linux-kernel - Re: [PATCH v2 1/2] staging: comedi: cast function output to assigned variable type

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 18 Feb 2021 18:51:59 +0300
From:   Dan Carpenter <dan.carpenter@...cle.com>
To:     Atul Gopinathan <atulgopinathan@...il.com>
Cc:     Greg KH <gregkh@...uxfoundation.org>, devel@...verdev.osuosl.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 1/2] staging: comedi: cast function output to assigned
 variable type

On Thu, Feb 18, 2021 at 06:22:20PM +0530, Atul Gopinathan wrote:
> On Thu, Feb 18, 2021 at 01:21:05PM +0100, Greg KH wrote:
> > On Thu, Feb 18, 2021 at 04:17:55PM +0530, Atul Gopinathan wrote:
> > > On Thu, Feb 18, 2021 at 10:31:15AM +0000, Ian Abbott wrote:
> > > > On 18/02/2021 08:44, Atul Gopinathan wrote:
> > > > > Fix the following warning generated by sparse:
> > > > > 
> > > > > drivers/staging//comedi/comedi_fops.c:2956:23: warning: incorrect type in assignment (different address spaces)
> > > > > drivers/staging//comedi/comedi_fops.c:2956:23:    expected unsigned int *chanlist
> > > > > drivers/staging//comedi/comedi_fops.c:2956:23:    got void [noderef] <asn:1> *
> > > > > 
> > > > > compat_ptr() has a return type of "void __user *"
> > > > > as defined in "include/linux/compat.h"
> > > > > 
> > > > > cmd->chanlist is of type "unsigned int *" as defined
> > > > > in drivers/staging/comedi/comedi.h" in struct
> > > > > comedi_cmd.
> > > > > 
> > > > > Signed-off-by: Atul Gopinathan <atulgopinathan@...il.com>
> > > > > ---
> > > > >   drivers/staging/comedi/comedi_fops.c | 2 +-
> > > > >   1 file changed, 1 insertion(+), 1 deletion(-)
> > > > > 
> > > > > diff --git a/drivers/staging/comedi/comedi_fops.c b/drivers/staging/comedi/comedi_fops.c
> > > > > index e85a99b68f31..fc4ec38012b4 100644
> > > > > --- a/drivers/staging/comedi/comedi_fops.c
> > > > > +++ b/drivers/staging/comedi/comedi_fops.c
> > > > > @@ -2953,7 +2953,7 @@ static int get_compat_cmd(struct comedi_cmd *cmd,
> > > > >   	cmd->scan_end_arg = v32.scan_end_arg;
> > > > >   	cmd->stop_src = v32.stop_src;
> > > > >   	cmd->stop_arg = v32.stop_arg;
> > > > > -	cmd->chanlist = compat_ptr(v32.chanlist);
> > > > > +	cmd->chanlist = (unsigned int __force *)compat_ptr(v32.chanlist);
> > > > >   	cmd->chanlist_len = v32.chanlist_len;
> > > > >   	cmd->data = compat_ptr(v32.data);
> > > > >   	cmd->data_len = v32.data_len;
> > > > > 
> > > > 
> > > > This patch and the other one in your series clash with commit 9d5d041eebe3
> > > > ("staging: comedi: comedi_fops.c: added casts to get rid of sparse
> > > > warnings") by B K Karthik.
> > > 
> > > Oh I see. Not sure if this is the right place to ask, but which tree and
> > > branch should one work with when messing with the code in staging/
> > > directory? (wanted to avoid such clashes in future)
> > 
> > staging-next is the best one to use from the staging.git tree.  But as
> > the above commit was merged in 5.9-rc1, way back in July of last year, I
> > have no idea what tree you are currently using to not hit that...
> 
> I'm using the staging tree alright, cloned it yesterday. Except I used the
> --depth parameter. I believe that is the culprit. How bad a mistake is
> that?
> 
> (Why depth? I'm currently staying in a remote area where internet
> download speeds are less than 100Kbps. I tried a normal git clone of
> the staging tree and it's estimated time was more than half a day. Not
> to mention, it fails due to loss of connection midway every time)

I live in Africa and used to have this problem before Africa got
connected to fibre optic.  Greg's suggestion of using `wget -c` to
download a git bundle is a good one.  However, in my experience
`wget -c` is not 100% accurate on super flakey internet.  Then if it
fails the verify will fail and you'll have to re-download the entire
thing.

If you want I have a different option.  What I made a clone of Linus's
tree.  Then I did:

tar cvvf linux.tar linux/.git
for i in $(seq 0 116) ; do dd if=linux.tar of=linux.tar.${i} ibs=1M count=20 skip=$((20 * $i)) ; done

Now I have put those on a private webserver I have.  You can download
them by doing:

for i in $(seq 0 116) ; do wget -c http://51.158.124.72/kernel_split/linux.tar.${i} ; done
md5sum * > my_md5s
wget http://51.158.124.72/kernel_split/md5sums
diff -u md5sum my_md5s

Re-download files which don't match

for i in $(seq 0 116) ; do cat linux.tar.${i} >> linux.tar ; done
cat linux.tar | tar xv
cd linux
git reset --hard

Let me know if this works for you.  I will leave the source up for a
week or two before I delete it.

regards,
dan carpenter