bpf/selftests: improve arg parsing in test_verifier
Subsystems (3)
PATCH 1/1 bpf/selftests: improve arg parsing in test_verifier
Current test_verifier provides little feedback or argument validation, instead silently falling back to running all tests in case of user error or even expected use cases. Trying to do manual exploratory testing, switching between kernel versions (e.g. with varying tests), or working around problematic tests (e.g. kernel hangs/crashes) can be a frustrating experience.
Rework argument parsing to be more robust and strict, and provide basic help on errors. Clamp test ranges to valid values and add an option to list available built-in tests ("-l"). Default "test_verifier" behaviour (run all tests) is unchanged and backwards-compatible. Updated examples:
$ test_verifier die die die # previously ran all tests
Usage: test_verifier -l | [-v|-vv] [<tst_lo> [<tst_hi>]]$ test_verifier 700 9999 # runs test subset from 700 to end
--- tools/testing/selftests/bpf/test_verifier.c | 54 ++++++++++++--------- 1 file changed, 30 insertions(+), 24 deletions(-)
tools/testing/selftests/bpf/test_verifier.c +30 -24
@@ -10,9 +10,11 @@#include <endian.h> #include <asm/types.h> #include <linux/types.h> +#include <linux/minmax.h> #include <stdint.h> #include <stdio.h> #include <stdlib.h> +#include <ctype.h> #include <unistd.h> #include <errno.h> #include <string.h>
@@ -1848,36 +1850,40 @@ int main(int argc, char **argv) {unsigned int from =0 , to =ARRAY_SIZE (tests);bool unpriv = !is_admin (); -int arg =1 ; - -if (argc >1 &&strcmp (argv[1 ],"-v" ) ==0 ) { +int i, arg =1 ; + +while (argc >1 && *argv[arg] =='-' ) { +if (strcmp (argv[arg],"-l" ) ==0 ) { +for (i = from; i < to; i++) +printf ("#%d %s\n" , i, tests[i].descr); +return EXIT_SUCCESS ; + }else if (strcmp (argv[arg],"-v" ) ==0 ) { + verbose = true; + verif_log_level =1 ; + }else if (strcmp (argv[arg],"-vv" ) ==0 ) { + verbose = true; + verif_log_level =2 ; + }else + goto out_help; arg++; - verbose = true; - verif_log_level =1 ; argc--; } -if (argc >1 &&strcmp (argv[1 ],"-vv" ) ==0 ) { - arg++; - verbose = true; - verif_log_level =2 ; - argc--; - } - -if (argc ==3 ) { -unsigned int l =atoi (argv[arg]); -unsigned int u =atoi (argv[arg +1 ]); -if (l < to && u < to) { - from = l; - to = u +1 ; - } - }else if (argc ==2 ) { -unsigned int t =atoi (argv[arg]); +for (i =1 ; i <=2 && argc >1 ; i++, arg++, argc--) { +unsigned int t =min (atoi (argv[arg]),ARRAY_SIZE (tests) -1 ); -if (t < to) { +if (!isdigit (*argv[arg])) + goto out_help; +if (i ==1 ) from = t; - to = t +1 ; - } + to = t +1 ; + } + +if (argc >1 ) { +out_help: +printf ("Usage: %s -l | [-v|-vv] [<tst_lo> [<tst_hi>]]\n" , + argv[0 ]); +return EXIT_FAILURE ; } unpriv_disabled =get_unpriv_disabled ();
Signature · 1 line
2.34.1
Quoted 27 lines · Tony Ambardar
Current
test_verifierprovides little feedback or argument validation, instead silently falling back to running all tests in case of user error or even expected use cases. Trying to do manual exploratory testing, switching between kernel versions (e.g. with varying tests), or working around problematic tests (e.g. kernel hangs/crashes) can be a frustrating experience.Rework argument parsing to be more robust and strict, and provide basic help on errors. Clamp test ranges to valid values and add an option to list available built-in tests ("-l"). Default "
test_verifier" behaviour (run all tests) is unchanged and backwards-compatible. Updated examples:$ test_verifier die die die # previously ran all tests Usage: test_verifier -l | [-v|-vv] [<tst_lo> [<tst_hi>]]$ test_verifier 700 9999 # runs test subset from 700 to endSigned-off-by Tony Ambardar <[email protected]>--- tools/testing/selftests/bpf/test_verifier.c | 54 ++++++++++++--------- 1 file changed, 30 insertions(+), 24 deletions(-)tools/testing/selftests/bpf/test_verifier.c +0
-0
tools/testing/selftests/bpf/test_verifier.c line 13
Quoted 4 lines
tools/testing/selftests/bpf/test_verifier.c +1
-0@@ -10,9 +10,11 @@#include <endian.h> #include <asm/types.h> #include <linux/types.h> +#include <linux/minmax.h>
this fails to compile
BINARY test_verifier
test_verifier.c:13:10: fatal error: linux/minmax.h: No such file or directory
13 | #include <linux/minmax.h>
| ^~~~~~~~~~~~~~~~
looks like you could use perhaps <linux/kernel.h> instead?
jirka
Quoted 76 lines
tools/testing/selftests/bpf/test_verifier.c +29
-24#include <stdint.h> #include <stdio.h> #include <stdlib.h> +#include <ctype.h> #include <unistd.h> #include <errno.h> #include <string.h> @@ -1848,36 +1850,40 @@ int main(int argc, char **argv) {unsigned int from =0 , to =ARRAY_SIZE (tests);bool unpriv = !is_admin (); -int arg =1 ; - -if (argc >1 &&strcmp (argv[1 ],"-v" ) ==0 ) { +int i, arg =1 ; + +while (argc >1 && *argv[arg] =='-' ) { +if (strcmp (argv[arg],"-l" ) ==0 ) { +for (i = from; i < to; i++) +printf ("#%d %s\n" , i, tests[i].descr); +return EXIT_SUCCESS ; + }else if (strcmp (argv[arg],"-v" ) ==0 ) { + verbose = true; + verif_log_level =1 ; + }else if (strcmp (argv[arg],"-vv" ) ==0 ) { + verbose = true; + verif_log_level =2 ; + }else + goto out_help; arg++; - verbose = true; - verif_log_level =1 ; argc--; } -if (argc >1 &&strcmp (argv[1 ],"-vv" ) ==0 ) { - arg++; - verbose = true; - verif_log_level =2 ; - argc--; - } - -if (argc ==3 ) { -unsigned int l =atoi (argv[arg]); -unsigned int u =atoi (argv[arg +1 ]); -if (l < to && u < to) { - from = l; - to = u +1 ; - } - }else if (argc ==2 ) { -unsigned int t =atoi (argv[arg]); +for (i =1 ; i <=2 && argc >1 ; i++, arg++, argc--) { +unsigned int t =min (atoi (argv[arg]),ARRAY_SIZE (tests) -1 ); -if (t < to) { +if (!isdigit (*argv[arg])) + goto out_help; +if (i ==1 ) from = t; - to = t +1 ; - } + to = t +1 ; + } + +if (argc >1 ) { +out_help: +printf ("Usage: %s -l | [-v|-vv] [<tst_lo> [<tst_hi>]]\n" , + argv[0 ]); +return EXIT_FAILURE ; } unpriv_disabled =get_unpriv_disabled ();Signature · 1 line
2.34.1
On Mon, Sep 25, 2023 at 04:37:02PM -0700, Tony Ambardar wrote:
SNIP
Quoted 36 lines
tools/testing/selftests/bpf/test_verifier.c +15
-14@@ -1848,36 +1850,40 @@ int main(int argc, char **argv) {unsigned int from =0 , to =ARRAY_SIZE (tests);bool unpriv = !is_admin (); -int arg =1 ; - -if (argc >1 &&strcmp (argv[1 ],"-v" ) ==0 ) { +int i, arg =1 ; + +while (argc >1 && *argv[arg] =='-' ) { +if (strcmp (argv[arg],"-l" ) ==0 ) { +for (i = from; i < to; i++) +printf ("#%d %s\n" , i, tests[i].descr); +return EXIT_SUCCESS ; + }else if (strcmp (argv[arg],"-v" ) ==0 ) { + verbose = true; + verif_log_level =1 ; + }else if (strcmp (argv[arg],"-vv" ) ==0 ) { + verbose = true; + verif_log_level =2 ; + }else + goto out_help; arg++; - verbose = true; - verif_log_level =1 ; argc--; } -if (argc >1 &&strcmp (argv[1 ],"-vv" ) ==0 ) { - arg++; - verbose = true; - verif_log_level =2 ; - argc--; - } - -if (argc ==3 ) { -unsigned int l =atoi (argv[arg]);
tools/testing/selftests/bpf/test_verifier.c int main(int argc, char **argv) · line 1873
Quoted 10 lines
tools/testing/selftests/bpf/test_verifier.c +2
-7-unsigned int u =atoi (argv[arg +1 ]); -if (l < to && u < to) { - from = l; - to = u +1 ; - } - }else if (argc ==2 ) { -unsigned int t =atoi (argv[arg]); +for (i =1 ; i <=2 && argc >1 ; i++, arg++, argc--) { +unsigned int t =min (atoi (argv[arg]),ARRAY_SIZE (tests) -1 );
this looks like unnecessary loop, the code before is easy to understand, could we just do the args check on isdigit and valid index value in there?
jirka
Quoted 23 lines
tools/testing/selftests/bpf/test_verifier.c +11
-3-if (t < to) { +if (!isdigit (*argv[arg])) + goto out_help; +if (i ==1 ) from = t; - to = t +1 ; - } + to = t +1 ; + } + +if (argc >1 ) { +out_help: +printf ("Usage: %s -l | [-v|-vv] [<tst_lo> [<tst_hi>]]\n" , + argv[0 ]); +return EXIT_FAILURE ; } unpriv_disabled =get_unpriv_disabled ();Signature · 1 line
2.34.1