It looks fine to me. I agree with the general perspective that a user
can't explicitly control down to the last byte their stack usage, so
my complaints are not real problems.
On Thu, Mar 4, 2021 at 2:17 AM Sebastian Huber
wrote:
>
> Make sure that a user-provided stack size is the minimum size a
-Original Message-
From: Sebastian Huber
Sent: Thursday, March 4, 2021 09:23
To: Kinsey Moore ; devel@rtems.org
Subject: Re: [PATCH v2] score: Ensure stack alignment requirement
> On 04/03/2021 16:20, Kinsey Moore wrote:
>>> + alignment_overhead = CPU_STACK_ALIGNMENT
On 04/03/2021 16:20, Kinsey Moore wrote:
+ /*
+ * In order to make sure that a user-provided stack size is the minimum which
+ * can be allocated for the stack, we have to align it up to the next stack
+ * boundary.
+ */
+ alignment_overhead = CPU_STACK_ALIGNMENT - 1;
This should be C
Looks good other than a minor tweak.
Kinsey
-Original Message-
From: devel On Behalf Of Sebastian Huber
Sent: Thursday, March 4, 2021 03:17
To: devel@rtems.org
Subject: [PATCH v2] score: Ensure stack alignment requirement
> Make sure that a user-provided stack size is the minimum size a
