Hi Christophe,
No, they were seen when bootstrapping on arm-linux-gnueabihf with the patch to
to make -mthumb optional. The patch store flags in an arm_feature_set array and
GCC -Wnarrowing complained about the difference of type.
Best regards,
Thomas
On 14/11/16 17:00, Christophe Lyon wrote:
Hi,
On 14 November 2016 at 15:07, Thomas Preudhomme
<thomas.preudho...@foss.arm.com> wrote:
Hi,
Currently arm_feature_set is defined in gcc/config/arm/arm-flags as an array
of 2 unsigned long. However, the flags stored in these two entries are
(signed) int, being combinations of bits set via expression of the form 1 <<
bitno. This creates 3 issues:
1) undefined behavior when setting the msb (1 << 31)
2) undefined behavior when storing a flag with msb set (negative int) into
one of the unsigned array entries (positive int)
Just curious: are these problems seen when building with ubsan enabled?
3) waste of space since the top 32 bits of each entry is not used
This patch changes the definition of FL_* macro to be unsigned int by using
the form 1U << bitno instead and changes the definition of arm_feature_set
to be an array of 2 unsigned (int) entries.
Bootstrapped on arm-linux-gnueabihf targeting Thumb-2 state.
Is this ok for trunk?
Best regards,
Thomas
