Message ID | 1424725642-26270-1-git-send-email-computersforpeace@gmail.com |
---|---|
State | Accepted |
Headers | show |
Am 23.02.2015 um 22:07 schrieb Brian Norris: > commit 0e707ae79ba3 ("UBI: do propagate positive error codes up") seems > to have produced an unintended change in the control flow here. > > Completely untested, but it looks obvious. > > Caught by Coverity, which didn't like the indentation. CID 1271184. > > Signed-off-by: Brian Norris <computersforpeace@gmail.com> > Cc: Dan Carpenter <dan.carpenter@oracle.com> > --- > Should go into 4.0, I expect. Good catch, patch tested and applied! Thank you Brian! //richard
On Mon, Feb 23, 2015 at 10:21:53PM +0100, Richard Weinberger wrote: > Am 23.02.2015 um 22:07 schrieb Brian Norris: > > commit 0e707ae79ba3 ("UBI: do propagate positive error codes up") seems > > to have produced an unintended change in the control flow here. > > > > Completely untested, but it looks obvious. > > > > Caught by Coverity, which didn't like the indentation. CID 1271184. > > > > Signed-off-by: Brian Norris <computersforpeace@gmail.com> > > Cc: Dan Carpenter <dan.carpenter@oracle.com> > > --- > > Should go into 4.0, I expect. > > Good catch, patch tested and applied! Is this going in 4.0? It fixes a typo in a hastily-applied patch that made it to 4.0-rc1. I'm also not sure I understand the role of the +linux-next and +master branches in linux-ubifs.git. Typically 'next' means for the current+1 release (i.e., 4.1), while 'not-next' (i.e., your master branch?) would be for the current release (4.0). But you have +master based on top of +linux-next. Brian
Hi! Am 16.03.2015 um 19:08 schrieb Brian Norris: > On Mon, Feb 23, 2015 at 10:21:53PM +0100, Richard Weinberger wrote: >> Am 23.02.2015 um 22:07 schrieb Brian Norris: >>> commit 0e707ae79ba3 ("UBI: do propagate positive error codes up") seems >>> to have produced an unintended change in the control flow here. >>> >>> Completely untested, but it looks obvious. >>> >>> Caught by Coverity, which didn't like the indentation. CID 1271184. >>> >>> Signed-off-by: Brian Norris <computersforpeace@gmail.com> >>> Cc: Dan Carpenter <dan.carpenter@oracle.com> >>> --- >>> Should go into 4.0, I expect. >> >> Good catch, patch tested and applied! > > Is this going in 4.0? It fixes a typo in a hastily-applied patch that > made it to 4.0-rc1. That's the plan. > I'm also not sure I understand the role of the +linux-next and +master > branches in linux-ubifs.git. Typically 'next' means for the current+1 > release (i.e., 4.1), while 'not-next' (i.e., your master branch?) would > be for the current release (4.0). But you have +master based on top of > +linux-next. I'm using Artem's scheme. next is the branch Linus pulls from. Artem, why are the two UBIFS fixes from master not in next? I thought you want to send a pull request to Linus? If you want, I can do that. I'm anyway preparing some UBI fixes. Thanks, //richard
On Mon, 2015-03-16 at 19:21 +0100, Richard Weinberger wrote: > Hi! > > Am 16.03.2015 um 19:08 schrieb Brian Norris: > > On Mon, Feb 23, 2015 at 10:21:53PM +0100, Richard Weinberger wrote: > >> Am 23.02.2015 um 22:07 schrieb Brian Norris: > >>> commit 0e707ae79ba3 ("UBI: do propagate positive error codes up") seems > >>> to have produced an unintended change in the control flow here. > >>> > >>> Completely untested, but it looks obvious. > >>> > >>> Caught by Coverity, which didn't like the indentation. CID 1271184. > >>> > >>> Signed-off-by: Brian Norris <computersforpeace@gmail.com> > >>> Cc: Dan Carpenter <dan.carpenter@oracle.com> > >>> --- > >>> Should go into 4.0, I expect. > >> > >> Good catch, patch tested and applied! > > > > Is this going in 4.0? It fixes a typo in a hastily-applied patch that > > made it to 4.0-rc1. > > That's the plan. > > > I'm also not sure I understand the role of the +linux-next and +master > > branches in linux-ubifs.git. Typically 'next' means for the current+1 > > release (i.e., 4.1), while 'not-next' (i.e., your master branch?) would > > be for the current release (4.0). But you have +master based on top of > > +linux-next. > > I'm using Artem's scheme. next is the branch Linus pulls from. > Artem, why are the two UBIFS fixes from master not in next? Just wanted to keep them there for several days. > I thought you want to send a pull request to Linus? No, I was not going to send this for 4.0. There is one fix which looks important, but on the other hand, I did not test it, and no one reported this problem and said this patch helps, so I was not going to send it earlier. > If you want, I can do that. I'm anyway preparing some UBI fixes. Yes, please, merge master to linux-next.
diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c index da4c79259f67..16e34b37d134 100644 --- a/drivers/mtd/ubi/eba.c +++ b/drivers/mtd/ubi/eba.c @@ -425,9 +425,10 @@ retry: ubi_warn(ubi, "corrupted VID header at PEB %d, LEB %d:%d", pnum, vol_id, lnum); err = -EBADMSG; - } else + } else { err = -EINVAL; ubi_ro_mode(ubi); + } } goto out_free; } else if (err == UBI_IO_BITFLIPS)
commit 0e707ae79ba3 ("UBI: do propagate positive error codes up") seems to have produced an unintended change in the control flow here. Completely untested, but it looks obvious. Caught by Coverity, which didn't like the indentation. CID 1271184. Signed-off-by: Brian Norris <computersforpeace@gmail.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> --- Should go into 4.0, I expect. drivers/mtd/ubi/eba.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)