[efi:urgent 1/1] drivers/firmware/efi/libstub/fdt.c:303:9: warning: cast from pointer to integer of different size

[efi:urgent 1/1] drivers/firmware/efi/libstub/fdt.c:303:9: warning: cast from pointer to integer of different size

[kbuild test robot]

[-- Attachment #1: Type: text/plain, Size: 2056 bytes --]

tree:   https://git.kernel.org/pub/scm/linux/kernel/git/mfleming/efi.git urgent
head:   5615742edc64c99bbdec5e98bf9d9d872756d99c
commit: 5615742edc64c99bbdec5e98bf9d9d872756d99c [1/1] efi/libstub: arm*: Pass latest memory map to the kernel
config: arm-multi_v7_defconfig (attached as .config)
compiler: arm-linux-gnueabi-gcc (Debian 6.1.1-9) 6.1.1 20160705
reproduce:
        wget https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        git checkout 5615742edc64c99bbdec5e98bf9d9d872756d99c
        # save the attached .config to linux build tree
        make.cross ARCH=arm 

All warnings (new ones prefixed by >>):

   drivers/firmware/efi/libstub/fdt.c: In function 'allocate_new_fdt_and_exit_boot':
>> drivers/firmware/efi/libstub/fdt.c:303:9: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
            (u64)memory_map, (u32)map_size);
            ^

vim +303 drivers/firmware/efi/libstub/fdt.c

   287				new_fdt_size += EFI_PAGE_SIZE;
   288			} else {
   289				pr_efi_err(sys_table, "Unable to construct new device tree.\n");
   290				goto fail_free_new_fdt;
   291			}
   292		}
   293	
   294		priv.runtime_map = runtime_map;
   295		priv.runtime_entry_count = &runtime_entry_count;
   296		status = efi_exit_boot_services(sys_table, handle, &map, &priv,
   297						exit_boot_func);
   298	
   299		if (status == EFI_SUCCESS) {
   300			efi_set_virtual_address_map_t *svam;
   301	
   302			status = update_fdt_memmap((void *)*new_fdt_addr,
 > 303						   (u64)memory_map, (u32)map_size);
   304			if (status != EFI_SUCCESS) {
   305				/*
   306				 * The kernel won't get far without the memory map, but
   307				 * may still be able to print something meaningful so
   308				 * return success here.
   309				 */
   310				return EFI_SUCCESS;
   311			}

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 39253 bytes --]

Re: [efi:urgent 1/1] drivers/firmware/efi/libstub/fdt.c:303:9: warning: cast from pointer to integer of different size

[Matt Fleming]

On Tue, 20 Dec, at 07:04:42AM, Fengguang Wu wrote:
> tree:   https://git.kernel.org/pub/scm/linux/kernel/git/mfleming/efi.git urgent
> head:   5615742edc64c99bbdec5e98bf9d9d872756d99c
> commit: 5615742edc64c99bbdec5e98bf9d9d872756d99c [1/1] efi/libstub: arm*: Pass latest memory map to the kernel
> config: arm-multi_v7_defconfig (attached as .config)
> compiler: arm-linux-gnueabi-gcc (Debian 6.1.1-9) 6.1.1 20160705
> reproduce:
>         wget https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross -O ~/bin/make.cross
>         chmod +x ~/bin/make.cross
>         git checkout 5615742edc64c99bbdec5e98bf9d9d872756d99c
>         # save the attached .config to linux build tree
>         make.cross ARCH=arm 
> 
> All warnings (new ones prefixed by >>):
> 
>    drivers/firmware/efi/libstub/fdt.c: In function 'allocate_new_fdt_and_exit_boot':
> >> drivers/firmware/efi/libstub/fdt.c:303:9: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
>             (u64)memory_map, (u32)map_size);
>             ^
> 
> vim +303 drivers/firmware/efi/libstub/fdt.c
> 
>    287				new_fdt_size += EFI_PAGE_SIZE;
>    288			} else {
>    289				pr_efi_err(sys_table, "Unable to construct new device tree.\n");
>    290				goto fail_free_new_fdt;
>    291			}
>    292		}
>    293	
>    294		priv.runtime_map = runtime_map;
>    295		priv.runtime_entry_count = &runtime_entry_count;
>    296		status = efi_exit_boot_services(sys_table, handle, &map, &priv,
>    297						exit_boot_func);
>    298	
>    299		if (status == EFI_SUCCESS) {
>    300			efi_set_virtual_address_map_t *svam;
>    301	
>    302			status = update_fdt_memmap((void *)*new_fdt_addr,
>  > 303						   (u64)memory_map, (u32)map_size);
>    304			if (status != EFI_SUCCESS) {
>    305				/*
>    306				 * The kernel won't get far without the memory map, but
>    307				 * may still be able to print something meaningful so
>    308				 * return success here.
>    309				 */
>    310				return EFI_SUCCESS;
>    311			}
> 

James, could you take care of fixing this up? You need to cast
memory_map to unsigned long before u64 on ARM.

Re: [efi:urgent 1/1] drivers/firmware/efi/libstub/fdt.c:303:9: warning: cast from pointer to integer of different size

[James Morse]

Hi Matt,

On 20/12/16 11:39, Matt Fleming wrote:
> James, could you take care of fixing this up? You need to cast
> memory_map to unsigned long before u64 on ARM.

Yes, sure - sorry I didn't spot this. It turns out my ARM config didn't build
the EFI stub. Oops.

Fixup patch incoming,


Thanks,

James

[PATCH] efi/libstub: arm*: double cast memory_map to hide compiler warning

[James Morse]

"efi/libstub: arm*: Pass latest memory map to the kernel" added a new
call to update fdt in-place. This caused a new warning on 32bit ARM:

../drivers/firmware/efi/libstub/fdt.c: In function 'allocate_new_fdt_and_exit_boot':
../drivers/firmware/efi/libstub/fdt.c:303:9: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
         (u64)memory_map, (u32)map_size);
         ^

Cast to an integer of native pointer size first.

Reported-by: kbuild test robot <fengguang.wu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: James Morse <james.morse-5wv7dgnIgG8@public.gmane.org>
---
I guess this too needs to go to stable unless you intend to squash it?

 drivers/firmware/efi/libstub/fdt.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/firmware/efi/libstub/fdt.c b/drivers/firmware/efi/libstub/fdt.c
index 6d33d70..9b11b05 100644
--- a/drivers/firmware/efi/libstub/fdt.c
+++ b/drivers/firmware/efi/libstub/fdt.c
@@ -300,7 +300,8 @@ efi_status_t allocate_new_fdt_and_exit_boot(efi_system_table_t *sys_table,
 		efi_set_virtual_address_map_t *svam;
 
 		status = update_fdt_memmap((void *)*new_fdt_addr,
-					   (u64)memory_map, (u32)map_size);
+					   (u64)(unsigned long)memory_map,
+					   (u32)map_size);
 		if (status != EFI_SUCCESS) {
 			/*
 			 * The kernel won't get far without the memory map, but
-- 
2.10.1

Re: [PATCH] efi/libstub: arm*: double cast memory_map to hide compiler warning

[Matt Fleming]

On Tue, 20 Dec, at 01:52:07PM, James Morse wrote:
> "efi/libstub: arm*: Pass latest memory map to the kernel" added a new
> call to update fdt in-place. This caused a new warning on 32bit ARM:
> 
> ../drivers/firmware/efi/libstub/fdt.c: In function 'allocate_new_fdt_and_exit_boot':
> ../drivers/firmware/efi/libstub/fdt.c:303:9: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
>          (u64)memory_map, (u32)map_size);
>          ^
> 
> Cast to an integer of native pointer size first.
> 
> Reported-by: kbuild test robot <fengguang.wu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> Signed-off-by: James Morse <james.morse-5wv7dgnIgG8@public.gmane.org>
> ---
> I guess this too needs to go to stable unless you intend to squash it?
> 
>  drivers/firmware/efi/libstub/fdt.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)

Thanks James, I've folded this into your previous patch.

Re: [PATCH] efi/libstub: arm*: double cast memory_map to hide compiler warning

[Jeffrey Hugo]

On 12/20/2016 7:07 AM, Matt Fleming wrote:
> On Tue, 20 Dec, at 01:52:07PM, James Morse wrote:
>> "efi/libstub: arm*: Pass latest memory map to the kernel" added a new
>> call to update fdt in-place. This caused a new warning on 32bit ARM:
>>
>> ../drivers/firmware/efi/libstub/fdt.c: In function 'allocate_new_fdt_and_exit_boot':
>> ../drivers/firmware/efi/libstub/fdt.c:303:9: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
>>          (u64)memory_map, (u32)map_size);
>>          ^
>>
>> Cast to an integer of native pointer size first.
>>
>> Reported-by: kbuild test robot <fengguang.wu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
>> Signed-off-by: James Morse <james.morse-5wv7dgnIgG8@public.gmane.org>
>> ---
>> I guess this too needs to go to stable unless you intend to squash it?
>>
>>  drivers/firmware/efi/libstub/fdt.c | 3 ++-
>>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> Thanks James, I've folded this into your previous patch.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-efi" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

Thanks James.  I'm happy to see this got cleaned up and sorted out while 
I was on vacation.  I see Ard submitted a pull request already, but for 
what its worth, the latest version on the EFI urgent branch tested ok on 
the platforms I have available.

-- 
Jeffrey Hugo
Qualcomm Datacenter Technologies as an affiliate of Qualcomm 
Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the
Code Aurora Forum, a Linux Foundation Collaborative Project.

Re: [PATCH] efi/libstub: arm*: double cast memory_map to hide compiler warning

[Ard Biesheuvel]

On 3 January 2017 at 21:11, Jeffrey Hugo <jhugo-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org> wrote:
> On 12/20/2016 7:07 AM, Matt Fleming wrote:
>>
>> On Tue, 20 Dec, at 01:52:07PM, James Morse wrote:
>>>
>>> "efi/libstub: arm*: Pass latest memory map to the kernel" added a new
>>> call to update fdt in-place. This caused a new warning on 32bit ARM:
>>>
>>> ../drivers/firmware/efi/libstub/fdt.c: In function
>>> 'allocate_new_fdt_and_exit_boot':
>>> ../drivers/firmware/efi/libstub/fdt.c:303:9: warning: cast from pointer
>>> to integer of different size [-Wpointer-to-int-cast]
>>>          (u64)memory_map, (u32)map_size);
>>>          ^
>>>
>>> Cast to an integer of native pointer size first.
>>>
>>> Reported-by: kbuild test robot <fengguang.wu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
>>> Signed-off-by: James Morse <james.morse-5wv7dgnIgG8@public.gmane.org>
>>> ---
>>> I guess this too needs to go to stable unless you intend to squash it?
>>>
>>>  drivers/firmware/efi/libstub/fdt.c | 3 ++-
>>>  1 file changed, 2 insertions(+), 1 deletion(-)
>>
>>
>> Thanks James, I've folded this into your previous patch.
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-efi" in
>> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
>
> Thanks James.  I'm happy to see this got cleaned up and sorted out while I
> was on vacation.  I see Ard submitted a pull request already, but for what
> its worth, the latest version on the EFI urgent branch tested ok on the
> platforms I have available.
>

Thanks for confirming

Cheers,
Ard.