Le 28/08/2018 à 17:22, Shivaprasad G Bhat a écrit :
> If the hostpage size is greater than the TARGET_PAGESIZE, the
> target-pages of size TARGET_PAGESIZE are marked valid only till the
> length requested during the elfload. The glibc attempts to consume unused
> space in the last page of data segment(__libc_memalign() in
> elf/dl-minimal.c). If PT_LOAD p_align is greater than or
> equal to hostpage size, the GLRO(dl_pagesize) is actually the host pagesize
> as set in the auxillary vectors. So, there is no explicit mmap request for
> the remaining target-pages on the last hostpage. The glibc assumes that
> particular space as available and subsequent attempts to use
> those addresses lead to crash as the target_mmap has not marked them valid
> for target-pages as valid.
>
> The issue is seen when trying to chroot to 16.04-x86_64 ubuntu on a PPC64
> host where the fork fails to access the thread_id as it is allocated on a
> page not marked valid. The recent glibc doesnt have checks for thread-id in
> fork, but the issue can manifest somewhere else, none the less.
>
> The fix here is to map all the target-pages of the hostpage during the
> elfload if the p_align is greater than or equal to hostpage size, for
> data segment to allow the glibc for proper consumption.
>
> Signed-off-by: Shivaprasad G Bhat <sb...@linux.vnet.ibm.com>
> ---
> v1: https://lists.gnu.org/archive/html/qemu-devel/2018-08/msg05730.html
> Changes from v1:
> - Made the conditionals consistent with the commit "33143c446e" and
> changed the commit message accordingly.
>
> linux-user/elfload.c | 37 +++++++++++++++++++++++++++++--------
> 1 file changed, 29 insertions(+), 8 deletions(-)
>
> diff --git a/linux-user/elfload.c b/linux-user/elfload.c
> index 8638612aec..cced43f45c 100644
> --- a/linux-user/elfload.c
> +++ b/linux-user/elfload.c
> @@ -1438,9 +1438,23 @@ struct exec
>
> /* Necessary parameters */
> #define TARGET_ELF_EXEC_PAGESIZE TARGET_PAGE_SIZE
> -#define TARGET_ELF_PAGESTART(_v) ((_v) & \
> - ~(abi_ulong)(TARGET_ELF_EXEC_PAGESIZE-1))
> -#define TARGET_ELF_PAGEOFFSET(_v) ((_v) & (TARGET_ELF_EXEC_PAGESIZE-1))
> +#define TARGET_ELF_PAGESTART(_v, _a, _s, _m) \
> + (((_a & ~_m) != 0) ? \
> + (_v) & ~(abi_ulong)(TARGET_ELF_EXEC_PAGESIZE - 1) : \
> + ((TARGET_ELF_EXEC_PAGESIZE > _s) ? \
> + (_v) & ~(abi_ulong)(TARGET_ELF_EXEC_PAGESIZE - 1) : \
> + (_v) & ~(abi_ulong)(_s - 1)));
> +#define TARGET_ELF_PAGEOFFSET(_v, _a, _s, _m) \
> + (((_a & ~_m) != 0) ? \
> + (_v) & (TARGET_ELF_EXEC_PAGESIZE - 1) : \
> + ((TARGET_ELF_EXEC_PAGESIZE > _s) ? \
> + (_v) & (TARGET_ELF_EXEC_PAGESIZE - 1) : \
> + (_v) & (_s - 1)));
> +#define TARGET_ELF_PAGELENGTH(_v, _a, _s, _m) \
> + (((_a & ~_m) != 0) ? \
> + TARGET_PAGE_ALIGN(_v) : \
> + ((TARGET_ELF_EXEC_PAGESIZE > _s) ? \
> + TARGET_PAGE_ALIGN(_v) : HOST_PAGE_ALIGN(_v)));
I think you should use eppnt->p_align instead of info->alignment and
simplify this to:
#define TARGET_ELF_EXEC_PAGESIZE \
(((eppnt->p_align & ~qemu_host_page_mask) != 0) ? \
TARGET_PAGE_SIZE : MAX(qemu_host_page_size, TARGET_PAGE_SIZE))
#define TARGET_ELF_PAGELENGTH(_v) ROUND_UP((_v), TARGET_ELF_EXEC_PAGESIZE)
Thanks,
Laurent
