[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <55936976.7090001@intel.com>
Date: Wed, 01 Jul 2015 12:15:50 +0800
From: Pan Xinhui <xinhuix.pan@...el.com>
To: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
CC: Yury Norov <yury.norov@...il.com>,
Andrew Morton <akpm@...ux-foundation.org>,
Rasmus Villemoes <linux@...musvillemoes.dk>, tj@...nel.org,
peterz@...radead.org, sudeep.holla@....com, mina86@...a86.com,
"mnipxh@....com" <mnipxh@....com>,
Alexey Klimov <klimov.linux@...il.com>,
"yanmin_zhang@...ux.intel.com" <yanmin_zhang@...ux.intel.com>
Subject: [PATCH V2] lib/bitmap.c: fix some parsing issues and code style
In __bitmap_parselist we can accept whitespaces on head or tail
during every parsing procedure.
If input has valid ranges, there is no reason to reject the user.
fixes are:
1) if input ends with ',', bit 0 might be set unexpectedly.
now we check if any digit is available after every loop.
2) if input has '0-', bit 0 might be set unexpectedly,
now we return -EINVAL as this kind of input is definitely wrong.
3) minor code style fix in __bitmap_parse.
and avoid in-loop incrementation of ndigits.
commit 2528a8b also add some check, but it's still not enough.
it only correct the result in fix 1 above.
Signed-off-by: Pan Xinhui <xinhuix.pan@...el.com>
---
lib/bitmap.c | 43 ++++++++++++++++++++++++++-----------------
1 file changed, 26 insertions(+), 17 deletions(-)
---
change log
V2 from v1: solve the codes conflicts
---
diff --git a/lib/bitmap.c b/lib/bitmap.c
index a578a01..b0e3afc 100644
--- a/lib/bitmap.c
+++ b/lib/bitmap.c
@@ -367,7 +367,8 @@ int __bitmap_parse(const char *buf, unsigned int buflen,
nchunks = nbits = totaldigits = c = 0;
do {
- chunk = ndigits = 0;
+ chunk = 0;
+ ndigits = totaldigits;
/* Get the next chunk of the bitmap */
while (buflen) {
@@ -406,9 +407,9 @@ int __bitmap_parse(const char *buf, unsigned int buflen,
return -EOVERFLOW;
chunk = (chunk << 4) | hex_to_bin(c);
- ndigits++; totaldigits++;
+ totaldigits++;
}
- if (ndigits == 0)
+ if (unlikely(ndigits == totaldigits))
return -EINVAL;
if (nchunks == 0 && chunk == 0)
continue;
@@ -505,7 +506,7 @@ static int __bitmap_parselist(const char *buf, unsigned int buflen,
int nmaskbits)
{
unsigned a, b;
- int c, old_c, totaldigits;
+ int c, old_c, totaldigits, ndigits;
const char __user __force *ubuf = (const char __user __force *)buf;
int at_start, in_range;
@@ -515,6 +516,7 @@ static int __bitmap_parselist(const char *buf, unsigned int buflen,
at_start = 1;
in_range = 0;
a = b = 0;
+ ndigits = totaldigits;
/* Get the next cpu# or a range of cpu#'s */
while (buflen) {
@@ -528,23 +530,27 @@ static int __bitmap_parselist(const char *buf, unsigned int buflen,
if (isspace(c))
continue;
- /*
- * If the last character was a space and the current
- * character isn't '\0', we've got embedded whitespace.
- * This is a no-no, so throw an error.
- */
- if (totaldigits && c && isspace(old_c))
- return -EINVAL;
-
/* A '\0' or a ',' signal the end of a cpu# or range */
if (c == '\0' || c == ',')
break;
+ /*
+ * whitespaces between digits are not allowed,
+ * but it's ok if whitespaces are on head or tail.
+ * when old_c is whilespace,
+ * if totaldigits == ndigits, whitespace is on head.
+ * if whitespace is on tail, it should not run here.
+ * as c was ',' or '\0',
+ * the last code line has broken the current loop.
+ */
+ if ((totaldigits != ndigits) && isspace(old_c))
+ return -EINVAL;
if (c == '-') {
if (at_start || in_range)
return -EINVAL;
b = 0;
in_range = 1;
+ at_start = 1;
continue;
}
@@ -557,15 +563,18 @@ static int __bitmap_parselist(const char *buf, unsigned int buflen,
at_start = 0;
totaldigits++;
}
+ if (unlikely(ndigits == totaldigits))
+ continue;
+ /* if no digit is after '-', it's wrong*/
+ if (unlikely(at_start && in_range))
+ return -EINVAL;
if (!(a <= b))
return -EINVAL;
if (b >= nmaskbits)
return -ERANGE;
- if (!at_start) {
- while (a <= b) {
- set_bit(a, maskp);
- a++;
- }
+ while (a <= b) {
+ set_bit(a, maskp);
+ a++;
}
} while (buflen && c == ',');
return 0;
--
1.9.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists