Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
#include "../../kernel.h"
|
|
|
|
|
#include "../../proc.h"
|
2008-11-19 13:26:10 +01:00
|
|
|
#include "../../vm.h"
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
|
|
|
|
|
#include <minix/type.h>
|
2008-11-19 13:26:10 +01:00
|
|
|
#include <minix/syslib.h>
|
|
|
|
|
#include <minix/sysutil.h>
|
2009-05-15 19:07:36 +02:00
|
|
|
#include <minix/cpufeature.h>
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
#include <string.h>
|
|
|
|
|
|
2008-11-19 13:26:10 +01:00
|
|
|
#include <sys/vm_i386.h>
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
|
|
|
|
|
#include <minix/portio.h>
|
|
|
|
|
|
|
|
|
|
#include "proto.h"
|
2008-11-19 13:26:10 +01:00
|
|
|
#include "../../proto.h"
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
#include "../../proto.h"
|
2008-11-19 13:26:10 +01:00
|
|
|
#include "../../debug.h"
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
PRIVATE int psok = 0;
|
|
|
|
|
|
|
|
|
|
#define PROCPDEPTR(pr, pi) ((u32_t *) ((u8_t *) vm_pagedirs +\
|
|
|
|
|
I386_PAGE_SIZE * pr->p_nr + \
|
|
|
|
|
I386_VM_PT_ENT_SIZE * pi))
|
|
|
|
|
|
|
|
|
|
u8_t *vm_pagedirs = NULL;
|
|
|
|
|
|
|
|
|
|
#define NOPDE -1
|
|
|
|
|
#define PDEMASK(n) (1L << (n))
|
|
|
|
|
PUBLIC u32_t dirtypde;
|
|
|
|
|
#define WANT_FREEPDES (sizeof(dirtypde)*8-5)
|
|
|
|
|
PRIVATE int nfreepdes = 0, freepdes[WANT_FREEPDES], inusepde = NOPDE;
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
|
2009-05-12 13:35:01 +02:00
|
|
|
#define HASPT(procptr) ((procptr)->p_seg.p_cr3 != 0)
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
FORWARD _PROTOTYPE( u32_t phys_get32, (vir_bytes v) );
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
FORWARD _PROTOTYPE( void set_cr3, (void) );
|
|
|
|
|
FORWARD _PROTOTYPE( void vm_enable_paging, (void) );
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
/* *** Internal VM Functions *** */
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
PUBLIC void vm_init(struct proc *newptproc)
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
{
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
int i;
|
|
|
|
|
if(vm_running)
|
|
|
|
|
minix_panic("vm_init: vm_running", NO_NUM);
|
|
|
|
|
vm_set_cr3(newptproc);
|
|
|
|
|
level0(vm_enable_paging);
|
|
|
|
|
vm_running = 1;
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
}
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
#define TYPEDIRECT 0
|
|
|
|
|
#define TYPEPROCMAP 1
|
|
|
|
|
#define TYPEPHYS 2
|
|
|
|
|
|
|
|
|
|
/* This macro sets up a mapping from within the kernel's address
|
|
|
|
|
* space to any other area of memory, either straight physical
|
|
|
|
|
* memory (PROC == NULL) or a process view of memory, in 4MB chunks.
|
|
|
|
|
* It recognizes PROC having kernel address space as a special case.
|
|
|
|
|
*
|
|
|
|
|
* It sets PTR to the pointer within kernel address space at the start
|
|
|
|
|
* of the 4MB chunk, and OFFSET to the offset within that chunk
|
|
|
|
|
* that corresponds to LINADDR.
|
|
|
|
|
*
|
|
|
|
|
* It needs FREEPDE (available and addressable PDE within kernel
|
|
|
|
|
* address space), SEG (hardware segment), VIRT (in-datasegment
|
|
|
|
|
* address if known).
|
|
|
|
|
*/
|
|
|
|
|
#define CREATEPDE(PROC, PTR, LINADDR, REMAIN, BYTES, PDE, TYPE) { \
|
|
|
|
|
u32_t *pdeptr = NULL; \
|
|
|
|
|
int proc_pde_index; \
|
|
|
|
|
proc_pde_index = I386_VM_PDE(LINADDR); \
|
|
|
|
|
PDE = NOPDE; \
|
|
|
|
|
if((PROC) && (((PROC) == ptproc) || !HASPT(PROC))) { \
|
|
|
|
|
PTR = LINADDR; \
|
|
|
|
|
TYPE = TYPEDIRECT; \
|
|
|
|
|
} else { \
|
|
|
|
|
int fp; \
|
|
|
|
|
int mustinvl; \
|
|
|
|
|
u32_t pdeval, *pdevalptr, mask; \
|
|
|
|
|
phys_bytes offset; \
|
|
|
|
|
vmassert(psok); \
|
|
|
|
|
if(PROC) { \
|
|
|
|
|
TYPE = TYPEPROCMAP; \
|
|
|
|
|
vmassert(!iskernelp(PROC)); \
|
|
|
|
|
vmassert(HASPT(PROC)); \
|
|
|
|
|
pdeptr = PROCPDEPTR(PROC, proc_pde_index); \
|
|
|
|
|
pdeval = *pdeptr; \
|
|
|
|
|
} else { \
|
|
|
|
|
TYPE = TYPEPHYS; \
|
|
|
|
|
pdeval = (LINADDR & I386_VM_ADDR_MASK_4MB) | \
|
|
|
|
|
I386_VM_BIGPAGE | I386_VM_PRESENT | \
|
|
|
|
|
I386_VM_WRITE | I386_VM_USER; \
|
|
|
|
|
} \
|
|
|
|
|
for(fp = 0; fp < nfreepdes; fp++) { \
|
|
|
|
|
int k = freepdes[fp]; \
|
|
|
|
|
if(inusepde == k) \
|
|
|
|
|
continue; \
|
|
|
|
|
*PROCPDEPTR(ptproc, k) = 0; \
|
|
|
|
|
PDE = k; \
|
|
|
|
|
vmassert(k >= 0); \
|
|
|
|
|
vmassert(k < sizeof(dirtypde)*8); \
|
|
|
|
|
mask = PDEMASK(PDE); \
|
|
|
|
|
if(dirtypde & mask) \
|
|
|
|
|
continue; \
|
|
|
|
|
break; \
|
|
|
|
|
} \
|
|
|
|
|
vmassert(PDE != NOPDE); \
|
|
|
|
|
vmassert(mask); \
|
|
|
|
|
if(dirtypde & mask) { \
|
|
|
|
|
mustinvl = 1; \
|
|
|
|
|
} else { \
|
|
|
|
|
mustinvl = 0; \
|
|
|
|
|
} \
|
|
|
|
|
inusepde = PDE; \
|
|
|
|
|
*PROCPDEPTR(ptproc, PDE) = pdeval; \
|
|
|
|
|
offset = LINADDR & I386_VM_OFFSET_MASK_4MB; \
|
|
|
|
|
PTR = I386_BIG_PAGE_SIZE*PDE + offset; \
|
|
|
|
|
REMAIN = MIN(REMAIN, I386_BIG_PAGE_SIZE - offset); \
|
2009-09-28 17:47:01 +02:00
|
|
|
if(1 || mustinvl) { \
|
|
|
|
|
level0(reload_cr3); \
|
|
|
|
|
} \
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
} \
|
|
|
|
|
}
|
2009-05-12 13:35:01 +02:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
#define DONEPDE(PDE) { \
|
|
|
|
|
if(PDE != NOPDE) { \
|
|
|
|
|
vmassert(PDE > 0); \
|
|
|
|
|
vmassert(PDE < sizeof(dirtypde)*8); \
|
|
|
|
|
dirtypde |= PDEMASK(PDE); \
|
|
|
|
|
} \
|
|
|
|
|
}
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
#define WIPEPDE(PDE) { \
|
|
|
|
|
if(PDE != NOPDE) { \
|
|
|
|
|
vmassert(PDE > 0); \
|
|
|
|
|
vmassert(PDE < sizeof(dirtypde)*8); \
|
|
|
|
|
*PROCPDEPTR(ptproc, PDE) = 0; \
|
|
|
|
|
} \
|
|
|
|
|
}
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
/*===========================================================================*
|
|
|
|
|
* lin_lin_copy *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
int lin_lin_copy(struct proc *srcproc, vir_bytes srclinaddr,
|
|
|
|
|
struct proc *dstproc, vir_bytes dstlinaddr, vir_bytes bytes)
|
|
|
|
|
{
|
|
|
|
|
u32_t addr;
|
|
|
|
|
int o1, o2;
|
|
|
|
|
int procslot;
|
|
|
|
|
int firstloop = 1;
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
NOREC_ENTER(linlincopy);
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
vmassert(vm_running);
|
|
|
|
|
vmassert(nfreepdes >= 3);
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
vmassert(ptproc);
|
|
|
|
|
vmassert(proc_ptr);
|
|
|
|
|
vmassert(read_cr3() == ptproc->p_seg.p_cr3);
|
|
|
|
|
|
|
|
|
|
procslot = ptproc->p_nr;
|
|
|
|
|
|
|
|
|
|
vmassert(procslot >= 0 && procslot < I386_VM_DIR_ENTRIES);
|
|
|
|
|
|
|
|
|
|
while(bytes > 0) {
|
|
|
|
|
phys_bytes srcptr, dstptr;
|
|
|
|
|
vir_bytes chunk = bytes;
|
|
|
|
|
int srcpde, dstpde;
|
|
|
|
|
int srctype, dsttype;
|
|
|
|
|
|
|
|
|
|
/* Set up 4MB ranges. */
|
|
|
|
|
inusepde = NOPDE;
|
|
|
|
|
CREATEPDE(srcproc, srcptr, srclinaddr, chunk, bytes, srcpde, srctype);
|
|
|
|
|
CREATEPDE(dstproc, dstptr, dstlinaddr, chunk, bytes, dstpde, dsttype);
|
|
|
|
|
|
|
|
|
|
/* Copy pages. */
|
|
|
|
|
PHYS_COPY_CATCH(srcptr, dstptr, chunk, addr);
|
|
|
|
|
|
|
|
|
|
DONEPDE(srcpde);
|
|
|
|
|
DONEPDE(dstpde);
|
|
|
|
|
|
|
|
|
|
if(addr) {
|
|
|
|
|
/* If addr is nonzero, a page fault was caught. */
|
|
|
|
|
|
|
|
|
|
if(addr >= srcptr && addr < (srcptr + chunk)) {
|
|
|
|
|
WIPEPDE(srcpde);
|
|
|
|
|
WIPEPDE(dstpde);
|
|
|
|
|
NOREC_RETURN(linlincopy, EFAULT_SRC);
|
|
|
|
|
}
|
|
|
|
|
if(addr >= dstptr && addr < (dstptr + chunk)) {
|
|
|
|
|
WIPEPDE(srcpde);
|
|
|
|
|
WIPEPDE(dstpde);
|
|
|
|
|
NOREC_RETURN(linlincopy, EFAULT_DST);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
minix_panic("lin_lin_copy fault out of range", NO_NUM);
|
|
|
|
|
|
|
|
|
|
/* Not reached. */
|
|
|
|
|
NOREC_RETURN(linlincopy, EFAULT);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
WIPEPDE(srcpde);
|
|
|
|
|
WIPEPDE(dstpde);
|
|
|
|
|
|
|
|
|
|
/* Update counter and addresses for next iteration, if any. */
|
|
|
|
|
bytes -= chunk;
|
|
|
|
|
srclinaddr += chunk;
|
|
|
|
|
dstlinaddr += chunk;
|
|
|
|
|
|
|
|
|
|
firstloop = 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
NOREC_RETURN(linlincopy, OK);
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
}
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
PRIVATE u32_t phys_get32(addr)
|
|
|
|
|
phys_bytes addr;
|
|
|
|
|
{
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
u32_t v;
|
|
|
|
|
int r;
|
|
|
|
|
|
|
|
|
|
if(!vm_running) {
|
|
|
|
|
phys_copy(addr, vir2phys(&v), sizeof(v));
|
|
|
|
|
return v;
|
|
|
|
|
}
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
if((r=lin_lin_copy(NULL, addr,
|
|
|
|
|
proc_addr(SYSTEM), vir2phys(&v), sizeof(v))) != OK) {
|
|
|
|
|
minix_panic("lin_lin_copy for phys_get32 failed", r);
|
|
|
|
|
}
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
return v;
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
}
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
PRIVATE u32_t vm_cr3; /* temp arg to level0() func */
|
|
|
|
|
|
|
|
|
|
PUBLIC void vm_set_cr3(struct proc *newptproc)
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
{
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
int u = 0;
|
|
|
|
|
if(!intr_disabled()) { lock; u = 1; }
|
|
|
|
|
vm_cr3= newptproc->p_seg.p_cr3;
|
|
|
|
|
if(vm_cr3) {
|
|
|
|
|
vmassert(intr_disabled());
|
|
|
|
|
level0(set_cr3);
|
|
|
|
|
vmassert(intr_disabled());
|
|
|
|
|
ptproc = newptproc;
|
|
|
|
|
vmassert(intr_disabled());
|
|
|
|
|
}
|
|
|
|
|
if(u) { unlock; }
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
PRIVATE void set_cr3()
|
|
|
|
|
{
|
|
|
|
|
write_cr3(vm_cr3);
|
|
|
|
|
}
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
char *cr0_str(u32_t e)
|
|
|
|
|
{
|
|
|
|
|
static char str[80];
|
|
|
|
|
strcpy(str, "");
|
|
|
|
|
#define FLAG(v) do { if(e & (v)) { strcat(str, #v " "); e &= ~v; } } while(0)
|
|
|
|
|
FLAG(I386_CR0_PE);
|
|
|
|
|
FLAG(I386_CR0_MP);
|
|
|
|
|
FLAG(I386_CR0_EM);
|
|
|
|
|
FLAG(I386_CR0_TS);
|
|
|
|
|
FLAG(I386_CR0_ET);
|
|
|
|
|
FLAG(I386_CR0_PG);
|
|
|
|
|
FLAG(I386_CR0_WP);
|
|
|
|
|
if(e) { strcat(str, " (++)"); }
|
|
|
|
|
return str;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
char *cr4_str(u32_t e)
|
|
|
|
|
{
|
|
|
|
|
static char str[80];
|
|
|
|
|
strcpy(str, "");
|
|
|
|
|
FLAG(I386_CR4_VME);
|
|
|
|
|
FLAG(I386_CR4_PVI);
|
|
|
|
|
FLAG(I386_CR4_TSD);
|
|
|
|
|
FLAG(I386_CR4_DE);
|
|
|
|
|
FLAG(I386_CR4_PSE);
|
|
|
|
|
FLAG(I386_CR4_PAE);
|
|
|
|
|
FLAG(I386_CR4_MCE);
|
|
|
|
|
FLAG(I386_CR4_PGE);
|
|
|
|
|
if(e) { strcat(str, " (++)"); }
|
|
|
|
|
return str;
|
|
|
|
|
}
|
|
|
|
|
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
PRIVATE void vm_enable_paging(void)
|
|
|
|
|
{
|
2009-04-23 17:11:16 +02:00
|
|
|
u32_t cr0, cr4;
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
int pgeok;
|
2009-04-23 17:11:16 +02:00
|
|
|
|
2009-05-15 19:07:36 +02:00
|
|
|
psok = _cpufeature(_CPUF_I386_PSE);
|
|
|
|
|
pgeok = _cpufeature(_CPUF_I386_PGE);
|
2009-04-23 17:11:16 +02:00
|
|
|
|
|
|
|
|
cr0= read_cr0();
|
|
|
|
|
cr4= read_cr4();
|
|
|
|
|
|
|
|
|
|
/* First clear PG and PGE flag, as PGE must be enabled after PG. */
|
|
|
|
|
write_cr0(cr0 & ~I386_CR0_PG);
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
write_cr4(cr4 & ~(I386_CR4_PGE | I386_CR4_PSE));
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
|
|
|
|
|
cr0= read_cr0();
|
2009-04-23 17:11:16 +02:00
|
|
|
cr4= read_cr4();
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
/* Our first page table contains 4MB entries. */
|
|
|
|
|
if(psok)
|
|
|
|
|
cr4 |= I386_CR4_PSE;
|
|
|
|
|
|
|
|
|
|
write_cr4(cr4);
|
|
|
|
|
|
2009-04-23 17:11:16 +02:00
|
|
|
/* First enable paging, then enable global page flag. */
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
cr0 |= I386_CR0_PG;
|
|
|
|
|
write_cr0(cr0 );
|
|
|
|
|
cr0 |= I386_CR0_WP;
|
|
|
|
|
write_cr0(cr0);
|
2009-05-15 19:07:36 +02:00
|
|
|
|
|
|
|
|
/* May we enable these features? */
|
|
|
|
|
if(pgeok)
|
|
|
|
|
cr4 |= I386_CR4_PGE;
|
|
|
|
|
|
|
|
|
|
write_cr4(cr4);
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
PUBLIC vir_bytes alloc_remote_segment(u32_t *selector,
|
|
|
|
|
segframe_t *segments, int index, phys_bytes phys, vir_bytes size,
|
|
|
|
|
int priv)
|
|
|
|
|
{
|
|
|
|
|
phys_bytes offset = 0;
|
|
|
|
|
/* Check if the segment size can be recorded in bytes, that is, check
|
|
|
|
|
* if descriptor's limit field can delimited the allowed memory region
|
|
|
|
|
* precisely. This works up to 1MB. If the size is larger, 4K pages
|
|
|
|
|
* instead of bytes are used.
|
|
|
|
|
*/
|
|
|
|
|
if (size < BYTE_GRAN_MAX) {
|
|
|
|
|
init_dataseg(&segments->p_ldt[EXTRA_LDT_INDEX+index],
|
|
|
|
|
phys, size, priv);
|
|
|
|
|
*selector = ((EXTRA_LDT_INDEX+index)*0x08) | (1*0x04) | priv;
|
|
|
|
|
offset = 0;
|
|
|
|
|
} else {
|
|
|
|
|
init_dataseg(&segments->p_ldt[EXTRA_LDT_INDEX+index],
|
|
|
|
|
phys & ~0xFFFF, 0, priv);
|
|
|
|
|
*selector = ((EXTRA_LDT_INDEX+index)*0x08) | (1*0x04) | priv;
|
|
|
|
|
offset = phys & 0xFFFF;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return offset;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
PUBLIC phys_bytes umap_remote(struct proc* rp, int seg,
|
|
|
|
|
vir_bytes vir_addr, vir_bytes bytes)
|
|
|
|
|
{
|
|
|
|
|
/* Calculate the physical memory address for a given virtual address. */
|
|
|
|
|
struct far_mem *fm;
|
|
|
|
|
|
2008-11-19 13:26:10 +01:00
|
|
|
#if 0
|
|
|
|
|
if(rp->p_misc_flags & MF_FULLVM) return 0;
|
|
|
|
|
#endif
|
|
|
|
|
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
if (bytes <= 0) return( (phys_bytes) 0);
|
|
|
|
|
if (seg < 0 || seg >= NR_REMOTE_SEGS) return( (phys_bytes) 0);
|
|
|
|
|
|
|
|
|
|
fm = &rp->p_priv->s_farmem[seg];
|
|
|
|
|
if (! fm->in_use) return( (phys_bytes) 0);
|
|
|
|
|
if (vir_addr + bytes > fm->mem_len) return( (phys_bytes) 0);
|
|
|
|
|
|
|
|
|
|
return(fm->mem_phys + (phys_bytes) vir_addr);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*===========================================================================*
|
|
|
|
|
* umap_local *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
PUBLIC phys_bytes umap_local(rp, seg, vir_addr, bytes)
|
|
|
|
|
register struct proc *rp; /* pointer to proc table entry for process */
|
|
|
|
|
int seg; /* T, D, or S segment */
|
|
|
|
|
vir_bytes vir_addr; /* virtual address in bytes within the seg */
|
|
|
|
|
vir_bytes bytes; /* # of bytes to be copied */
|
|
|
|
|
{
|
|
|
|
|
/* Calculate the physical memory address for a given virtual address. */
|
|
|
|
|
vir_clicks vc; /* the virtual address in clicks */
|
|
|
|
|
phys_bytes pa; /* intermediate variables as phys_bytes */
|
|
|
|
|
phys_bytes seg_base;
|
|
|
|
|
|
2008-11-19 13:26:10 +01:00
|
|
|
if(seg != T && seg != D && seg != S)
|
|
|
|
|
minix_panic("umap_local: wrong seg", seg);
|
|
|
|
|
|
Split of architecture-dependent and -independent functions for i386,
mainly in the kernel and headers. This split based on work by
Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture
port.
. kernel does not program the interrupt controller directly, do any
other architecture-dependent operations, or contain assembly any more,
but uses architecture-dependent functions in arch/$(ARCH)/.
. architecture-dependent constants and types defined in arch/$(ARCH)/include.
. <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now,
architecture-independent functions.
. int86, sdevio, readbios, and iopenable are now i386-specific kernel calls
and live in arch/i386/do_* now.
. i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have
gone, and 'machine.protected' is gone (and always taken to be 1 in i386).
If 86 support is to return, it should be a new architecture.
. prototypes for the architecture-dependent functions defined in
kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h
. /etc/make.conf included in makefiles and shell scripts that need to
know the building architecture; it defines ARCH=<arch>, currently only
i386.
. some basic per-architecture build support outside of the kernel (lib)
. in clock.c, only dequeue a process if it was ready
. fixes for new include files
files deleted:
. mpx/klib.s - only for choosing between mpx/klib86 and -386
. klib86.s - only for 86
i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/:
. mpx386.s (entry point)
. klib386.s
. sconst.h
. exception.c
. protect.c
. protect.h
. i8269.c
2006-12-22 16:22:27 +01:00
|
|
|
if (bytes <= 0) return( (phys_bytes) 0);
|
|
|
|
|
if (vir_addr + bytes <= vir_addr) return 0; /* overflow */
|
|
|
|
|
vc = (vir_addr + bytes - 1) >> CLICK_SHIFT; /* last click of data */
|
|
|
|
|
|
|
|
|
|
if (seg != T)
|
|
|
|
|
seg = (vc < rp->p_memmap[D].mem_vir + rp->p_memmap[D].mem_len ? D : S);
|
|
|
|
|
|
|
|
|
|
if ((vir_addr>>CLICK_SHIFT) >= rp->p_memmap[seg].mem_vir +
|
|
|
|
|
rp->p_memmap[seg].mem_len) return( (phys_bytes) 0 );
|
|
|
|
|
|
|
|
|
|
if (vc >= rp->p_memmap[seg].mem_vir +
|
|
|
|
|
rp->p_memmap[seg].mem_len) return( (phys_bytes) 0 );
|
|
|
|
|
|
|
|
|
|
seg_base = (phys_bytes) rp->p_memmap[seg].mem_phys;
|
|
|
|
|
seg_base = seg_base << CLICK_SHIFT; /* segment origin in bytes */
|
|
|
|
|
pa = (phys_bytes) vir_addr;
|
|
|
|
|
pa -= rp->p_memmap[seg].mem_vir << CLICK_SHIFT;
|
|
|
|
|
return(seg_base + pa);
|
|
|
|
|
}
|
|
|
|
|
|
2008-11-19 13:26:10 +01:00
|
|
|
/*===========================================================================*
|
|
|
|
|
* umap_virtual *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
PUBLIC phys_bytes umap_virtual(rp, seg, vir_addr, bytes)
|
|
|
|
|
register struct proc *rp; /* pointer to proc table entry for process */
|
|
|
|
|
int seg; /* T, D, or S segment */
|
|
|
|
|
vir_bytes vir_addr; /* virtual address in bytes within the seg */
|
|
|
|
|
vir_bytes bytes; /* # of bytes to be copied */
|
|
|
|
|
{
|
|
|
|
|
vir_bytes linear;
|
|
|
|
|
u32_t phys = 0;
|
|
|
|
|
|
|
|
|
|
if(seg == MEM_GRANT) {
|
|
|
|
|
phys = umap_grant(rp, vir_addr, bytes);
|
|
|
|
|
} else {
|
|
|
|
|
if(!(linear = umap_local(rp, seg, vir_addr, bytes))) {
|
|
|
|
|
kprintf("SYSTEM:umap_virtual: umap_local failed\n");
|
|
|
|
|
phys = 0;
|
|
|
|
|
} else {
|
|
|
|
|
if(vm_lookup(rp, linear, &phys, NULL) != OK) {
|
|
|
|
|
kprintf("SYSTEM:umap_virtual: vm_lookup of %s: seg 0x%lx: 0x%lx failed\n", rp->p_name, seg, vir_addr);
|
|
|
|
|
phys = 0;
|
|
|
|
|
}
|
|
|
|
|
if(phys == 0)
|
|
|
|
|
minix_panic("vm_lookup returned phys", phys);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if(phys == 0) {
|
|
|
|
|
kprintf("SYSTEM:umap_virtual: lookup failed\n");
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* Now make sure addresses are contiguous in physical memory
|
|
|
|
|
* so that the umap makes sense.
|
|
|
|
|
*/
|
|
|
|
|
if(bytes > 0 && !vm_contiguous(rp, linear, bytes)) {
|
|
|
|
|
kprintf("umap_virtual: %s: %d at 0x%lx (vir 0x%lx) not contiguous\n",
|
|
|
|
|
rp->p_name, bytes, linear, vir_addr);
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* phys must be larger than 0 (or the caller will think the call
|
|
|
|
|
* failed), and address must not cross a page boundary.
|
|
|
|
|
*/
|
|
|
|
|
vmassert(phys);
|
|
|
|
|
|
|
|
|
|
return phys;
|
|
|
|
|
}
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
|
2008-11-19 13:26:10 +01:00
|
|
|
/*===========================================================================*
|
|
|
|
|
* vm_lookup *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
PUBLIC int vm_lookup(struct proc *proc, vir_bytes virtual, vir_bytes *physical, u32_t *ptent)
|
|
|
|
|
{
|
|
|
|
|
u32_t *root, *pt;
|
|
|
|
|
int pde, pte;
|
|
|
|
|
u32_t pde_v, pte_v;
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
NOREC_ENTER(vmlookup);
|
2008-11-19 13:26:10 +01:00
|
|
|
|
|
|
|
|
vmassert(proc);
|
|
|
|
|
vmassert(physical);
|
|
|
|
|
vmassert(!(proc->p_rts_flags & SLOT_FREE));
|
|
|
|
|
|
2009-05-12 13:35:01 +02:00
|
|
|
if(!HASPT(proc)) {
|
|
|
|
|
*physical = virtual;
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
NOREC_RETURN(vmlookup, OK);
|
2009-05-12 13:35:01 +02:00
|
|
|
}
|
|
|
|
|
|
2008-11-19 13:26:10 +01:00
|
|
|
/* Retrieve page directory entry. */
|
|
|
|
|
root = (u32_t *) proc->p_seg.p_cr3;
|
|
|
|
|
vmassert(!((u32_t) root % I386_PAGE_SIZE));
|
|
|
|
|
pde = I386_VM_PDE(virtual);
|
|
|
|
|
vmassert(pde >= 0 && pde < I386_VM_DIR_ENTRIES);
|
|
|
|
|
pde_v = phys_get32((u32_t) (root + pde));
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
|
2008-11-19 13:26:10 +01:00
|
|
|
if(!(pde_v & I386_VM_PRESENT)) {
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
NOREC_RETURN(vmlookup, EFAULT);
|
2008-11-19 13:26:10 +01:00
|
|
|
}
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
/* We don't expect to ever see this. */
|
|
|
|
|
if(pde_v & I386_VM_BIGPAGE) {
|
|
|
|
|
*physical = pde_v & I386_VM_ADDR_MASK_4MB;
|
|
|
|
|
if(ptent) *ptent = pde_v;
|
|
|
|
|
*physical += virtual & I386_VM_OFFSET_MASK_4MB;
|
|
|
|
|
} else {
|
|
|
|
|
/* Retrieve page table entry. */
|
|
|
|
|
pt = (u32_t *) I386_VM_PFA(pde_v);
|
|
|
|
|
vmassert(!((u32_t) pt % I386_PAGE_SIZE));
|
|
|
|
|
pte = I386_VM_PTE(virtual);
|
|
|
|
|
vmassert(pte >= 0 && pte < I386_VM_PT_ENTRIES);
|
|
|
|
|
pte_v = phys_get32((u32_t) (pt + pte));
|
|
|
|
|
if(!(pte_v & I386_VM_PRESENT)) {
|
|
|
|
|
NOREC_RETURN(vmlookup, EFAULT);
|
|
|
|
|
}
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
if(ptent) *ptent = pte_v;
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
/* Actual address now known; retrieve it and add page offset. */
|
|
|
|
|
*physical = I386_VM_PFA(pte_v);
|
|
|
|
|
*physical += virtual % I386_PAGE_SIZE;
|
|
|
|
|
}
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
NOREC_RETURN(vmlookup, OK);
|
2008-11-19 13:26:10 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* From virtual address v in process p,
|
|
|
|
|
* lookup physical address and assign it to d.
|
|
|
|
|
* If p is NULL, assume it's already a physical address.
|
|
|
|
|
*/
|
|
|
|
|
#define LOOKUP(d, p, v, flagsp) { \
|
|
|
|
|
int r; \
|
|
|
|
|
if(!(p)) { (d) = (v); } \
|
|
|
|
|
else { \
|
|
|
|
|
if((r=vm_lookup((p), (v), &(d), flagsp)) != OK) { \
|
|
|
|
|
kprintf("vm_copy: lookup failed of 0x%lx in %d (%s)\n"\
|
|
|
|
|
"kernel stacktrace: ", (v), (p)->p_endpoint, \
|
|
|
|
|
(p)->p_name); \
|
|
|
|
|
util_stacktrace(); \
|
|
|
|
|
return r; \
|
|
|
|
|
} } }
|
|
|
|
|
|
|
|
|
|
/*===========================================================================*
|
|
|
|
|
* vm_contiguous *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
PUBLIC int vm_contiguous(struct proc *targetproc, u32_t vir_buf, size_t bytes)
|
|
|
|
|
{
|
|
|
|
|
int first = 1, r, boundaries = 0;
|
|
|
|
|
u32_t prev_phys, po;
|
|
|
|
|
u32_t prev_vir;
|
|
|
|
|
|
|
|
|
|
vmassert(targetproc);
|
|
|
|
|
vmassert(bytes > 0);
|
2009-05-12 13:35:01 +02:00
|
|
|
|
|
|
|
|
if(!HASPT(targetproc))
|
|
|
|
|
return 1;
|
2008-11-19 13:26:10 +01:00
|
|
|
|
|
|
|
|
/* Start and end at page boundary to make logic simpler. */
|
|
|
|
|
po = vir_buf % I386_PAGE_SIZE;
|
|
|
|
|
if(po > 0) {
|
|
|
|
|
bytes += po;
|
|
|
|
|
vir_buf -= po;
|
|
|
|
|
}
|
|
|
|
|
po = (vir_buf + bytes) % I386_PAGE_SIZE;
|
|
|
|
|
if(po > 0)
|
|
|
|
|
bytes += I386_PAGE_SIZE - po;
|
|
|
|
|
|
|
|
|
|
/* Keep going as long as we cross a page boundary. */
|
|
|
|
|
while(bytes > 0) {
|
|
|
|
|
u32_t phys;
|
|
|
|
|
|
|
|
|
|
if((r=vm_lookup(targetproc, vir_buf, &phys, NULL)) != OK) {
|
|
|
|
|
kprintf("vm_contiguous: vm_lookup failed, %d\n", r);
|
|
|
|
|
kprintf("kernel stack: ");
|
|
|
|
|
util_stacktrace();
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if(!first) {
|
|
|
|
|
if(prev_phys+I386_PAGE_SIZE != phys) {
|
|
|
|
|
kprintf("vm_contiguous: no (0x%lx, 0x%lx)\n",
|
|
|
|
|
prev_phys, phys);
|
|
|
|
|
kprintf("kernel stack: ");
|
|
|
|
|
util_stacktrace();
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
first = 0;
|
|
|
|
|
|
|
|
|
|
prev_phys = phys;
|
|
|
|
|
prev_vir = vir_buf;
|
|
|
|
|
vir_buf += I386_PAGE_SIZE;
|
|
|
|
|
bytes -= I386_PAGE_SIZE;
|
|
|
|
|
boundaries++;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*===========================================================================*
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
* vm_suspend *
|
2008-11-19 13:26:10 +01:00
|
|
|
*===========================================================================*/
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
PUBLIC int vm_suspend(struct proc *caller, struct proc *target,
|
|
|
|
|
vir_bytes linaddr, vir_bytes len, int wrflag, int type)
|
2008-11-19 13:26:10 +01:00
|
|
|
{
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
/* This range is not OK for this process. Set parameters
|
|
|
|
|
* of the request and notify VM about the pending request.
|
|
|
|
|
*/
|
|
|
|
|
vmassert(!RTS_ISSET(caller, VMREQUEST));
|
|
|
|
|
vmassert(!RTS_ISSET(target, VMREQUEST));
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
RTS_LOCK_SET(caller, VMREQUEST);
|
2008-12-18 15:30:55 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
#if DEBUG_VMASSERT
|
|
|
|
|
caller->p_vmrequest.stacktrace[0] = '\0';
|
|
|
|
|
util_stacktrace_strcat(caller->p_vmrequest.stacktrace);
|
2008-11-19 13:26:10 +01:00
|
|
|
#endif
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
caller->p_vmrequest.writeflag = 1;
|
|
|
|
|
caller->p_vmrequest.start = linaddr;
|
|
|
|
|
caller->p_vmrequest.length = len;
|
|
|
|
|
caller->p_vmrequest.who = target->p_endpoint;
|
|
|
|
|
caller->p_vmrequest.type = type;
|
|
|
|
|
|
|
|
|
|
/* Connect caller on vmrequest wait queue. */
|
|
|
|
|
if(!(caller->p_vmrequest.nextrequestor = vmrequest))
|
|
|
|
|
mini_notify(proc_addr(SYSTEM), VM_PROC_NR);
|
|
|
|
|
vmrequest = caller;
|
|
|
|
|
}
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
/*===========================================================================*
|
|
|
|
|
* delivermsg *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
int delivermsg(struct proc *rp)
|
|
|
|
|
{
|
|
|
|
|
phys_bytes addr;
|
|
|
|
|
int r;
|
|
|
|
|
NOREC_ENTER(deliver);
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
vmassert(rp->p_misc_flags & MF_DELIVERMSG);
|
|
|
|
|
vmassert(rp->p_delivermsg.m_source != NONE);
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
vmassert(rp->p_delivermsg_lin);
|
|
|
|
|
#if DEBUG_VMASSERT
|
|
|
|
|
if(rp->p_delivermsg_lin !=
|
|
|
|
|
umap_local(rp, D, rp->p_delivermsg_vir, sizeof(message))) {
|
|
|
|
|
printf("vir: 0x%lx lin was: 0x%lx umap now: 0x%lx\n",
|
|
|
|
|
rp->p_delivermsg_vir, rp->p_delivermsg_lin,
|
|
|
|
|
umap_local(rp, D, rp->p_delivermsg_vir, sizeof(message)));
|
|
|
|
|
minix_panic("that's wrong", NO_NUM);
|
|
|
|
|
}
|
2008-11-19 13:26:10 +01:00
|
|
|
|
2008-12-18 15:30:55 +01:00
|
|
|
#endif
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
vm_set_cr3(rp);
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
PHYS_COPY_CATCH(vir2phys(&rp->p_delivermsg),
|
|
|
|
|
rp->p_delivermsg_lin, sizeof(message), addr);
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
if(addr) {
|
|
|
|
|
vm_suspend(rp, rp, rp->p_delivermsg_lin, sizeof(message), 1,
|
|
|
|
|
VMSTYPE_DELIVERMSG);
|
|
|
|
|
r = VMSUSPEND;
|
|
|
|
|
} else {
|
|
|
|
|
#if DEBUG_VMASSERT
|
|
|
|
|
rp->p_delivermsg.m_source = NONE;
|
|
|
|
|
rp->p_delivermsg_lin = 0;
|
|
|
|
|
#endif
|
|
|
|
|
rp->p_misc_flags &= ~MF_DELIVERMSG;
|
|
|
|
|
r = OK;
|
2008-11-19 13:26:10 +01:00
|
|
|
}
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
NOREC_RETURN(deliver, r);
|
2008-11-19 13:26:10 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
char *flagstr(u32_t e, int dir)
|
|
|
|
|
{
|
|
|
|
|
static char str[80];
|
|
|
|
|
strcpy(str, "");
|
|
|
|
|
FLAG(I386_VM_PRESENT);
|
|
|
|
|
FLAG(I386_VM_WRITE);
|
|
|
|
|
FLAG(I386_VM_USER);
|
|
|
|
|
FLAG(I386_VM_PWT);
|
|
|
|
|
FLAG(I386_VM_PCD);
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
FLAG(I386_VM_GLOBAL);
|
2008-11-19 13:26:10 +01:00
|
|
|
if(dir)
|
|
|
|
|
FLAG(I386_VM_BIGPAGE); /* Page directory entry only */
|
|
|
|
|
else
|
|
|
|
|
FLAG(I386_VM_DIRTY); /* Page table entry only */
|
|
|
|
|
return str;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void vm_pt_print(u32_t *pagetable, u32_t v)
|
|
|
|
|
{
|
|
|
|
|
int pte, l = 0;
|
|
|
|
|
int col = 0;
|
|
|
|
|
|
|
|
|
|
vmassert(!((u32_t) pagetable % I386_PAGE_SIZE));
|
|
|
|
|
|
|
|
|
|
for(pte = 0; pte < I386_VM_PT_ENTRIES; pte++) {
|
|
|
|
|
u32_t pte_v, pfa;
|
|
|
|
|
pte_v = phys_get32((u32_t) (pagetable + pte));
|
|
|
|
|
if(!(pte_v & I386_VM_PRESENT))
|
|
|
|
|
continue;
|
|
|
|
|
pfa = I386_VM_PFA(pte_v);
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
kprintf("%4d:%08lx:%08lx %2s ",
|
|
|
|
|
pte, v + I386_PAGE_SIZE*pte, pfa,
|
|
|
|
|
(pte_v & I386_VM_WRITE) ? "rw":"RO");
|
2008-11-19 13:26:10 +01:00
|
|
|
col++;
|
|
|
|
|
if(col == 3) { kprintf("\n"); col = 0; }
|
|
|
|
|
}
|
|
|
|
|
if(col > 0) kprintf("\n");
|
|
|
|
|
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void vm_print(u32_t *root)
|
|
|
|
|
{
|
|
|
|
|
int pde;
|
|
|
|
|
|
|
|
|
|
vmassert(!((u32_t) root % I386_PAGE_SIZE));
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
printf("page table 0x%lx:\n", root);
|
|
|
|
|
|
|
|
|
|
for(pde = 10; pde < I386_VM_DIR_ENTRIES; pde++) {
|
2008-11-19 13:26:10 +01:00
|
|
|
u32_t pde_v;
|
|
|
|
|
u32_t *pte_a;
|
|
|
|
|
pde_v = phys_get32((u32_t) (root + pde));
|
|
|
|
|
if(!(pde_v & I386_VM_PRESENT))
|
|
|
|
|
continue;
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
if(pde_v & I386_VM_BIGPAGE) {
|
|
|
|
|
kprintf("%4d: 0x%lx, flags %s\n",
|
|
|
|
|
pde, I386_VM_PFA(pde_v), flagstr(pde_v, 1));
|
|
|
|
|
} else {
|
|
|
|
|
pte_a = (u32_t *) I386_VM_PFA(pde_v);
|
|
|
|
|
kprintf("%4d: pt %08lx %s\n",
|
|
|
|
|
pde, pte_a, flagstr(pde_v, 1));
|
|
|
|
|
vm_pt_print(pte_a, pde * I386_VM_PT_ENTRIES * I386_PAGE_SIZE);
|
|
|
|
|
kprintf("\n");
|
|
|
|
|
}
|
2008-11-19 13:26:10 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
u32_t thecr3;
|
|
|
|
|
|
|
|
|
|
u32_t read_cr3(void)
|
|
|
|
|
{
|
|
|
|
|
level0(getcr3val);
|
|
|
|
|
return thecr3;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/*===========================================================================*
|
|
|
|
|
* lin_memset *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
int vm_phys_memset(phys_bytes ph, u8_t c, phys_bytes bytes)
|
|
|
|
|
{
|
|
|
|
|
char *v;
|
|
|
|
|
u32_t p;
|
|
|
|
|
NOREC_ENTER(physmemset);
|
|
|
|
|
|
|
|
|
|
p = c | (c << 8) | (c << 16) | (c << 24);
|
|
|
|
|
|
|
|
|
|
if(!vm_running) {
|
|
|
|
|
phys_memset(ph, p, bytes);
|
|
|
|
|
NOREC_RETURN(physmemset, OK);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
vmassert(nfreepdes >= 3);
|
|
|
|
|
|
|
|
|
|
/* With VM, we have to map in the physical memory.
|
|
|
|
|
* We can do this 4MB at a time.
|
|
|
|
|
*/
|
|
|
|
|
while(bytes > 0) {
|
|
|
|
|
int pde, t;
|
|
|
|
|
vir_bytes chunk = bytes;
|
|
|
|
|
phys_bytes ptr;
|
|
|
|
|
inusepde = NOPDE;
|
|
|
|
|
CREATEPDE(((struct proc *) NULL), ptr, ph, chunk, bytes, pde, t);
|
|
|
|
|
/* We can memset as many bytes as we have remaining,
|
|
|
|
|
* or as many as remain in the 4MB chunk we mapped in.
|
|
|
|
|
*/
|
|
|
|
|
phys_memset(ptr, p, chunk);
|
|
|
|
|
DONEPDE(pde);
|
|
|
|
|
bytes -= chunk;
|
|
|
|
|
ph += chunk;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
NOREC_RETURN(physmemset, OK);
|
|
|
|
|
}
|
|
|
|
|
|
2008-11-19 13:26:10 +01:00
|
|
|
/*===========================================================================*
|
|
|
|
|
* virtual_copy_f *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
PUBLIC int virtual_copy_f(src_addr, dst_addr, bytes, vmcheck)
|
|
|
|
|
struct vir_addr *src_addr; /* source virtual address */
|
|
|
|
|
struct vir_addr *dst_addr; /* destination virtual address */
|
|
|
|
|
vir_bytes bytes; /* # of bytes to copy */
|
|
|
|
|
int vmcheck; /* if nonzero, can return VMSUSPEND */
|
|
|
|
|
{
|
|
|
|
|
/* Copy bytes from virtual address src_addr to virtual address dst_addr.
|
|
|
|
|
* Virtual addresses can be in ABS, LOCAL_SEG, REMOTE_SEG, or BIOS_SEG.
|
|
|
|
|
*/
|
|
|
|
|
struct vir_addr *vir_addr[2]; /* virtual source and destination address */
|
|
|
|
|
phys_bytes phys_addr[2]; /* absolute source and destination */
|
|
|
|
|
int seg_index;
|
|
|
|
|
int i, r;
|
|
|
|
|
struct proc *procs[2];
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
NOREC_ENTER(virtualcopy);
|
2008-11-19 13:26:10 +01:00
|
|
|
|
|
|
|
|
/* Check copy count. */
|
|
|
|
|
if (bytes <= 0) return(EDOM);
|
|
|
|
|
|
|
|
|
|
/* Do some more checks and map virtual addresses to physical addresses. */
|
|
|
|
|
vir_addr[_SRC_] = src_addr;
|
|
|
|
|
vir_addr[_DST_] = dst_addr;
|
|
|
|
|
|
|
|
|
|
for (i=_SRC_; i<=_DST_; i++) {
|
|
|
|
|
int proc_nr, type;
|
|
|
|
|
struct proc *p;
|
|
|
|
|
|
|
|
|
|
type = vir_addr[i]->segment & SEGMENT_TYPE;
|
|
|
|
|
if((type != PHYS_SEG && type != BIOS_SEG) &&
|
|
|
|
|
isokendpt(vir_addr[i]->proc_nr_e, &proc_nr))
|
|
|
|
|
p = proc_addr(proc_nr);
|
|
|
|
|
else
|
|
|
|
|
p = NULL;
|
|
|
|
|
|
|
|
|
|
procs[i] = p;
|
|
|
|
|
|
|
|
|
|
/* Get physical address. */
|
|
|
|
|
switch(type) {
|
|
|
|
|
case LOCAL_SEG:
|
|
|
|
|
case LOCAL_VM_SEG:
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
if(!p) {
|
|
|
|
|
NOREC_RETURN(virtualcopy, EDEADSRCDST);
|
|
|
|
|
}
|
2008-11-19 13:26:10 +01:00
|
|
|
seg_index = vir_addr[i]->segment & SEGMENT_INDEX;
|
|
|
|
|
if(type == LOCAL_SEG)
|
|
|
|
|
phys_addr[i] = umap_local(p, seg_index, vir_addr[i]->offset,
|
|
|
|
|
bytes);
|
|
|
|
|
else
|
|
|
|
|
phys_addr[i] = umap_virtual(p, seg_index, vir_addr[i]->offset,
|
|
|
|
|
bytes);
|
|
|
|
|
if(phys_addr[i] == 0) {
|
|
|
|
|
kprintf("virtual_copy: map 0x%x failed for %s seg %d, "
|
|
|
|
|
"offset %lx, len %d, i %d\n",
|
|
|
|
|
type, p->p_name, seg_index, vir_addr[i]->offset,
|
|
|
|
|
bytes, i);
|
|
|
|
|
}
|
|
|
|
|
break;
|
|
|
|
|
case REMOTE_SEG:
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
if(!p) {
|
|
|
|
|
NOREC_RETURN(virtualcopy, EDEADSRCDST);
|
|
|
|
|
}
|
2008-11-19 13:26:10 +01:00
|
|
|
seg_index = vir_addr[i]->segment & SEGMENT_INDEX;
|
|
|
|
|
phys_addr[i] = umap_remote(p, seg_index, vir_addr[i]->offset, bytes);
|
|
|
|
|
break;
|
|
|
|
|
#if _MINIX_CHIP == _CHIP_INTEL
|
|
|
|
|
case BIOS_SEG:
|
|
|
|
|
phys_addr[i] = umap_bios(vir_addr[i]->offset, bytes );
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
case PHYS_SEG:
|
|
|
|
|
phys_addr[i] = vir_addr[i]->offset;
|
|
|
|
|
break;
|
|
|
|
|
default:
|
|
|
|
|
kprintf("virtual_copy: strange type 0x%x\n", type);
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
NOREC_RETURN(virtualcopy, EINVAL);
|
2008-11-19 13:26:10 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* Check if mapping succeeded. */
|
|
|
|
|
if (phys_addr[i] <= 0 && vir_addr[i]->segment != PHYS_SEG) {
|
|
|
|
|
kprintf("virtual_copy EFAULT\n");
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
NOREC_RETURN(virtualcopy, EFAULT);
|
2008-11-19 13:26:10 +01:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
if(vm_running) {
|
|
|
|
|
int r;
|
|
|
|
|
struct proc *caller;
|
2008-11-19 13:26:10 +01:00
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
caller = proc_addr(who_p);
|
|
|
|
|
|
|
|
|
|
if(RTS_ISSET(caller, VMREQUEST)) {
|
|
|
|
|
struct proc *target;
|
|
|
|
|
int pn;
|
|
|
|
|
vmassert(caller->p_vmrequest.vmresult != VMSUSPEND);
|
|
|
|
|
RTS_LOCK_UNSET(caller, VMREQUEST);
|
|
|
|
|
if(caller->p_vmrequest.vmresult != OK) {
|
|
|
|
|
printf("virtual_copy: returning VM error %d\n",
|
|
|
|
|
caller->p_vmrequest.vmresult);
|
|
|
|
|
NOREC_RETURN(virtualcopy, caller->p_vmrequest.vmresult);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if((r=lin_lin_copy(procs[_SRC_], phys_addr[_SRC_],
|
|
|
|
|
procs[_DST_], phys_addr[_DST_], bytes)) != OK) {
|
|
|
|
|
struct proc *target;
|
|
|
|
|
int wr;
|
|
|
|
|
phys_bytes lin;
|
|
|
|
|
if(r != EFAULT_SRC && r != EFAULT_DST)
|
|
|
|
|
minix_panic("lin_lin_copy failed", r);
|
|
|
|
|
if(!vmcheck) {
|
|
|
|
|
NOREC_RETURN(virtualcopy, r);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
vmassert(procs[_SRC_] && procs[_DST_]);
|
|
|
|
|
|
|
|
|
|
if(r == EFAULT_SRC) {
|
|
|
|
|
lin = phys_addr[_SRC_];
|
|
|
|
|
target = procs[_SRC_];
|
|
|
|
|
wr = 0;
|
|
|
|
|
} else if(r == EFAULT_DST) {
|
|
|
|
|
lin = phys_addr[_DST_];
|
|
|
|
|
target = procs[_DST_];
|
|
|
|
|
wr = 1;
|
|
|
|
|
} else {
|
|
|
|
|
minix_panic("r strange", r);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#if 0
|
|
|
|
|
printf("virtual_copy: suspending caller %d / %s, target %d / %s\n",
|
|
|
|
|
caller->p_endpoint, caller->p_name,
|
|
|
|
|
target->p_endpoint, target->p_name);
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
|
vmassert(k_reenter == -1);
|
|
|
|
|
vmassert(proc_ptr->p_endpoint == SYSTEM);
|
|
|
|
|
vm_suspend(caller, target, lin, bytes, wr, VMSTYPE_KERNELCALL);
|
|
|
|
|
|
|
|
|
|
NOREC_RETURN(virtualcopy, VMSUSPEND);
|
2008-11-19 13:26:10 +01:00
|
|
|
}
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
|
|
|
|
|
NOREC_RETURN(virtualcopy, OK);
|
2008-11-19 13:26:10 +01:00
|
|
|
}
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
vmassert(!vm_running);
|
|
|
|
|
|
|
|
|
|
/* can't copy to/from process with PT without VM */
|
|
|
|
|
#define NOPT(p) (!(p) || !HASPT(p))
|
|
|
|
|
if(!NOPT(procs[_SRC_])) {
|
|
|
|
|
kprintf("ignoring page table src: %s / %d at 0x%lx\n",
|
|
|
|
|
procs[_SRC_]->p_name, procs[_SRC_]->p_endpoint, procs[_SRC_]->p_seg.p_cr3);
|
|
|
|
|
}
|
|
|
|
|
if(!NOPT(procs[_DST_])) {
|
|
|
|
|
kprintf("ignoring page table dst: %s / %d at 0x%lx\n",
|
|
|
|
|
procs[_DST_]->p_name, procs[_DST_]->p_endpoint,
|
|
|
|
|
procs[_DST_]->p_seg.p_cr3);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* Now copy bytes between physical addresseses. */
|
|
|
|
|
if(phys_copy(phys_addr[_SRC_], phys_addr[_DST_], (phys_bytes) bytes))
|
|
|
|
|
NOREC_RETURN(virtualcopy, EFAULT);
|
|
|
|
|
|
|
|
|
|
NOREC_RETURN(virtualcopy, OK);
|
2008-11-19 13:26:10 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*===========================================================================*
|
|
|
|
|
* data_copy *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
PUBLIC int data_copy(
|
|
|
|
|
endpoint_t from_proc, vir_bytes from_addr,
|
|
|
|
|
endpoint_t to_proc, vir_bytes to_addr,
|
|
|
|
|
size_t bytes)
|
|
|
|
|
{
|
|
|
|
|
struct vir_addr src, dst;
|
|
|
|
|
|
|
|
|
|
src.segment = dst.segment = D;
|
|
|
|
|
src.offset = from_addr;
|
|
|
|
|
dst.offset = to_addr;
|
|
|
|
|
src.proc_nr_e = from_proc;
|
|
|
|
|
dst.proc_nr_e = to_proc;
|
|
|
|
|
|
|
|
|
|
return virtual_copy(&src, &dst, bytes);
|
|
|
|
|
}
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
/*===========================================================================*
|
|
|
|
|
* data_copy_vmcheck *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
PUBLIC int data_copy_vmcheck(
|
|
|
|
|
endpoint_t from_proc, vir_bytes from_addr,
|
|
|
|
|
endpoint_t to_proc, vir_bytes to_addr,
|
|
|
|
|
size_t bytes)
|
|
|
|
|
{
|
|
|
|
|
struct vir_addr src, dst;
|
|
|
|
|
|
|
|
|
|
src.segment = dst.segment = D;
|
|
|
|
|
src.offset = from_addr;
|
|
|
|
|
dst.offset = to_addr;
|
|
|
|
|
src.proc_nr_e = from_proc;
|
|
|
|
|
dst.proc_nr_e = to_proc;
|
|
|
|
|
|
|
|
|
|
return virtual_copy_vmcheck(&src, &dst, bytes);
|
|
|
|
|
}
|
|
|
|
|
|
2008-11-19 13:26:10 +01:00
|
|
|
/*===========================================================================*
|
|
|
|
|
* arch_pre_exec *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
PUBLIC int arch_pre_exec(struct proc *pr, u32_t ip, u32_t sp)
|
|
|
|
|
{
|
|
|
|
|
/* wipe extra LDT entries, set program counter, and stack pointer. */
|
|
|
|
|
memset(pr->p_seg.p_ldt + EXTRA_LDT_INDEX, 0,
|
|
|
|
|
sizeof(pr->p_seg.p_ldt[0]) * (LDT_SIZE - EXTRA_LDT_INDEX));
|
|
|
|
|
pr->p_reg.pc = ip;
|
|
|
|
|
pr->p_reg.sp = sp;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*===========================================================================*
|
|
|
|
|
* arch_umap *
|
|
|
|
|
*===========================================================================*/
|
|
|
|
|
PUBLIC int arch_umap(struct proc *pr, vir_bytes offset, vir_bytes count,
|
|
|
|
|
int seg, phys_bytes *addr)
|
|
|
|
|
{
|
|
|
|
|
switch(seg) {
|
|
|
|
|
case BIOS_SEG:
|
|
|
|
|
*addr = umap_bios(offset, count);
|
|
|
|
|
return OK;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* This must be EINVAL; the umap fallback function in
|
|
|
|
|
* lib/syslib/alloc_util.c depends on it to detect an
|
|
|
|
|
* older kernel (as opposed to mapping error).
|
|
|
|
|
*/
|
|
|
|
|
return EINVAL;
|
|
|
|
|
}
|
|
|
|
|
|
Primary goal for these changes is:
- no longer have kernel have its own page table that is loaded
on every kernel entry (trap, interrupt, exception). the primary
purpose is to reduce the number of required reloads.
Result:
- kernel can only access memory of process that was running when
kernel was entered
- kernel must be mapped into every process page table, so traps to
kernel keep working
Problem:
- kernel must often access memory of arbitrary processes (e.g. send
arbitrary processes messages); this can't happen directly any more;
usually because that process' page table isn't loaded at all, sometimes
because that memory isn't mapped in at all, sometimes because it isn't
mapped in read-write.
So:
- kernel must be able to map in memory of any process, in its own
address space.
Implementation:
- VM and kernel share a range of memory in which addresses of
all page tables of all processes are available. This has two purposes:
. Kernel has to know what data to copy in order to map in a range
. Kernel has to know where to write the data in order to map it in
That last point is because kernel has to write in the currently loaded
page table.
- Processes and kernel are separated through segments; kernel segments
haven't changed.
- The kernel keeps the process whose page table is currently loaded
in 'ptproc.'
- If it wants to map in a range of memory, it writes the value of the
page directory entry for that range into the page directory entry
in the currently loaded map. There is a slot reserved for such
purposes. The kernel can then access this memory directly.
- In order to do this, its segment has been increased (and the
segments of processes start where it ends).
- In the pagefault handler, detect if the kernel is doing
'trappable' memory access (i.e. a pagefault isn't a fatal
error) and if so,
- set the saved instruction pointer to phys_copy_fault,
breaking out of phys_copy
- set the saved eax register to the address of the page
fault, both for sanity checking and for checking in
which of the two ranges that phys_copy was called
with the fault occured
- Some boot-time processes do not have their own page table,
and are mapped in with the kernel, and separated with
segments. The kernel detects this using HASPT. If such a
process has to be scheduled, any page table will work and
no page table switch is done.
Major changes in kernel are
- When accessing user processes memory, kernel no longer
explicitly checks before it does so if that memory is OK.
It simply makes the mapping (if necessary), tries to do the
operation, and traps the pagefault if that memory isn't present;
if that happens, the copy function returns EFAULT.
So all of the CHECKRANGE_OR_SUSPEND macros are gone.
- Kernel no longer has to copy/read and parse page tables.
- A message copying optimisation: when messages are copied, and
the recipient isn't mapped in, they are copied into a buffer
in the kernel. This is done in QueueMess. The next time
the recipient is scheduled, this message is copied into
its memory. This happens in schedcheck().
This eliminates the mapping/copying step for messages, and makes
it easier to deliver messages. This eliminates soft_notify.
- Kernel no longer creates a page table at all, so the vm_setbuf
and pagetable writing in memory.c is gone.
Minor changes in kernel are
- ipc_stats thrown out, wasn't used
- misc flags all renamed to MF_*
- NOREC_* macros to enter and leave functions that should not
be called recursively; just sanity checks really
- code to fully decode segment selectors and descriptors
to print on exceptions
- lots of vmassert()s added, only executed if DEBUG_VMASSERT is 1
2009-09-21 16:31:52 +02:00
|
|
|
/* VM reports page directory slot we're allowed to use freely. */
|
|
|
|
|
void i386_freepde(int pde)
|
|
|
|
|
{
|
|
|
|
|
if(nfreepdes >= WANT_FREEPDES)
|
|
|
|
|
return;
|
|
|
|
|
freepdes[nfreepdes++] = pde;
|
|
|
|
|
}
|