Re: [PATCH] mtd: Add check for kcalloc()
From: Jiasheng Jiang
Date: Mon Feb 03 2025 - 21:34:57 EST
Hi Miquel,
On Mon, Feb 3, 2025 at 3:32 AM Miquel Raynal <miquel.raynal@xxxxxxxxxxx> wrote:
>
> On 02/02/2025 at 20:54:56 GMT, Jiasheng Jiang <jiashengjiangcool@xxxxxxxxx> wrote:
>
> > Add a check for kcalloc() and print an error message if it fails.
>
> Checking, yes, but logging, no. IIRC the core does it already if required.
>
> > Fixes: 78c08247b9d3 ("mtd: Support kmsg dumper based on pstore/blk")
>
> Cc: stable is missing
>
> > Signed-off-by: Jiasheng Jiang <jiashengjiangcool@xxxxxxxxx>
> > ---
> > drivers/mtd/mtdpstore.c | 5 +++++
> > 1 file changed, 5 insertions(+)
> >
> > diff --git a/drivers/mtd/mtdpstore.c b/drivers/mtd/mtdpstore.c
> > index 7ac8ac901306..368997155c07 100644
> > --- a/drivers/mtd/mtdpstore.c
> > +++ b/drivers/mtd/mtdpstore.c
> > @@ -423,6 +423,11 @@ static void mtdpstore_notify_add(struct mtd_info *mtd)
> > longcnt = BITS_TO_LONGS(div_u64(mtd->size, mtd->erasesize));
> > cxt->badmap = kcalloc(longcnt, sizeof(long), GFP_KERNEL);
> >
> > + if (!cxt->rmmap || !cxt->usedmap || !cxt->badmap) {
> > + dev_err(&mtd->dev, "failed to allocate memory for map\n");
> > + return;
> > + }
> > +
> > /* just support dmesg right now */
> > cxt->dev.flags = PSTORE_FLAGS_DMESG;
> > cxt->dev.zone.read = mtdpstore_read;
>
> What if register_pstore_device() fails? Your only partially fixing the
> problem if you don't handle the free()s there as well.
>
> Thanks,
> Miquèl
Thanks, I have submitted a v2 to fix the problems above.
-Jiasheng