Thanks Alison . Lesson learned !!!
On 2026-11-05 15:21 -0800, Alison Schofield wrote: > On Wed, Nov 04, 2026 at 09:48:25PM +0530, [email protected] wrote: > > From: Mohammad Shahid <[email protected]> > > > > pmem_attach_disk() allocates a disk using blk_alloc_disk() > > before initializing badblocks. > > > > If devm_init_badblocks() fails, the function return directly > > without releasing the previously allocated disk. > > > > Use the existing cleanup path to release the disk and preserve > > the original error code returned by devm_init_badblocks(). > > Hi Mohammad, > > Your patch looks right but this leak is fixed in v7.4-rc1 by: > 855a46681e0b ("nvdimm/pmem: Release gendisk on probe failure") > > Basing and building your patches on the latest tree, usually > the most recent mainline -rc before sending helps avoid this. > Sometimes a patch is also just on the mailing list, so check > the list too. > > -- Alison > > > > > Signed-off-by: Mohammad Shahid <[email protected]> > > --- > > drivers/nvdimm/pmem.c | 5 +++-- > > 1 file changed, 3 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c > > index 5fb86595e8bd..76e80eccc8e7 100644 > > --- a/drivers/nvdimm/pmem.c > > +++ b/drivers/nvdimm/pmem.c > > @@ -563,8 +563,9 @@ static int pmem_attach_disk(struct device *dev, > > nvdimm_namespace_disk_name(ndns, disk->disk_name); > > set_capacity(disk, (pmem->size - pmem->pfn_pad - pmem->data_offset) > > / 512); > > - if (devm_init_badblocks(dev, &pmem->bb)) > > - return -ENOMEM; > > + rc = devm_init_badblocks(dev, &pmem->bb); > > + if (rc) > > + goto out; > > nvdimm_badblocks_populate(nd_region, &pmem->bb, &bb_range); > > disk->bb = &pmem->bb; > > > > -- > > 2.43.0 > > > >
