clk: ti: clkctrl: Fix failed to enable error with double udelay timeout

Commit 3d8598fb9c5a ("clk: ti: clkctrl: use fallback udelay approach if timekeeping is suspended") added handling for cases when timekeeping is suspended. But looks like we can still get occasional "failed to enable" errors on the PM runtime resume path with udelay() returning faster than expected.

With ti-sysc interconnect target module driver this leads into device failure with PM runtime failing with "failed to enable" clkctrl error.

Let's fix the issue with a delay of two times the desired delay as in often done for udelay() to account for the inaccuracy.

Fixes 3d8598fb9c5a ("clk: ti: clkctrl: use fallback udelay approach if timekeeping is suspended")

Cc Keerthy <[email protected]>

Cc Tero Kristo <[email protected]>

Signed-off-by Tony Lindgren <[email protected]>

---
  drivers/clk/ti/clkctrl.c | 5 +++--
  1 file changed, 3 insertions(+), 2 deletions(-)

drivers/clk/ti/clkctrl.c +3 -2

@@ -100,11 +100,12 @@ static bool _omap4_is_timeout(union omap4_timeout *time, u32 timeout)
  	 * can be from a timer that requires pm_runtime access, which
  	 * will eventually bring us here with timekeeping_suspended,
  	 * during both suspend entry and resume paths. This happens
-	 * at least on am43xx platform.
+	 * at least on am43xx platform. Account for flakeyness
+	 * with udelay() by multiplying the timeout value by 2.
  	 */
  	if (unlikely(_early_timeout || timekeeping_suspended)) {
  		if (time->cycles++ < timeout) {
-			udelay(1);
+			udelay(1 * 2);
  			return false;
  		}
  	} else {

Quoted 9 lines · Tony Lindgren

Commit 3d8598fb9c5a ("clk: ti: clkctrl: use fallback udelay approach if timekeeping is suspended") added handling for cases when timekeeping is suspended. But looks like we can still get occasional "failed to enable" errors on the PM runtime resume path with udelay() returning faster than expected.

With ti-sysc interconnect target module driver this leads into device failure with PM runtime failing with "failed to enable" clkctrl error.

Quoted 2 lines

Let's fix the issue with a delay of two times the desired delay as in often done for udelay() to account for the inaccuracy.

Tested for DS0 and rtc+ddr modes on am43 and ds0 on am33.

Tested-by Keerthy <[email protected]>

Quoted 27 lines

Fixes 3d8598fb9c5a ("clk: ti: clkctrl: use fallback udelay approach if timekeeping is suspended")

Cc Keerthy <[email protected]>

Cc Tero Kristo <[email protected]>

Signed-off-by Tony Lindgren <[email protected]>

---
  drivers/clk/ti/clkctrl.c | 5 +++--
  1 file changed, 3 insertions(+), 2 deletions(-)

drivers/clk/ti/clkctrl.c +3 -2

@@ -100,11 +100,12 @@ static bool _omap4_is_timeout(union omap4_timeout *time, u32 timeout)
  	 * can be from a timer that requires pm_runtime access, which
  	 * will eventually bring us here with timekeeping_suspended,
  	 * during both suspend entry and resume paths. This happens
-	 * at least on am43xx platform.
+	 * at least on am43xx platform. Account for flakeyness
+	 * with udelay() by multiplying the timeout value by 2.
  	 */
  	if (unlikely(_early_timeout || timekeeping_suspended)) {
  		if (time->cycles++ < timeout) {
-			udelay(1);
+			udelay(1 * 2);
  			return false;
  		}
  	} else {

Quoted 15 lines · Keerthy

Quoted 11 lines · Tony Lindgren

Commit 3d8598fb9c5a ("clk: ti: clkctrl: use fallback udelay approach if timekeeping is suspended") added handling for cases when timekeeping is suspended. But looks like we can still get occasional "failed to enable" errors on the PM runtime resume path with udelay() returning faster than expected.

With ti-sysc interconnect target module driver this leads into device failure with PM runtime failing with "failed to enable" clkctrl error.

Let's fix the issue with a delay of two times the desired delay as in often done for udelay() to account for the inaccuracy.

Quoted 3 lines

Tested for DS0 and rtc+ddr modes on am43 and ds0 on am33.

Tested-by Keerthy <[email protected]>

Thanks for testing. This one should be applied into v5.4-rc series please if no more comments.

Regards,

Tony

Quoted 19 lines

Quoted 19 lines

Fixes 3d8598fb9c5a ("clk: ti: clkctrl: use fallback udelay approach if timekeeping is suspended")

Cc Keerthy <[email protected]>

Cc Tero Kristo <[email protected]>

Signed-off-by Tony Lindgren <[email protected]>

---
   drivers/clk/ti/clkctrl.c | 5 +++--
   1 file changed, 3 insertions(+), 2 deletions(-)

drivers/clk/ti/clkctrl.c +2 -1

@@ -100,11 +100,12 @@ static bool _omap4_is_timeout(union omap4_timeout *time, u32 timeout)
   	 * can be from a timer that requires pm_runtime access, which
   	 * will eventually bring us here with timekeeping_suspended,
   	 * during both suspend entry and resume paths. This happens
-	 * at least on am43xx platform.
+	 * at least on am43xx platform. Account for flakeyness
+	 * with udelay() by multiplying the timeout value by 2.
   	 */

Quoted 8 lines

drivers/clk/ti/clkctrl.c static bool _omap4_is_timeout(union omap4_timeout *time, u32 timeout) · line 111

Quoted 8 lines

drivers/clk/ti/clkctrl.c +1 -1

   	if (unlikely(_early_timeout || timekeeping_suspended)) {
   		if (time->cycles++ < timeout) {
-			udelay(1);
+			udelay(1 * 2);
   			return false;
   		}
   	} else {

Commit 3d8598fb9c5a ("clk: ti: clkctrl: use fallback udelay approach if timekeeping is suspended") added handling for cases when timekeeping is suspended. But looks like we can still get occasional "failed to enable" errors on the PM runtime resume path with udelay() returning faster than expected.

With ti-sysc interconnect target module driver this leads into device failure with PM runtime failing with "failed to enable" clkctrl error.

Let's fix the issue with a delay of two times the desired delay as in often done for udelay() to account for the inaccuracy.

Fixes 3d8598fb9c5a ("clk: ti: clkctrl: use fallback udelay approach if timekeeping is suspended")

Cc Keerthy <[email protected]>

Cc Tero Kristo <[email protected]>

Signed-off-by Tony Lindgren <[email protected]>

---