On Fri, Mar 02, 2018 at 01:19:56PM +0800, Dave Young wrote:
> On 02/27/18 at 01:48pm, AKASHI Takahiro wrote:
> > The code guarded by CONFIG_X86_64 is necessary on some architectures
> > which have a dedicated kernel mapping outside of linear memory mapping.
> > (arm64 is among those.)
> >
> > In this patch, an additional argument, kernel_map, is added to enable/
> > disable the code removing #ifdef.
> >
> > Signed-off-by: AKASHI Takahiro <takahiro.akashi@xxxxxxxxxx>
> > Cc: Dave Young <dyoung@xxxxxxxxxx>
> > Cc: Vivek Goyal <vgoyal@xxxxxxxxxx>
> > Cc: Baoquan He <bhe@xxxxxxxxxx>
> > ---
> > arch/x86/kernel/crash.c | 25 +++++++++++++------------
> > 1 file changed, 13 insertions(+), 12 deletions(-)
> >
> > diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
> > index 2123fa0efc17..913fd8021f8a 100644
> > --- a/arch/x86/kernel/crash.c
> > +++ b/arch/x86/kernel/crash.c
> > @@ -347,7 +347,7 @@ static int prepare_elf64_ram_headers_callback(struct resource *res, void *arg)
> > return 0;
> > }
> >
> > -static int prepare_elf64_headers(struct crash_elf_data *ced,
> > +static int prepare_elf64_headers(struct crash_elf_data *ced, int kernel_map,
> > void **addr, unsigned long *sz)
> > {
> > Elf64_Ehdr *ehdr;
> > @@ -414,17 +414,17 @@ static int prepare_elf64_headers(struct crash_elf_data *ced,
> > phdr->p_filesz = phdr->p_memsz = VMCOREINFO_NOTE_SIZE;
> > (ehdr->e_phnum)++;
> >
> > -#ifdef CONFIG_X86_64
> > /* Prepare PT_LOAD type program header for kernel text region */
> > - phdr = (Elf64_Phdr *)bufp;
> > - bufp += sizeof(Elf64_Phdr);
> > - phdr->p_type = PT_LOAD;
> > - phdr->p_flags = PF_R|PF_W|PF_X;
> > - phdr->p_vaddr = (Elf64_Addr)_text;
> > - phdr->p_filesz = phdr->p_memsz = _end - _text;
> > - phdr->p_offset = phdr->p_paddr = __pa_symbol(_text);
> > - (ehdr->e_phnum)++;
> > -#endif
> > + if (kernel_map) {
> > + phdr = (Elf64_Phdr *)bufp;
> > + bufp += sizeof(Elf64_Phdr);
> > + phdr->p_type = PT_LOAD;
> > + phdr->p_flags = PF_R|PF_W|PF_X;
> > + phdr->p_vaddr = (Elf64_Addr)_text;
> > + phdr->p_filesz = phdr->p_memsz = _end - _text;
> > + phdr->p_offset = phdr->p_paddr = __pa_symbol(_text);
> > + (ehdr->e_phnum)++;
> > + }
> >
> > /* Go through all the ranges in cmem->ranges[] and prepare phdr */
> > for (i = 0; i < cmem->nr_ranges; i++) {
> > @@ -477,7 +477,8 @@ static int prepare_elf_headers(struct kimage *image, void **addr,
> > goto out;
> >
> > /* By default prepare 64bit headers */
> > - ret = prepare_elf64_headers(ced, addr, sz);
> > + ret = prepare_elf64_headers(ced,
> > + (int)IS_ENABLED(CONFIG_X86_64), addr, sz);
>
> A bool would be enough for kernel_map
Yeah, for now.
What I thought of is that we might want to extend its functionality
in the future as I did here for kernel_map without changing its interface.
But I'd defer to you.
-Takahiro AKASHI
> > if (ret)
> > goto out;
> >
>
> Thanks
> Dave
> > --
> > 2.16.2
> >
_______________________________________________
kexec mailing list
kexec@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/kexec
