From: Sean Christopherson <sean.j.christopherson@intel.com>
To: Bill Wendling <morbo@google.com>
Cc: kvm@vger.kernel.org, jmattson@google.com
Subject: Re: [kvm-unit-tests PATCH] x86: setjmp: check expected value of "i" to give better feedback
Date: Thu, 19 Sep 2019 13:39:19 -0700 [thread overview]
Message-ID: <20190919203919.GF30495@linux.intel.com> (raw)
In-Reply-To: <20190911023142.85970-1-morbo@google.com>
On Tue, Sep 10, 2019 at 07:31:42PM -0700, Bill Wendling wrote:
> Use a list of expected values instead of printing out numbers, which
> aren't very meaningful. This prints only if the expected and actual
> values differ.
>
> Signed-off-by: Bill Wendling <morbo@google.com>
> ---
> x86/setjmp.c | 19 +++++++++++++++----
> 1 file changed, 15 insertions(+), 4 deletions(-)
>
> diff --git a/x86/setjmp.c b/x86/setjmp.c
> index 976a632..c0b25ec 100644
> --- a/x86/setjmp.c
> +++ b/x86/setjmp.c
> @@ -1,19 +1,30 @@
> #include "libcflat.h"
> #include "setjmp.h"
>
> +int expected[] = {
This should be 'static const'.
> + 0, 1, 2, 3, 4, 5, 6, 7, 8, 9
> +};
> +
> +#define NUM_EXPECTED (sizeof(expected) / sizeof(int))
How about NUM_LONGJMPS? And you can use ARRAY_SIZE.
> +
> int main(void)
> {
> - volatile int i;
> + volatile int i = -1, index = 0;
> + volatile bool had_errors = false;
> jmp_buf j;
>
> if (setjmp(j) == 0) {
> i = 0;
> }
> - printf("%d\n", i);
> - if (++i < 10) {
> + if (expected[index++] != i) {
> + printf("FAIL: actual %d / expected %d\n", i, expected[index]);
This will print the wrong expected value on failure since index was
incremented above.
> + had_errors = true;
> + }
had_errors seems like overkill. If we fail once, what's the pointing of
continuing on?
> + if (index < NUM_EXPECTED) {
> + i++;
> longjmp(j, 1);
> }
And we should pass i to longjmp(), otherwise there isn't really any point
to having separate index and i variables, e.g.:
volatile int index = 0;
jmp_buf j;
int i;
i = setjmp(j);
if (expected[index] != i) {
printf("FAIL: actual %d / expected %d\n", i, expected[index]);
return -1;
}
index++;
if (i < NUM_LONGJMPS)
longjmp(jmp_buf, i + 1);
return 0;
>
> - printf("done\n");
> + printf("Test %s\n", had_errors ? "FAILED" : "PASSED");
> return 0;
> }
> --
> 2.23.0.162.g0b9fbb3734-goog
>
next prev parent reply other threads:[~2019-09-19 20:39 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-11 2:31 [kvm-unit-tests PATCH] x86: setjmp: check expected value of "i" to give better feedback Bill Wendling
2019-09-19 20:10 ` Jim Mattson
2019-09-19 20:39 ` Sean Christopherson [this message]
2019-09-19 22:14 ` [kvm-unit-tests PATCH 1/1] " Bill Wendling
2019-09-19 22:16 ` Bill Wendling
2019-09-24 14:03 ` [kvm-unit-tests PATCH] " Paolo Bonzini
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190919203919.GF30495@linux.intel.com \
--to=sean.j.christopherson@intel.com \
--cc=jmattson@google.com \
--cc=kvm@vger.kernel.org \
--cc=morbo@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.