database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-11-04 20:11:56 +03:00
Code Activity

Refactor EXEC_BACKEND code so that postmaster child processes reattach

Browse Source 
to shared memory as soon as possible, ie, right after read_backend_variables.
The effective difference from the original code is that this happens
before instead of after read_nondefault_variables(), which loads GUC
information and is apparently capable of expanding the backend's memory
allocation more than you'd think it should.  This should fix the
failure-to-attach-to-shared-memory reports we've been seeing on Windows.
Also clean up a few bits of unnecessarily grotty EXEC_BACKEND code.
This commit is contained in:
Tom Lane
2004-12-29 21:36:09 +00:00
parent e14018dc5e
commit eee5abce46
7 changed files with 179 additions and 124 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
src/backend/storage/ipc/ipci.c
View File

@@ -8,7 +8,7 @@
*
*
* IDENTIFICATION
* $PostgreSQL: pgsql/src/backend/storage/ipc/ipci.c,v 1.72 2004/09/29 15:15:55 tgl Exp $
* $PostgreSQL: pgsql/src/backend/storage/ipc/ipci.c,v 1.73 2004/12/29 21:36:06 tgl Exp $
*
*-------------------------------------------------------------------------
*/
@@ -19,6 +19,7 @@
#include "access/xlog.h"
#include "miscadmin.h"
#include "postmaster/bgwriter.h"
#include "postmaster/postmaster.h"
#include "storage/bufmgr.h"
#include "storage/freespace.h"
#include "storage/ipc.h"
@@ -37,14 +38,15 @@
* Creates and initializes shared memory and semaphores.
*
* This is called by the postmaster or by a standalone backend.
* It is also called by a backend forked from the postmaster under
* the EXEC_BACKEND case. (In the non EXEC_BACKEND case, backends
* start life already attached to shared memory.) The initialization
* functions are set up to simply "attach" to pre-existing shared memory
* structures in the latter case. We have to do that in order to
* initialize pointers in local memory that reference the shared structures.
* (In the non EXEC_BACKEND case, these pointer values are inherited via
* fork() from the postmaster.)
* It is also called by a backend forked from the postmaster in the
* EXEC_BACKEND case. In the latter case, the shared memory segment
* already exists and has been physically attached to, but we have to
* initialize pointers in local memory that reference the shared structures,
* because we didn't inherit the correct pointer values from the postmaster
* as we do in the fork() scenario. The easiest way to do that is to run
* through the same code as before. (Note that the called routines mostly
* check IsUnderPostmaster, rather than EXEC_BACKEND, to detect this case.
* This is a bit code-wasteful and could be cleaned up.)
*
* If "makePrivate" is true then we only need private memory, not shared
* memory. This is true for a standalone backend, false for a postmaster.
@@ -101,14 +103,16 @@ CreateSharedMemoryAndSemaphores(bool makePrivate,
else
{
/*
* Attach to the shmem segment. (this should only ever be reached
* by EXEC_BACKEND code, and only then with makePrivate == false)
* We are reattaching to an existing shared memory segment.
* This should only be reached in the EXEC_BACKEND case, and
* even then only with makePrivate == false.
*/
#ifdef EXEC_BACKEND
Assert(!makePrivate);
seghdr = PGSharedMemoryCreate(-1, makePrivate, 0);
Assert(UsedShmemSegAddr != NULL);
seghdr = UsedShmemSegAddr;
#else
Assert(false);
elog(PANIC, "should be attached to shared memory already");
#endif
}