Thanks! Actually, I do't think this function is useful on the client. >From the name of it, it would have been used in the server code, but even there it was recently removed as I now see. I'll submit a removal patch. On Sep 8, 2014, at 6:38 PM, Dan Carpenter wrote: > On Tue, Sep 09, 2014 at 12:12:01AM +0200, Hugues wrote: >> Fix potential buffer underflow >> >> When the string 'name' start with a '-' ptr == name and so --ptr will >> underflow, this patch corrected this by checking ptr before decrementing. > > Good eye. Please fix up the changelog and send it inline etc and > resend. I've added Oleg and Andreas to the CC list so that this doesn't > get lost. > > regards, > dan carpenter > >>> From 5786040323bd1dc3149fb4afbea9f01e2bf4cfdf Mon Sep 17 00:00:00 2001 >> From: Hugues Morisset <morisset.hugues@xxxxxxxxx> >> Date: Mon, 8 Sep 2014 22:59:04 +0200 >> Subject: [PATCH 4/4] staging: lustre: Fix potential buffer underflow >> >> Signed-off-by: Hugues Morisset <morisset.hugues@xxxxxxxxx> >> --- >> drivers/staging/lustre/lustre/include/obd.h | 7 ++++++- >> 1 file changed, 6 insertions(+), 1 deletion(-) >> >> diff --git a/drivers/staging/lustre/lustre/include/obd.h b/drivers/staging/lustre/lustre/include/obd.h >> index 409eae6..bb67e40 100644 >> --- a/drivers/staging/lustre/lustre/include/obd.h >> +++ b/drivers/staging/lustre/lustre/include/obd.h >> @@ -677,7 +677,12 @@ static inline int is_osp_on_mdt(char *name) >> if (strncmp(ptr + 1, "MDT", 3) != 0) >> return 0; >> >> - while (*(--ptr) != '-' && ptr != name); >> + if (ptr == name) >> + return 0; >> + >> + --ptr; >> + while (ptr != '-' && ptr != name) >> + --ptr; >> >> if (ptr == name) >> return 0; >> -- >> 2.1.0 >> > >> _______________________________________________ >> devel mailing list >> devel@xxxxxxxxxxxxxxxxxxxxxx >> http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel > _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel