On Thu, Jun 6, 2024 at 6:30 AM Cupertino Miranda
<cupertino.miranda@xxxxxxxxxx> wrote:
>
> Add support for __regex and __regex_unpriv macros to check the test
> execution output against a regular expression. This is similar to __msg
> and __msg_unpriv, however those expect full text matching.
>
> Signed-off-by: Cupertino Miranda <cupertino.miranda@xxxxxxxxxx>
> Cc: jose.marchesi@xxxxxxxxxx
> Cc: david.faust@xxxxxxxxxx
> Cc: Yonghong Song <yonghong.song@xxxxxxxxx>
> Cc: Eduard Zingerman <eddyz87@xxxxxxxxx>
> Cc: Andrii Nakryiko <andrii.nakryiko@xxxxxxxxx>
> ---
> tools/testing/selftests/bpf/progs/bpf_misc.h | 11 +-
> tools/testing/selftests/bpf/test_loader.c | 143 +++++++++++++++----
> 2 files changed, 121 insertions(+), 33 deletions(-)
>
> diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
> index fb2f5513e29e..c0280bd2f340 100644
> --- a/tools/testing/selftests/bpf/progs/bpf_misc.h
> +++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
> @@ -7,9 +7,9 @@
> *
> * The test_loader sequentially loads each program in a skeleton.
> * Programs could be loaded in privileged and unprivileged modes.
> - * - __success, __failure, __msg imply privileged mode;
> - * - __success_unpriv, __failure_unpriv, __msg_unpriv imply
> - * unprivileged mode.
> + * - __success, __failure, __msg, __regex imply privileged mode;
> + * - __success_unpriv, __failure_unpriv, __msg_unpriv, __regex_unpriv
> + * imply unprivileged mode.
> * If combination of privileged and unprivileged attributes is present
> * both modes are used. If none are present privileged mode is implied.
> *
> @@ -24,6 +24,9 @@
> * Multiple __msg attributes could be specified.
> * __msg_unpriv Same as __msg but for unprivileged mode.
> *
> + * __regex Same as __msg, but using a regular expression.
> + * __regex_unpriv Same as __msg_unpriv but using a regular expression.
> + *
> * __success Expect program load success in privileged mode.
> * __success_unpriv Expect program load success in unprivileged mode.
> *
> @@ -59,10 +62,12 @@
> * __auxiliary_unpriv Same, but load program in unprivileged mode.
> */
> #define __msg(msg) __attribute__((btf_decl_tag("comment:test_expect_msg=" msg)))
> +#define __regex(regex) __attribute__((btf_decl_tag("comment:test_expect_regex=" regex)))
> #define __failure __attribute__((btf_decl_tag("comment:test_expect_failure")))
> #define __success __attribute__((btf_decl_tag("comment:test_expect_success")))
> #define __description(desc) __attribute__((btf_decl_tag("comment:test_description=" desc)))
> #define __msg_unpriv(msg) __attribute__((btf_decl_tag("comment:test_expect_msg_unpriv=" msg)))
> +#define __regex_unpriv(regex) __attribute__((btf_decl_tag("comment:test_expect_regex_unpriv=" regex)))
> #define __failure_unpriv __attribute__((btf_decl_tag("comment:test_expect_failure_unpriv")))
> #define __success_unpriv __attribute__((btf_decl_tag("comment:test_expect_success_unpriv")))
> #define __log_level(lvl) __attribute__((btf_decl_tag("comment:test_log_level="#lvl)))
> diff --git a/tools/testing/selftests/bpf/test_loader.c b/tools/testing/selftests/bpf/test_loader.c
> index 524c38e9cde4..a9a7f5f55855 100644
> --- a/tools/testing/selftests/bpf/test_loader.c
> +++ b/tools/testing/selftests/bpf/test_loader.c
> @@ -2,6 +2,7 @@
> /* Copyright (c) 2022 Meta Platforms, Inc. and affiliates. */
> #include <linux/capability.h>
> #include <stdlib.h>
> +#include <regex.h>
> #include <test_progs.h>
> #include <bpf/btf.h>
>
> @@ -17,9 +18,11 @@
> #define TEST_TAG_EXPECT_FAILURE "comment:test_expect_failure"
> #define TEST_TAG_EXPECT_SUCCESS "comment:test_expect_success"
> #define TEST_TAG_EXPECT_MSG_PFX "comment:test_expect_msg="
> +#define TEST_TAG_EXPECT_REGEX_PFX "comment:test_expect_regex="
> #define TEST_TAG_EXPECT_FAILURE_UNPRIV "comment:test_expect_failure_unpriv"
> #define TEST_TAG_EXPECT_SUCCESS_UNPRIV "comment:test_expect_success_unpriv"
> #define TEST_TAG_EXPECT_MSG_PFX_UNPRIV "comment:test_expect_msg_unpriv="
> +#define TEST_TAG_EXPECT_REGEX_PFX_UNPRIV "comment:test_expect_regex_unpriv="
> #define TEST_TAG_LOG_LEVEL_PFX "comment:test_log_level="
> #define TEST_TAG_PROG_FLAGS_PFX "comment:test_prog_flags="
> #define TEST_TAG_DESCRIPTION_PFX "comment:test_description="
> @@ -46,10 +49,26 @@ enum mode {
> UNPRIV = 2
> };
>
> +enum message_type {
> + SUBSTRING = 0,
> + REGEX
> +};
> +struct expect_msg {
> + union {
> + const char *substring;
> + struct {
> + const char *expr;
> + regex_t regex;
> + bool failed_to_compile;
you are not even using this field
> + } regex;
> + };
> + enum message_type type;
I think this is an overkill. We are unlikely to have third type of
"match message" spec, so let's do just simple:
struct expect_msg {
const char *substr; /* substring match */
const char *regex_str; /* regex-based match */
regex_t regex;
};
No unions, no nesting. substr is NULL for regex, otherwise regex_str
is NULL (for substring case). Simple.
> +};
> +
> struct test_subspec {
> char *name;
> bool expect_failure;
> - const char **expect_msgs;
> + struct expect_msg *expect_msg;
it's still messages (plural), why renaming the field?
> size_t expect_msg_cnt;
> int retval;
> bool execute;
> @@ -89,28 +108,58 @@ void test_loader_fini(struct test_loader *tester)
>
> static void free_test_spec(struct test_spec *spec)
> {
> + int i;
> +
> + /* Delalocate regex from expect_msg array. */
typo: deallocate
> + for (i = 0; i < spec->priv.expect_msg_cnt; i++)
> + if (spec->priv.expect_msg[i].type == REGEX)
> + regfree(&spec->priv.expect_msg[i].regex.regex);
and then:
if (spec->priv.expect_msg[i].regex_str)
regfree(...);
Also, where is unpriv handling?
> +
> free(spec->priv.name);
> free(spec->unpriv.name);
> - free(spec->priv.expect_msgs);
> - free(spec->unpriv.expect_msgs);
> + free(spec->priv.expect_msg);
> + free(spec->unpriv.expect_msg);
>
> spec->priv.name = NULL;
> spec->unpriv.name = NULL;
> - spec->priv.expect_msgs = NULL;
> - spec->unpriv.expect_msgs = NULL;
> + spec->priv.expect_msg = NULL;
> + spec->unpriv.expect_msg = NULL;
> }
>
> -static int push_msg(const char *msg, struct test_subspec *subspec)
> +static int push_msg(const char *match, enum message_type msg_type, struct test_subspec *subspec)
here we can just pass two mutually exclusive strings to specify what case it is:
static int push_msg(const char *substr, const char *regex_str, struct
test_subspec *subspec)
> {
> void *tmp;
> + int regcomp_res;
> + char error_msg[100];
> + struct expect_msg *em;
we don't need both tmp and em, let's keep just
struct expect_msg *msg;
>
> - tmp = realloc(subspec->expect_msgs, (1 + subspec->expect_msg_cnt) * sizeof(void *));
> + tmp = realloc(subspec->expect_msg,
> + (1 + subspec->expect_msg_cnt) * sizeof(struct expect_msg));
then `msg = realloc(..)` here
> if (!tmp) {
> ASSERT_FAIL("failed to realloc memory for messages\n");
> return -ENOMEM;
> }
> - subspec->expect_msgs = tmp;
> - subspec->expect_msgs[subspec->expect_msg_cnt++] = msg;
> + subspec->expect_msg = tmp;
> + em = &subspec->expect_msg[subspec->expect_msg_cnt];
and reuse msg = &subspec->... here after assigning `subspec->expected_msg = msg`
> + subspec->expect_msg_cnt += 1;
> +
> + em->type = msg_type;
> + switch (msg_type) {
> + case SUBSTRING:
> + em->substring = match;
> + break;
> + case REGEX:
> + em->regex.expr = match;
> + regcomp_res = regcomp(&em->regex.regex, match, REG_EXTENDED|REG_NEWLINE);
> + if (regcomp_res != 0) {
> + regerror(regcomp_res, &em->regex.regex, error_msg, 100);
> + fprintf(stderr, "Regexp compilation error in '%s': '%s'\n",
> + match, error_msg);
> + ASSERT_FAIL("failed to compile regex\n");
> + return -EINVAL;
> + }
> + break;
> + }
and then
if (substr) {
msg->substr = substr;
} else {
char err_msg[100];
int err;
err = regcomp(...);
if (err) {
...
}
}
Note I moved err_msg/err closer to the only place where they are being
used (and kept names less verbose).
>
> return 0;
> }
> @@ -233,13 +282,25 @@ static int parse_test_spec(struct test_loader *tester,
> spec->mode_mask |= UNPRIV;
> } else if (str_has_pfx(s, TEST_TAG_EXPECT_MSG_PFX)) {
> msg = s + sizeof(TEST_TAG_EXPECT_MSG_PFX) - 1;
> - err = push_msg(msg, &spec->priv);
> + err = push_msg(msg, SUBSTRING, &spec->priv);
> if (err)
> goto cleanup;
> spec->mode_mask |= PRIV;
> } else if (str_has_pfx(s, TEST_TAG_EXPECT_MSG_PFX_UNPRIV)) {
> msg = s + sizeof(TEST_TAG_EXPECT_MSG_PFX_UNPRIV) - 1;
> - err = push_msg(msg, &spec->unpriv);
> + err = push_msg(msg, SUBSTRING, &spec->unpriv);
> + if (err)
> + goto cleanup;
> + spec->mode_mask |= UNPRIV;
> + } else if (str_has_pfx(s, TEST_TAG_EXPECT_REGEX_PFX)) {
> + msg = s + sizeof(TEST_TAG_EXPECT_REGEX_PFX) - 1;
> + err = push_msg(msg, REGEX, &spec->priv);
> + if (err)
> + goto cleanup;
> + spec->mode_mask |= PRIV;
> + } else if (str_has_pfx(s, TEST_TAG_EXPECT_REGEX_PFX_UNPRIV)) {
> + msg = s + sizeof(TEST_TAG_EXPECT_REGEX_PFX_UNPRIV) - 1;
> + err = push_msg(msg, REGEX, &spec->unpriv);
> if (err)
> goto cleanup;
> spec->mode_mask |= UNPRIV;
> @@ -336,16 +397,16 @@ static int parse_test_spec(struct test_loader *tester,
> spec->unpriv.execute = spec->priv.execute;
> }
>
> - if (!spec->unpriv.expect_msgs) {
> - size_t sz = spec->priv.expect_msg_cnt * sizeof(void *);
> + if (!spec->unpriv.expect_msg) {
> + size_t sz = spec->priv.expect_msg_cnt * sizeof(struct expect_msg);
>
> - spec->unpriv.expect_msgs = malloc(sz);
> - if (!spec->unpriv.expect_msgs) {
> - PRINT_FAIL("failed to allocate memory for unpriv.expect_msgs\n");
> + spec->unpriv.expect_msg = malloc(sz);
> + if (!spec->unpriv.expect_msg) {
> + PRINT_FAIL("failed to allocate memory for unpriv.expect\n");
> err = -ENOMEM;
> goto cleanup;
> }
> - memcpy(spec->unpriv.expect_msgs, spec->priv.expect_msgs, sz);
> + memcpy(spec->unpriv.expect_msg, spec->priv.expect_msg, sz);
hmm.. this is problematic given regcomp() and regfree(). We don't know
what sort of internal state libc will maintain in regex_t, so it's
unsafe to just memcpy() priv into unpriv like this. We probably need
to refactor this into an explicit push_msg() calls for spec->unpriv,
which is, luckily, trivial to do.
> spec->unpriv.expect_msg_cnt = spec->priv.expect_msg_cnt;
> }
> }
> @@ -402,27 +463,49 @@ static void validate_case(struct test_loader *tester,
> struct bpf_program *prog,
> int load_err)
> {
> - int i, j;
> + int i, j, reg_error;
"err" is totally fine, no need to be so verbose here
> + char *match;
> + regmatch_t reg_match[1];
>
> for (i = 0; i < subspec->expect_msg_cnt; i++) {
> - char *match;
> - const char *expect_msg;
> -
> - expect_msg = subspec->expect_msgs[i];
> + struct expect_msg *em = &subspec->expect_msg[i];
msg to say consistent?
> +
> + match = NULL;
> + switch (em->type) {
> + case SUBSTRING:
> + match = strstr(tester->log_buf + tester->next_match_pos, em->substring);
> + tester->next_match_pos = match - tester->log_buf + strlen(em->substring);
match will be NULL if strstr() doesn't find substring
> + break;
> + case REGEX:
> + reg_error = regexec(&em->regex.regex,
> + tester->log_buf + tester->next_match_pos,
> + 1, reg_match, 0);
> + if (reg_error == 0)
> + match = tester->log_buf + tester->next_match_pos + reg_match[0].rm_so;
> + tester->next_match_pos += reg_match[0].rm_eo;
Similarly, if we didn't find a match, we should not update position
> + break;
> + }
>
> - match = strstr(tester->log_buf + tester->next_match_pos, expect_msg);
> if (!ASSERT_OK_PTR(match, "expect_msg")) {
> - /* if we are in verbose mode, we've already emitted log */
> if (env.verbosity == VERBOSE_NONE)
> emit_verifier_log(tester->log_buf, true /*force*/);
> - for (j = 0; j < i; j++)
> - fprintf(stderr,
> - "MATCHED MSG: '%s'\n", subspec->expect_msgs[j]);
> - fprintf(stderr, "EXPECTED MSG: '%s'\n", expect_msg);
> + for (j = 0; j <= i; j++) {
> + const char *header = (j < i) ? "MATCHED" : "EXPECTED";
> + struct expect_msg *tmp = &subspec->expect_msg[j];
reuse msg variable?
> +
> + switch (tmp->type) {
> + case SUBSTRING:
> + fprintf(stderr,
> + "%s MSG: '%s'\n", header, tmp->substring);
> + break;
> + case REGEX:
> + fprintf(stderr,
> + "%s REGEX: '%s'\n", header, tmp->regex.expr);
> + break;
> + }
it's probably better to have just one fprintf handling everything, it
will be easier to follow
fprintf(stderr, "%s %s: '%s'\n",
j < i ? "MATCHED " : "EXPECTED",
msg->substr ? "SUBSTR" : " REGEX",
msg->substr ?: msg->regex_str);
(note spaces for alignment)
pw-bot: cr
> + }
> return;
> }
> -
> - tester->next_match_pos = match - tester->log_buf + strlen(expect_msg);
> }
> }
>
> --
> 2.39.2
>
