POSIX "aio_return/2-1" failed.

LTP Mitani 2 participants 7 messages Oct 8, 2010 – Oct 27, 2010

Share & tools

Mitani

author

2010-10-08 05:29 UTC #0 lore raw

Hi,

"conformance/interfaces/aio_return/2-1" failed with following message:
------------
conformance/interfaces/aio_return/2-1: execution: FAILED: Output:
aio_return/2-1.c Second call to aio_return() should return -1 : 111
------------

Same problem occurs in "aio_return/3-2.c" and "aio_return/4-1".

Environment is as follows:
- RHEL5.5 --- (x86, x86_64, ia64)
- RHEL4.8 --- (x86, x86_64, ia64)

This testset seems to be the error root test for "aio_return()".

First "aio_return()" returned with 111. This return value is the length of buffer of "aio_write()".
But second "aio_return()" returned with 111, too. It's unexpected result for this test set.
Therefore, this test failed.

Man page says that
"This function should be called only once for any given request, after aio_error(2) returns something other than EINPROGRESS.":
------------
# LANG=C man 3 aio_return

AIO_RETURN(3) Linux Programmer's Manual
AIO_RETURN(3)

NAME
aio_return - get return status of asynchronous I/O operation

SYNOPSIS

       #include <aio.h>

       ssize_t aio_return(struct aiocb *aiocbp);

DESCRIPTION
The aio_return function returns the final return status for the
asynchronous I/O
request with control block pointed to by aiocbp.

This function should be called only once for any given
request, after
aio_error(2) returns something other than EINPROGRESS.

RETURN VALUE
If the asynchronous I/O operation has completed, this function
returns the value
that would have been returned in case of a synchronous read, write,
or fsync
request. Otherwise the return value is undefined. On error, the
error value is
returned.

ERRORS
EINVAL aiocbp does not point at a control block for an asynchronous
I/O request
of which the return status has not been retrieved yet.

CONFORMING TO

       POSIX 1003.1-2003

Garrett Cooper

2010-10-12 20:23 UTC #1 lore raw

Quoted 84 lines · Mitani

Hi,
"conformance/interfaces/aio_return/2-1" failed with following message:
------------
conformance/interfaces/aio_return/2-1: execution: FAILED: Output:
aio_return/2-1.c Second call to aio_return() should return -1 : 111
------------
Same problem occurs in "aio_return/3-2.c" and "aio_return/4-1".
Environment is as follows:
- RHEL5.5 --- (x86, x86_64, ia64)
- RHEL4.8 --- (x86, x86_64, ia64)
This testset seems to be the error root test for "aio_return()".
First "aio_return()" returned with 111. This return value is the length of buffer of "aio_write()".
But second "aio_return()" returned with 111, too. It's unexpected result for this test set.
Therefore, this test failed.
Man page says that
"This function should be called only once for any given request, after aio_error(2) returns something other than EINPROGRESS.":
------------
# LANG=C man 3 aio_return
AIO_RETURN(3) Linux Programmer's Manual AIO_RETURN(3)
NAME
aio_return - get return status of asynchronous I/O operation
SYNOPSIS
#include <aio.h>
ssize_t aio_return(struct aiocb *aiocbp);
DESCRIPTION
The aio_return function returns the final return status for the asynchronous I/O
request with control block pointed to by aiocbp.
This function should be called only once for any given request, after
aio_error(2) returns something other than EINPROGRESS.
RETURN VALUE
If the asynchronous I/O operation has completed, this function returns the value
that would have been returned in case of a synchronous read, write, or fsync
request. Otherwise the return value is undefined. On error, the error value is
returned.
ERRORS
EINVAL aiocbp does not point at a control block for an asynchronous I/O request
of which the return status has not been retrieved yet.
CONFORMING TO
POSIX 1003.1-2003
SEE ALSO
aio_cancel(3), aio_error(3), aio_fsync(3), aio_read(3), aio_suspend(3),
aio_write(3)
2003-11-14
AIO_RETURN(3)
#
------------
And, it says that
"If the asynchronous I/O operation has completed, this function returns the value that would have been returned in case of a synchronous read, write, or fsync request. Otherwise the return value is undefined. On error, the error value is returned.".

Quoted 6 lines

It can be supposed that the return value of second "aio_return()" is undefined.
Therefore, it is not mistake that return value of the second "aio_return()" is one at success, I think.
And, I think that "UNTESTED" or "UNRESOLVED" may be is more appropriate for this test.

I'll respond to this sometime in the next couple of days after I
do some more investigation on the issue.
Thanks,
-Garrett

Garrett Cooper

2010-10-18 12:00 UTC #2 lore raw

In reply to Mitani

Quoted 84 lines · Mitani

Hi,
"conformance/interfaces/aio_return/2-1" failed with following message:
------------
conformance/interfaces/aio_return/2-1: execution: FAILED: Output:
aio_return/2-1.c Second call to aio_return() should return -1 : 111
------------
Same problem occurs in "aio_return/3-2.c" and "aio_return/4-1".
Environment is as follows:
- RHEL5.5 --- (x86, x86_64, ia64)
- RHEL4.8 --- (x86, x86_64, ia64)
This testset seems to be the error root test for "aio_return()".
First "aio_return()" returned with 111. This return value is the length of buffer of "aio_write()".
But second "aio_return()" returned with 111, too. It's unexpected result for this test set.
Therefore, this test failed.
Man page says that
"This function should be called only once for any given request, after aio_error(2) returns something other than EINPROGRESS.":
------------
# LANG=C man 3 aio_return
AIO_RETURN(3) Linux Programmer's Manual AIO_RETURN(3)
NAME
aio_return - get return status of asynchronous I/O operation
SYNOPSIS
#include <aio.h>
ssize_t aio_return(struct aiocb *aiocbp);
DESCRIPTION
The aio_return function returns the final return status for the asynchronous I/O
request with control block pointed to by aiocbp.
This function should be called only once for any given request, after
aio_error(2) returns something other than EINPROGRESS.
RETURN VALUE
If the asynchronous I/O operation has completed, this function returns the value
that would have been returned in case of a synchronous read, write, or fsync
request. Otherwise the return value is undefined. On error, the error value is
returned.
ERRORS
EINVAL aiocbp does not point at a control block for an asynchronous I/O request
of which the return status has not been retrieved yet.
CONFORMING TO
POSIX 1003.1-2003
SEE ALSO
aio_cancel(3), aio_error(3), aio_fsync(3), aio_read(3), aio_suspend(3),
aio_write(3)
2003-11-14
AIO_RETURN(3)
#
------------
And, it says that
"If the asynchronous I/O operation has completed, this function returns the value that would have been returned in case of a synchronous read, write, or fsync request. Otherwise the return value is undefined. On error, the error value is returned.".

Quoted 6 lines

It can be supposed that the return value of second "aio_return()" is undefined.
Therefore, it is not mistake that return value of the second "aio_return()" is one at success, I think.
And, I think that "UNTESTED" or "UNRESOLVED" may be is more appropriate for this test.

No. POSIX clearly states:

"If the asynchronous I/O operation has completed, then the return status, as described for read(), write(), and fsync(), shall be returned. If the asynchronous I/O operation has not yet completed, the results of aio_return() are undefined.

If the aio_return() function fails, it shall return -1 and set errno to indicate the error."

ERRORS

"The aio_return() function may fail if:

[EINVAL]
The aiocbp argument does not refer to an asynchronous operation
whose return status has not yet been retrieved."

What the test implementers were looking to do was test out that -1
/ EINVAL were returned, and maybe that fact that they were testing was in fact implementation defined, as opposed to what's stated in the standard, as the spec doesn't state that the behavior for aio_return would be s.t. the state of the AIO stream is reset when aio_return is called.
So yes, I agree this should be fixed, but let me check with the
POSIX folks what the best course of action is.
Thanks,
-Garrett

------------------------------------------------------------------------------
Download new Adobe(R) Flash(R) Builder(TM) 4
The new Adobe(R) Flex(R) 4 and Flash(R) Builder(TM) 4 (formerly Flex(R) Builder(TM)) enable the development of rich applications that run across multiple browsers and platforms. Download your free trials today!
http://p.sf.net/sfu/adobe-dev2dev
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Garrett Cooper

2010-10-21 09:41 UTC #3 lore raw

Quoted 110 lines · Garrett Cooper

Quoted 84 lines · Mitani
Hi,
"conformance/interfaces/aio_return/2-1" failed with following message:
------------
conformance/interfaces/aio_return/2-1: execution: FAILED: Output:
aio_return/2-1.c Second call to aio_return() should return -1 : 111
------------
Same problem occurs in "aio_return/3-2.c" and "aio_return/4-1".
Environment is as follows:
- RHEL5.5 --- (x86, x86_64, ia64)
- RHEL4.8 --- (x86, x86_64, ia64)
This testset seems to be the error root test for "aio_return()".
First "aio_return()" returned with 111. This return value is the length of buffer of "aio_write()".
But second "aio_return()" returned with 111, too. It's unexpected result for this test set.
Therefore, this test failed.
Man page says that
"This function should be called only once for any given request, after aio_error(2) returns something other than EINPROGRESS.":
------------
# LANG=C man 3 aio_return
AIO_RETURN(3) Linux Programmer's Manual AIO_RETURN(3)
NAME
aio_return - get return status of asynchronous I/O operation
SYNOPSIS
#include <aio.h>
ssize_t aio_return(struct aiocb *aiocbp);
DESCRIPTION
The aio_return function returns the final return status for the asynchronous I/O
request with control block pointed to by aiocbp.
This function should be called only once for any given request, after
aio_error(2) returns something other than EINPROGRESS.
RETURN VALUE
If the asynchronous I/O operation has completed, this function returns the value
that would have been returned in case of a synchronous read, write, or fsync
request. Otherwise the return value is undefined. On error, the error value is
returned.
ERRORS
EINVAL aiocbp does not point at a control block for an asynchronous I/O request
of which the return status has not been retrieved yet.
CONFORMING TO
POSIX 1003.1-2003
SEE ALSO
aio_cancel(3), aio_error(3), aio_fsync(3), aio_read(3), aio_suspend(3),
aio_write(3)
2003-11-14
AIO_RETURN(3)
#
------------
And, it says that
"If the asynchronous I/O operation has completed, this function returns the value that would have been returned in case of a synchronous read, write, or fsync request. Otherwise the return value is undefined. On error, the error value is returned.".
Quoted 6 lines
It can be supposed that the return value of second "aio_return()" is undefined.
Therefore, it is not mistake that return value of the second "aio_return()" is one at success, I think.
And, I think that "UNTESTED" or "UNRESOLVED" may be is more appropriate for this test.
No. POSIX clearly states:
"If the asynchronous I/O operation has completed, then the return status, as described for read(), write(), and fsync(), shall be returned. If the asynchronous I/O operation has not yet completed, the results of aio_return() are undefined.
If the aio_return() function fails, it shall return -1 and set errno to indicate the error."
ERRORS
"The aio_return() function may fail if:
[EINVAL]
The aiocbp argument does not refer to an asynchronous operation whose return status has not yet been retrieved."

Quoted 8 lines

What the test implementers were looking to do was test out that -1 / EINVAL were returned, and maybe that fact that they were testing was in fact implementation defined, as opposed to what's stated in the standard, as the spec doesn't state that the behavior for aio_return would be s.t. the state of the AIO stream is reset when aio_return is called.
So yes, I agree this should be fixed, but let me check with the POSIX folks what the best course of action is.

I've read into this a bit more, and I think that the aio_return
testcases were written incorrectly. Please try out this testcase... if it works for you, then I'll apply necessary logic to other testcases as well.
It worked for me at least on FreeBSD.
Thanks,
-Garrett

testcases/open_posix_testsuite/conformance/interfaces/aio_return/1-1.c +8 ~~-14~~

index bfc8e12..090387c 100644
@@ -43,8 +43,10 @@ int main(void)
 	struct aiocb aiocb;
 	int fd, retval;
 
+#if 0
 	if (sysconf(_SC_ASYNCHRONOUS_IO) != 200112L)
 		return PTS_UNSUPPORTED;
+#endif
 
 	snprintf(tmpfname, sizeof(tmpfname), "/tmp/pts_aio_return_1_1_%d", 
 		  getpid());

@@ -73,26 +75,18 @@ int main(void)
 		exit(PTS_FAIL);
 	}
 
-	do {
-		retval = aio_error(&aiocb);
-		if (retval == -1) {
-			close(fd);
-			printf(TNAME " Error at aio_error(): %s\n",
-				strerror(errno));
-			exit(PTS_FAIL);
-		}
-	} while (retval == EINPROGRESS);
-
+	while (aio_error(&aiocb) == EINPROGRESS)
+		sleep(1);
 	retval = aio_return(&aiocb);
 
-	if (retval != BUF_SIZE) {
-		close(fd);
+	if (retval == -1) {
 		printf(TNAME " Error at aio_return(): %s\n",
-		       strerror(errno));
+		       strerror(aio_error(&aiocb)));
 		exit(PTS_FAIL);
 	}
-	
+
 	close(fd);
+	printf("aio_return(&aiocb) = %d\n", retval);
 	printf("Test PASSED\n");
 	exit(PTS_PASS);
 }

------------------------------------------------------------------------------
Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store
http://p.sf.net/sfu/nokia-dev2dev

_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Mitani

author

2010-10-21 11:50 UTC #4 lore raw

Quoted 143 lines

-----Original Message-----
From: Garrett Cooper [mailto:[email protected]]
Sent: Thursday, October 21, 2010 6:42 PM
To: Mitani
Cc: [email protected]
Subject: Re: [LTP] POSIX "aio_return/2-1" failed.
Quoted 5 lines · Garrett Cooper
Quoted 4 lines · Mitani
Hi,
"conformance/interfaces/aio_return/2-1" failed with following
message:
Quoted 33 lines
Quoted 33 lines
------------
conformance/interfaces/aio_return/2-1: execution: FAILED: Output:
aio_return/2-1.c Second call to aio_return() should return -1 : 111
------------
Same problem occurs in "aio_return/3-2.c" and "aio_return/4-1".
Environment is as follows:
- RHEL5.5 --- (x86, x86_64, ia64)
- RHEL4.8 --- (x86, x86_64, ia64)
This testset seems to be the error root test for "aio_return()".
First "aio_return()" returned with 111. This return value is the length of buffer of "aio_write()".
But second "aio_return()" returned with 111, too. It's unexpected result for this test set.
Therefore, this test failed.
Man page says that
"This function should be called only once for any given request, after
aio_error(2) returns something other than EINPROGRESS.":
------------
# LANG=C man 3 aio_return
AIO_RETURN(3) Linux Programmer's Manual AIO_RETURN(3)
NAME
aio_return - get return status of asynchronous I/O operation
Quoted 7 lines
Quoted 7 lines
SYNOPSIS
#include <aio.h>
ssize_t aio_return(struct aiocb *aiocbp);
DESCRIPTION
The aio_return function returns the final return status for
the
Quoted 4 lines
Quoted 4 lines
asynchronous I/O
request with control block pointed to by aiocbp.
This function should be called only once for an
y given
Quoted 5 lines
Quoted 5 lines
request, after
aio_error(2) returns something other than EINPROGRESS.
RETURN VALUE
If the asynchronous I/O operation has completed, this
function
Quoted 2 lines
Quoted 2 lines
returns the value
that would have been returned in case of a
synchronous read,
Quoted 2 lines
Quoted 2 lines
write, or fsync
request. Otherwise the return value is undefined. On
error,
Quoted 7 lines
Quoted 7 lines
the error value is
returned.
ERRORS
EINVAL aiocbp does not point at a control block for an asynchronous I/O request
of which the return status has not been retrieved
yet.
Quoted 6 lines
Quoted 6 lines
CONFORMING TO
POSIX 1003.1-2003
SEE ALSO
aio_cancel(3), aio_error(3), aio_fsync(3), aio_r
ead(3),
Quoted 8 lines
Quoted 8 lines
aio_suspend(3),
aio_write(3)
2003-11-14
AIO_RETURN(3)
#
------------
Quoted 8 lines
Quoted 8 lines
And, it says that
"If the asynchronous I/O operation has completed, this function returns the value that would have been returned in case of a synchronous read, write, or fsync request. Otherwise the return value is undefined. On error, the error value is returned.".
It can be supposed that the return value of second "aio_return()"
is
Quoted 2 lines
Quoted 2 lines
undefined.
Therefore, it is not mistake that return value of the second
"aio_return()"
Quoted 9 lines
Quoted 3 lines
is one at success, I think.
And, I think that "UNTESTED" or "UNRESOLVED" may be is more appropriate for this test.
No. POSIX clearly states:
"If the asynchronous I/O operation has completed, then the return status, as described for read(), write(), and fsync(), shall be returned. If the asynchronous I/O operation has not yet completed,
the
Quoted 7 lines
results of aio_return() are undefined.
If the aio_return() function fails, it shall return -1 and set errno to indicate the error."
ERRORS
Quoted 7 lines
"The aio_return() function may fail if:
[EINVAL]
The aiocbp argument does not refer to an asynchronous operation whose return status has not yet been retrieved."
What the test implementers were looking to do was test out that
-1
Quoted 1 line
/ EINVAL were returned, and maybe that fact that they were testing
was
Quoted 3 lines
in fact implementation defined, as opposed to what's stated in the standard, as the spec doesn't state that the behavior for aio_return would be s.t. the state of the AIO stream is reset when aio_return
is
Quoted 3 lines
called.
So yes, I agree this should be fixed, but let me check with the POSIX folks what the best course of action is.

Quoted 7 lines

I've read into this a bit more, and I think that the aio_return
testcases were written incorrectly. Please try out this testcase... if it works for you, then I'll apply necessary logic to other testcases as well.
It worked for me at least on FreeBSD.
Thanks,
-Garrett

I confirmed that your revision worked in my environment.
I also look forward to other testcases (2-1, 3-2, 4-1) being revised.

Thank you--

-Tomonori Mitani

Garrett Cooper

2010-10-21 12:00 UTC #5 lore raw

Quoted 152 lines · Mitani

Quoted 143 lines
-----Original Message-----
From: Garrett Cooper [mailto:[email protected]]
Sent: Thursday, October 21, 2010 6:42 PM
To: Mitani
Cc: [email protected]
Subject: Re: [LTP] POSIX "aio_return/2-1" failed.
Quoted 5 lines · Garrett Cooper
Quoted 4 lines · Mitani
Hi,
"conformance/interfaces/aio_return/2-1" failed with following
message:
Quoted 33 lines
Quoted 33 lines
------------
conformance/interfaces/aio_return/2-1: execution: FAILED: Output:
aio_return/2-1.c Second call to aio_return() should return -1 : 111
------------
Same problem occurs in "aio_return/3-2.c" and "aio_return/4-1".
Environment is as follows:
- RHEL5.5 --- (x86, x86_64, ia64)
- RHEL4.8 --- (x86, x86_64, ia64)
This testset seems to be the error root test for "aio_return()".
First "aio_return()" returned with 111. This return value is the length of buffer of "aio_write()".
But second "aio_return()" returned with 111, too. It's unexpected result for this test set.
Therefore, this test failed.
Man page says that
"This function should be called only once for any given request, after
aio_error(2) returns something other than EINPROGRESS.":
------------
# LANG=C man 3 aio_return
AIO_RETURN(3) Linux Programmer's Manual AIO_RETURN(3)
NAME
aio_return - get return status of asynchronous I/O operation
Quoted 7 lines
Quoted 7 lines
SYNOPSIS
#include <aio.h>
ssize_t aio_return(struct aiocb *aiocbp);
DESCRIPTION
The aio_return function returns the final return status for
the
Quoted 4 lines
Quoted 4 lines
asynchronous I/O
request with control block pointed to by aiocbp.
This function should be called only once for an
y given
Quoted 5 lines
Quoted 5 lines
request, after
aio_error(2) returns something other than EINPROGRESS.
RETURN VALUE
If the asynchronous I/O operation has completed, this
function
Quoted 2 lines
Quoted 2 lines
returns the value
that would have been returned in case of a
synchronous read,
Quoted 2 lines
Quoted 2 lines
write, or fsync
request. Otherwise the return value is undefined. On
error,
Quoted 7 lines
Quoted 7 lines
the error value is
returned.
ERRORS
EINVAL aiocbp does not point at a control block for an asynchronous I/O request
of which the return status has not been retrieved
yet.
Quoted 6 lines
Quoted 6 lines
CONFORMING TO
POSIX 1003.1-2003
SEE ALSO
aio_cancel(3), aio_error(3), aio_fsync(3), aio_r
ead(3),
Quoted 8 lines
Quoted 8 lines
aio_suspend(3),
aio_write(3)
2003-11-14
AIO_RETURN(3)
#
------------
Quoted 8 lines
Quoted 8 lines
And, it says that
"If the asynchronous I/O operation has completed, this function returns the value that would have been returned in case of a synchronous read, write, or fsync request. Otherwise the return value is undefined. On error, the error value is returned.".
It can be supposed that the return value of second "aio_return()"
is
Quoted 2 lines
Quoted 2 lines
undefined.
Therefore, it is not mistake that return value of the second
"aio_return()"
Quoted 9 lines
Quoted 3 lines
is one at success, I think.
And, I think that "UNTESTED" or "UNRESOLVED" may be is more appropriate for this test.
No. POSIX clearly states:
"If the asynchronous I/O operation has completed, then the return status, as described for read(), write(), and fsync(), shall be returned. If the asynchronous I/O operation has not yet completed,
the
Quoted 7 lines
results of aio_return() are undefined.
If the aio_return() function fails, it shall return -1 and set errno to indicate the error."
ERRORS
Quoted 7 lines
"The aio_return() function may fail if:
[EINVAL]
The aiocbp argument does not refer to an asynchronous operation whose return status has not yet been retrieved."
What the test implementers were looking to do was test out that
-1
Quoted 1 line
/ EINVAL were returned, and maybe that fact that they were testing
was
Quoted 3 lines
in fact implementation defined, as opposed to what's stated in the standard, as the spec doesn't state that the behavior for aio_return would be s.t. the state of the AIO stream is reset when aio_return
is
Quoted 3 lines
called.
So yes, I agree this should be fixed, but let me check with the POSIX folks what the best course of action is.
Quoted 7 lines
I've read into this a bit more, and I think that the aio_return testcases were written incorrectly. Please try out this testcase... if it works for you, then I'll apply necessary logic to other testcases as well.
It worked for me at least on FreeBSD.
Thanks,
-Garrett

Quoted 2 lines

I confirmed that your revision worked in my environment.
I also look forward to other testcases (2-1, 3-2, 4-1) being revised.

Please update your git, rebuild, and test :).
Let me know if you run into any issues.
Cheers,
-Garrett

Mitani

author

2010-10-27 03:59 UTC #6 lore raw

Quoted 173 lines

-----Original Message-----
From: Garrett Cooper [mailto:[email protected]]
Sent: Thursday, October 21, 2010 9:00 PM
To: Mitani
Cc: [email protected]
Subject: Re: [LTP] POSIX "aio_return/2-1" failed.
Quoted 9 lines · Mitani
Quoted 9 lines
-----Original Message-----
From: Garrett Cooper [mailto:[email protected]]
Sent: Thursday, October 21, 2010 6:42 PM
To: Mitani
Cc: [email protected]
Subject: Re: [LTP] POSIX "aio_return/2-1" failed.
On Mon, Oct 18, 2010 at 5:00 AM, Garrett Cooper <[email protected]> wrote:
Quoted 8 lines
Quoted 8 lines
Quoted 5 lines · Mitani
Quoted 4 lines
Hi,
"conformance/interfaces/aio_return/2-1" failed with following
message:
Quoted 2 lines
Quoted 2 lines
------------
conformance/interfaces/aio_return/2-1: execution: FAILED:
Output:
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
aio_return/2-1.c Second call to aio_return() should return -1 :
111
Quoted 24 lines
Quoted 24 lines
Quoted 24 lines
Quoted 24 lines
------------
Same problem occurs in "aio_return/3-2.c" and "aio_return/4-1".
Environment is as follows:
- RHEL5.5 --- (x86, x86_64, ia64)
- RHEL4.8 --- (x86, x86_64, ia64)
This testset seems to be the error root test for "aio_return()".
First "aio_return()" returned with 111. This return value is the length of buffer of "aio_write()".
But second "aio_return()" returned with 111, too. It's unexpected result for this test set.
Therefore, this test failed.
Man page says that
"This function should be called only once for any given request, after
aio_error(2) returns something other than EINPROGRESS.":
------------
# LANG=C man 3 aio_return
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
AIO_RETURN(3) Linux Programmer's Manual AIO_RETURN(3)
NAME
aio_return - get return status of asynchronous I/O
operation
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
SYNOPSIS
#include <aio.h>
ssize_t aio_return(struct aiocb *aiocbp);
DESCRIPTION
The aio_return function returns the final return status
for
Quoted 61 lines
Quoted 61 lines
the
Quoted 3 lines
Quoted 3 lines
asynchronous I/O
request with control block pointed to by aiocbp.
This function should be called only once for an y given
Quoted 5 lines
Quoted 5 lines
request, after
aio_error(2) returns something other than EINPROGRESS.
RETURN VALUE
If the asynchronous I/O operation has completed, this
function
Quoted 2 lines
Quoted 2 lines
returns the value
that would have been returned in case of a
synchronous read,
Quoted 2 lines
Quoted 2 lines
write, or fsync
request. Otherwise the return value is undefined. On
error,
Quoted 7 lines
Quoted 7 lines
the error value is
returned.
ERRORS
EINVAL aiocbp does not point at a control block for an asynchronous I/O request
of which the return status has not been retrieved
yet.
Quoted 5 lines
Quoted 5 lines
CONFORMING TO
POSIX 1003.1-2003
SEE ALSO
aio_cancel(3), aio_error(3), aio_fsync(3), aio_r ead(3),
Quoted 8 lines
Quoted 8 lines
aio_suspend(3),
aio_write(3)
2003-11-14
AIO_RETURN(3)
#
------------
Quoted 8 lines
Quoted 8 lines
And, it says that
"If the asynchronous I/O operation has completed, this function returns the value that would have been returned in case of a synchronous read, write, or fsync request. Otherwise the return value is undefined. On error, the error value is returned.".
It can be supposed that the return value of second "aio_return()"
is
Quoted 2 lines
Quoted 2 lines
undefined.
Therefore, it is not mistake that return value of the second
"aio_return()"
Quoted 6 lines
Quoted 3 lines
is one at success, I think.
And, I think that "UNTESTED" or "UNRESOLVED" may be is more appropriate for this test.
No. POSIX clearly states:
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
"If the asynchronous I/O operation has completed, then the return status, as described for read(), write(), and fsync(), shall be returned. If the asynchronous I/O operation has not yet completed,
the
Quoted 3 lines
results of aio_return() are undefined.
If the aio_return() function fails, it shall return -1 and set
errno
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
to indicate the error."
ERRORS
"The aio_return() function may fail if:
[EINVAL]
The aiocbp argument does not refer to an asynchronous
operation
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
whose return status has not yet been retrieved."
What the test implementers were looking to do was test out
that
Quoted 4 lines
Quoted 4 lines
-1
Quoted 1 line
/ EINVAL were returned, and maybe that fact that they were testing
was
Quoted 1 line
in fact implementation defined, as opposed to what's stated in
the
Quoted 1 line
Quoted 1 line
Quoted 1 line
standard, as the spec doesn't state that the behavior for
aio_return
Quoted 4 lines
Quoted 4 lines
Quoted 1 line
would be s.t. the state of the AIO stream is reset when aio_return
is
Quoted 2 lines
called.
So yes, I agree this should be fixed, but let me check with
the
Quoted 5 lines
Quoted 5 lines
Quoted 1 line
POSIX folks what the best course of action is.
I've read into this a bit more, and I think that the aio_return testcases were written incorrectly. Please try out this testcase... if it works for you, then I'll apply necessary logic to other
testcases
Quoted 6 lines
Quoted 4 lines
as well.
It worked for me at least on FreeBSD.
Thanks,
-Garrett

Quoted 7 lines

Quoted 2 lines
I confirmed that your revision worked in my environment.
I also look forward to other testcases (2-1, 3-2, 4-1) being revised.
Please update your git, rebuild, and test :).
Let me know if you run into any issues.
Cheers,
-Garrett

I tried to run "aio_return" tests, but some of them failed with following message:
------------
conformance/interfaces/aio_return/3-2: execution: FAILED: Output:
aio_return/3-2.c aio_return() should fail with (-1, 22); failed with (4096,
0) instead
conformance/interfaces/aio_return/2-1: execution: FAILED: Output:
aio_return/2-1.c Second call to aio_return() should return -1; aio_return() returned 111
------------

Environment is as follows:
- RHEL4.8 --- (x86, x86_64, ia64)
- RHEL5.5 --- (x86, x86_64, ia64)
- kernel --- kernel-2.6.9-89.EL
- glibc --- glibc-2.3.4-2.43

The reason of above failures is that second "aio_return()" doesn't return "-1".

The manpage of "aio_return()" of "FreeBSD 9-current" says:
--------------------
RETURN VALUES
If the asynchronous I/O request has completed, the status is returned
as
described in read(2), write(2), or fsync(2). Otherwise, aio_return() returns -1 and sets errno to indicate the error condition.
--------------------
http://www.freebsd.org/cgi/man.cgi?query=aio_return&apropos=0&sektion=0&manp
ath=FreeBSD+9-current&format=html

On the other hand, the manpage of RHEL4.8/RHEL5.5 says:
--------------------
RETURN VALUE
If the asynchronous I/O operation has completed, this function
returns
the value that would have been returned in case of a synchronous
read,
write, or fsync request. Otherwise the return value is undefined.
On
error, the error value is returned.
--------------------

According to this manual, the return value may not be "-1" in Redhat systems.
Perhaps "aio_return()" doesn't have guarantee to return "-1" even in other Linux systems, I think.

Regards--

-Tomonori Mitani