Re: [PATCH v13 10/15] x86/kexec: Use crash_prepare_headers() helper to simplify code

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Calling `crash_prepare_headers()` from the memory hotplug notifier causes a self-deadlock on `mem_hotplug_lock`.
--

commit 3ffc33a6d9cd3e7840883325b10a8172439b4b8f
Author: Jinjie Ruan <ruanjinjie@huawei.com>

x86/kexec: Use crash_prepare_headers() helper to simplify code

This commit replaces x86's custom prepare_elf_headers() with the generic
crash_prepare_headers() helper to handle cmem allocation, population, and
exclusion of crash ranges. The change implements the required
architecture-specific callbacks to support the generic core, reducing
code duplication.

> diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
> index 7fa6d45ebe3f7..10ef24611f2af 100644
> --- a/arch/x86/kernel/crash.c
> +++ b/arch/x86/kernel/crash.c

[ ... ]

> @@ -532,7 +465,7 @@ void arch_crash_handle_hotplug_event(struct kimage *image, void *arg)
>  	 * Create the new elfcorehdr reflecting the changes to CPU and/or
>  	 * memory resources.
>  	 */
> -	if (prepare_elf_headers(&elfbuf, &elfsz, &nr_mem_ranges)) {
> +	if (crash_prepare_headers(IS_ENABLED(CONFIG_X86_64), &elfbuf, &elfsz, NULL)) {

Will this cause a self-deadlock when called from the memory hotplug notifier?

During memory hotplug operations, the memory hotplug lock is held for
write before notifying the chain. For instance:

online_pages()
    mem_hotplug_begin() // takes mem_hotplug_lock for write
    ...
    memory_notify(MEM_ONLINE, ...)
        crash_memhp_notifier()
            arch_crash_handle_hotplug_event()
                crash_prepare_headers()

Then inside crash_prepare_headers() from kernel/crash_core.c:

int crash_prepare_headers(int need_kernel_map, void **addr, unsigned long *sz,
			  unsigned long *nr_mem_ranges)
{
	unsigned int max_nr_ranges;
	struct crash_mem *cmem;
	int ret;

	get_online_mems();

get_online_mems() attempts to acquire the mem_hotplug_lock for read. 

Since the percpu_rw_semaphore does not support recursive read-after-write
locking by the same thread, won't this block indefinitely and result in a
deadlock?

>  		pr_err("unable to create new elfcorehdr");
>  		goto out;
>  	}

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260511030454.1730881-1-ruanjinjie@huawei.com?part=10