The final step in regulator_register() is to call regulator_resolve_supply() for each registered regulator (including the one in the process of being registered). The regulator_resolve_supply() function first checks if rdev->supply is NULL, then it performs various steps to try to find the supply. If successful, rdev->supply is set inside of set_supply(). This procedure can encounter a race condition if two concurrent tasks call regulator_register() near to each other on separate CPUs and one of the regulators has rdev->supply_name specified. There is currently nothing guaranteeing atomicity between the rdev->supply check and set steps. Thus, both tasks can observe rdev->supply==NULL in their regulator_resolve_supply() calls. This then results in both creating a struct regulator for the supply. One ends up actually stored in rdev->supply and the other is lost (though still present in the supply's consumer_list). Here is a kernel log snippet showing the issue: [ 12.421768] gpu_cc_gx_gdsc: supplied by pm8350_s5_level [ 12.425854] gpu_cc_gx_gdsc: supplied by pm8350_s5_level [ 12.429064] debugfs: Directory 'regulator.4-SUPPLY' with parent '17a00000.rsc:rpmh-regulator-gfxlvl-pm8350_s5_level' already present! Avoid this race condition by holding the rdev->mutex lock inside of regulator_resolve_supply() while checking and setting rdev->supply. Signed-off-by: David Collins <collinsd@xxxxxxxxxxxxxx> --- drivers/regulator/core.c | 39 ++++++++++++++++++++++++++++----------- 1 file changed, 28 insertions(+), 11 deletions(-) diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index fee9241..3ae5ccd 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -1813,23 +1813,34 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) { struct regulator_dev *r; struct device *dev = rdev->dev.parent; - int ret; + int ret = 0; /* No supply to resolve? */ if (!rdev->supply_name) return 0; - /* Supply already resolved? */ + /* Supply already resolved? (fast-path without locking contention) */ if (rdev->supply) return 0; + /* + * Recheck rdev->supply with rdev->mutex lock held to avoid a race + * between rdev->supply null check and setting rdev->supply in + * set_supply() from concurrent tasks. + */ + regulator_lock(rdev); + + /* Supply just resolved by a concurrent task? */ + if (rdev->supply) + goto out; + r = regulator_dev_lookup(dev, rdev->supply_name); if (IS_ERR(r)) { ret = PTR_ERR(r); /* Did the lookup explicitly defer for us? */ if (ret == -EPROBE_DEFER) - return ret; + goto out; if (have_full_constraints()) { r = dummy_regulator_rdev; @@ -1837,15 +1848,18 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) } else { dev_err(dev, "Failed to resolve %s-supply for %s\n", rdev->supply_name, rdev->desc->name); - return -EPROBE_DEFER; + ret = -EPROBE_DEFER; + goto out; } } if (r == rdev) { dev_err(dev, "Supply for %s (%s) resolved to itself\n", rdev->desc->name, rdev->supply_name); - if (!have_full_constraints()) - return -EINVAL; + if (!have_full_constraints()) { + ret = -EINVAL; + goto out; + } r = dummy_regulator_rdev; get_device(&r->dev); } @@ -1859,7 +1873,8 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) if (r->dev.parent && r->dev.parent != rdev->dev.parent) { if (!device_is_bound(r->dev.parent)) { put_device(&r->dev); - return -EPROBE_DEFER; + ret = -EPROBE_DEFER; + goto out; } } @@ -1867,13 +1882,13 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) ret = regulator_resolve_supply(r); if (ret < 0) { put_device(&r->dev); - return ret; + goto out; } ret = set_supply(rdev, r); if (ret < 0) { put_device(&r->dev); - return ret; + goto out; } /* @@ -1886,11 +1901,13 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) if (ret < 0) { _regulator_put(rdev->supply); rdev->supply = NULL; - return ret; + goto out; } } - return 0; +out: + regulator_unlock(rdev); + return ret; } /* Internal regulator request function */ -- The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project