s390/vtime: correct scaled cputime for SMT
The scaled cputime is supposed to be derived from the normal per-thread cputime by dividing it with the average thread density in the last interval. The calculation of the scaling values for the average thread density is incorrect. The current, incorrect calculation: Ci = cycle count with i active threads T = unscaled cputime, sT = scaled cputime sT = T * (C1 + C2 + ... + Cn) / (1*C1 + 2*C2 + ... + n*Cn) The calculation happens to yield the correct numbers for the simple cases with only one Ci value not zero. But for cases with multiple Ci values not zero it fails. E.g. on a SMT-2 system with one thread active half the time and two threads active for the other half of the time it fails, the scaling factor should be 3/4 but the formula gives 2/3. The correct formula is sT = T * (C1/1 + C2/2 + ... + Cn/n) / (C1 + C2 + ... + Cn) Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
This commit is contained in:
Родитель
58f8e9da33
Коммит
61cc37906b
|
@ -89,17 +89,21 @@ static int do_account_vtime(struct task_struct *tsk, int hardirq_offset)
|
|||
if (smp_cpu_mtid &&
|
||||
time_after64(jiffies_64, __this_cpu_read(mt_scaling_jiffies))) {
|
||||
u64 cycles_new[32], *cycles_old;
|
||||
u64 delta, mult, div;
|
||||
u64 delta, fac, mult, div;
|
||||
|
||||
cycles_old = this_cpu_ptr(mt_cycles);
|
||||
if (stcctm5(smp_cpu_mtid + 1, cycles_new) < 2) {
|
||||
fac = 1;
|
||||
mult = div = 0;
|
||||
for (i = 0; i <= smp_cpu_mtid; i++) {
|
||||
delta = cycles_new[i] - cycles_old[i];
|
||||
mult += delta;
|
||||
div += (i + 1) * delta;
|
||||
div += delta;
|
||||
mult *= i + 1;
|
||||
mult += delta * fac;
|
||||
fac *= i + 1;
|
||||
}
|
||||
if (mult > 0) {
|
||||
div *= fac;
|
||||
if (div > 0) {
|
||||
/* Update scaling factor */
|
||||
__this_cpu_write(mt_scaling_mult, mult);
|
||||
__this_cpu_write(mt_scaling_div, div);
|
||||
|
|
Загрузка…
Ссылка в новой задаче