Re: [PATCH 088/190] Revert "mmc_spi: add a status check for spi_sync_locked"

[Laurent Pinchart]

Hi Greg,

On Wed, Apr 28, 2021 at 09:18:03AM +0200, Greg Kroah-Hartman wrote:
> On Thu, Apr 22, 2021 at 10:08:45AM +0200, Ulf Hansson wrote:
> > On Wed, 21 Apr 2021 at 15:19, Laurent Pinchart wrote:
> > > On Wed, Apr 21, 2021 at 02:59:23PM +0200, Greg Kroah-Hartman wrote:
> > > > This reverts commit 611025983b7976df0183390a63a2166411d177f1.
> > > >
> > > > Commits from @umn.edu addresses have been found to be submitted in "bad
> > > > faith" to try to test the kernel community's ability to review "known
> > > > malicious" changes.  The result of these submissions can be found in a
> > > > paper published at the 42nd IEEE Symposium on Security and Privacy
> > > > entitled, "Open Source Insecurity: Stealthily Introducing
> > > > Vulnerabilities via Hypocrite Commits" written by Qiushi Wu (University
> > > > of Minnesota) and Kangjie Lu (University of Minnesota).
> > > >
> > > > Because of this, all submissions from this group must be reverted from
> > > > the kernel tree and will need to be re-reviewed again to determine if
> > > > they actually are a valid fix.  Until that work is complete, remove this
> > > > change to ensure that no problems are being introduced into the
> > > > codebase.
> > > >
> > > > Cc: Kangjie Lu <kjlu@xxxxxxx>
> > > > Cc: Laurent Pinchart <laurent.pinchart@xxxxxxxxxxxxxxxx>
> > > > Cc: Ulf Hansson <ulf.hansson@xxxxxxxxxx>
> > > > Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
> > >
> > > Acked-by: Laurent Pinchart <laurent.pinchart@xxxxxxxxxxxxxxxx>
> > >
> > > I don't spot an obvious issue with the original patch though.
> > >
> > > > ---
> > > >  drivers/mmc/host/mmc_spi.c | 4 ----
> > > >  1 file changed, 4 deletions(-)
> > > >
> > > > diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
> > > > index 02f4fd26e76a..cc40b050e302 100644
> > > > --- a/drivers/mmc/host/mmc_spi.c
> > > > +++ b/drivers/mmc/host/mmc_spi.c
> > > > @@ -800,10 +800,6 @@ mmc_spi_readblock(struct mmc_spi_host *host, struct spi_transfer *t,
> > > >       }
> > > >
> > > >       status = spi_sync_locked(spi, &host->m);
> > > > -     if (status < 0) {
> > > > -             dev_dbg(&spi->dev, "read error %d\n", status);
> > > > -             return status;
> > > > -     }
> > 
> > Returning here means we never give back the ownership of the buffer to
> > the CPU. Can that be considered as vulnerability?
> 
> It's a "resource leak", which is a bug.  If you want to declare that as
> a "vulnerability" or not, I do not know.  Personally I do not think it
> is...

How is that a resource leak ? The dma_sync_single_for_device() calls
above this block don't take the buffer ownership away from the CPU in a
way that leaks it.

> > If that is that a problem, I can point out that there is already one
> > more case in this file, where this pattern is repeated. See
> > mmc_spi_writeblock(). This code has been there since 2007.
> 
> Yeah, these error paths are impossible to hit anyway.
> 
> I'll go drop this patch as it is not correct and will create a "correct"
> patch for this as well.

-- 
Regards,

Laurent Pinchart