On Fri, Dec 17, 2021 at 12:26:56AM +0530, Siddhesh Poyarekar wrote:
> On 12/16/21 21:19, Jakub Jelinek wrote:
> > Yes, but please fix up formatting, wholeval should go below old_wholeval.
> > Though, perhaps it would be better to:
> >
> >if (tree_int_cst_compare (oldval, val))
> > return
On 12/16/21 21:19, Jakub Jelinek wrote:
Yes, but please fix up formatting, wholeval should go below old_wholeval.
Though, perhaps it would be better to:
if (tree_int_cst_compare (oldval, val))
return true;
gcc_checking_assert (tree_int_cst_compare (old_wholeval, wholeval) == 0);
r
On Thu, Dec 16, 2021 at 06:11:27AM +0530, Siddhesh Poyarekar wrote:
> On 12/16/21 00:13, Jakub Jelinek wrote:
> > On Wed, Dec 15, 2021 at 10:42:29PM +0530, Siddhesh Poyarekar wrote:
> > > On 12/15/21 20:51, Jakub Jelinek wrote:
> > > > Shouldn't this also tree_int_cst_compare (old_wholeval, wholeva
On 12/16/21 00:13, Jakub Jelinek wrote:
On Wed, Dec 15, 2021 at 10:42:29PM +0530, Siddhesh Poyarekar wrote:
On 12/15/21 20:51, Jakub Jelinek wrote:
Shouldn't this also tree_int_cst_compare (old_wholeval, wholeval) ?
AFAICT, there is no situation where wholeval changes but val doesn't, so I
b
On Wed, Dec 15, 2021 at 10:42:29PM +0530, Siddhesh Poyarekar wrote:
> On 12/15/21 20:51, Jakub Jelinek wrote:
> > Shouldn't this also tree_int_cst_compare (old_wholeval, wholeval) ?
> >
>
> AFAICT, there is no situation where wholeval changes but val doesn't, so I
> believe the val check should b
On 12/15/21 20:51, Jakub Jelinek wrote:
Shouldn't this also tree_int_cst_compare (old_wholeval, wholeval) ?
AFAICT, there is no situation where wholeval changes but val doesn't, so
I believe the val check should be sufficient. Do you think otherwise?
Siddhesh
On Wed, Dec 01, 2021 at 07:57:52PM +0530, Siddhesh Poyarekar wrote:
> static inline bool
> -object_sizes_set (struct object_size_info *osi, unsigned varno,
> - unsigned HOST_WIDE_INT val)
> +object_sizes_set (struct object_size_info *osi, unsigned varno, tree val,
> +
Transform tree-object-size to operate on tree objects instead of host
wide integers. This makes it easier to extend to dynamic expressions
for object sizes.
The compute_builtin_object_size interface also now returns a tree
expression instead of HOST_WIDE_INT, so callers have been adjusted to
acco
