Re: [Crash-utility] [PATCH v2 1/2] diskdump/netdump: fix segmentation fault caused by failure of stopping CPUs
On Wed, May 31, 2023 at 12:52 PM HAGIO KAZUHITO(萩尾 一仁) <k-hagio-ab(a)nec.com>
wrote:
> void
> map_cpus_to_prstatus_kdump_cmprs(void)
> {
> @@ -131,7 +180,7 @@ map_cpus_to_prstatus_kdump_cmprs(void)
> nrcpus = (kt->kernel_NR_CPUS ? kt->kernel_NR_CPUS : NR_CPUS);
>
> for (i = 0, j = 0; i < nrcpus; i++) {
> - if (in_cpu_map(ONLINE_MAP, i)) {
> + if (in_cpu_map(ONLINE_MAP, i) &&
(!symbol_exists("crash_notes") || have_crash_notes(i))) {
Thank you for the update.
kernel_symbol_exists() is better.
And it might be also good to check it out of the loop only once.
Lianbo, what do you think?
int crash_notes_exists = kernel_symbol_exists("crash_notes");
Fine to me, we can take the above changes. Kazu.
And also thank you for sharing the steps on how to reproduce, HATAYAMA.
This definitely helps me a lot.
https://listman.redhat.com/archives/crash-utility/2023-May/010709.html
The v2 series looks good to me, and works fine on aarch64, so: Ack.
Thanks.
Lianbo
Anyway, we can change it when merging, so for the series,
Acked-by: Kazuhito Hagio <k-hagio-ab(a)nec.com>
Thanks,
Kazu
Attachments:
- attachment.html (text/html — 2.0 KB)