Commit ac12a18b by Arnaud Charlet

[multiple changes]

2010-09-10  Bob Duff  <duff@adacore.com>

	* s-os_lib.ads, g-expect.ads: Add comments.

2010-09-10  Robert Dewar  <dewar@adacore.com>

	* exp_ch5.adb: Minor reformatting.

From-SVN: r164189
parent e5536c09
2010-09-10 Bob Duff <duff@adacore.com>
* s-os_lib.ads, g-expect.ads: Add comments.
2010-09-10 Robert Dewar <dewar@adacore.com>
* exp_ch5.adb: Minor reformatting.
2010-09-10 Thomas Quinot <quinot@adacore.com>
* scos.ads: Add comments.
......
......@@ -1010,14 +1010,7 @@ package body Exp_Ch5 is
function Build_Step (J : Nat) return Node_Id;
-- The increment step for the index of the right-hand side is written
-- as an attribute reference (Succ or Pred). This function returns
-- the corresponding node, which is placed at the end of theloop body.
-- Note that on the last iteration of the loop, the index is increased
-- (or decreased) past the corresponding bound. This is consistent with
-- the C semantics of the back-end, where such an off-by-one value on a
-- dead index variable is OK. However, in CodePeer mode this leads to
-- spurious warnings, and thus we place a guard around the attribute
-- reference. For obvious reasons we only do this for CodePeer.
-- the corresponding node, which is placed at the end of the loop body.
----------------
-- Build_Step --
......@@ -1045,6 +1038,13 @@ package body Exp_Ch5 is
Expressions => New_List (
New_Occurrence_Of (Rnn (J), Loc))));
-- Note that on the last iteration of the loop, the index is increased
-- (or decreased) past the corresponding bound. This is consistent with
-- the C semantics of the back-end, where such an off-by-one value on a
-- dead index variable is OK. However, in CodePeer mode this leads to
-- spurious warnings, and thus we place a guard around the attribute
-- reference. For obvious reasons we only do this for CodePeer.
if CodePeer_Mode then
Step :=
Make_If_Statement (Loc,
......
......@@ -184,6 +184,9 @@ package GNAT.Expect is
-- subprograms to also match on output on standard error.
--
-- Invalid_Process is raised if the process could not be spawned.
--
-- For information about spawning processes from tasking programs, see the
-- "NOTE: Spawn in tasking programs" in System.OS_Lib (s-os_lib.ads).
procedure Close (Descriptor : in out Process_Descriptor);
-- Terminate the process and close the pipes to it. It implicitly does the
......
......@@ -688,9 +688,8 @@ package System.OS_Lib is
-- (notably Unix systems) a simple file name may also work (if the
-- executable can be located in the path).
--
-- "Spawn" should be avoided in tasking applications, since there are
-- subtle interactions between creating a process and signals/locks
-- that can cause troubles.
-- Spawning processes from tasking programs is not recommended. See
-- "NOTE: Spawn in tasking programs" below.
--
-- Note: Arguments in Args that contain spaces and/or quotes such as
-- "--GCC=gcc -v" or "--GCC=""gcc -v""" are not portable across all
......@@ -717,7 +716,8 @@ package System.OS_Lib is
-- by the operating system, or -1 under VxWorks and any other similar
-- operating systems which have no notion of separately spawnable programs.
--
-- "Spawn" should not be used in tasking applications.
-- Spawning processes from tasking programs is not recommended. See
-- "NOTE: Spawn in tasking programs" below.
procedure Spawn
(Program_Name : String;
......@@ -730,7 +730,8 @@ package System.OS_Lib is
-- Standard Error output is also redirected.
-- Return_Code is set to the status code returned by the operating system
--
-- "Spawn" should not be used in tasking applications.
-- Spawning processes from tasking programs is not recommended. See
-- "NOTE: Spawn in tasking programs" below.
procedure Spawn
(Program_Name : String;
......@@ -747,7 +748,8 @@ package System.OS_Lib is
-- will be set to the status code returned by the operating system.
-- Otherwise, Return_Code is undefined.
--
-- "Spawn" should not be used in tasking applications.
-- Spawning processes from tasking programs is not recommended. See
-- "NOTE: Spawn in tasking programs" below.
type Process_Id is private;
-- A private type used to identify a process activated by the following
......@@ -768,7 +770,8 @@ package System.OS_Lib is
-- returned. Parameters are to be used as in Spawn. If Invalid_Pid is
-- returned the program could not be spawned.
--
-- "Non_Blocking_Spawn" should not be used in tasking applications.
-- Spawning processes from tasking programs is not recommended. See
-- "NOTE: Spawn in tasking programs" below.
--
-- This function will always return Invalid_Pid under VxWorks, since there
-- is no notion of executables under this OS.
......@@ -783,7 +786,8 @@ package System.OS_Lib is
-- Standard Error output is also redirected. Invalid_Pid is returned
-- if the program could not be spawned successfully.
--
-- "Non_Blocking_Spawn" should not be used in tasking applications.
-- Spawning processes from tasking programs is not recommended. See
-- "NOTE: Spawn in tasking programs" below.
--
-- This function will always return Invalid_Pid under VxWorks, since there
-- is no notion of executables under this OS.
......@@ -801,7 +805,8 @@ package System.OS_Lib is
-- file could not be created or if the program could not be spawned
-- successfully.
--
-- "Non_Blocking_Spawn" should not be used in tasking applications.
-- Spawning processes from tasking programs is not recommended. See
-- "NOTE: Spawn in tasking programs" below.
--
-- This function will always return Invalid_Pid under VxWorks, since there
-- is no notion of executables under this OS.
......@@ -827,6 +832,70 @@ package System.OS_Lib is
-- be freed by the programmer (when it is no longer needed) to avoid
-- memory leaks.
-------------------------------------
-- NOTE: Spawn in Tasking Programs --
-------------------------------------
-- Spawning processes in tasking programs using the above Spawn and
-- Non_Blocking_Spawn subprograms is not recommended, because there are
-- subtle interactions between creating a process and signals/locks that
-- can cause trouble. These issues are not specific to Ada; they depend
-- primarily on the operating system.
-- If you need to spawn processes in a tasking program, you will need to
-- understand the semantics of your operating system, and you are likely to
-- write non-portable code, because operating systems differ in this area.
-- The Spawn and Non_Blocking_Spawn subprograms call the following
-- operating system functions:
-- On Windows: spawnvp (blocking) or CreateProcess (non-blocking)
-- On Solaris: fork1, followed in the child process by execv
-- On other Unix-like systems, and on VMS: fork, followed in the child
-- process by execv.
-- On vxworks, nucleus, and RTX, spawning of processes is not supported
-- For details, look at the functions __gnat_portable_spawn and
-- __gnat_portable_no_block_spawn in adaint.c.
-- You should read the operating-system-specific documentation for the
-- above functions, paying special attention to subtle interactions with
-- threading, signals, locks, and file descriptors. Most of the issues are
-- related to the fact that on Unix, there is a window of time between fork
-- and execv; Windows does not have this problem, because spawning is done
-- in a single operation.
-- On Posix-compliant systems, such as Linux, fork duplicates just the
-- calling thread. (On Solaris, fork1 is the Posix-compliant version of
-- fork.)
-- You should avoid using signals while spawning. This includes signals
-- used internally by the Ada run-time system, such as timer signals used
-- to implement delay statements.
-- It is best to spawn any subprocesses very early, before the parent
-- process creates tasks, locks, or installs signal handlers. Certainly
-- avoid doing simultaneous spawns from multiple threads of the same
-- process.
-- There is no problem spawning a subprocess that uses tasking: the
-- problems are caused only by tasking in the parent.
-- If the parent is using tasking, and needs to spawn subprocesses at
-- arbitrary times, one technique is for the parent to spawn (very early)
-- a particular spawn-manager subprocess whose job is to spawn other
-- processes. The spawn-manager avoids tasking. The parent sends messages
-- to the spawn-manager requesting it to spawn processes, using whatever
-- inter-process communication mechanism you like, such as sockets.
-- In short, mixing spawning of subprocesses with tasking is a tricky
-- business, and should be avoided if possible, but if it is necessary,
-- the above guidelines should be followed, and you should beware of
-- portability problems.
-------------------
-- Miscellaneous --
-------------------
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment