Hi,
I've pulled Linus' tree this morning and, after running 'make oldconfig', tried a build. During that build I got the
following warnings, which look to me like they should be fixed. 'git describe' shows v5.2-915-g5ad18b2e60b7 and my
compiler is the 20190706 snapshot of gcc 9.
In file included from arch/x86/kernel/head64.c:35:
In function 'sanitize_boot_params',
inlined from 'copy_bootdata' at arch/x86/kernel/head64.c:391:2:
./arch/x86/include/asm/bootparam_utils.h:40:3: warning: 'memset' offset [197, 448] from the object at 'boot_params' is
out of the bounds of referenced subobject 'ext_ramdisk_image' with type 'unsigned int' at offset 192 [-Warray-bounds]
40 | memset(&boot_params->ext_ramdisk_image, 0,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
41 | (char *)&boot_params->efi_info -
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
42 | (char *)&boot_params->ext_ramdisk_image);
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./arch/x86/include/asm/bootparam_utils.h:43:3: warning: 'memset' offset [493, 497] from the object at 'boot_params' is
out of the bounds of referenced subobject 'kbd_status' with type 'unsigned char' at offset 491 [-Warray-bounds]
43 | memset(&boot_params->kbd_status, 0,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
44 | (char *)&boot_params->hdr -
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~
45 | (char *)&boot_params->kbd_status);
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Happy to test any patches, but please cc me as I'm not subscribed to LKML.
Chris
On 09.07.19 08:06, Chris Clayton wrote:
Hi,
> I've pulled Linus' tree this morning and, after running 'make oldconfig', tried a build. During that build I got the
> following warnings, which look to me like they should be fixed. 'git describe' shows v5.2-915-g5ad18b2e60b7 and my
> compiler is the 20190706 snapshot of gcc 9.
Thanks for the report. I'm rebuilding right know anyways, so I'll look
out for it.
> In file included from arch/x86/kernel/head64.c:35:
> In function 'sanitize_boot_params',
> inlined from 'copy_bootdata' at arch/x86/kernel/head64.c:391:2:
> ./arch/x86/include/asm/bootparam_utils.h:40:3: warning: 'memset' offset [197, 448] from the object at 'boot_params' is
> out of the bounds of referenced subobject 'ext_ramdisk_image' with type 'unsigned int' at offset 192 [-Warray-bounds]
> 40 | memset(&boot_params->ext_ramdisk_image, 0,
> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 41 | (char *)&boot_params->efi_info -
> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 42 | (char *)&boot_params->ext_ramdisk_image);
> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ./arch/x86/include/asm/bootparam_utils.h:43:3: warning: 'memset' offset [493, 497] from the object at 'boot_params' is
> out of the bounds of referenced subobject 'kbd_status' with type 'unsigned char' at offset 491 [-Warray-bounds]
> 43 | memset(&boot_params->kbd_status, 0,
> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 44 | (char *)&boot_params->hdr -
> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 45 | (char *)&boot_params->kbd_status);
> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Can you check older versions, too ? Maybe also trying older gcc ?
--mtx
--
Enrico Weigelt, metux IT consult
Free software and Linux embedded engineering
[email protected] -- +49-151-27565287
On 09/07/2019 11:37, Enrico Weigelt, metux IT consult wrote:
> On 09.07.19 08:06, Chris Clayton wrote:
>
> Hi,
>
>> I've pulled Linus' tree this morning and, after running 'make oldconfig', tried a build. During that build I got the
>> following warnings, which look to me like they should be fixed. 'git describe' shows v5.2-915-g5ad18b2e60b7 and my
>> compiler is the 20190706 snapshot of gcc 9.
>
> Thanks for the report. I'm rebuilding right know anyways, so I'll look
> out for it.
Thanks for the reply.
>> In file included from arch/x86/kernel/head64.c:35:
>> In function 'sanitize_boot_params',
>> inlined from 'copy_bootdata' at arch/x86/kernel/head64.c:391:2:
>> ./arch/x86/include/asm/bootparam_utils.h:40:3: warning: 'memset' offset [197, 448] from the object at 'boot_params' is
>> out of the bounds of referenced subobject 'ext_ramdisk_image' with type 'unsigned int' at offset 192 [-Warray-bounds]
>> 40 | memset(&boot_params->ext_ramdisk_image, 0,
>> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> 41 | (char *)&boot_params->efi_info -
>> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> 42 | (char *)&boot_params->ext_ramdisk_image);
>> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> ./arch/x86/include/asm/bootparam_utils.h:43:3: warning: 'memset' offset [493, 497] from the object at 'boot_params' is
>> out of the bounds of referenced subobject 'kbd_status' with type 'unsigned char' at offset 491 [-Warray-bounds]
>> 43 | memset(&boot_params->kbd_status, 0,
>> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> 44 | (char *)&boot_params->hdr -
>> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> 45 | (char *)&boot_params->kbd_status);
>> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> Can you check older versions, too ? Maybe also trying older gcc ?
>
I see the same warnings building linux-5.2.0 with gcc9. However, I don't see the warnings building linux-5.2.0 with the
the 20190705 of gcc8. So the warnings could result from an improvement (i.e. the problem was in the kernel, but
undiscovered by gcc8) or from a regression in gcc9.
>
> --mtx
>
On 09/07/2019 12:39, Chris Clayton wrote:
>
>
> On 09/07/2019 11:37, Enrico Weigelt, metux IT consult wrote:
>> On 09.07.19 08:06, Chris Clayton wrote:
>>
>> Hi,
>>
>>> I've pulled Linus' tree this morning and, after running 'make oldconfig', tried a build. During that build I got the
>>> following warnings, which look to me like they should be fixed. 'git describe' shows v5.2-915-g5ad18b2e60b7 and my
>>> compiler is the 20190706 snapshot of gcc 9.
>>
>> Thanks for the report. I'm rebuilding right know anyways, so I'll look
>> out for it.
>
> Thanks for the reply.
>
>>> In file included from arch/x86/kernel/head64.c:35:
>>> In function 'sanitize_boot_params',
>>> inlined from 'copy_bootdata' at arch/x86/kernel/head64.c:391:2:
>>> ./arch/x86/include/asm/bootparam_utils.h:40:3: warning: 'memset' offset [197, 448] from the object at 'boot_params' is
>>> out of the bounds of referenced subobject 'ext_ramdisk_image' with type 'unsigned int' at offset 192 [-Warray-bounds]
>>> 40 | memset(&boot_params->ext_ramdisk_image, 0,
>>> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>> 41 | (char *)&boot_params->efi_info -
>>> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>> 42 | (char *)&boot_params->ext_ramdisk_image);
>>> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>> ./arch/x86/include/asm/bootparam_utils.h:43:3: warning: 'memset' offset [493, 497] from the object at 'boot_params' is
>>> out of the bounds of referenced subobject 'kbd_status' with type 'unsigned char' at offset 491 [-Warray-bounds]
>>> 43 | memset(&boot_params->kbd_status, 0,
>>> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>> 44 | (char *)&boot_params->hdr -
>>> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>> 45 | (char *)&boot_params->kbd_status);
>>> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>
>> Can you check older versions, too ? Maybe also trying older gcc ?
>>
>
> I see the same warnings building linux-5.2.0 with gcc9. However, I don't see the warnings building linux-5.2.0 with the
> the 20190705 of gcc8. So the warnings could result from an improvement (i.e. the problem was in the kernel, but
> undiscovered by gcc8) or from a regression in gcc9.
>
From the discussion starting at https://marc.info/?l=linux-kernel&m=156401014023908, it would appear that the problem is
undiscovered by gcc8. Building a fresh pull of Linus' tree this morning (v5.3-rc3-282-g33920f1ec5bf), I see that the
warnings are still being emitted. Adding the participants in the other discussion to this one.
>>
>> --mtx
>>
On 8/6/19 11:30 PM, Chris Clayton wrote:
> On 09/07/2019 12:39, Chris Clayton wrote:
>> On 09/07/2019 11:37, Enrico Weigelt, metux IT consult wrote:
>>> On 09.07.19 08:06, Chris Clayton wrote:
...
>>> Can you check older versions, too ? Maybe also trying older gcc ?
>>>
>>
>> I see the same warnings building linux-5.2.0 with gcc9. However, I don't see the warnings building linux-5.2.0 with the
>> the 20190705 of gcc8. So the warnings could result from an improvement (i.e. the problem was in the kernel, but
>> undiscovered by gcc8) or from a regression in gcc9.
>>
>
> From the discussion starting at https://marc.info/?l=linux-kernel&m=156401014023908, it would appear that the problem is
> undiscovered by gcc8. Building a fresh pull of Linus' tree this morning (v5.3-rc3-282-g33920f1ec5bf), I see that the
> warnings are still being emitted. Adding the participants in the other discussion to this one.
>
The warnings are still there because the fix has not been committed to any
tree yet.
If you could try out my proposed fix [1], and reply to that thread with perhaps a
Tested-by tag, that would help encourage the maintainers to accept it.
So far it hasn't made it to the top of their inboxes, but I'm hoping... :)
[1] https://lore.kernel.org/r/[email protected]
("x86/boot: save fields explicitly, zero out everything else")
thanks,
--
John Hubbard
NVIDIA