5.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tony Lindgren <tony@xxxxxxxxxxx>
commit f71f6ff8c1f682a1cae4e8d7bdeed9d7f76b8f75 upstream.
Commit 34539b442b3b ("bus: ti-sysc: Flush posted write on enable before
reset") caused a regression reproducable on omap4 duovero where the ISS
target module can produce interconnect errors on boot. Turns out the
registers are not accessible until after a delay for devices needing
a ti,sysc-delay-us value.
Let's fix this by flushing the posted write only after the reset delay.
We do flushing also for ti,sysc-delay-us using devices as that should
trigger an interconnect error if the delay is not properly configured.
Let's also add some comments while at it.
Fixes: 34539b442b3b ("bus: ti-sysc: Flush posted write on enable before reset")
Cc: stable@xxxxxxxxxxxxxxx
Signed-off-by: Tony Lindgren <tony@xxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
drivers/bus/ti-sysc.c | 18 ++++++++++++++----
1 file changed, 14 insertions(+), 4 deletions(-)
--- a/drivers/bus/ti-sysc.c
+++ b/drivers/bus/ti-sysc.c
@@ -2104,13 +2104,23 @@ static int sysc_reset(struct sysc *ddata
sysc_val = sysc_read_sysconfig(ddata);
sysc_val |= sysc_mask;
sysc_write(ddata, sysc_offset, sysc_val);
- /* Flush posted write */
+
+ /*
+ * Some devices need a delay before reading registers
+ * after reset. Presumably a srst_udelay is not needed
+ * for devices that use a rstctrl register reset.
+ */
+ if (ddata->cfg.srst_udelay)
+ fsleep(ddata->cfg.srst_udelay);
+
+ /*
+ * Flush posted write. For devices needing srst_udelay
+ * this should trigger an interconnect error if the
+ * srst_udelay value is needed but not configured.
+ */
sysc_val = sysc_read_sysconfig(ddata);
}
- if (ddata->cfg.srst_udelay)
- fsleep(ddata->cfg.srst_udelay);
-
if (ddata->post_reset_quirk)
ddata->post_reset_quirk(ddata);
