All of lore.kernel.org
 help / color / mirror / Atom feed
From: Cyril Hrubis <chrubis@suse.cz>
To: ltp@lists.linux.it
Subject: [LTP] [PATCH v3] BPF: Regression test for 64bit arithmetic
Date: Fri, 13 Sep 2019 13:46:14 +0200	[thread overview]
Message-ID: <20190913114613.GA7939@rei> (raw)
In-Reply-To: <1568372120.3306.13.camel@suse.de>

Hi!
> > +		BPF_LD_IMM64(BPF_REG_4, A64INT),        /* 21: r4 =
> > 2^61 */
>   ^
>   I think 2^61 is different to 1<<60

Agreed, 1<<60 is 2^60

> > +		BPF_ALU64_REG(BPF_SUB, BPF_REG_4, BPF_REG_6), /* 23:
> > r4 -= r6 */
> > +		BPF_STX_MEM(BPF_DW, BPF_REG_3, BPF_REG_4, 0), /* 24:
> > *r3 = r4 */
> > +
> > +		BPF_MOV64_IMM(BPF_REG_0, 0),            /* 25: r0 =
> > 0 */
> > +		BPF_EXIT_INSN(),		        /* 26:
> > return r0 */
> > +	};
> > +
> > +	/* Leaks memory when -i is specified */
> > +	prog = tst_alloc(sizeof(insn));
> > +	memcpy(prog, insn, sizeof(insn));
> > +
> > +	memset(attr, 0, sizeof(*attr));
> > +	attr->prog_type = BPF_PROG_TYPE_SOCKET_FILTER;
> > +	attr->insns = ptr_to_u64(prog);
> > +	attr->insn_cnt = ARRAY_SIZE(insn);
> > +	attr->license = ptr_to_u64("GPL");
> > +	attr->log_buf = ptr_to_u64(log);
> > +	attr->log_size = BUFSIZ;
> > +	attr->log_level = 1;
> > +
> > +	TEST(bpf(BPF_PROG_LOAD, attr, sizeof(*attr)));
> > +	if (TST_RET == -1) {
> > +		if (log[0] != 0) {
> > +			tst_res(TINFO, "Verification log:");
> > +			fputs(log, stderr);
> > +			tst_brk(TBROK | TTERRNO, "Failed
> > verification");
> > +		} else {
> > +			tst_brk(TBROK | TTERRNO, "Failed to load
> > program");
> > +		}
> > +	}
> > +
> > +	return TST_RET;
> > +}
> > +
> > +static void setup(void)
> > +{
> > +	memcpy(msg, MSG, sizeof(MSG));
> > +}
> > +
> > +static void run(void)
> > +{
> > +	int map_fd, prog_fd;
> > +	int sk[2];
> > +
> > +	memset(attr, 0, sizeof(*attr));
> > +	attr->map_type = BPF_MAP_TYPE_ARRAY;
> > +	attr->key_size = 4;
> > +	attr->value_size = 8;
> > +	attr->max_entries = 2;
> > +
> > +	TEST(bpf(BPF_MAP_CREATE, attr, sizeof(*attr)));
> > +	if (TST_RET == -1) {
> > +		if (TST_ERR == EPERM) {
> > +			tst_brk(TCONF | TTERRNO,
> > +				"bpf() requires CAP_SYS_ADMIN on
> > this system");
> > +		} else {
> > +			tst_brk(TBROK | TTERRNO, "Failed to create
> > array map");
> > +		}
> > +	}
> > +	map_fd = TST_RET;
> > +
> > +	prog_fd = load_prog(map_fd);
> > +
> > +	SAFE_SOCKETPAIR(AF_UNIX, SOCK_DGRAM, 0, sk);
> > +	SAFE_SETSOCKOPT(sk[1], SOL_SOCKET, SO_ATTACH_BPF,
> > +			&prog_fd, sizeof(prog_fd));
> > +
> > +	SAFE_WRITE(1, sk[0], msg, sizeof(MSG));
> > +
> > +	memset(attr, 0, sizeof(*attr));
> > +	attr->map_fd = map_fd;
> > +	attr->key = ptr_to_u64(key);
> > +	attr->value = ptr_to_u64(val);
> > +	*key = 0;
> > +
> > +	TEST(bpf(BPF_MAP_LOOKUP_ELEM, attr, sizeof(*attr)));
> > +	if (TST_RET == -1) {
> > +		tst_res(TFAIL | TTERRNO, "array map lookup");
> > +	} else if (*val != A64INT + 1) {
> > +		tst_res(TFAIL,
> > +			"val = %lu, but should be val = %lu + 1",
>                                 ^
> Not sure if it is really needed but I would use %llu here. If so, other
> places as well.

Technically we should be using PRIu64 which expands to %lu on 64 bit and
to %llu on 32 bit. But given that sizeof(long long) == sizeof(long) on
64 bit on Linux %llu should work as well.

-- 
Cyril Hrubis
chrubis@suse.cz

  reply	other threads:[~2019-09-13 11:46 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-09-11  9:54 [LTP] [PATCH v3] BPF: Regression test for 64bit arithmetic Richard Palethorpe
2019-09-12 14:56 ` Cyril Hrubis
2019-09-17  7:26   ` Richard Palethorpe
2019-09-13 10:55 ` Clemens Famulla-Conrad
2019-09-13 11:46   ` Cyril Hrubis [this message]
2019-09-17  7:27     ` Richard Palethorpe
2019-09-27 11:08       ` Cyril Hrubis

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=20190913114613.GA7939@rei \
    --to=chrubis@suse.cz \
    --cc=ltp@lists.linux.it \
    /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.