[RFC,07/10] clk: rockchip: set clk-ddr to GET_RATE_NOCACHE.

Commit Message

Enric Balletbo i Serra May 14, 2018, 9:16 p.m. UTC

From: Derek Basehore <[e-mail]>

This adds the flag to the clk-ddr in rockchip to not use the cached
rate for get_rate. This is to handle timeout error conditions in SMC
for the set rate function.

Signed-off-by: Derek Basehore <[e-mail]>
Signed-off-by: Enric Balletbo i Serra <[e-mail]>
---

 drivers/clk/rockchip/clk-ddr.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Stephen Boyd May 15, 2018, 8:41 p.m. UTC | #1

Quoting Enric Balletbo i Serra (2018-05-14 14:16:07)
> From: Derek Basehore <[e-mail]>
> 
> This adds the flag to the clk-ddr in rockchip to not use the cached
> rate for get_rate. This is to handle timeout error conditions in SMC
> for the set rate function.

We need some more information here. Why does timeout error condition in
set_rate() matter for get_rate()?

Patch

diff --git a/drivers/clk/rockchip/clk-ddr.c b/drivers/clk/rockchip/clk-ddr.c
index de00590af167..dffd5ac5847b 100644
--- a/drivers/clk/rockchip/clk-ddr.c
+++ b/drivers/clk/rockchip/clk-ddr.c
@@ -226,7 +226,7 @@  struct clk *rockchip_clk_register_ddrclk(const char *name, int flags,
 	init.num_parents = num_parents;
 
 	init.flags = flags;
-	init.flags |= CLK_SET_RATE_NO_REPARENT;
+	init.flags |= CLK_SET_RATE_NO_REPARENT | CLK_GET_RATE_NOCACHE;
 
 	switch (ddr_flag) {
 	case ROCKCHIP_DDRCLK_SIP: