[PATCH] arm64/efi: prefer AllocatePages() over efi_low_alloc() for vmlinux

[Ard Biesheuvel]

On 24 July 2015 at 12:49, Mark Rutland <mark.rutland at arm.com> wrote:
> Hi Ard,
>
> On Fri, Jul 24, 2015 at 10:41:53AM +0100, Ard Biesheuvel wrote:
>> When allocating memory for the kernel image, try the AllocatePages()
>> boot service to obtain memory at the preferred offset of
>> 'dram_base + TEXT_OFFSET', and only revert to efi_low_alloc() if that
>> fails. This is the only way to allocate at the base of DRAM if DRAM
>> starts at 0x0, since efi_low_alloc() refuses to allocate at 0x0.
>>
>> Tested-by: Haojian Zhuang <haojian.zhuang at linaro.org>
>> Signed-off-by: Ard Biesheuvel <ard.biesheuvel at linaro.org>
>> ---
>>  arch/arm64/kernel/efi-stub.c | 47 ++++++++++++++++++++++++++++++++++++--------
>>  1 file changed, 39 insertions(+), 8 deletions(-)
>>
>> diff --git a/arch/arm64/kernel/efi-stub.c b/arch/arm64/kernel/efi-stub.c
>> index f5374065ad53..c8df74d14368 100644
>> --- a/arch/arm64/kernel/efi-stub.c
>> +++ b/arch/arm64/kernel/efi-stub.c
>> @@ -13,7 +13,7 @@
>>  #include <asm/efi.h>
>>  #include <asm/sections.h>
>>
>> -efi_status_t __init handle_kernel_image(efi_system_table_t *sys_table,
>> +efi_status_t __init handle_kernel_image(efi_system_table_t *sys_table_arg,
>
> Any reason for the _arg addition?
>

Yes. Unfortunately, the efi_call_early() macro has a hidden
'efi_system_table_t *' parameter which it refers to by the name
'sys_table_arg'

>>                                       unsigned long *image_addr,
>>                                       unsigned long *image_size,
>>                                       unsigned long *reserve_addr,
>> @@ -23,21 +23,52 @@ efi_status_t __init handle_kernel_image(efi_system_table_t *sys_table,
>>  {
>>       efi_status_t status;
>>       unsigned long kernel_size, kernel_memsize = 0;
>> +     unsigned long nr_pages;
>>
>>       /* Relocate the image, if required. */
>>       kernel_size = _edata - _text;
>>       if (*image_addr != (dram_base + TEXT_OFFSET)) {
>>               kernel_memsize = kernel_size + (_end - _edata);
>> -             status = efi_low_alloc(sys_table, kernel_memsize + TEXT_OFFSET,
>> -                                    SZ_2M, reserve_addr);
>> +
>> +             //
>> +             // First, try a straight allocation at the preferred offset.
>> +             // This will work around the issue where, if dram_base == 0x0,
>> +             // efi_low_alloc() refuses to allocate at 0x0 (to prevent the
>> +             // address of the allocation to be mistaken for a FAIL return
>> +             // value or a NULL pointer). It will also ensure that, on
>> +             // platforms where the [dram_base, dram_base + TEXT_OFFSET)
>> +             // interval is partially occupied by the firmware (like on APM
>> +             // Mustang), we can still place the kernel at the address
>> +             // 'dram_base + TEXT_OFFSET'.
>> +             //
>
> /*
>  * Nit: please use the standard comment style
>  */
>

Absolutely. EDK2 development must be really screwing with my brain if
I am unconsciously using C++ style comments in C code. Time to move
back to the light side for a while :-)

>> +             *reserve_addr = dram_base + TEXT_OFFSET;
>> +             nr_pages = round_up(kernel_memsize, EFI_ALLOC_ALIGN) /
>> +                        EFI_PAGE_SIZE;
>> +             status = efi_call_early(allocate_pages, EFI_ALLOCATE_ADDRESS,
>> +                                     EFI_LOADER_DATA, nr_pages,
>> +                                     (efi_physical_addr_t *)reserve_addr);
>> +             if (status == EFI_SUCCESS) {
>> +                     memcpy((void *)*reserve_addr, (void *)*image_addr,
>> +                            kernel_size);
>> +                     *image_addr = *reserve_addr;
>> +                     *reserve_size = kernel_memsize;
>> +             } else {
>> +                     status = efi_low_alloc(sys_table_arg,
>> +                                            kernel_memsize + TEXT_OFFSET,
>> +                                            SZ_2M, reserve_addr);
>> +
>> +                     if (status == EFI_SUCCESS) {
>> +                             memcpy((void *)*reserve_addr + TEXT_OFFSET,
>> +                                    (void *)*image_addr,
>> +                                    kernel_size);
>> +                             *image_addr = *reserve_addr + TEXT_OFFSET;
>> +                             *reserve_size = kernel_memsize + TEXT_OFFSET;
>> +                     }
>> +             }
>>               if (status != EFI_SUCCESS) {
>> -                     pr_efi_err(sys_table, "Failed to relocate kernel\n");
>> +                     pr_efi_err(sys_table_arg, "Failed to relocate kernel\n");
>>                       return status;
>>               }
>> -             memcpy((void *)*reserve_addr + TEXT_OFFSET, (void *)*image_addr,
>> -                    kernel_size);
>
> Could we have a new_image_addr assigned in each case, and keep the
> common memcpy here, followed by assignment to *image_addr? That would
> save a couple of lines and guarantee the two cases stay in sync.
>

Well, the memcpy() occurs before the assignment of *image_addr, which
is also used as the src arg. So I could record the value of
*image_addr in a temp, I suppose. I will do that in the next version.

> Otherwise this looks good to me.
>

Thanks,
Ard.