Skip to contentSkip to conversation

clk: versaclock5: restore cache_only on regcache_sync() failure in resume

PATCH Applied [email protected] 2 participants 4 messages Jul 24 – Sep 29

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 clk: versaclock5: restore cache_only on regcache_sync() failure in resume

From: bui duc phuc <[email protected]>

If regcache_sync() fails in vc5_resume(), the function only logs the error, leaving regmap with cache_only disabled even though the register cache has not been synchronized with the hardware.
Re-enable cache_only on failure to restore the state established by vc5_suspend().

---
 drivers/clk/clk-versaclock5.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/clk/clk-versaclock5.c +3 -1
index 913fcc5675f1..d87899118905 100644
@@ -1240,8 +1240,10 @@ static int __maybe_unused vc5_resume(struct device *dev)
 
 	regcache_cache_only(vc5->regmap, false);
 	ret = regcache_sync(vc5->regmap);
-	if (ret)
+	if (ret) {
+		regcache_cache_only(vc5->regmap, true);
 		dev_err(dev, "Failed to restore register map: %d\n", ret);
+	}
 	return ret;
 }
 
Signature · 1 line
2.43.0
#1 lore raw

Hi bui,

Quoted 18 lines · On Fri, Jul 24, 2026 at 02:46:03PM +0700, [email protected] wrote:

From: bui duc phuc <[email protected]>

If regcache_sync() fails in vc5_resume(), the function only logs the error, leaving regmap with cache_only disabled even though the register cache has not been synchronized with the hardware.
Re-enable cache_only on failure to restore the state established by vc5_suspend().

---
 drivers/clk/clk-versaclock5.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/clk/clk-versaclock5.c +0 -0
index 913fcc5675f1..d87899118905 100644

drivers/clk/clk-versaclock5.c static int __maybe_unused vc5_resume(struct device *dev) · line 1248

Quoted 10 lines
drivers/clk/clk-versaclock5.c +3 -1
@@ -1240,8 +1240,10 @@ static int __maybe_unused vc5_resume(struct device *dev)
 
 	regcache_cache_only(vc5->regmap, false);
 	ret = regcache_sync(vc5->regmap);
-	if (ret)
+	if (ret) {
+		regcache_cache_only(vc5->regmap, true);
 		dev_err(dev, "Failed to restore register map: %d\n", ret);
+	}
 	return ret;
 }

I don't see precedent for this pattern anywhere else in the tree. Is this an observed problem or something that could theoretically happen?

If this fails in the code today, the pm core will propagate the error, and the device won't be marked as resumed.

Brian

#2 lore raw

Hi Brian,

Thank you for your feedback.

Quoted 6 lines

I don't see precedent for this pattern anywhere else in the tree. Is this an observed problem or something that could theoretically happen?

If this fails in the code today, the pm core will propagate the error, and the device won't be marked as resumed.

Fair point, there's no precedent in system sleep PM,
though runtime PM has a few cases doing the same thing, so it's not an entirely new pattern.
This is a defensive fix for a theoretical case; I don't have an actual bug report for it.

But after a failed resume, what guarantees that no further regmap accesses can occur
before the next successful resume? If there is such a guarantee, I'm happy to drop the patch ,
otherwise I'd lean towards keeping it.

Best regards,
Phuc

Applied
#3 lore raw
In reply to [email protected]
Quoted 1 line · On Fri, 24 Jul 2026 14:46:03 +0700, [email protected] wrote:

clk: versaclock5: restore cache_only on regcache_sync() failure in resume

Applied, thanks!

[1/1] clk: versaclock5: restore cache_only on regcache_sync() failure in resume

      commit: 7b283b8ce37d40d22f41bdc20b407cd76fe86e68

Best regards,

Signature · 1 line
Brian Masney <[email protected]>