On Sun, 15 Apr 2018, Carsten Emde wrote:
> Hi Clark and John,
>
> Cylictest's --smi option allows to continuously display SMIs that occur
> during the measurement. However, the SMI count was not yet included in the
> summary at the end of the program run. Added this.
>
> While we are at it, update the list of Intel family 6 models that support
> retrieving the SMI count by copying newly added items from kernel sources
> at tools/power/x86/turbostat.
>
> Signed-off-by: Carsten Emde <C.Emde@xxxxxxxxx>
>
> ---
> src/cyclictest/cyclictest.c | 11 +++++++++++
> 1 file changed, 11 insertions(+)
>
> Index: rt-tests/src/cyclictest/cyclictest.c
> ===================================================================
> --- rt-tests.orig/src/cyclictest/cyclictest.c
> +++ rt-tests/src/cyclictest/cyclictest.c
> @@ -931,6 +931,10 @@ static int has_smi_counter(void)
> case 0x4D: /* AVN */
> case 0x4C: /* AMT */
> case 0x57: /* PHI */
> + case 0x5C: /* BXT */
> + case 0x5F: /* DNV */
> + case 0x7A: /* Gemini Lake */
> + case 0x85: /* Knights Mill */
> break;
> default:
> return 0;
> @@ -2032,6 +2036,13 @@ static void print_hist(struct thread_par
> fprintf(fd, " # %05lu others", par[i]->stats->hist_overflow - par[i]->stats->num_outliers);
> fprintf(fd, "\n");
> }
> + if (smi) {
> + fprintf(fd, "# SMIs:");
> + for (i = 0; i < nthreads; i++)
> + fprintf(fd, " %05lu", par[i]->stats->smi_count);
> + fprintf(fd, "\n");
> + }
> +
> fprintf(fd, "\n");
>
> if (use_histfile)
>
Hi Carsten
Thanks for the patch. It does seem to to conflict a bit with
49c61dcd3de7 cyclictest: new CPUs with SMI counter support
My guess is that you generated your patch for stable/v1.0
Could you regenerate the patch for this branch instead?
stable/devel/v1.0.1
Thanks
John
--
To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
