Hello Simon,

what's up with this patch? Do you plan to apply it, or is there still
an issue?

Regards,
Petr Tesarik

On Thu, 24 Sep 2015 08:48:52 +0200
Petr Tesarik <[email protected]> wrote:

> There may be more than one crash kernel regions on x86. Currently,
> kexec-tools picks the largest one. If high reservation is smaller
> than low, it will try to load panic kernel low. However, the kexec
> syscall checks that target address is within crashk_res boundaries,
> so attempts to load crash kernel low result in -EADDRNOTAVAIL, and
> kexec prints out this error message:
> 
> kexec_load failed: Cannot assign requested address
> 
> Looking at the logic in arch/x86/kernel/setup.c, there are only two
> possible layouts:
> 
>   1. crashk_res is below 4G, and there is only one region,
>   2. crashk_res is above 4G, and crashk_low_res is below 4G
> 
> In either case, kexec-tools must pick the highest region.
> 
> Changelog:
> 
>   * v3: rename function to get_crash_kernel_load_range
>   * v2: remove unnecessary local variables
> 
> Signed-off-by: Petr Tesarik <[email protected]>
> 
> ---
>  kexec/arch/i386/crashdump-x86.c    | 21 +++++++--------------
>  kexec/arch/i386/kexec-x86-common.c |  4 ++--
>  kexec/kexec.h                      |  2 +-
>  3 files changed, 10 insertions(+), 17 deletions(-)
> 
> diff --git a/kexec/arch/i386/crashdump-x86.c b/kexec/arch/i386/crashdump-x86.c
> index 63959b7..d4969c5 100644
> --- a/kexec/arch/i386/crashdump-x86.c
> +++ b/kexec/arch/i386/crashdump-x86.c
> @@ -1026,24 +1026,17 @@ int load_crashdump_segments(struct kexec_info *info, 
> char* mod_cmdline,
>       return 0;
>  }
>  
> -int get_max_crash_kernel_limit(uint64_t *start, uint64_t *end)
> +/* On x86, the kernel may make a low reservation in addition to the
> + * normal reservation. However, the kernel refuses to load the panic
> + * kernel to low memory, so always choose the highest range.
> + */
> +int get_crash_kernel_load_range(uint64_t *start, uint64_t *end)
>  {
> -     int i, idx = -1;
> -     unsigned long sz_max = 0, sz;
> -
>       if (!crash_reserved_mem_nr)
>               return -1;
>  
> -     for (i = crash_reserved_mem_nr - 1; i >= 0; i--) {
> -             sz = crash_reserved_mem[i].end - crash_reserved_mem[i].start +1;
> -             if (sz <= sz_max)
> -                     continue;
> -             sz_max = sz;
> -             idx = i;
> -     }
> -
> -     *start = crash_reserved_mem[idx].start;
> -     *end = crash_reserved_mem[idx].end;
> +     *start = crash_reserved_mem[crash_reserved_mem_nr - 1].start;
> +     *end = crash_reserved_mem[crash_reserved_mem_nr - 1].end;
>  
>       return 0;
>  }
> diff --git a/kexec/arch/i386/kexec-x86-common.c 
> b/kexec/arch/i386/kexec-x86-common.c
> index 041c8ec..484d712 100644
> --- a/kexec/arch/i386/kexec-x86-common.c
> +++ b/kexec/arch/i386/kexec-x86-common.c
> @@ -371,9 +371,9 @@ int get_memory_ranges(struct memory_range **range, int 
> *ranges,
>           !(kexec_flags & KEXEC_PRESERVE_CONTEXT)) {
>               uint64_t start, end;
>  
> -             ret = get_max_crash_kernel_limit(&start, &end);
> +             ret = get_crash_kernel_load_range(&start, &end);
>               if (ret != 0) {
> -                     fprintf(stderr, "get_max_crash_kernel_limit failed.\n");
> +                     fprintf(stderr, "get_crash_kernel_load_range 
> failed.\n");
>                       return -1;
>               }
>  
> diff --git a/kexec/kexec.h b/kexec/kexec.h
> index 0fa977f..7c97b25 100644
> --- a/kexec/kexec.h
> +++ b/kexec/kexec.h
> @@ -288,7 +288,7 @@ int arch_process_options(int argc, char **argv);
>  int arch_compat_trampoline(struct kexec_info *info);
>  void arch_update_purgatory(struct kexec_info *info);
>  int is_crashkernel_mem_reserved(void);
> -int get_max_crash_kernel_limit(uint64_t *start, uint64_t *end);
> +int get_crash_kernel_load_range(uint64_t *start, uint64_t *end);
>  char *get_command_line(void);
>  
>  int kexec_iomem_for_each_line(char *match,


_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec

Reply via email to