On Thu, May 25 2023 at 01:56, Michael Ellerman wrote:
A subsequent patch will enable partial SMT states, ie. when not all SMT
threads are brought online.
Nitpick. I stumbled over this 'subsequent patch' theme a couple of times
now because it's very similar to the 'This patch does' phrase.
Just explain what you want to achieve at the end.
quoted hunk ↗ jump to hunk
#else
#define topology_max_packages() (1)
static inline int
@@ -159,6 +160,7 @@ static inline int topology_max_smt_threads(void) { return 1; }
static inline bool topology_is_primary_thread(unsigned int cpu) { return true; }
static inline bool topology_smt_supported(void) { return false; }
static inline bool topology_smt_threads_supported(unsigned int threads) { return false; }
+static inline bool topology_smt_thread_allowed(unsigned int cpu) { return false; }
Not all these functions need a !SMP stub. Think about the context in
which they are called. There is probably precedence for pointless ones,
but that does not make an argument to add more.
quoted hunk ↗ jump to hunk
+/**
+ * topology_smt_thread_allowed - When enabling SMT check whether this particular
+ * CPU thread is allowed to be brought online.
+ * @cpu: CPU to check
+ */
+bool topology_smt_thread_allowed(unsigned int cpu)
+{
+ /*
+ * No extra logic s required here to support different thread values
+ * because threads will always == 1 or smp_num_siblings because of
+ * topology_smt_threads_supported().
+ */
+ return true;
+}
+
As x86 only supoorts the on/off model there is no need for this function
if you pick up the CONFIG_SMT_NUM_THREADS_DYNAMIC idea.
You still need something like that for your PPC use case, but that
reduces the overall impact, right?
Thanks,
tglx