Hi Alison, Thank you for pointing this out! I'll drop my patch since Muchun's version covers both cases. I'll review Muchun's patch and add my Reviewed-by if it looks good.
Best regards, Jianyun On Tue, Oct 13, 2026 at 7:21 AM Alison Schofield <[email protected]> wrote: > > On Sun, Oct 11, 2026 at 09:18:57PM +0800, [email protected] wrote: > > From: Jianyun Gao <[email protected]> > > > > When the region driver is not loaded, mapping_store() returns rc > > which is 0 at that point, instead of an error code. Set it to > > -ENXIO to match the equivalent check in size_store(). > > > Hi Jianyun, > > The change looks correct, but there is another path in mapping_store() with > the same problem. When alloc_is_aligned() fails, rc is still zero, so the > sysfs write reports success without allocating the requested range. > > Muchun Song already posted a patch addressing both cases. > https://lore.kernel.org/nvdimm/[email protected]/ > > Please take a look and add your reviewed-by tag to Muchun's patchset > or comment if it doesn't meet your needs. > > -- Alison > > > > > Fixes: 9567da0b408a ("device-dax: Introduce bus + driver model") > > Signed-off-by: Jianyun Gao <[email protected]> > > --- > > drivers/dax/bus.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/drivers/dax/bus.c b/drivers/dax/bus.c > > index b809e1a264af..262f97b5e7aa 100644 > > --- a/drivers/dax/bus.c > > +++ b/drivers/dax/bus.c > > @@ -1192,7 +1192,7 @@ static ssize_t mapping_store(struct device *dev, > > struct device_attribute *attr, > > return rc; > > if (!dax_region->dev->driver) { > > up_write(&dax_region_rwsem); > > - return rc; > > + return -ENXIO; > > } > > rc = down_write_killable(&dax_dev_rwsem); > > if (rc) { > > -- > > 2.34.1 > >
