diff mbox series

[2/5] parse-options: introduce precision handling for `OPTION_INTEGER`

Message ID 20250401-b4-pks-parse-options-integers-v1-2-a628ad40c3b4@pks.im (mailing list archive)
State New
Headers show
Series parse-options: harden handling of integer values | expand

Commit Message

Patrick Steinhardt April 1, 2025, 3:01 p.m. UTC
The `OPTION_INTEGER` option type accepts a signed integer. The type of
the underlying integer is a simple `int`, which restricts the range of
values accepted by such options. But there is a catch: because the
caller provides a pointer to the value via the `.value` field, which is
a simple void pointer. This has two consequences:

  - There is no check whether the passed value is sufficiently long to
    store the entire range of `int`. This can lead to integer wraparound
    in the best case and out-of-bounds writes in the worst case.

  - Even when a caller knows that they want to store a value larger than
    `INT_MAX` they don't have a way to do so.

Funny enough, even if the caller gets everything correct the parsing
logic is still insufficient because we use `strtol()` to parse the
argument, which returns a `long`. But as that value is implicitly cast
when assigning it to the `int` field we may still get invalid results.

In practice this doesn't tend to be a huge issue because users typically
don't end up passing huge values to most commands. But the parsing logic
is demonstrably broken, and it is too easy to get the calling convention
wrong.

Improve the situation by introducing a new `precision` field into the
structure. This field gets assigned automatically by `OPT_INTEGER_F()`
and tracks the size of the passed value. Like this it becomes possible
for the caller to pass arbitrarily-sized integers and the underlying
logic knows to handle it correctly by doing range checks. Furthermore,
convert the code to use `strtoimax()` intstead of `strtol()` so that we
can also parse values larger than `LONG_MAX`.

Note that we do not yet assert signedness of the passed variable, which
is another source of bugs. This will be handled in a subsequent commit.

Signed-off-by: Patrick Steinhardt <ps@pks.im>
---
 builtin/fmt-merge-msg.c       |  2 ++
 builtin/merge.c               |  1 +
 builtin/show-branch.c         |  1 +
 builtin/tag.c                 |  1 +
 parse-options.c               | 53 +++++++++++++++++++++++++++++++------------
 parse-options.h               |  6 +++++
 t/helper/test-parse-options.c |  3 +++
 t/t0040-parse-options.sh      | 23 ++++++++++++++++++-
 8 files changed, 75 insertions(+), 15 deletions(-)

Comments

René Scharfe April 1, 2025, 6:47 p.m. UTC | #1
Am 01.04.25 um 17:01 schrieb Patrick Steinhardt:
> diff --git a/parse-options.c b/parse-options.c
> index 35fbb3b0d63..dbda9b7cfe7 100644
> --- a/parse-options.c
> +++ b/parse-options.c
> @@ -172,25 +172,50 @@ static enum parse_opt_result do_get_value(struct parse_opt_ctx_t *p,
>  			return (*opt->ll_callback)(p, opt, p_arg, p_unset);
>  	}
>  	case OPTION_INTEGER:
> +	{
> +		intmax_t upper_bound = (((intmax_t) 1 << (opt->precision * 8 - 1)) - 1);

Ugh, how does this not overflow?  The macro maximum_signed_value_of_type
does a similar calculation better.

> +		intmax_t lower_bound = -upper_bound - 1;

This depends on two's complement being used, which is bad for purity and
portability to obsolete machines, but probably OK in practice.

> +		intmax_t value;
> +
>  		if (unset) {
> -			*(int *)opt->value = 0;
> -			return 0;
> -		}
> -		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
> -			*(int *)opt->value = opt->defval;
> -			return 0;
> -		}
> -		if (get_arg(p, opt, flags, &arg))
> +			value = 0;
> +		} else if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
> +			value = opt->defval;
> +		} else if (get_arg(p, opt, flags, &arg)) {
>  			return -1;
> -		if (!*arg)
> +		} else if (!*arg) {
>  			return error(_("%s expects a numerical value"),
>  				     optname(opt, flags));
> -		*(int *)opt->value = strtol(arg, (char **)&s, 10);
> -		if (*s)
> -			return error(_("%s expects a numerical value"),
> -				     optname(opt, flags));
> -		return 0;
> +		} else {
> +			value = strtoimax(arg, (char **)&s, 10);
> +			if (*s)
> +				return error(_("%s expects a numerical value"),
> +					     optname(opt, flags));
> +
> +		}
>
> +		if (value < lower_bound || value > upper_bound)
> +			return error(_("value %"PRIdMAX" for %s not in range [%"PRIdMAX",%"PRIdMAX"]"),
> +				     value, optname(opt, flags), lower_bound, upper_bound);
> +
> +		switch (opt->precision) {
> +		case 1:
> +			*(int8_t *)opt->value = value;
> +			return 0;
> +		case 2:
> +			*(int16_t *)opt->value = value;
> +			return 0;
> +		case 4:
> +			*(int32_t *)opt->value = value;
> +			return 0;
> +		case 8:
> +			*(int64_t *)opt->value = value;
> +			return 0;

Do we even need all these sizes?  Or can we whittle it down to ssize_t?

And for which quantities do we need to accept negative values anyway?

René
diff mbox series

Patch

diff --git a/builtin/fmt-merge-msg.c b/builtin/fmt-merge-msg.c
index 240cdb474bc..3b6aac2cf7f 100644
--- a/builtin/fmt-merge-msg.c
+++ b/builtin/fmt-merge-msg.c
@@ -24,6 +24,7 @@  int cmd_fmt_merge_msg(int argc,
 			.type = OPTION_INTEGER,
 			.long_name = "log",
 			.value = &shortlog_len,
+			.precision = sizeof(shortlog_len),
 			.argh = N_("n"),
 			.help = N_("populate log with at most <n> entries from shortlog"),
 			.flags = PARSE_OPT_OPTARG,
@@ -33,6 +34,7 @@  int cmd_fmt_merge_msg(int argc,
 			.type = OPTION_INTEGER,
 			.long_name = "summary",
 			.value = &shortlog_len,
+			.precision = sizeof(shortlog_len),
 			.argh = N_("n"),
 			.help = N_("alias for --log (deprecated)"),
 			.flags = PARSE_OPT_OPTARG | PARSE_OPT_HIDDEN,
diff --git a/builtin/merge.c b/builtin/merge.c
index 21787d45165..9ab10c7db0a 100644
--- a/builtin/merge.c
+++ b/builtin/merge.c
@@ -254,6 +254,7 @@  static struct option builtin_merge_options[] = {
 		.type = OPTION_INTEGER,
 		.long_name = "log",
 		.value = &shortlog_len,
+		.precision = sizeof(shortlog_len),
 		.argh = N_("n"),
 		.help = N_("add (at most <n>) entries from shortlog to merge commit message"),
 		.flags = PARSE_OPT_OPTARG,
diff --git a/builtin/show-branch.c b/builtin/show-branch.c
index dab37019d29..b549d8c3f5b 100644
--- a/builtin/show-branch.c
+++ b/builtin/show-branch.c
@@ -671,6 +671,7 @@  int cmd_show_branch(int ac,
 			.type = OPTION_INTEGER,
 			.long_name = "more",
 			.value = &extra,
+			.precision = sizeof(extra),
 			.argh = N_("n"),
 			.help = N_("show <n> more commits after the common ancestor"),
 			.flags = PARSE_OPT_OPTARG,
diff --git a/builtin/tag.c b/builtin/tag.c
index b266f12bb48..7597d93c71b 100644
--- a/builtin/tag.c
+++ b/builtin/tag.c
@@ -483,6 +483,7 @@  int cmd_tag(int argc,
 			.type = OPTION_INTEGER,
 			.short_name = 'n',
 			.value = &filter.lines,
+			.precision = sizeof(filter.lines),
 			.argh = N_("n"),
 			.help = N_("print <n> lines of each tag message"),
 			.flags = PARSE_OPT_OPTARG,
diff --git a/parse-options.c b/parse-options.c
index 35fbb3b0d63..dbda9b7cfe7 100644
--- a/parse-options.c
+++ b/parse-options.c
@@ -172,25 +172,50 @@  static enum parse_opt_result do_get_value(struct parse_opt_ctx_t *p,
 			return (*opt->ll_callback)(p, opt, p_arg, p_unset);
 	}
 	case OPTION_INTEGER:
+	{
+		intmax_t upper_bound = (((intmax_t) 1 << (opt->precision * 8 - 1)) - 1);
+		intmax_t lower_bound = -upper_bound - 1;
+		intmax_t value;
+
 		if (unset) {
-			*(int *)opt->value = 0;
-			return 0;
-		}
-		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
-			*(int *)opt->value = opt->defval;
-			return 0;
-		}
-		if (get_arg(p, opt, flags, &arg))
+			value = 0;
+		} else if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+			value = opt->defval;
+		} else if (get_arg(p, opt, flags, &arg)) {
 			return -1;
-		if (!*arg)
+		} else if (!*arg) {
 			return error(_("%s expects a numerical value"),
 				     optname(opt, flags));
-		*(int *)opt->value = strtol(arg, (char **)&s, 10);
-		if (*s)
-			return error(_("%s expects a numerical value"),
-				     optname(opt, flags));
-		return 0;
+		} else {
+			value = strtoimax(arg, (char **)&s, 10);
+			if (*s)
+				return error(_("%s expects a numerical value"),
+					     optname(opt, flags));
+
+		}
 
+		if (value < lower_bound || value > upper_bound)
+			return error(_("value %"PRIdMAX" for %s not in range [%"PRIdMAX",%"PRIdMAX"]"),
+				     value, optname(opt, flags), lower_bound, upper_bound);
+
+		switch (opt->precision) {
+		case 1:
+			*(int8_t *)opt->value = value;
+			return 0;
+		case 2:
+			*(int16_t *)opt->value = value;
+			return 0;
+		case 4:
+			*(int32_t *)opt->value = value;
+			return 0;
+		case 8:
+			*(int64_t *)opt->value = value;
+			return 0;
+		default:
+			BUG("invalid precision for option %s",
+			    optname(opt, flags));
+		}
+	}
 	case OPTION_MAGNITUDE:
 		if (unset) {
 			*(unsigned long *)opt->value = 0;
diff --git a/parse-options.h b/parse-options.h
index 997ffbee805..8d5f9c95f9c 100644
--- a/parse-options.h
+++ b/parse-options.h
@@ -92,6 +92,10 @@  typedef int parse_opt_subcommand_fn(int argc, const char **argv,
  * `value`::
  *   stores pointers to the values to be filled.
  *
+ * `precision`::
+ *   precision of the integer pointed to by `value`. Should typically be its
+ *   `sizeof()`.
+ *
  * `argh`::
  *   token to explain the kind of argument this option wants. Does not
  *   begin in capital letter, and does not end with a full stop.
@@ -151,6 +155,7 @@  struct option {
 	int short_name;
 	const char *long_name;
 	void *value;
+	size_t precision;
 	const char *argh;
 	const char *help;
 
@@ -214,6 +219,7 @@  struct option {
 	.short_name = (s), \
 	.long_name = (l), \
 	.value = (v), \
+	.precision = sizeof(*v), \
 	.argh = N_("n"), \
 	.help = (h), \
 	.flags = (f), \
diff --git a/t/helper/test-parse-options.c b/t/helper/test-parse-options.c
index 997f55fd45b..b1275dfade4 100644
--- a/t/helper/test-parse-options.c
+++ b/t/helper/test-parse-options.c
@@ -120,6 +120,7 @@  int cmd__parse_options(int argc, const char **argv)
 	};
 	struct string_list expect = STRING_LIST_INIT_NODUP;
 	struct string_list list = STRING_LIST_INIT_NODUP;
+	int16_t i16 = 0;
 
 	struct option options[] = {
 		OPT_BOOL(0, "yes", &boolean, "get a boolean"),
@@ -139,6 +140,7 @@  int cmd__parse_options(int argc, const char **argv)
 		OPT_NEGBIT(0, "neg-or4", &boolean, "same as --no-or4", 4),
 		OPT_GROUP(""),
 		OPT_INTEGER('i', "integer", &integer, "get a integer"),
+		OPT_INTEGER(0, "i16", &i16, "get a 16 bit integer"),
 		OPT_INTEGER('j', NULL, &integer, "get a integer, too"),
 		OPT_MAGNITUDE('m', "magnitude", &magnitude, "get a magnitude"),
 		OPT_SET_INT(0, "set23", &integer, "set integer to 23", 23),
@@ -210,6 +212,7 @@  int cmd__parse_options(int argc, const char **argv)
 	}
 	show(&expect, &ret, "boolean: %d", boolean);
 	show(&expect, &ret, "integer: %d", integer);
+	show(&expect, &ret, "i16: %"PRIdMAX, (intmax_t) i16);
 	show(&expect, &ret, "magnitude: %lu", magnitude);
 	show(&expect, &ret, "timestamp: %"PRItime, timestamp);
 	show(&expect, &ret, "string: %s", string ? string : "(not set)");
diff --git a/t/t0040-parse-options.sh b/t/t0040-parse-options.sh
index 2fe3522305f..e3ca7a27738 100755
--- a/t/t0040-parse-options.sh
+++ b/t/t0040-parse-options.sh
@@ -22,6 +22,7 @@  usage: test-tool parse-options <options>
 
     -i, --[no-]integer <n>
                           get a integer
+    --[no-]i16 <n>        get a 16 bit integer
     -j <n>                get a integer, too
     -m, --magnitude <n>   get a magnitude
     --[no-]set23          set integer to 23
@@ -136,6 +137,7 @@  test_expect_success 'OPT_MAGNITUDE() 3giga' '
 cat >expect <<\EOF
 boolean: 2
 integer: 1729
+i16: 0
 magnitude: 16384
 timestamp: 0
 string: 123
@@ -156,6 +158,7 @@  test_expect_success 'short options' '
 cat >expect <<\EOF
 boolean: 2
 integer: 1729
+i16: 9000
 magnitude: 16384
 timestamp: 0
 string: 321
@@ -167,7 +170,7 @@  file: prefix/fi.le
 EOF
 
 test_expect_success 'long options' '
-	test-tool parse-options --boolean --integer 1729 --magnitude 16k \
+	test-tool parse-options --boolean --integer 1729 --i16 9000 --magnitude 16k \
 		--boolean --string2=321 --verbose --verbose --no-dry-run \
 		--abbrev=10 --file fi.le --obsolete \
 		>output 2>output.err &&
@@ -179,6 +182,7 @@  test_expect_success 'abbreviate to something longer than SHA1 length' '
 	cat >expect <<-EOF &&
 	boolean: 0
 	integer: 0
+	i16: 0
 	magnitude: 0
 	timestamp: 0
 	string: (not set)
@@ -253,6 +257,7 @@  test_expect_success 'superfluous value provided: cmdmode' '
 cat >expect <<\EOF
 boolean: 1
 integer: 13
+i16: 0
 magnitude: 0
 timestamp: 0
 string: 123
@@ -276,6 +281,7 @@  test_expect_success 'intermingled arguments' '
 cat >expect <<\EOF
 boolean: 0
 integer: 2
+i16: 0
 magnitude: 0
 timestamp: 0
 string: (not set)
@@ -343,6 +349,7 @@  cat >expect <<\EOF
 Callback: "four", 0
 boolean: 5
 integer: 4
+i16: 0
 magnitude: 0
 timestamp: 0
 string: (not set)
@@ -368,6 +375,7 @@  test_expect_success 'OPT_CALLBACK() and callback errors work' '
 cat >expect <<\EOF
 boolean: 1
 integer: 23
+i16: 0
 magnitude: 0
 timestamp: 0
 string: (not set)
@@ -447,6 +455,7 @@  test_expect_success 'OPT_NUMBER_CALLBACK() works' '
 cat >expect <<\EOF
 boolean: 0
 integer: 0
+i16: 0
 magnitude: 0
 timestamp: 0
 string: (not set)
@@ -783,4 +792,16 @@  test_expect_success 'magnitude with units but no numbers' '
 	test_must_be_empty out
 '
 
+test_expect_success 'i16 limits range' '
+	test-tool parse-options --i16 32767 >out &&
+	test_grep "i16: 32767" out &&
+	test_must_fail test-tool parse-options --i16 32768 2>err &&
+	test_grep "value 32768 for option .i16. not in range \[-32768,32767\]" err &&
+
+	test-tool parse-options --i16 -32768 >out &&
+	test_grep "i16: -32768" out &&
+	test_must_fail test-tool parse-options --i16 -32769 2>err &&
+	test_grep "value -32769 for option .i16. not in range \[-32768,32767\]" err
+'
+
 test_done