View Issue Details

Jump to NotesJump to History
IDProjectCategoryView StatusDate SubmittedLast Update
00001061003.1(2008)/Issue 7System Interfacespublic2009-06-30 19:162013-04-16 13:06
ReporterDon CragunAssigned Toajosey  
PrioritynormalSeverityObjectionTypeOmission
Status ClosedResolutionAccepted As Marked 
NameDon Cragun
OrganizationSelf
User Reference
Sectiontime()
Page Number2107
Line Number66673
Interp StatusApproved
Final Accepted Text0000106:0000197
Summary0000106: time() missing error
Description _____________________________________________________________________________
 OBJECTION Enhancement Request Number 35
 dcragun:xxxxxxxxx Defect in XSH time() (rdvk# 3)
 {time() missing error} Sat, 25 Apr 2009 06:36:18 +0100 (BST)
 _____________________________________________________________________________


Many systems today have 32-bit time_t types. At some point in
 the future, these systems will be unable to represent seconds
 since the Epoch (the return value from time()) in an object of
 type time_t, but the ERRORS section says "No errors are
 defined." As with other interfaces where returned information
 might not fit into the space reserved for the return value,
 time() should specify that when this happens, (time_t)-1 will
 be returned and errno will be set to EOVERFLOW.
Desired ActionSend this issue down the interpretations track: the standard is
 silent on this issue; concerns have been forwarded to the sponsor.

 Change:
     "shall return the value of time."
 on P2107, L66670 in the RETURN VALUE section to:
     "shall return the value of time without changing the
      value of errno."

 Change:
     "No errors are defined."
 on P2107, L66673 in the ERRORS sections to:
     "The time() function shall fail if:
      [EOVERFLOW] The current value of time will not fit in an
                     object of type time_t."

 Change:
     "None."
 on P2108, L66710 in the APPLICATION USAGE section to:
     "As (time_t)-1 is a permissible return value in a successful
      situation, an application wishing to check for error
      conditions should set errno to 0, then call time(), and, if
      it returns (time_t)-1, check to see if errno is non-zero."

Tagstc1-2008

Activities

msbrown

2009-06-30 19:18

manager   bugnote:0000140

Add to clock_gettime Page 668 after line 22669

The clock_gettime() function shall fail if:

  [EOVERFLOW] The current value of time will not fit in an
                     object of type time_t."

On the time() page:

Add a may fail case.

 Change:
     "No errors are defined."
 on P2107, L66673 in the ERRORS sections to:
     "The time() function may fail if:
      [EOVERFLOW] The current value of time will not fit in an
                     object of type time_t."

Add to RATIONALE on time()

On some systems the time() function is implemented using a system call that
does not return an error condition in addition to the return value.
In this case it is impossible to differentiate between a valid and
invalid return value and hence overflow conditions cannot be
detected.

geoffclare

2009-07-10 11:30

manager   bugnote:0000157

Last edited: 2009-07-10 11:31

 When we progress this down the interpretations track, the accepted text
needs to be updated to account for the comments made in reply to the
minutes of the May 14 2009 teleconference:

From: Geoff Clare <gwc@xxxxxxxxx>

> On the time() page:
>
> Add a may  fail case.
>
>  Change:
>      "No errors are defined."
>  on P2107, L66673 in the ERRORS sections to:
>      "The time() function may fail if:
>       [EOVERFLOW]    The current value of time will not fit in an
>                      object of type time_t."

With CX shading.

From: Nick Stoughton <nick@xxxxxxxxx>

> Add to clock_gettime Page 668 after line 22669
>
> The clock_gettime() function shall fail if:
>
>   [EOVERFLOW]    The current value of time will not fit in an
>                      object of type time_t."

I think error condition this would be better worded as something like:
"[EOVERFLOW]   The number of seconds will not fit in an object of type
               time_t."

The current value of time for this call includes the nanoseconds
field ... so "the current value of time" will never fit in a time_t!


Don Cragun

2009-08-13 18:28

viewer   bugnote:0000197

Last edited: 2009-10-12 05:55

 Interpretation response
------------------------
The standard does not speak to this issue, and as such
no conformance distinction can be made between alternative
implementations based on this. This is being referred to the
sponsor.

Rationale:
-------------
None.

Notes to the Editor (not part of this interpretation):
-------------------------------------------------------

Add to the ERRORS section of clock_gettime() after P668, L22669:
        The clock_gettime() function shall fail if:

        [EOVERFLOW] The number of seconds will not fit in an object
                        of type time_t."

Change:
        No errors are defined.
on P2107, L66673 in the ERRORS section of time() to:
        The time() function may fail if:

        [EOVERFLOW] The number of seconds since the Epoch will not
                        fit in an object of type time_t.
with CX shading.

Add to the RATIONALE section of time() after P2108, L66718:
        On some systems the time() function is implemented using a
        system call that does not return an error condition in addition
        to the return value. On these systems it is impossible to
        differentiate between valid and invalid return values and hence
        overflow conditions cannot be reliably detected.

Issue History

Date Modified Username Field Change
2009-06-30 19:16 msbrown New Issue
2009-06-30 19:16 msbrown Status New => Under Review
2009-06-30 19:16 msbrown Assigned To => ajosey
2009-06-30 19:16 msbrown Name => Mark Brown
2009-06-30 19:16 msbrown Organization => IBM
2009-06-30 19:16 msbrown Section => time()
2009-06-30 19:16 msbrown Page Number => 2107
2009-06-30 19:16 msbrown Line Number => 66673
2009-06-30 19:18 msbrown Note Added: 0000140
2009-06-30 19:18 msbrown Status Under Review => Interpretation Required
2009-06-30 19:18 msbrown Resolution Open => Accepted As Marked
2009-06-30 19:18 msbrown Final Accepted Text => 0000106:0000140
2009-07-01 16:16 Don Cragun Name Mark Brown => Don Cragun
2009-07-01 16:16 Don Cragun Organization IBM => Self
2009-07-01 16:16 Don Cragun Reporter msbrown => Don Cragun
2009-07-10 11:30 geoffclare Note Added: 0000157
2009-07-10 11:31 geoffclare Note Edited: 0000157
2009-08-13 18:28 Don Cragun Interp Status => ---
2009-08-13 18:28 Don Cragun Note Added: 0000197
2009-08-13 18:29 Don Cragun Final Accepted Text 0000106:0000140 => 0000106:0000197
2009-08-13 18:31 Don Cragun Interp Status --- => Pending
2009-08-13 18:38 Don Cragun Note Edited: 0000197
2009-09-17 15:41 nick Interp Status Pending => Proposed
2009-10-12 05:55 ajosey Note Edited: 0000197
2009-10-12 05:55 ajosey Interp Status Proposed => Approved
2010-09-21 11:03 geoffclare Tag Attached: tc1-2008
2013-04-16 13:06 ajosey Status Interpretation Required => Closed