Current OMAP2PLUS CPUfreq tagret() functions returns when all
the CPU's are not online. This breaks CPUfreq when secondary CPUs
are offlined on SMP system.

The intention of that check was just avoid CPU frequency change
during the window when CPU becomes online but it's cpufreq_init is
not done yet. Otherwise it can lead to notifiers being sent on
a CPU which is not yet registered to the governor.

But this race conditions is already managed by the CPUfreq
core driver by updating the available cpumask accordingly.

OMAP CPUFReq driver make use same cpumask for the notifiers
so the above problem doesn't exist. In my initial implementation
of the OMAP4 CPUFreq driver, I was using 'for_each_online_cpu()'
for notifiers which lead me to add that check. Later I fixed
the notifies but didn't realise that the check has become
redundant then.

Fix it by removing the superfluous check in target().

Thanks for Nishant Menon <[email protected]> for reporting issue
with hot-plug and Kevin Hilman <[email protected]> for his
comment on excessive check in target().

Signed-off-by: Santosh Shilimkar <[email protected]>
Reported-by: Nishanth Menon <[email protected]>
Tested-by: Vishwanath BS <[email protected]>
Cc: Kevin Hilman <[email protected]>
---
This change is an outcome of discussion on another patch to fix
the hot-plug issue. For details refer below thread:
http://www.mail-archive.com/[email protected]/msg50348.html

 arch/arm/mach-omap2/omap2plus-cpufreq.c |    4 ----
 1 files changed, 0 insertions(+), 4 deletions(-)

diff --git a/arch/arm/mach-omap2/omap2plus-cpufreq.c 
b/arch/arm/mach-omap2/omap2plus-cpufreq.c
index 33a91ec..3bf1704 100644
--- a/arch/arm/mach-omap2/omap2plus-cpufreq.c
+++ b/arch/arm/mach-omap2/omap2plus-cpufreq.c
@@ -81,10 +81,6 @@ static int omap_target(struct cpufreq_policy *policy,
        int i, ret = 0;
        struct cpufreq_freqs freqs;
 
-       /* Changes not allowed until all CPUs are online */
-       if (is_smp() && (num_online_cpus() < NR_CPUS))
-               return ret;
-
        /* Ensure desired rate is within allowed range.  Some govenors
         * (ondemand) will just pass target_freq=0 to get the minimum. */
        if (target_freq < policy->min)
-- 
1.6.0.4

--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to