Re: [PATCH] Makefile: Unset stack-protector gcc flag if it is not supported

From: Richard Weinberger
Date: Thu Feb 20 2014 - 08:58:06 EST


Am 20.02.2014 14:53, schrieb Markos Chandras:
> On 02/20/2014 12:53 PM, Richard Weinberger wrote:
>> On Thu, Feb 20, 2014 at 1:46 PM, Markos Chandras
>> <Markos.Chandras@xxxxxxxxxx> wrote:
>>> On 02/07/2014 05:15 PM, Markos Chandras wrote:
>>>>
>>>> In case the compiler does not support the stack-protector option,
>>>> unset the flag to avoid build failures. Printing a warning is enough
>>>> to let the user know that this flag will not be used.
>>>> Fixes the following build problem when using a toolchain which does
>>>> not support the -fstack-protector-strong flag:
>>>>
>>>> Makefile:614: Cannot use CONFIG_CC_STACKPROTECTOR_STRONG:
>>>> -fstack-protector-strong not supported by compiler
>>>> mips-linux-gnu-gcc: error: unrecognized command line option
>>>> '-fstack-protector-strong'
>>>> scripts/Makefile.build:308: recipe for target 'scripts/mod/empty.o' failed
>>>> make[2]: *** [scripts/mod/empty.o] Error 1
>>>> make[2]: *** Waiting for unfinished jobs....
>>>> scripts/Makefile.build:455: recipe for target 'scripts/mod' failed
>>>>
>>>> Cc: Michal Marek <mmarek@xxxxxxx>
>>>> Cc: linux-kbuild@xxxxxxxxxxxxxxx
>>>> Signed-off-by: Markos Chandras <markos.chandras@xxxxxxxxxx>
>>>> ---
>>>> Makefile | 2 ++
>>>> 1 file changed, 2 insertions(+)
>>>>
>>>> diff --git a/Makefile b/Makefile
>>>> index 606ef7c..63a5333 100644
>>>> --- a/Makefile
>>>> +++ b/Makefile
>>>> @@ -607,12 +607,14 @@ ifdef CONFIG_CC_STACKPROTECTOR_REGULAR
>>>> ifeq ($(call cc-option, $(stackp-flag)),)
>>>> $(warning Cannot use CONFIG_CC_STACKPROTECTOR: \
>>>> -fstack-protector not supported by compiler))
>>>> + stackp-flag :=
>>>> endif
>>>> else ifdef CONFIG_CC_STACKPROTECTOR_STRONG
>>>> stackp-flag := -fstack-protector-strong
>>>> ifeq ($(call cc-option, $(stackp-flag)),)
>>>> $(warning Cannot use CONFIG_CC_STACKPROTECTOR_STRONG: \
>>>> -fstack-protector-strong not supported by compiler)
>>>> + stackp-flag :=
>>>> endif
>>>> else
>>>> # Force off for distro compilers that enable stack protector by
>>>> default.
>>>>
>>>
>>> Ping again?
>>
>> This is by design. We want the build fail if
>> CONFIG_CC_STACKPROTECTOR_STRONG is set
>> but the toolchain does not support it.
>> Otherwise we could end up with a kernel without stackprotector even if
>> the config symbol is set.
>>
>
> Hi Richard,
>
> Then maybe the $(warning...) should be converted to $(error...)?
> Because it's not really a warning if the build failure is expected to happen.

No. This would break silentoldconfig.

Thanks,
//richard
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/