View Issue Details
ID | Project | Category | View Status | Date Submitted | Last Update |
---|---|---|---|---|---|
0000733 | 1003.1(2013)/Issue7+TC1 | Base Definitions and Headers | public | 2013-08-12 12:35 | 2019-06-10 08:55 |
Reporter | steffen | Assigned To | |||
Priority | normal | Severity | Editorial | Type | Clarification Requested |
Status | Closed | Resolution | Accepted As Marked | ||
Name | Steffen Nurpmeso | ||||
Organization | |||||
User Reference | |||||
Section | |||||
Page Number | 331, 493, 546, 1951 | ||||
Line Number | 11054, 16840, 18989-90, 62637 | ||||
Interp Status | --- | ||||
Final Accepted Text | 0000733:0001719 | ||||
Summary | 0000733: Volatile qualification of sig_atomic_t is ambigiuous | ||||
Description | The standard is ambigiuous on wether the sig_atomic_t typedef includes the `volatile' qualifier or not. Whereas p. 331 l. 11054 says it is "Possibly volatile-qualified integer type", and p. 546 l. 18989-90 state "Integer type of an object that can be accessed as an atomic entity, even in the presence of asynchronous interrupts.") further references (p. 493, l. 16840; p. 1951 l. 62637) enforce the user to provide the qualifier explicitly. Because the type is async-safe by definition the `volatile' qualifier must necessarily be part of the typedef if such a qualification is necessary to enforce the necessary constraints. Note this issue is related to issue 728. | ||||
Desired Action | On page 493, line 16840, change assigning a value to an object declared as volatile sig_atomic_t, to assigning a value to an object declared as sig_atomic_t, On page 1951, line 62637, change assigning a value to an object declared as volatile sig_atomic_t, to assigning a value to an object declared as sig_atomic_t, | ||||
Tags | tc2-2008 |
related to | 0000728 | Closed | Restrictions on signal handlers are both excessive and insufficient |
|
At page 546 line 18989 section 2.12.1 in the description of sig_atomic_t change: Integer type to Possibly volatile-qualified integer type |
|
C99 and C11 say that the implementation can choose whether to make sig_atomic_t a typedef of a volatile- or non-volatile-qualified integer type, but where the standard explicitly uses volatile as a qualifier we require it to be volatile. Thus, the "possibly" when discussing the definition of the type does not conflict with the need to declare 'volatile sig_atomic_t'. |
Date Modified | Username | Field | Change |
---|---|---|---|
2013-08-12 12:35 | steffen | New Issue | |
2013-08-12 12:35 | steffen | Name | => Steffen Nurpmeso |
2013-08-12 12:35 | steffen | Page Number | => 331, 493, 546, 1951 |
2013-08-12 12:35 | steffen | Line Number | => 11054, 16840, 18989-90, 62637 |
2013-08-15 15:36 | eblake | Relationship added | related to 0000728 |
2013-08-15 16:00 | geoffclare | Note Added: 0001719 | |
2013-08-15 16:02 | geoffclare | Interp Status | => --- |
2013-08-15 16:02 | geoffclare | Final Accepted Text | => 0000733:0001719 |
2013-08-15 16:02 | geoffclare | Status | New => Resolved |
2013-08-15 16:02 | geoffclare | Resolution | Open => Accepted As Marked |
2013-08-15 16:02 | geoffclare | Tag Attached: tc2-2008 | |
2013-08-15 16:11 | nick | Note Added: 0001720 | |
2013-08-15 16:13 | nick | Note Edited: 0001720 | |
2013-08-15 16:13 | eblake | Note Edited: 0001720 | |
2013-08-15 16:13 | nick | Note Edited: 0001720 | |
2019-06-10 08:55 | agadmin | Status | Resolved => Closed |