[PATCH] sandbox: Fix LTO to work with STACKPROTECTOR

[Andrew Goodbody]

On 01/07/2024 14:57, Simon Glass wrote:
> Hi Andrew,
> 
> On Mon, 24 Jun 2024 at 10:29, Andrew Goodbody
> <andrew.goodbody at linaro.org> wrote:
>>
>> Add the STACKPROTECTOR symbols to the script that generates the
>> symbols that should not be removed by the use of LTO when linking
>> a shared object. This prevents a fail to build due to link errors.
>>
>> https://source.denx.de/u-boot/u-boot/-/issues/35
>>
>> Signed-off-by: Andrew Goodbody <andrew.goodbody at linaro.org>
>> ---
>>
>>   scripts/gen_ll_addressable_symbols.sh | 2 ++
>>   1 file changed, 2 insertions(+)
>>
>> diff --git a/scripts/gen_ll_addressable_symbols.sh b/scripts/gen_ll_addressable_symbols.sh
>> index d0864804aa..ebf89b04bf 100755
>> --- a/scripts/gen_ll_addressable_symbols.sh
>> +++ b/scripts/gen_ll_addressable_symbols.sh
>> @@ -13,3 +13,5 @@ set -e
>>   echo '#include <common.h>'
>>   $@ 2>/dev/null | grep -oe '_u_boot_list_2_[a-zA-Z0-9_]*_2_[a-zA-Z0-9_]*' | \
>>          sort -u | sed -e 's/^\(.*\)/extern char \1[];\n__ADDRESSABLE(\1);/'
>> +$@ 2>/dev/null | grep -oe '__stack_chk_.*' | \
>> +       sort -u | sed -e 's/^\(.*\)/extern char \1[];\n__ADDRESSABLE(\1);/'
>> --
> 
> It is OK to add these new ones, but here you seem to be adding lots of
> other things also. So far we are only allowing linker lists. So can
> you update this to allow just linker lists and stack_chk? Perhaps
> egrep (_u_boot_list_2_|__stack_chk_) or similar?
> 
> Regards,
> Simon

I am sorry but I do not follow you here. I am not sure what you mean by 
'linker lists'. The script is matching on symbols from object files as 
output by nm. I have not changed that original match expression.

My addition only adds the following 3 symbols to the output of 
gen_ll_addressable_symbols.sh

__stack_chk_guard
__stack_chk_fail
__stack_chk_fail_local

These all come from the same file, common/stackprot.c

I can reduce the match to just search for '__stack_chk_guard' and that 
still works and I can put that as another expression in the grep instead 
of doing the whole nm | grep | sort | sed thing a second time. Would 
that address your concerns?

Thanks,
Andrew