
cribari@darwin7 ~ $ gcc -O3 -S -o coin3.S coin.c
cribari@darwin7 ~ $ diff coin0.S coin3.S
3c3
<       .section        .rodata
---
>       .section        .rodata.str1.8,"aMS",@progbits,1
5c5
< .LC1:
---
> .LC2:
7a8
>       .p2align 4
11c12
< .LFB6:
---
> .LFB22:
13c14
<       pushq   %rbp
---
>       pushq   %r12
15,21c16,17
<       .cfi_offset 6, -16
<       movq    %rsp, %rbp
<       .cfi_def_cfa_register 6
<       subq    $48, %rsp
<       movl    %edi, -36(%rbp)
<       movl    $1073741823, -16(%rbp)
<       movl    $0, -24(%rbp)
---
>       .cfi_offset 12, -16
>       movl    %edi, %r12d
22a19,24
>       pushq   %rbp
>       .cfi_def_cfa_offset 24
>       .cfi_offset 6, -24
>       pushq   %rbx
>       .cfi_def_cfa_offset 32
>       .cfi_offset 3, -32
24,36c26,31
<       movl    $0, -20(%rbp)
<       jmp     .L2
< .L5:
<       call    rand@PLT
<       movl    %eax, -12(%rbp)
<       movl    -12(%rbp), %eax
<       cmpl    %eax, -16(%rbp)
<       jb      .L3
<       movl    -24(%rbp), %eax
<       addl    $1, %eax
<       jmp     .L4
< .L3:
<       movl    -24(%rbp), %eax
---
>       testl   %r12d, %r12d
>       je      .L9
>       xorl    %ebx, %ebx
>       xorl    %ebp, %ebp
>       .p2align 4,,10
>       .p2align 3
38,71c33,38
<       movl    %eax, -24(%rbp)
<       addl    $1, -20(%rbp)
< .L2:
<       movl    -20(%rbp), %eax
<       cmpl    -36(%rbp), %eax
<       jb      .L5
<       movl    -24(%rbp), %eax
<       testq   %rax, %rax
<       js      .L6
<       pxor    %xmm0, %xmm0
<       cvtsi2sdq       %rax, %xmm0
<       jmp     .L7
< .L6:
<       movq    %rax, %rdx
<       shrq    %rdx
<       andl    $1, %eax
<       orq     %rax, %rdx
<       pxor    %xmm0, %xmm0
<       cvtsi2sdq       %rdx, %xmm0
<       addsd   %xmm0, %xmm0
< .L7:
<       movsd   .LC0(%rip), %xmm1
<       mulsd   %xmm0, %xmm1
<       movl    -36(%rbp), %eax
<       testq   %rax, %rax
<       js      .L8
<       pxor    %xmm0, %xmm0
<       cvtsi2sdq       %rax, %xmm0
<       jmp     .L9
< .L8:
<       movq    %rax, %rdx
<       shrq    %rdx
<       andl    $1, %eax
<       orq     %rax, %rdx
---
>       call    rand@PLT
>       cmpl    $1073741824, %eax
>       adcl    $0, %ebp
>       addl    $1, %ebx
>       cmpl    %ebx, %r12d
>       jne     .L4
73,83c40,51
<       cvtsi2sdq       %rdx, %xmm0
<       addsd   %xmm0, %xmm0
< .L9:
<       divsd   %xmm0, %xmm1
<       movsd   %xmm1, -8(%rbp)
<       movq    -8(%rbp), %rdx
<       movl    -36(%rbp), %eax
<       movq    %rdx, %xmm0
<       movl    %eax, %esi
<       leaq    .LC1(%rip), %rax
<       movq    %rax, %rdi
---
>       cvtsi2sdq       %rbp, %xmm0
>       mulsd   .LC1(%rip), %xmm0
> .L2:
>       movl    %r12d, %eax
>       pxor    %xmm1, %xmm1
>       popq    %rbx
>       .cfi_remember_state
>       .cfi_def_cfa_offset 24
>       movl    %r12d, %esi
>       cvtsi2sdq       %rax, %xmm1
>       popq    %rbp
>       .cfi_def_cfa_offset 16
85,89c53,63
<       call    printf@PLT
<       nop
<       leave
<       .cfi_def_cfa 7, 8
<       ret
---
>       popq    %r12
>       .cfi_def_cfa_offset 8
>       leaq    .LC2(%rip), %rdi
>       divsd   %xmm1, %xmm0
>       jmp     printf@PLT
>       .p2align 4,,10
>       .p2align 3
> .L9:
>       .cfi_restore_state
>       pxor    %xmm0, %xmm0
>       jmp     .L2
91c65
< .LFE6:
---
> .LFE22:
93,94c67,68
<       .section        .rodata
< .LC3:
---
>       .section        .rodata.str1.1,"aMS",@progbits,1
> .LC4:
96c70
< .LC5:
---
> .LC6:
98c72,73
<       .text
---
>       .section        .text.startup,"ax",@progbits
>       .p2align 4
102c77
< .LFB7:
---
> .LFB23:
104c79
<       pushq   %rbp
---
>       pushq   %r15
106,126c81,101
<       .cfi_offset 6, -16
<       movq    %rsp, %rbp
<       .cfi_def_cfa_register 6
<       subq    $32, %rsp
<       call    clock@PLT
<       movq    %rax, -16(%rbp)
<       movl    $10, -20(%rbp)
<       jmp     .L11
< .L12:
<       movl    -20(%rbp), %eax
<       movl    %eax, %edi
<       call    coin
<       movl    -20(%rbp), %edx
<       movl    %edx, %eax
<       sall    $2, %eax
<       addl    %edx, %eax
<       addl    %eax, %eax
<       movl    %eax, -20(%rbp)
< .L11:
<       cmpl    $10000000, -20(%rbp)
<       jbe     .L12
---
>       .cfi_offset 15, -16
>       pushq   %r14
>       .cfi_def_cfa_offset 24
>       .cfi_offset 14, -24
>       pushq   %r13
>       .cfi_def_cfa_offset 32
>       .cfi_offset 13, -32
>       leaq    .LC2(%rip), %r13
>       pushq   %r12
>       .cfi_def_cfa_offset 40
>       .cfi_offset 12, -40
>       movl    $7, %r12d
>       pushq   %rbp
>       .cfi_def_cfa_offset 48
>       .cfi_offset 6, -48
>       movl    $10, %ebp
>       pushq   %rbx
>       .cfi_def_cfa_offset 56
>       .cfi_offset 3, -56
>       subq    $24, %rsp
>       .cfi_def_cfa_offset 80
128,133c103,129
<       movq    %rax, -8(%rbp)
<       movq    -8(%rbp), %rax
<       subq    -16(%rbp), %rax
<       pxor    %xmm0, %xmm0
<       cvtsi2sdq       %rax, %xmm0
<       movsd   .LC2(%rip), %xmm1
---
>       movq    %rax, %r14
>       .p2align 4,,10
>       .p2align 3
> .L18:
>       movl    $1750, %edi
>       xorl    %r15d, %r15d
>       xorl    %ebx, %ebx
>       call    srand@PLT
>       .p2align 4,,10
>       .p2align 3
> .L15:
>       call    rand@PLT
>       cmpl    $1073741824, %eax
>       adcl    $0, %ebx
>       addl    $1, %r15d
>       cmpl    %ebp, %r15d
>       jne     .L15
>       pxor    %xmm0, %xmm0
>       pxor    %xmm1, %xmm1
>       movl    %r15d, %esi
>       movq    %r13, %rdi
>       cvtsi2sdq       %rbx, %xmm0
>       mulsd   .LC1(%rip), %xmm0
>       movl    $1, %eax
>       cvtsi2sdl       %r15d, %xmm1
>       leal    (%r15,%r15,4), %ebp
>       addl    %ebp, %ebp
135,138c131,138
<       movq    %xmm0, %rax
<       movq    %rax, %xmm0
<       leaq    .LC3(%rip), %rax
<       movq    %rax, %rdi
---
>       call    printf@PLT
>       subl    $1, %r12d
>       jne     .L18
>       call    clock@PLT
>       pxor    %xmm1, %xmm1
>       leaq    .LC4(%rip), %rdi
>       subq    %r14, %rax
>       cvtsi2sdq       %rax, %xmm1
139a140,142
>       movapd  %xmm1, %xmm0
>       movsd   %xmm1, 8(%rsp)
>       divsd   .LC3(%rip), %xmm0
141,150c144
<       movq    -8(%rbp), %rax
<       subq    -16(%rbp), %rax
<       pxor    %xmm0, %xmm0
<       cvtsi2sdq       %rax, %xmm0
<       movsd   .LC4(%rip), %xmm1
<       divsd   %xmm1, %xmm0
<       movq    %xmm0, %rax
<       movq    %rax, %xmm0
<       leaq    .LC5(%rip), %rax
<       movq    %rax, %rdi
---
>       movsd   8(%rsp), %xmm1
151a146,148
>       divsd   .LC5(%rip), %xmm1
>       leaq    .LC6(%rip), %rdi
>       movapd  %xmm1, %xmm0
153,155c150,164
<       movl    $0, %eax
<       leave
<       .cfi_def_cfa 7, 8
---
>       addq    $24, %rsp
>       .cfi_def_cfa_offset 56
>       xorl    %eax, %eax
>       popq    %rbx
>       .cfi_def_cfa_offset 48
>       popq    %rbp
>       .cfi_def_cfa_offset 40
>       popq    %r12
>       .cfi_def_cfa_offset 32
>       popq    %r13
>       .cfi_def_cfa_offset 24
>       popq    %r14
>       .cfi_def_cfa_offset 16
>       popq    %r15
>       .cfi_def_cfa_offset 8
158c167
< .LFE7:
---
> .LFE23:
160c169
<       .section        .rodata
---
>       .section        .rodata.cst8,"aM",@progbits,8
162c171
< .LC0:
---
> .LC1:
166c175
< .LC2:
---
> .LC3:
170c179
< .LC4:
---
> .LC5:
cribari@darwin7 ~ $

Linux, street photography and Baroque music enthusiast.

Offline

#9 2022-04-05 15:42:03

Scimmia
Fellow
Registered: 2012-09-01
Posts: 13,729

Re: gcc optimization (for speed) flags not working

So you're using ccache. "Compiler cache that speeds up recompilation by caching previous compilations"

Online

#10 2022-04-05 15:42:22

loqs
Member
Registered: 2014-03-06
Posts: 18,970

Re: gcc optimization (for speed) flags not working

/usr/lib/ccache/bin/gcc

ccache would be a suspect if not for the later outputs showing gcc produces different assembler at different optimization levels.
The question remains why that is not translating into different timings for you.
Edit:
If you invoke /usr/bin/gcc to avoid ccache any difference?

Last edited by loqs (2022-04-05 15:46:44)

Offline

#11 2022-04-05 16:03:34

cribari
Member
From: Recife, Brazil
Registered: 2016-01-28
Posts: 84
Website

Re: gcc optimization (for speed) flags not working

Scimmia wrote:

So you're using ccache. "Compiler cache that speeds up recompilation by caching previous compilations"

I removed ccache from my system, but the problem persisted. Suggestions are welcome.

Linux, street photography and Baroque music enthusiast.

Offline

#12 2022-04-05 16:04:40

cribari
Member
From: Recife, Brazil
Registered: 2016-01-28
Posts: 84
Website

Re: gcc optimization (for speed) flags not working

loqs wrote:
/usr/lib/ccache/bin/gcc

ccache would be a suspect if not for the later outputs showing gcc produces different assembler at different optimization levels.
The question remains why that is not translating into different timings for you.
Edit:
If you invoke /usr/bin/gcc to avoid ccache any difference?

Unfortunately, not.

Linux, street photography and Baroque music enthusiast.

Offline

#13 2022-04-05 17:29:28

cribari
Member
From: Recife, Brazil
Registered: 2016-01-28
Posts: 84
Website

Re: gcc optimization (for speed) flags not working

I have a backup laptop that runs Arch. The problem also happens there.

I have a second backup laptop (an older machine) that runs Manjaro. I hadn't updated it for a while. I compiled and executed the C program in it using different optimization flags. gcc version was 11.1.0. The optimization flags worked as expected. I then updated all packages and gcc was updated to version 11.2.0. I rebooted, compiled the C program with different optimization flags, and such flags no longer worked as expected; see below. Somehow, the issue must be related to gcc version 11.2. 

[cribari@darwin4 ~]$ gcc --version
gcc (GCC) 11.1.0
Copyright (C) 2021 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

[cribari@darwin4 ~]$ gcc -O3 coin.c
[cribari@darwin4 ~]$ ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.09 seconds
EXECUTION TIME: 0.00 minutes
[cribari@darwin4 ~]$ gcc -O0 coin.c
[cribari@darwin4 ~]$ ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.16 seconds
EXECUTION TIME: 0.00 minutes
[cribari@darwin4 ~]$ 

[cribari@darwin4 ~]$ gcc --version
gcc (GCC) 11.2.0
Copyright (C) 2021 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

[cribari@darwin4 ~]$ gcc -O3 coin.c
[cribari@darwin4 ~]$ ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.22 seconds
EXECUTION TIME: 0.00 minutes
[cribari@darwin4 ~]$ gcc -O0 coin.c
[cribari@darwin4 ~]$ ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.22 seconds
EXECUTION TIME: 0.00 minutes
[cribari@darwin4 ~]$

Linux, street photography and Baroque music enthusiast.

Offline

#14 2022-04-05 21:01:23

latalante1
Member
Registered: 2018-08-30
Posts: 111

Re: gcc optimization (for speed) flags not working

I am not a programmer.
You omitted from this the most important C base library - glibc (its various versions, archlinux, manjaro, ubuntu).
Especially those functions that are the main duty of this program.

78.08%  coin    libc.so.6             [.] random
 3.44%  coin    libc.so.6             [.] rand
12.50%  coin    libc.so.6             [.] random_r

Perf will tell you the truth.
Certainly the version compiled with -O0 will follow many more instructions than the version -O2.

perf stat --repeat 10 -e instructions,cycles,cpu-clock --sync ./coin_O0
perf stat --repeat 10 -e instructions,cycles,cpu-clock --sync ./coin_O2

Alternatively, build a static version. Test it on different systems. From ububtu under archlinux and vice versa.

Maybe the programmer will explain it much more clearly.

Offline

#15 2022-04-05 21:09:23

rowdog
Member
From: East Texas
Registered: 2009-08-19
Posts: 118

Re: gcc optimization (for speed) flags not working

I ran the code with both -O0 and -O3 a number of times and got results that vary by as much as 40%. My conclusion is that the sample size is too small and the test is overwhelmed by the noise of the system (because no system is actually idle).

When I add an extra zero on COIN_MAX, I see about a 10% increase for -O3.

There's another thing I noticed too. Arch with gcc 11.2.0 and Gentoo with gcc 11.2.1 both show about 10% speed up while Debian with gcc 10.2.1 shows a large speed up of as much as 30% in both a virtual machine and on a Raspberry Pi. I'm not sure what's going on there but I suspect improvements gcc and/or glibc.

I also noted that your code rounds the result to %.2f which can round away the 10% increase on such a small sample.

My suggestion is to try it again with 

#define COIN_MAX 100000000

There can also be a difference in chips so this was tested on AMD 3900xt and RPI 4B.

Offline

#16 2022-04-05 21:21:04

cribari
Member
From: Recife, Brazil
Registered: 2016-01-28
Posts: 84
Website

Re: gcc optimization (for speed) flags not working

rowdog wrote:

I ran the code with both -O0 and -O3 a number of times and got results that vary by as much as 40%. My conclusion is that the sample size is too small and the test is overwhelmed by the noise of the system (because no system is actually idle).

When I add an extra zero on COIN_MAX, I see about a 10% increase for -O3.

There's another thing I noticed too. Arch with gcc 11.2.0 and Gentoo with gcc 11.2.1 both show about 10% speed up while Debian with gcc 10.2.1 shows a large speed up of as much as 30% in both a virtual machine and on a Raspberry Pi. I'm not sure what's going on there but I suspect improvements gcc and/or glibc.

I also noted that your code rounds the result to %.2f which can round away the 10% increase on such a small sample.

My suggestion is to try it again with 

#define COIN_MAX 100000000

There can also be a difference in chips so this was tested on AMD 3900xt and RPI 4B.

1) Up to gcc 11.1 this code ran in 0.08 second on my Arch laptop when compiled with -O3. 2) As per your suggestion, I added an extra zero to COIN_MAX; the results are below. No major change. 

cribari@darwin7 ~ $ gcc --version
gcc (GCC) 11.2.0
Copyright (C) 2021 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

cribari@darwin7 ~ $ gcc -O0 coin.c && ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
Percentage of tails from 100000000 tosses: 50.00
EXECUTION TIME: 1.68 seconds
EXECUTION TIME: 0.03 minutes
cribari@darwin7 ~ $ gcc -O3 coin.c && ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
Percentage of tails from 100000000 tosses: 50.00
EXECUTION TIME: 1.65 seconds
EXECUTION TIME: 0.03 minutes

Linux, street photography and Baroque music enthusiast.

Offline

#17 2022-04-05 21:24:20

cribari
Member
From: Recife, Brazil
Registered: 2016-01-28
Posts: 84
Website

Re: gcc optimization (for speed) flags not working

latalante1 wrote:

You omitted from this the most important C base library - glibc (its various versions, archlinux, manjaro, ubuntu).

Here is what I have: 

cribari@darwin7 ~ $ pacman -Qi glibc
Name            : glibc
Version         : 2.35-3
Description     : GNU C Library
Architecture    : x86_64
URL             : https://www.gnu.org/software/libc
Licenses        : GPL  LGPL
Groups          : None
Provides        : None
Depends On      : linux-api-headers>=4.10  tzdata  filesystem
Optional Deps   : gd: for memusagestat [installed]
                  perl: for mtrace [installed]
Required By     : a52dec  acpi  alsa-lib  alsa-plugins  alsa-utils  argon2
                  attr  audacity  audit  babl  base  bash  bind  binutils
                  bison  bluez-libs  brotli  btrfs-progs  bubblewrap  bzip2
                  c-ares  cabextract  ccache  cdparanoia  ceph-libs
                  chromaprint  confuse  coreutils  cracklib  dav1d
                  debuginfod  device-mapper  diffutils  ding-libs  discount
                  dmidecode  dosfstools  dotconf  efibootmgr  efivar
                  elfutils  enscript  ethtool  exfatprogs  expat  faac
                  faad2  fakeroot  fftw  file  findutils  flex  fluidsynth
                  fribidi  fuse2  fuse3  gavl  gawk  gcc-libs  gdb  gdbm
                  giflib  gnupg  grep  gsl  gsm  gvim  gzip  haveged  hdparm
                  hiredis  hspell  http-parser  hwloc  hyphen  ijs
                  iniparser  iproute2  jack2  jansson  jasper  jdk-openjdk
                  jdk11-openjdk  jemalloc  jpegoptim  jre-openjdk
                  jre-openjdk-headless  jre11-openjdk
                  jre11-openjdk-headless  json-c  kbd  keyutils  kmod  krb5
                  l-smash  ladspa  less  libaec  libao  libass  libassuan
                  libasyncns  libatomic_ops  libavif  libavtp  libbsd
                  libburn  libcap  libcap-ng  libcbor  libcddb  libdatrie
                  libdmtx  libdvbpsi  libdvdcss  libdvdread  libebur128
                  libedit  libepoxy  libevdev  libexif  libexttextcat
                  libffi  libfido2  libfreeaptx  libfreexl  libgit2
                  libgpg-error  libical  libice  libid3tag  libidn  libinih
                  libinstpatch  libjpeg-turbo  libkate  libksba  libldac
                  liblouis  libltc  libmad  libmaxminddb  libmd  libmicrodns
                  libmms  libmnl  libmpcdec  libmpeg2  libmypaint  libndp
                  libnet  libnfnetlink  libnghttp2  libnl  libnsl  libogg
                  libomxil-bellagio  libopenmpt  libp11-kit  libpcap
                  libpciaccess  libpgm  libpipeline  libplacebo  libraw1394
                  libsamplerate  libseccomp  libsigsegv  libsndfile
                  libsodium  libstemmer  libsynctex  libtasn1  libtiff
                  libtommath  libtool  libunistring  libupnp  liburing
                  libusb  libutempter  libuv  libvisual  libxau  libxcrypt
                  libxcvt  libxdmcp  libxshmfence  lilv  lmdb  lpsolve  lsof
                  lz4  lzo  m4  make  mbedtls  md4c  mdadm  metis
                  mkinitcpio-busybox  mlocate  mpfr  mtdev  mtools  ncurses
                  npth  nspr  nss-mdns  numactl  oath-toolkit  ocl-icd
                  opencore-amr  openmp  openmpi  openssh  openssl
                  opera-ffmpeg-codecs  optipng  opus  orc  ostree  packer
                  pacman  pam  patch  pciutils  perl  pinentry  pixman
                  pkgconf  popt  portaudio  portsmf  ppp  procps-ng
                  protobuf  psutils  raptor  rasqal  rav1e  readline  rhash
                  rnnoise  rpcsvc-proto  rtmpdump  run-parts  sane  sbc  sdl
                  sdl2  sed  serd  sg3_utils  shaderc  snappy  socat  sord
                  speexdsp  sratom  sudo  suil  svt-av1  sysfsutils
                  systemd-libs  talloc  tar  tdb  traceroute  tslib  twolame
                  usbutils  vamp-plugin-sdk  vid.stab  vivaldi-ffmpeg-codecs
                  vulkan-icd-loader  wavemon  wavpack  wayland  wget  which
                  xapian-core  xorg-bdftopcf  xvidcore  xxhash  yaml-cpp
                  zimg  zita-alsa-pcmi  zita-resampler  zlib  zstd
Optional For    : None
Conflicts With  : None
Replaces        : None
Installed Size  : 47.76 MiB
Packager        : Frederik Schwan <freswa@archlinux.org>
Build Date      : Sun 20 Mar 2022 05:45:03 PM -03
Install Date    : Wed 23 Mar 2022 08:36:15 AM -03
Install Reason  : Installed as a dependency for another package
Install Script  : Yes
Validated By    : Signature

Linux, street photography and Baroque music enthusiast.

Offline

#18 2022-04-05 21:41:14

latalante1
Member
Registered: 2018-08-30
Posts: 111

Re: gcc optimization (for speed) flags not working

@cribari
I tested this code under Archlinux with the following compilers.
gcc (GCC) 4.9.4
gcc (GCC) 9.3.1 20210319
gcc (GCC) 11.2.0 (default Archlinux)
gcc (GCC) 11.2.1 20211030
gcc (GCC) 12.0.1 20220220 (experimental)

Under my cpu there is no difference in the output of this program (different optimizations -O0 and -O2).

The differences between -O0 and -O2, measured with perf, are distinct.

In the case of my CPU. 94% of the functions during the execution of this program are in libc.so.6 (random, rand, random_r). As I guess, the compiler doesn't optimize them, because they come with glibc from Archlinux repository.

perf record -e cycles:pp --call-graph dwarf ./coin
perf report

Last edited by latalante1 (2022-04-05 22:13:04)

Offline

#19 2022-04-05 22:18:19

loqs
Member
Registered: 2014-03-06
Posts: 18,970

Re: gcc optimization (for speed) flags not working

latalante1 wrote:

Alternatively, build a static version. Test it on different systems. From ububtu under archlinux and vice versa.

Off topic but in theory with a static version the compiler could optimize through to just printing the results as everything is then known at compile time.

Offline

#20 2022-04-05 22:39:21

latalante1
Member
Registered: 2018-08-30
Posts: 111

Re: gcc optimization (for speed) flags not working

@loqs
I do not know. I pushed this idea to point glibc in the crosshair.

The same program compiled and run under Fedora shows identical output to Archlinux.

gcc (GCC) 11.2.1 20220127 (Red Hat 11.2.1-9)
glibc-2.34

Offline

#21 2022-04-05 23:34:03

cribari
Member
From: Recife, Brazil
Registered: 2016-01-28
Posts: 84
Website

Re: gcc optimization (for speed) flags not working

Ubuntu (my desktop after upgrade), the optimization flags work as expected:

cribari@edgeworth:~$ apt show gcc
Package: gcc
Version: 4:9.3.0-1ubuntu2
Priority: optional
Build-Essential: yes
Section: devel
Source: gcc-defaults (1.185.1ubuntu2)
Origin: Ubuntu
Maintainer: Ubuntu Developers <ubuntu-devel-discuss@lists.ubuntu.com>
Original-Maintainer: Debian GCC Maintainers <debian-gcc@lists.debian.org>
Bugs: https://bugs.launchpad.net/ubuntu/+filebug
Installed-Size: 51.2 kB
Provides: c-compiler, gcc-x86-64-linux-gnu (= 4:9.3.0-1ubuntu2)
Depends: cpp (= 4:9.3.0-1ubuntu2), gcc-9 (>= 9.3.0-3~)
Recommends: libc6-dev | libc-dev
Suggests: gcc-multilib, make, manpages-dev, autoconf, automake, libtool, flex, bison, gdb, gcc-doc
Conflicts: gcc-doc (<< 1:2.95.3)
Task: ubuntustudio-video, ubuntu-mate-core, ubuntu-mate-desktop
Download-Size: 5,208 B
APT-Manual-Installed: yes
APT-Sources: http://archive.ubuntu.com/ubuntu focal/main amd64 Packages
Description: GNU C compiler
 This is the GNU C compiler, a fairly portable optimizing compiler for C.
 .
 This is a dependency package providing the default GNU C compiler.

cribari@edgeworth:~$ apt show libc6
Package: libc6
Version: 2.31-0ubuntu9.7
Priority: required
Section: libs
Source: glibc
Origin: Ubuntu
Maintainer: Ubuntu Developers <ubuntu-devel-discuss@lists.ubuntu.com>
Original-Maintainer: GNU Libc Maintainers <debian-glibc@lists.debian.org>
Bugs: https://bugs.launchpad.net/ubuntu/+filebug
Installed-Size: 13.5 MB
Depends: libgcc-s1, libcrypt1 (>= 1:4.4.10-10ubuntu4)
Recommends: libidn2-0 (>= 2.0.5~)
Suggests: glibc-doc, debconf | debconf-2.0, locales
Conflicts: openrc (<< 0.27-2~)
Breaks: hurd (<< 1:0.9.git20170910-1), iraf-fitsutil (<< 2018.07.06-4), libtirpc1 (<< 0.2.3), locales (<< 2.31), locales-all (<< 2.31), nocache (<< 1.1-1~), nscd (<< 2.31), r-cran-later (<< 0.7.5+dfsg-2), wcc (<< 0.0.2+dfsg-3)
Replaces: libc6-amd64
Homepage: https://www.gnu.org/software/libc/libc.html
Task: minimal
Original-Vcs-Browser: https://salsa.debian.org/glibc-team/glibc
Original-Vcs-Git: https://salsa.debian.org/glibc-team/glibc.git
Download-Size: 2,714 kB
APT-Manual-Installed: yes
APT-Sources: http://archive.ubuntu.com/ubuntu focal-updates/main amd64 Packages
Description: GNU C Library: Shared libraries
 Contains the standard libraries that are used by nearly all programs on
 the system. This package includes shared versions of the standard C library
 and the standard math library, as well as many others.

N: There is 1 additional record. Please use the '-a' switch to see it

cribari@edgeworth:~$ gcc -O0 coin.c && ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.13 seconds
EXECUTION TIME: 0.00 minutes
cribari@edgeworth:~$ gcc -O3 coin.c && ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
EXECUTION TIME: 0.07 seconds
EXECUTION TIME: 0.00 minutes
cribari@edgeworth:~$

EDIT: Arch with #define COIN_MAX 1000000000: 

cribari@darwin7 ~ $ gcc -O0 coin.c && ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
Percentage of tails from 100000000 tosses: 50.00
Percentage of tails from 1000000000 tosses: 50.00
EXECUTION TIME: 16.43 seconds
EXECUTION TIME: 0.27 minutes
cribari@darwin7 ~ $ gcc -O3 coin.c && ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
Percentage of tails from 100000000 tosses: 50.00
Percentage of tails from 1000000000 tosses: 50.00
EXECUTION TIME: 15.87 seconds
EXECUTION TIME: 0.26 minutes

Ubuntu with #define COIN_MAX 1000000000: 

cribari@edgeworth:~$ gcc -O0 coin.c && ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
Percentage of tails from 100000000 tosses: 50.00
Percentage of tails from 1000000000 tosses: 50.00
EXECUTION TIME: 13.05 seconds
EXECUTION TIME: 0.22 minutes
cribari@edgeworth:~$ gcc -O3 coin.c && ./a.out
Percentage of tails from 10 tosses: 60.00
Percentage of tails from 100 tosses: 54.00
Percentage of tails from 1000 tosses: 48.50
Percentage of tails from 10000 tosses: 49.55
Percentage of tails from 100000 tosses: 49.87
Percentage of tails from 1000000 tosses: 49.90
Percentage of tails from 10000000 tosses: 49.98
Percentage of tails from 100000000 tosses: 50.00
Percentage of tails from 1000000000 tosses: 50.00
EXECUTION TIME: 7.39 seconds
EXECUTION TIME: 0.12 minutes

Last edited by cribari (2022-04-06 00:46:32)

Linux, street photography and Baroque music enthusiast.

Offline

#22 2022-04-05 23:41:21

loqs
Member
Registered: 2014-03-06
Posts: 18,970

Re: gcc optimization (for speed) flags not working

cribari wrote:

the optimization flags work as expected:

Given the results in the thread so far the remaining issue is that your expectations are not matching the results due to the comparison being flawed,  too many changed variables between samples and sample size is too small.

Offline

Pages: 1

Board footer

Powered by FluxBB