On Thu, May 26, 2016 at 09:49:42PM +0800, Zhangjian (Bamvor) wrote:
> Hi, yury
>
> The coredump is usable in our platform. It miss the following definition:
> +#define compat_elf_greg_telf_greg_t
> +#define compat_elf_gregset_t elf_gregset_t
>
> And it leads to the wrong register save in core
Hi, again
I found another issue in binfmt_ilp32.c. We are using the ELF_ET_DYN_BASE
for ilp32 application. The default ELF_ET_DYN_BASE is calculated by
TASK_SIZE_64. IIUC, we should define the following things in binfmt_ilp32.c
which is the same value as aarch32:
+#undef ELF_ET_DYN_BASE
+#define
Hi, Yury
On 2016/5/24 8:04, Yury Norov wrote:
to handle ILP32 binaries
Signed-off-by: Yury Norov
---
arch/arm64/kernel/Makefile | 1 +
arch/arm64/kernel/binfmt_ilp32.c | 91
2 files changed, 92 insertions(+)
create mode 100644 arch/arm64/ke
On Thu, May 26, 2016 at 09:49:42PM +0800, Zhangjian (Bamvor) wrote:
> Hi, yury
>
> The coredump is usable in our platform. It miss the following definition:
> +#define compat_elf_greg_telf_greg_t
> +#define compat_elf_gregset_t elf_gregset_t
>
> And it leads to the wrong register save in core
Hi, yury
The coredump is usable in our platform. It miss the following definition:
+#define compat_elf_greg_t elf_greg_t
+#define compat_elf_gregset_t elf_gregset_t
And it leads to the wrong register save in core dump. After apply this patch,
gdb could debug core dump files.
Here is the
