[PATCH v6] Implement decoding of statx syscall

[Eugene Syromyatnikov]

On Fri, Mar 17, 2017 at 08:08:29PM +0300, Victor Krapivensky wrote:
<...>
> --- /dev/null
> +++ b/statx.c
> @@ -0,0 +1,111 @@
> +/*
> + * Copyright (c) 2017 The strace developers.
> + * All rights reserved.
> + *
> + * Redistribution and use in source and binary forms, with or without
> + * modification, are permitted provided that the following conditions
> + * are met:
> + * 1. Redistributions of source code must retain the above copyright
> + *    notice, this list of conditions and the following disclaimer.
> + * 2. Redistributions in binary form must reproduce the above copyright
> + *    notice, this list of conditions and the following disclaimer in the
> + *    documentation and/or other materials provided with the distribution.
> + * 3. The name of the author may not be used to endorse or promote products
> + *    derived from this software without specific prior written permission.
> + *
> + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
> + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
> + * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
> + * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
> + * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
> + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
> + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
> + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
> + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
> + * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
> + */
> +
> +#include "defs.h"
> +#include "statx.h"
> +
> +#include <sys/stat.h>
> +
> +#include "xlat/statx_masks.h"
> +#include "xlat/statx_attrs.h"
> +#include "xlat/at_statx_sync_types.h"
> +
> +SYS_FUNC(statx)
> +{
> +	if (entering(tcp)) {
> +		print_dirfd(tcp, tcp->u_arg[0]);
> +		printpath(tcp, tcp->u_arg[1]);
> +		tprints(", ");
> +		if (printflags(at_flags, tcp->u_arg[2] & ~AT_STATX_SYNC_TYPE,
> +		               NULL))
> +		{
> +			tprints("|");
I haven't look at v7 of the patch yet, but coverage [1] shows that at least
this line is not covered by the test.

> +		}
> +		printxval(at_statx_sync_types,
> +		          tcp->u_arg[2] & AT_STATX_SYNC_TYPE, "AT_STATX_???");
> +		tprints(", ");
> +		printflags(statx_masks, tcp->u_arg[3], "STATX_???");
> +		tprints(", ");
> +	} else {
> +#define PRINT_FIELD_U(field) \
> +	tprintf(", %s=%llu", #field, (unsigned long long) stx.field)
> +
> +#define PRINT_FIELD_TIME(field)						\
> +	do {								\
> +		tprints(", " #field "=");				\
> +		tprints(sprinttime(stx.field.sec));			\
> +		if (stx.field.nsec)					\
Only one branch in regards of of stx.field.nsec value is covered in all
instances of this macro.

> +			tprintf(".%09" PRId32, stx.field.nsec);		\
> +	} while (0)
> +
> +		struct_statx stx;
> +		if (umove_or_printaddr(tcp, tcp->u_arg[4], &stx)) {
> +			return 0;
> +		}
> +
> +		tprints("{stx_mask=");
> +		printflags(statx_masks, stx.stx_mask, "STATX_???");
> +
> +		if (!abbrev(tcp)) {
Only branch with abbrev(tcp) == 0 is covered.

> +			PRINT_FIELD_U(stx_blksize);
> +		}
> +
> +		tprints(", stx_attributes=");
> +		printflags(statx_attrs, stx.stx_attributes, "STATX_ATTR_???");
> +
> +		if (!abbrev(tcp)) {
> +			PRINT_FIELD_U(stx_nlink);
> +			printuid(", stx_uid=", stx.stx_uid);
> +			printuid(", stx_gid=", stx.stx_gid);
> +		}
> +
> +		tprints(", stx_mode=");
> +		print_symbolic_mode_t(stx.stx_mode);
> +
> +		if (!abbrev(tcp)) {
> +			PRINT_FIELD_U(stx_ino);
> +		}
> +
> +		PRINT_FIELD_U(stx_size);
> +
> +		if (!abbrev(tcp)) {
> +			PRINT_FIELD_U(stx_blocks);
> +			PRINT_FIELD_TIME(stx_atime);
> +			PRINT_FIELD_TIME(stx_btime);
> +			PRINT_FIELD_TIME(stx_ctime);
> +			PRINT_FIELD_TIME(stx_mtime);
> +			PRINT_FIELD_U(stx_rdev_major);
> +			PRINT_FIELD_U(stx_rdev_minor);
> +			PRINT_FIELD_U(stx_dev_major);
> +			PRINT_FIELD_U(stx_dev_minor);
> +		} else {
> +			tprints(", ...");
... and this one, but this is a common problem of all current *stat*
checks.

> +		}
> +		tprints("}");
> +	}
> +	return 0;
> +}

[1] http://esyr.org/dump/strace-4.16.0.69.f1ea-dirty-coverage/strace_git/statx.c.gcov.html