sched: avoid migrating when softint on tgt cpu should be short

The scheduling change (bug 31501544) to avoid putting RT threads on cores that
are handling softint's was catching cases where there was no reason
to believe the softint would take a long time, resulting in unnecessary
migration overhead. This patch reduces the migration to cases where
the core has a softint that is actually likely to take a long time,
as opposed to the RCU, SCHED, and TIMER softints that are rather quick.
Bug: 31752786

Change-Id: Ib4e179f1e15c736b2fdba31070494e357e9fbbe2
Git-commit: ce05770bd37b8065b61ef650108ecef2b97b148b
Git-repo: https://android.googlesource.com/kernel/msm
[pkondeti@codeaurora.org: resolved minor merge conflicts]
Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org>
Signed-off-by: Satya Durga Srinivasu Prabhala <satyap@codeaurora.org>
This commit is contained in:
John Dias
2016-10-05 15:11:40 -07:00
committed by Satya Durga Srinivasu Prabhala
parent 54869f74a2
commit 615d2fba03
3 changed files with 24 additions and 4 deletions

View File

@@ -478,6 +478,12 @@ enum
};
#define SOFTIRQ_STOP_IDLE_MASK (~(1 << RCU_SOFTIRQ))
/* Softirq's where the handling might be long: */
#define LONG_SOFTIRQ_MASK ((1 << NET_TX_SOFTIRQ) | \
(1 << NET_RX_SOFTIRQ) | \
(1 << BLOCK_SOFTIRQ) | \
(1 << IRQ_POLL_SOFTIRQ) | \
(1 << TASKLET_SOFTIRQ))
/* map softirq index to softirq name. update 'softirq_to_name' in
* kernel/softirq.c when adding a new softirq.
@@ -513,6 +519,7 @@ extern void raise_softirq_irqoff(unsigned int nr);
extern void raise_softirq(unsigned int nr);
DECLARE_PER_CPU(struct task_struct *, ksoftirqd);
DECLARE_PER_CPU(__u32, active_softirqs);
static inline struct task_struct *this_cpu_ksoftirqd(void)
{