Message ID | 1364994565-16010-2-git-send-email-chenhui.zhao@freescale.com (mailing list archive) |
---|---|
State | Superseded, archived |
Delegated to: | Kumar Gala |
Headers | show |
On Apr 3, 2013, at 8:09 AM, Zhao Chenhui wrote: > From: Chen-Hui Zhao <chenhui.zhao@freescale.com> > > The bootloader have done time base sync for all cores, so skip > the synchronization process at boot time of kernel. > > Signed-off-by: Zhao Chenhui <chenhui.zhao@freescale.com> > Signed-off-by: Li Yang <leoli@freescale.com> > Signed-off-by: Andy Fleming <afleming@freescale.com> > --- > arch/powerpc/platforms/85xx/smp.c | 8 ++++++++ > 1 files changed, 8 insertions(+), 0 deletions(-) What harm is there in doing the sync? I'm sure there is another reason you want to skip the TB sync that should be conveyed in the commit message. - k
No other reason. Just avoid doing it again at boot time in kernel. -Chenhui
diff --git a/arch/powerpc/platforms/85xx/smp.c b/arch/powerpc/platforms/85xx/smp.c index 6a17599..35dae8e 100644 --- a/arch/powerpc/platforms/85xx/smp.c +++ b/arch/powerpc/platforms/85xx/smp.c @@ -62,6 +62,10 @@ static void mpc85xx_give_timebase(void) { unsigned long flags; + /* only do time base sync when system is running */ + if (system_state == SYSTEM_BOOTING) + return; + local_irq_save(flags); while (!tb_req) @@ -85,6 +89,10 @@ static void mpc85xx_take_timebase(void) { unsigned long flags; + /* only do time base sync when system is running */ + if (system_state == SYSTEM_BOOTING) + return; + local_irq_save(flags); tb_req = 1;