There are several places in kernel where modules unescapes input to convert
C-Style Escape Sequences into byte codes.
The patch provides generic implementation of such approach. Test cases are
also included into the patch.
Signed-off-by: Andy Shevchenko <[email protected]>
Cc: Samuel Thibault <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: Jason Baron <[email protected]>
Cc: Alexander Viro <[email protected]>
---
include/linux/string_helpers.h | 58 ++++++++++++++++++
lib/Kconfig.debug | 3 +
lib/Makefile | 4 +-
lib/string_helpers.c | 133 +++++++++++++++++++++++++++++++++++++++++
lib/test-string_helpers.c | 103 +++++++++++++++++++++++++++++++
5 files changed, 300 insertions(+), 1 deletion(-)
create mode 100644 lib/test-string_helpers.c
diff --git a/include/linux/string_helpers.h b/include/linux/string_helpers.h
index a3eb2f6..c18df71 100644
--- a/include/linux/string_helpers.h
+++ b/include/linux/string_helpers.h
@@ -13,4 +13,62 @@ enum string_size_units {
int string_get_size(u64 size, enum string_size_units units,
char *buf, int len);
+#define UNESCAPE_SPACE 0x01
+#define UNESCAPE_OCTAL 0x02
+#define UNESCAPE_HEX 0x04
+#define UNESCAPE_SPECIAL 0x08
+#define UNESCAPE_ANY \
+ (UNESCAPE_SPACE | UNESCAPE_OCTAL | UNESCAPE_HEX | UNESCAPE_SPECIAL)
+
+/**
+ * string_unescape - unquote characters in the given string
+ * @src: source buffer (escaped)
+ * @dst: destination buffer (unescaped)
+ * @size: size of the destination buffer (0 to unlimit)
+ * @flags: combination of the flags (bitwise OR):
+ * %UNESCAPE_SPACE:
+ * '\f' - form feed
+ * '\n' - new line
+ * '\r' - carriage return
+ * '\t' - horizontal tab
+ * '\v' - vertical tab
+ * %UNESCAPE_OCTAL:
+ * '\NNN' - byte with octal value NNN (1 to 3 digits)
+ * %UNESCAPE_HEX:
+ * '\xHH' - byte with hexadecimal value HH (1 to 2 digits)
+ * %UNESCAPE_SPECIAL:
+ * '\"' - double quote
+ * '\\' - backslash
+ * '\a' - alert (BEL)
+ * '\e' - escape
+ * %UNESCAPE_ANY:
+ * all previous together
+ *
+ * Returns amount of characters processed to the destination buffer excluding
+ * trailing '\0'.
+ *
+ * Due to size of output will be the same or smaller the translation could be
+ * done in place.
+ *
+ * Caller must provide valid source and destination pointers. Be aware that
+ * destination buffer will always be NULL-terminated. Source string must be
+ * NULL-terminated as well.
+ */
+int string_unescape(char *src, char *dst, size_t size, unsigned int flags);
+
+static inline int string_unescape_inplace(char *buf, unsigned int flags)
+{
+ return string_unescape(buf, buf, 0, flags);
+}
+
+static inline int string_unescape_any(char *src, char *dst, size_t size)
+{
+ return string_unescape(src, dst, size, UNESCAPE_ANY);
+}
+
+static inline int string_unescape_any_inplace(char *buf)
+{
+ return string_unescape_any(buf, buf, 0);
+}
+
#endif
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index ae80518..770a422 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -1481,5 +1481,8 @@ source "lib/Kconfig.kgdb"
source "lib/Kconfig.kmemcheck"
+config TEST_STRING_HELPERS
+ tristate "Test functions located in the string_helpers module at runtime"
+
config TEST_KSTRTOX
tristate "Test kstrto*() family of functions at runtime"
diff --git a/lib/Makefile b/lib/Makefile
index 10facc6..370a1d6 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -23,8 +23,10 @@ lib-y += kobject.o klist.o
obj-y += bcd.o div64.o sort.o parser.o halfmd4.o debug_locks.o random32.o \
bust_spinlocks.o hexdump.o kasprintf.o bitmap.o scatterlist.o \
- string_helpers.o gcd.o lcm.o list_sort.o uuid.o flex_array.o \
+ gcd.o lcm.o list_sort.o uuid.o flex_array.o \
bsearch.o find_last_bit.o find_next_bit.o llist.o memweight.o kfifo.o
+obj-y += string_helpers.o
+obj-$(CONFIG_TEST_STRING_HELPERS) += test-string_helpers.o
obj-y += kstrtox.o
obj-$(CONFIG_TEST_KSTRTOX) += test-kstrtox.o
diff --git a/lib/string_helpers.c b/lib/string_helpers.c
index 1cffc22..ed5c145 100644
--- a/lib/string_helpers.c
+++ b/lib/string_helpers.c
@@ -2,10 +2,12 @@
* Helpers for formatting and printing strings
*
* Copyright 31 August 2008 James Bottomley
+ * Copyright (C) 2013, Intel Corporation
*/
#include <linux/kernel.h>
#include <linux/math64.h>
#include <linux/export.h>
+#include <linux/ctype.h>
#include <linux/string_helpers.h>
/**
@@ -66,3 +68,134 @@ int string_get_size(u64 size, const enum string_size_units units,
return 0;
}
EXPORT_SYMBOL(string_get_size);
+
+static bool unescape_space(char **src, char **dst)
+{
+ char *p = *dst, *q = *src;
+
+ switch (*q) {
+ case 'n':
+ *p = '\n';
+ break;
+ case 'r':
+ *p = '\r';
+ break;
+ case 't':
+ *p = '\t';
+ break;
+ case 'v':
+ *p = '\v';
+ break;
+ case 'f':
+ *p = '\f';
+ break;
+ default:
+ return false;
+ }
+ *dst += 1;
+ *src += 1;
+ return true;
+}
+
+static bool unescape_octal(char **src, char **dst)
+{
+ char *p = *dst, *q = *src;
+ u8 num;
+
+ if (isodigit(*q) == 0)
+ return false;
+
+ num = (*q++) & 7;
+ while (num < 32 && isodigit(*q) && (q - *src < 3)) {
+ num <<= 3;
+ num += (*q++) & 7;
+ }
+ *p = num;
+ *dst += 1;
+ *src = q;
+ return true;
+}
+
+static bool unescape_hex(char **src, char **dst)
+{
+ char *p = *dst, *q = *src;
+ int digit;
+ u8 num;
+
+ if (*q++ != 'x')
+ return false;
+
+ num = digit = hex_to_bin(*q++);
+ if (digit < 0)
+ return false;
+
+ digit = hex_to_bin(*q);
+ if (digit >= 0) {
+ q++;
+ num = (num << 4) | digit;
+ }
+ *p = num;
+ *dst += 1;
+ *src = q;
+ return true;
+}
+
+static bool unescape_special(char **src, char **dst)
+{
+ char *p = *dst, *q = *src;
+
+ switch (*q) {
+ case '\"':
+ *p = '\"';
+ break;
+ case '\\':
+ *p = '\\';
+ break;
+ case 'a':
+ *p = '\a';
+ break;
+ case 'e':
+ *p = '\e';
+ break;
+ default:
+ return false;
+ }
+ *dst += 1;
+ *src += 1;
+ return true;
+}
+
+int string_unescape(char *src, char *dst, size_t size, unsigned int flags)
+{
+ char *out = dst;
+
+ while (*src && --size) {
+ if (src[0] == '\\' && src[1] != '\0' && size > 1) {
+ src++;
+ size--;
+
+ if (flags & UNESCAPE_SPACE &&
+ unescape_space(&src, &out))
+ continue;
+
+ if (flags & UNESCAPE_OCTAL &&
+ unescape_octal(&src, &out))
+ continue;
+
+ if (flags & UNESCAPE_HEX &&
+ unescape_hex(&src, &out))
+ continue;
+
+ if (flags & UNESCAPE_SPECIAL &&
+ unescape_special(&src, &out))
+ continue;
+
+ *out++ = '\\';
+ }
+ *out++ = *src++;
+ }
+ *out = '\0';
+
+ return out - dst;
+}
+EXPORT_SYMBOL(string_unescape);
diff --git a/lib/test-string_helpers.c b/lib/test-string_helpers.c
new file mode 100644
index 0000000..6ac48de
--- /dev/null
+++ b/lib/test-string_helpers.c
@@ -0,0 +1,103 @@
+/*
+ * Test cases for lib/string_helpers.c module.
+ */
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/random.h>
+#include <linux/string.h>
+#include <linux/string_helpers.h>
+
+struct test_string {
+ const char *in;
+ const char *out;
+ unsigned int flags;
+};
+
+static const struct test_string strings[] __initconst = {
+ {
+ .in = "\\f\\ \\n\\r\\t\\v",
+ .out = "\f\\ \n\r\t\v",
+ .flags = UNESCAPE_SPACE,
+ },
+ {
+ .in = "\\40\\1\\387\\0064\\05\\040\\8a\\110\\777",
+ .out = " \001\00387\0064\005 \\8aH?7",
+ .flags = UNESCAPE_OCTAL,
+ },
+ {
+ .in = "\\xv\\xa\\x2c\\xD\\x6f2",
+ .out = "\\xv\n,\ro2",
+ .flags = UNESCAPE_HEX,
+ },
+ {
+ .in = "\\h\\\\\\\"\\a\\e\\",
+ .out = "\\h\\\"\a\e\\",
+ .flags = UNESCAPE_SPECIAL,
+ },
+};
+
+static void __init test_string_unescape(unsigned int flags, bool inplace)
+{
+ char in[256];
+ char out_test[256];
+ char out_real[256];
+ int i, p = 0, q_test = 0, q_real = sizeof(out_real);
+
+ for (i = 0; i < ARRAY_SIZE(strings); i++) {
+ const char *s = strings[i].in;
+ int len = strlen(strings[i].in);
+
+ /* Copy string to in buffer */
+ memcpy(&in[p], s, len);
+ p += len;
+
+ /* Copy expected result for given flags */
+ if (flags & strings[i].flags) {
+ s = strings[i].out;
+ len = strlen(strings[i].out);
+ }
+ memcpy(&out_test[q_test], s, len);
+ q_test += len;
+ }
+ in[p++] = '\0';
+
+ /* Call string_unescape and compare result */
+ if (inplace) {
+ memcpy(out_real, in, p);
+ if (flags == UNESCAPE_ANY)
+ q_real = string_unescape_any_inplace(out_real);
+ else
+ q_real = string_unescape_inplace(out_real, flags);
+ } else if (flags == UNESCAPE_ANY) {
+ q_real = string_unescape_any(in, out_real, q_real);
+ } else {
+ q_real = string_unescape(in, out_real, q_real, flags);
+ }
+
+ if (q_real != q_test || memcmp(out_test, out_real, q_test)) {
+ pr_warn("Test failed: flags = %u\n", flags);
+ print_hex_dump(KERN_WARNING, "Input: ",
+ DUMP_PREFIX_NONE, 16, 1, in, p - 1, true);
+ print_hex_dump(KERN_WARNING, "Expected: ",
+ DUMP_PREFIX_NONE, 16, 1, out_test, q_test, true);
+ print_hex_dump(KERN_WARNING, "Got: ",
+ DUMP_PREFIX_NONE, 16, 1, out_real, q_real, true);
+ }
+}
+
+static int __init test_string_helpers_init(void)
+{
+ unsigned int i;
+
+ pr_info("Running tests...\n");
+ for (i = 0; i < UNESCAPE_ANY + 1; i++)
+ test_string_unescape(i, false);
+ test_string_unescape(get_random_int() % (UNESCAPE_ANY + 1), true);
+
+ return -EINVAL;
+}
+module_init(test_string_helpers_init);
+MODULE_LICENSE("Dual BSD/GPL");
--
1.8.2.rc0.22.gb3600c3
There is string_unescape_inplace() function which decodes strings in generic
way. Let's use it.
Signed-off-by: Andy Shevchenko <[email protected]>
Cc: Alexander Viro <[email protected]>
Cc: [email protected]
---
fs/binfmt_misc.c | 24 ++++--------------------
1 file changed, 4 insertions(+), 20 deletions(-)
diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index 751df5e..1c740e1 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -23,6 +23,7 @@
#include <linux/binfmts.h>
#include <linux/slab.h>
#include <linux/ctype.h>
+#include <linux/string_helpers.h>
#include <linux/file.h>
#include <linux/pagemap.h>
#include <linux/namei.h>
@@ -234,24 +235,6 @@ static char *scanarg(char *s, char del)
return s;
}
-static int unquote(char *from)
-{
- char c = 0, *s = from, *p = from;
-
- while ((c = *s++) != '\0') {
- if (c == '\\' && *s == 'x') {
- s++;
- c = toupper(*s++);
- *p = (c - (isdigit(c) ? '0' : 'A' - 10)) << 4;
- c = toupper(*s++);
- *p++ |= c - (isdigit(c) ? '0' : 'A' - 10);
- continue;
- }
- *p++ = c;
- }
- return p - from;
-}
-
static char * check_special_flags (char * sfs, Node * e)
{
char * p = sfs;
@@ -354,8 +337,9 @@ static Node *create_entry(const char __user *buffer, size_t count)
p[-1] = '\0';
if (!e->mask[0])
e->mask = NULL;
- e->size = unquote(e->magic);
- if (e->mask && unquote(e->mask) != e->size)
+ e->size = string_unescape_inplace(e->magic, UNESCAPE_HEX);
+ if (e->mask &&
+ string_unescape_inplace(e->mask, UNESCAPE_HEX) != e->size)
goto Einval;
if (e->size + e->offset > BINPRM_BUF_SIZE)
goto Einval;
--
1.8.2.rc0.22.gb3600c3
There is generic implementation of the function to unescape strings.
Signed-off-by: Andy Shevchenko <[email protected]>
Cc: Samuel Thibault <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: [email protected]
Cc: [email protected]
---
drivers/staging/speakup/kobjects.c | 6 +++--
drivers/staging/speakup/speakup.h | 1 -
drivers/staging/speakup/varhandlers.c | 46 -----------------------------------
3 files changed, 4 insertions(+), 49 deletions(-)
diff --git a/drivers/staging/speakup/kobjects.c b/drivers/staging/speakup/kobjects.c
index d6d9264..943b6c1 100644
--- a/drivers/staging/speakup/kobjects.c
+++ b/drivers/staging/speakup/kobjects.c
@@ -15,6 +15,7 @@
#include <linux/kernel.h>
#include <linux/kobject.h>
#include <linux/string.h>
+#include <linux/string_helpers.h>
#include <linux/sysfs.h>
#include <linux/ctype.h>
@@ -417,7 +418,7 @@ static ssize_t synth_direct_store(struct kobject *kobj,
bytes = min_t(size_t, len, 250);
strncpy(tmp, ptr, bytes);
tmp[bytes] = '\0';
- spk_xlate(tmp);
+ string_unescape_any_inplace(tmp);
synth_printf("%s", tmp);
ptr += bytes;
len -= bytes;
@@ -605,7 +606,8 @@ ssize_t spk_var_store(struct kobject *kobj, struct kobj_attribute *attr,
if (param->data == NULL)
return 0;
ret = 0;
- cp = spk_xlate((char *) buf);
+ cp = (char *)buf;
+ string_unescape_any_inplace(cp);
spk_lock(flags);
switch (param->var_type) {
diff --git a/drivers/staging/speakup/speakup.h b/drivers/staging/speakup/speakup.h
index c387a02..0126f71 100644
--- a/drivers/staging/speakup/speakup.h
+++ b/drivers/staging/speakup/speakup.h
@@ -54,7 +54,6 @@ void spk_get_index_count(int *linecount, int *sentcount);
extern int spk_set_key_info(const u_char *key_info, u_char *k_buffer);
extern char *spk_strlwr(char *s);
extern char *spk_s2uchar(char *start, char *dest);
-extern char *spk_xlate(char *s);
extern int speakup_kobj_init(void);
extern void speakup_kobj_exit(void);
extern int spk_chartab_get_value(char *keyword);
diff --git a/drivers/staging/speakup/varhandlers.c b/drivers/staging/speakup/varhandlers.c
index 0099cb1..7f6288f 100644
--- a/drivers/staging/speakup/varhandlers.c
+++ b/drivers/staging/speakup/varhandlers.c
@@ -328,49 +328,3 @@ char *spk_s2uchar(char *start, char *dest)
*dest = (u_char)val;
return start;
}
-
-char *spk_xlate(char *s)
-{
- static const char finds[] = "nrtvafe";
- static const char subs[] = "\n\r\t\013\001\014\033";
- static const char hx[] = "0123456789abcdefABCDEF";
- char *p = s, *p1, *p2, c;
- int num;
- while ((p = strchr(p, '\\'))) {
- p1 = p+1;
- p2 = strchr(finds, *p1);
- if (p2) {
- *p++ = subs[p2-finds];
- p1++;
- } else if (*p1 >= '0' && *p1 <= '7') {
- num = (*p1++)&7;
- while (num < 32 && *p1 >= '0' && *p1 <= '7') {
- num <<= 3;
- num += (*p1++)&7;
- }
- *p++ = num;
- } else if (*p1 == 'x' &&
- strchr(hx, p1[1]) && strchr(hx, p1[2])) {
- p1++;
- c = *p1++;
- if (c > '9')
- c = (c - '7') & 0x0f;
- else
- c -= '0';
- num = c << 4;
- c = *p1++;
- if (c > '9')
- c = (c-'7')&0x0f;
- else
- c -= '0';
- num += c;
- *p++ = num;
- } else
- *p++ = *p1++;
- p2 = p;
- while (*p1)
- *p2++ = *p1++;
- *p2 = '\0';
- }
- return s;
-}
--
1.8.2.rc0.22.gb3600c3
There is kernel function to do the job in generic way. Let's use it.
Signed-off-by: Andy Shevchenko <[email protected]>
Cc: Jason Baron <[email protected]>
---
lib/dynamic_debug.c | 48 +++++-------------------------------------------
1 file changed, 5 insertions(+), 43 deletions(-)
diff --git a/lib/dynamic_debug.c b/lib/dynamic_debug.c
index 4603245..99fec3a 100644
--- a/lib/dynamic_debug.c
+++ b/lib/dynamic_debug.c
@@ -24,6 +24,7 @@
#include <linux/sysctl.h>
#include <linux/ctype.h>
#include <linux/string.h>
+#include <linux/string_helpers.h>
#include <linux/uaccess.h>
#include <linux/dynamic_debug.h>
#include <linux/debugfs.h>
@@ -276,47 +277,6 @@ static inline int parse_lineno(const char *str, unsigned int *val)
return 0;
}
-/*
- * Undo octal escaping in a string, inplace. This is useful to
- * allow the user to express a query which matches a format
- * containing embedded spaces.
- */
-static char *unescape(char *str)
-{
- char *in = str;
- char *out = str;
-
- while (*in) {
- if (*in == '\\') {
- if (in[1] == '\\') {
- *out++ = '\\';
- in += 2;
- continue;
- } else if (in[1] == 't') {
- *out++ = '\t';
- in += 2;
- continue;
- } else if (in[1] == 'n') {
- *out++ = '\n';
- in += 2;
- continue;
- } else if (isodigit(in[1]) &&
- isodigit(in[2]) &&
- isodigit(in[3])) {
- *out++ = (((in[1] - '0') << 6) |
- ((in[2] - '0') << 3) |
- (in[3] - '0'));
- in += 4;
- continue;
- }
- }
- *out++ = *in++;
- }
- *out = '\0';
-
- return str;
-}
-
static int check_set(const char **dest, char *src, char *name)
{
int rc = 0;
@@ -370,8 +330,10 @@ static int ddebug_parse_query(char *words[], int nwords,
} else if (!strcmp(words[i], "module")) {
rc = check_set(&query->module, words[i+1], "module");
} else if (!strcmp(words[i], "format")) {
- rc = check_set(&query->format, unescape(words[i+1]),
- "format");
+ string_unescape_inplace(words[i+1], UNESCAPE_SPACE |
+ UNESCAPE_OCTAL |
+ UNESCAPE_SPECIAL);
+ rc = check_set(&query->format, words[i+1], "format");
} else if (!strcmp(words[i], "line")) {
char *first = words[i+1];
char *last = strchr(first, '-');
--
1.8.2.rc0.22.gb3600c3