. restore state depends on how saving of state was done;
also remember trap style in sig context
. actually set and restore TRACEBIT with new trap styles;
have to remove it once process enters kernel though, done
in debug trap exception handler
. introduce MF_STEP that makes arch-specific code
turn on trace bit instead of setting TRACEBIT directly,
a bit more arch-friendly and avoids keeping precious
state in per-process PSW arch-dependently
state is usually not in p_reg any more with sysenter/syscall trap entries,
so when saving/restarting do_ipc invocations the state has to be remembered
explicitly.
. add cpufeature detection of both
. use it for both ipc and kernelcall traps, using a register
for call number
. SYSENTER/SYSCALL does not save any context, therefore userland
has to save it
. to accomodate multiple kernel entry/exit types, the entry
type is recorded in the process struct. hitherto all types
were interrupt (soft int, exception, hard int); now SYSENTER/SYSCALL
is new, with the difference that context is not fully restored
from proc struct when running the process again. this can't be
done as some information is missing.
. complication: cases in which the kernel has to fully change
process context (i.e. sigreturn). in that case the exit type
is changed from SYSENTER/SYSEXIT to soft-int (i.e. iret) and
context is fully restored from the proc struct. this does mean
the PC and SP must change, as the sysenter/sysexit userland code
will otherwise try to restore its own context. this is true in the
sigreturn case.
. override all usage by setting libc_ipc=1
This commit removes all traces of Minix segments (the text/data/stack
memory map abstraction in the kernel) and significance of Intel segments
(hardware segments like CS, DS that add offsets to all addressing before
page table translation). This ultimately simplifies the memory layout
and addressing and makes the same layout possible on non-Intel
architectures.
There are only two types of addresses in the world now: virtual
and physical; even the kernel and processes have the same virtual
address space. Kernel and user processes can be distinguished at a
glance as processes won't use 0xF0000000 and above.
No static pre-allocated memory sizes exist any more.
Changes to booting:
. The pre_init.c leaves the kernel and modules exactly as
they were left by the bootloader in physical memory
. The kernel starts running using physical addressing,
loaded at a fixed location given in its linker script by the
bootloader. All code and data in this phase are linked to
this fixed low location.
. It makes a bootstrap pagetable to map itself to a
fixed high location (also in linker script) and jumps to
the high address. All code and data then use this high addressing.
. All code/data symbols linked at the low addresses is prefixed by
an objcopy step with __k_unpaged_*, so that that code cannot
reference highly-linked symbols (which aren't valid yet) or vice
versa (symbols that aren't valid any more).
. The two addressing modes are separated in the linker script by
collecting the unpaged_*.o objects and linking them with low
addresses, and linking the rest high. Some objects are linked
twice, once low and once high.
. The bootstrap phase passes a lot of information (e.g. free memory
list, physical location of the modules, etc.) using the kinfo
struct.
. After this bootstrap the low-linked part is freed.
. The kernel maps in VM into the bootstrap page table so that VM can
begin executing. Its first job is to make page tables for all other
boot processes. So VM runs before RS, and RS gets a fully dynamic,
VM-managed address space. VM gets its privilege info from RS as usual
but that happens after RS starts running.
. Both the kernel loading VM and VM organizing boot processes happen
using the libexec logic. This removes the last reason for VM to
still know much about exec() and vm/exec.c is gone.
Further Implementation:
. All segments are based at 0 and have a 4 GB limit.
. The kernel is mapped in at the top of the virtual address
space so as not to constrain the user processes.
. Processes do not use segments from the LDT at all; there are
no segments in the LDT any more, so no LLDT is needed.
. The Minix segments T/D/S are gone and so none of the
user-space or in-kernel copy functions use them. The copy
functions use a process endpoint of NONE to realize it's
a physical address, virtual otherwise.
. The umap call only makes sense to translate a virtual address
to a physical address now.
. Segments-related calls like newmap and alloc_segments are gone.
. All segments-related translation in VM is gone (vir2map etc).
. Initialization in VM is simpler as no moving around is necessary.
. VM and all other boot processes can be linked wherever they wish
and will be mapped in at the right location by the kernel and VM
respectively.
Other changes:
. The multiboot code is less special: it does not use mb_print
for its diagnostics any more but uses printf() as normal, saving
the output into the diagnostics buffer, only printing to the
screen using the direct print functions if a panic() occurs.
. The multiboot code uses the flexible 'free memory map list'
style to receive the list of free memory if available.
. The kernel determines the memory layout of the processes to
a degree: it tells VM where the kernel starts and ends and
where the kernel wants the top of the process to be. VM then
uses this entire range, i.e. the stack is right at the top,
and mmap()ped bits of memory are placed below that downwards,
and the break grows upwards.
Other Consequences:
. Every process gets its own page table as address spaces
can't be separated any more by segments.
. As all segments are 0-based, there is no distinction between
virtual and linear addresses, nor between userspace and
kernel addresses.
. Less work is done when context switching, leading to a net
performance increase. (8% faster on my machine for 'make servers'.)
. The layout and configuration of the GDT makes sysenter and syscall
possible.
- we must not deliver messages from/to unstable address spaces.
In such a case, we must postpone the delivery. To make sute
that a process which is expecting an asynchronous message does
not starve, we must remember that we skipped delivery of some
messages and we must try to deliver again once the source
address space is stable again.
- when kernel copies from userspace, it must be sure that the TLB
entries are not stale and thus the referenced memory is correct
- everytime we change a process' address space we set p_stale_tlb
bits for all CPUs.
- Whenever a cpu finds its bit set when it wants to access the
process' memory, it refreshes the TLB
- it is more conservative than it needs to be but it has low
overhead than checking precisely
- flush TLB of processes only if the page tables has been changed and
the page tables of this process are already loaded on this cpu which
means that there might be stale entries in TLB. Until now SMP was
always flushing TLB to make sure everything is consistent.
- when kernel profiles a process for the first time it saves an entry
describing the process [endpoint|name]
- every profile sample is only [endpoint|pc]
- profile utility creates a table of endpoint <-> name relations and
translates endpoints of samples into names and writing out the
results to comply with the processing tools
- "task" endpoints like KERNEL are negative thus we must cast it to
unsigned when hashing
- contributed by Bjorn Swift
- adds process accounting, for example counting the number of messages
sent, how often the process was preemted and how much time it spent
in the run queue. These statistics, along with the current cpu load,
are sent back to the user-space scheduler in the Out Of Quantum
message.
- the user-space scheduler may choose to make use of these statistics
when making scheduling decisions. For isntance the cpu load becomes
especially useful when scheduling on multiple cores.
- each CPU has its own runqueues
- processes on BSP are put on the runqueues later after a switch to
the final stack when cpuid works to avoid special cases
- enqueue() and dequeue() use the run queues of the cpu the process is
assigned to
- pick_proc() uses the local run queues
- printing of per-CPU run queues ('2') on serial console
- removes p_delivermsg_lin item from the process structure and code
related to it
- as the send part, the receive does not need to use the
PHYS_COPY_CATCH() and umap_local() couple.
- The address space of the target process is installed before
delivermsg() is called.
- unlike the linear address, the virtual address does not change when
paging is turned on nor after fork().
- FPU context is stored only if conflict between 2 FPU users or while
exporting context of a process to userspace while it is the active
user of FPU
- FPU has its owner (fpu_owner) which points to the process whose
state is currently loaded in FPU
- the FPU exception is only turned on when scheduling a process which
is not the owner of FPU
- FPU state is restored for the process that generated the FPU
exception. This process runs immediately without letting scheduler
to pick a new process to resolve the FPU conflict asap, to minimize
the FPU thrashing and FPU exception hadler execution
- faster all non-FPU-exception kernel entries as FPU state is not
checked nor saved
- removed MF_USED_FPU flag, only MF_FPU_INITIALIZED remains to signal
that a process has used FPU in the past
There seems to have been a broken assumption in the fpu context
restoring code. It restores the context of the running process, without
guarantee that the current process is the one that will be scheduled.
This caused fpu saving for a different process to be triggered without
fpu hardware being enabled, causing an fpu exception in the kernel. This
practically only shows up with DEBUG_RACE on. Fix my thruby+me.
The fix
. is to only set the fpu-in-use-by-this-process flag in the
exception handler, and then take care of fpu restoring when
actually returning to userspace
And the patch
. translates fpu saving and restoring to c in arch_system.c,
getting rid of a juicy chunk of assembly
. makes osfxsr_feature private to arch_system.c
. removes most of the arch dependent code from do_sigsend
- Currently the cpu time quantum is timer-ticks based. Thus the
remaining quantum is decreased only if the processes is interrupted
by a timer tick. As processes block a lot this typically does not
happen for normal user processes. Also the quantum depends on the
frequency of the timer.
- This change makes the quantum miliseconds based. Internally the
miliseconds are translated into cpu cycles. Everytime userspace
execution is interrupted by kernel the cycles just consumed by the
current process are deducted from the remaining quantum.
- It makes the quantum system timer frequency independent.
- The boot processes quantum is loosely derived from the tick-based
quantas and 60Hz timer and subject to future change
- the 64bit arithmetics is a little ugly, will be changes once we have
compiler support for 64bit integers (soon)
In this second phase, scheduling is moved from PM to its own
scheduler (see r6557 for phase one). In the next phase we hope to a)
include useful information in the "out of quantum" message and b)
create some simple scheduling policy that makes use of that
information.
When the system starts up, PM will iterate over its process table and
ask SCHED to take over scheduling unprivileged processes. This is
done by sending a SCHEDULING_START message to SCHED. This message
includes the processes endpoint, the parent's endpoint and its nice
level. The scheduler adds this process to its schedproc table, issues
a schedctl, and returns its own endpoint to PM - as the endpoint of
the effective scheduler. When a process terminates, a SCHEDULING_STOP
message is sent to the scheduler.
The reason for this effective endpoint is for future compatibility.
Some day, we may have a scheduler that, instead of scheduling the
process itself, forwards the SCHEDULING_START message on to another
scheduler.
PM has information on who schedules whom. As such, scheduling
messages from user-land are sent through PM. An example is when
processes change their priority, using nice(). In that case, a
getsetpriority message is sent to PM, which then sends a
SCHEDULING_SET_NICE to the process's effective scheduler.
When a process is forked through PM, it inherits its parent's
scheduler, but is spawned with an empty quantum. As before, a request
to fork a process flows through VM before returning to PM, which then
wakes up the child process. This flow has been modified slightly so
that PM notifies the scheduler of the new process, before waking up
the child process. If the scheduler fails to take over scheduling,
the child process is torn down and the fork fails with an erroneous
value.
Process priority is entirely decided upon using nice levels. PM
stores a copy of each process's nice level and when a child is
forked, its parent's nice level is sent in the SCHEDULING_START
message. How this level is mapped to a priority queue is up to the
scheduler. It should be noted that the nice level is used to
determine the max_priority and the parent could have been in a lower
priority when it was spawned. To prevent a CPU intensive process from
hawking the CPU by continuously forking children that get scheduled
in the max_priority, the scheduler should determine in which queue
the parent is currently scheduled, and schedule the child in that
same queue.
Other fixes: The USER_Q in kernel/proc.h was incorrectly defined as
NR_SCHED_QUEUES/2. That results in a "off by one" error when
converting priority->nice->priority for nice=0. This also had the
side effect that if someone were to set the MAX_USER_Q to something
else than 0, then USER_Q would be off.
this patch changes the way pagefaults are delivered to VM. It adopts
the same model as the out-of-quantum messages sent by kernel to a
scheduler.
- everytime a userspace pagefault occurs, kernel creates a message
which is sent to VM on behalf of the faulting process
- the process is blocked on delivery to VM in the standard IPC code
instead of waiting in a spacial in-kernel queue (stack) and is not
runnable until VM tell kernel that the pagefault is resolved and is
free to clear the RTS_PAGEFAULT flag.
- VM does not need call kernel and poll the pagefault information
which saves many (1/2?) calls and kernel calls that return "no more
data"
- VM notification by kernel does not need to use signals
- each entry in proc table is by 12 bytes smaller (~3k save)
- this is a small addition to the userspace scheduling.
proc_kernel_scheduler() tests whether to use the default scheduling
policy in kernel. It is true if the process' scheduler is NULL _or_
self. Currently none of the tests was complete.
- IPC_FLG_MSG_FROM_KERNEL status flag is returned to userspace if the
receive was satisfied by s message which was sent by the kernel on
behalf of a process. This perfectly reliale information.
- MF_SENDING_FROM_KERNEL flag added to processes to be able to set
IPC_FLG_MSG_FROM_KERNEL when finishing receive if the receiver
wasn't ready to receive immediately.
- PM is changed to use this information to confirm that the scheduling
messages are indeed from the kernel and not faked by a process.
PM uses sef_receive_status()
- get_work() is removed from PM to make the changes simpler
- cotributed by Bjorn Swift
- In this first phase, scheduling is moved from the kernel to the PM
server. The next steps are to a) moving scheduling to its own server
and b) include useful information in the "out of quantum" message,
so that the scheduler can make use of this information.
- The kernel process table now keeps record of who is responsible for
scheduling each process (p_scheduler). When this pointer is NULL,
the process will be scheduled by the kernel. If such a process runs
out of quantum, the kernel will simply renew its quantum an requeue
it.
- When PM loads, it will take over scheduling of all running
processes, except system processes, using sys_schedctl().
Essentially, this only results in taking over init. As children
inherit a scheduler from their parent, user space programs forked by
init will inherit PM (for now) as their scheduler.
- Once a process has been assigned a scheduler, and runs out of
quantum, its RTS_NO_QUANTUM flag will be set and the process
dequeued. The kernel will send a message to the scheduler, on the
process' behalf, informing the scheduler that it has run out of
quantum. The scheduler can take what ever action it pleases, based
on its policy, and then reschedule the process using the
sys_schedule() system call.
- Balance queues does not work as before. While the old in-kernel
function used to renew the quantum of processes in the highest
priority run queue, the user-space implementation only acts on
processes that have been bumped down to a lower priority queue.
This approach reacts slower to changes than the old one, but saves
us sending a sys_schedule message for each process every time we
balance the queues. Currently, when processes are moved up a
priority queue, their quantum is also renewed, but this can be
fiddled with.
- do_nice has been removed from kernel. PM answers to get- and
setpriority calls, updates it's own nice variable as well as the
max_run_queue. This will be refactored once scheduling is moved to a
separate server. We will probably have PM update it's local nice
value and then send a message to whoever is scheduling the process.
- changes to fix an issue in do_fork() where processes could run out
of quantum but bypassing the code path that handles it correctly.
The future plan is to remove the policy from do_fork() and implement
it in userspace too.
UPDATING INFO:
20100317:
/usr/src/etc/system.conf updated to ignore default kernel calls: copy
it (or merge it) to /etc/system.conf.
The hello driver (/dev/hello) added to the distribution:
# cd /usr/src/commands/scripts && make clean install
# cd /dev && MAKEDEV hello
KERNEL CHANGES:
- Generic signal handling support. The kernel no longer assumes PM as a signal
manager for every process. The signal manager of a given process can now be
specified in its privilege slot. When a signal has to be delivered, the kernel
performs the lookup and forwards the signal to the appropriate signal manager.
PM is the default signal manager for user processes, RS is the default signal
manager for system processes. To enable ptrace()ing for system processes, it
is sufficient to change the default signal manager to PM. This will temporarily
disable crash recovery, though.
- sys_exit() is now split into sys_exit() (i.e. exit() for system processes,
which generates a self-termination signal), and sys_clear() (i.e. used by PM
to ask the kernel to clear a process slot when a process exits).
- Added a new kernel call (i.e. sys_update()) to swap two process slots and
implement live update.
PM CHANGES:
- Posix signal handling is no longer allowed for system processes. System
signals are split into two fixed categories: termination and non-termination
signals. When a non-termination signaled is processed, PM transforms the signal
into an IPC message and delivers the message to the system process. When a
termination signal is processed, PM terminates the process.
- PM no longer assumes itself as the signal manager for system processes. It now
makes sure that every system signal goes through the kernel before being
actually processes. The kernel will then dispatch the signal to the appropriate
signal manager which may or may not be PM.
SYSLIB CHANGES:
- Simplified SEF init and LU callbacks.
- Added additional predefined SEF callbacks to debug crash recovery and
live update.
- Fixed a temporary ack in the SEF init protocol. SEF init reply is now
completely synchronous.
- Added SEF signal event type to provide a uniform interface for system
processes to deal with signals. A sef_cb_signal_handler() callback is
available for system processes to handle every received signal. A
sef_cb_signal_manager() callback is used by signal managers to process
system signals on behalf of the kernel.
- Fixed a few bugs with memory mapping and DS.
VM CHANGES:
- Page faults and memory requests coming from the kernel are now implemented
using signals.
- Added a new VM call to swap two process slots and implement live update.
- The call is used by RS at update time and in turn invokes the kernel call
sys_update().
RS CHANGES:
- RS has been reworked with a better functional decomposition.
- Better kernel call masks. com.h now defines the set of very basic kernel calls
every system service is allowed to use. This makes system.conf simpler and
easier to maintain. In addition, this guarantees a higher level of isolation
for system libraries that use one or more kernel calls internally (e.g. printf).
- RS is the default signal manager for system processes. By default, RS
intercepts every signal delivered to every system process. This makes crash
recovery possible before bringing PM and friends in the loop.
- RS now supports fast rollback when something goes wrong while initializing
the new version during a live update.
- Live update is now implemented by keeping the two versions side-by-side and
swapping the process slots when the old version is ready to update.
- Crash recovery is now implemented by keeping the two versions side-by-side
and cleaning up the old version only when the recovery process is complete.
DS CHANGES:
- Fixed a bug when the process doing ds_publish() or ds_delete() is not known
by DS.
- Fixed the completely broken support for strings. String publishing is now
implemented in the system library and simply wraps publishing of memory ranges.
Ideally, we should adopt a similar approach for other data types as well.
- Test suite fixed.
DRIVER CHANGES:
- The hello driver has been added to the Minix distribution to demonstrate basic
live update and crash recovery functionalities.
- Other drivers have been adapted to conform the new SEF interface.
process waiting for" logic, which is duplicated a few times in the
kernel. (For a new feature for top.)
Introducing it and throwing out ESRCDIED and EDSTDIED (replaced by
EDEADSRCDST - so we don't have to care which part of the blocking is
failing in system.c) simplifies some code in the kernel and callers that
check for E{DEADSRCDST,ESRCDIED,EDSTDIED}, but don't care about the
difference, a fair bit, and more significantly doesn't duplicate the
'blocked-on' logic.
- as thre are still KERNEL and IDLE entries, time accounting for
kernel and idle time works the same as for any other process
- everytime we stop accounting for the currently running process,
kernel or idle, we read the TSC counter and increment the p_cycles
entry.
- the process cycles inherently include some of the kernel cycles as
we can stop accounting for the process only after we save its
context and we start accounting just before we restore its context
- this assumes that the system does not scale the CPU frequency which
will be true for ... long time ;-)
- we don't need to test this in kernel as we always have interrupts
disabled
- if interrupts are enabled in kernel, it is only at very carefully
chosen places. There are no such places now.
* Userspace change to use the new kernel calls
- _taskcall(SYSTASK...) changed to _kernel_call(...)
- int 32 reused for the kernel calls
- _do_kernel_call() to make the trap to kernel
- kernel_call() to make the actuall kernel call from C using
_do_kernel_call()
- unlike ipc call the kernel call always succeeds as kernel is
always available, however, kernel may return an error
* Kernel side implementation of kernel calls
- the SYSTEm task does not run, only the proc table entry is
preserved
- every data_copy(SYSTEM is no data_copy(KERNEL
- "locking" is an empty operation now as everything runs in
kernel
- sys_task() is replaced by kernel_call() which copies the
message into kernel, dispatches the call to its handler and
finishes by either copying the results back to userspace (if
need be) or by suspending the process because of VM
- suspended processes are later made runnable once the memory
issue is resolved, picked up by the scheduler and only at
this time the call is resumed (in fact restarted) which does
not need to copy the message from userspace as the message
is already saved in the process structure.
- no ned for the vmrestart queue, the scheduler will restart
the system calls
- no special case in do_vmctl(), all requests remove the
RTS_VMREQUEST flag
Main changes:
- COW optimization for safecopy.
- safemap, a grant-based interface for sharing memory regions between processes.
- Integration with safemap and complete rework of DS, supporting new data types
natively (labels, memory ranges, memory mapped ranges).
- For further information:
http://wiki.minix3.org/en/SummerOfCode2009/MemoryGrants
Additional changes not included in the original Wu's branch:
- Fixed unhandled case in VM when using COW optimization for safecopy in case
of a block that has already been shared as SMAP.
- Better interface and naming scheme for sys_saferevmap and ds_retrieve_map
calls.
- Better input checking in syslib: check for page alignment when creating
memory mapping grants.
- DS notifies subscribers when an entry is deleted.
- Documented the behavior of indirect grants in case of memory mapping.
- Test suite in /usr/src/test/safeperf|safecopy|safemap|ds/* reworked
and extended.
- Minor fixes and general cleanup.
- TO-DO: Grant ids should be generated and managed the way endpoints are to make
sure grant slots are never misreused.
KERNEL CHANGES:
- The kernel only knows about privileges of kernel tasks and the root system
process (now RS).
- Kernel tasks and the root system process are the only processes that are made
schedulable by the kernel at startup. All the other processes in the boot image
don't get their privileges set at startup and are inhibited from running by the
RTS_NO_PRIV flag.
- Removed the assumption on the ordering of processes in the boot image table.
System processes can now appear in any order in the boot image table.
- Privilege ids can now be assigned both statically or dynamically. The kernel
assigns static privilege ids to kernel tasks and the root system process. Each
id is directly derived from the process number.
- User processes now all share the static privilege id of the root user
process (now INIT).
- sys_privctl split: we have more calls now to let RS set privileges for system
processes. SYS_PRIV_ALLOW / SYS_PRIV_DISALLOW are only used to flip the
RTS_NO_PRIV flag and allow / disallow a process from running. SYS_PRIV_SET_SYS /
SYS_PRIV_SET_USER are used to set privileges for a system / user process.
- boot image table flags split: PROC_FULLVM is the only flag that has been
moved out of the privilege flags and is still maintained in the boot image
table. All the other privilege flags are out of the kernel now.
RS CHANGES:
- RS is the only user-space process who gets to run right after in-kernel
startup.
- RS uses the boot image table from the kernel and three additional boot image
info table (priv table, sys table, dev table) to complete the initialization
of the system.
- RS checks that the entries in the priv table match the entries in the boot
image table to make sure that every process in the boot image gets schedulable.
- RS only uses static privilege ids to set privileges for system services in
the boot image.
- RS includes basic memory management support to allocate the boot image buffer
dynamically during initialization. The buffer shall contain the executable
image of all the system services we would like to restart after a crash.
- First step towards decoupling between resource provisioning and resource
requirements in RS: RS must know what resources it needs to restart a process
and what resources it has currently available. This is useful to tradeoff
reliability and resource consumption. When required resources are missing, the
process cannot be restarted. In that case, in the future, a system flag will
tell RS what to do. For example, if CORE_PROC is set, RS should trigger a
system-wide panic because the system can no longer function correctly without
a core system process.
PM CHANGES:
- The process tree built at initialization time is changed to have INIT as root
with pid 0, RS child of INIT and all the system services children of RS. This
is required to make RS in control of all the system services.
- PM no longer registers labels for system services in the boot image. This is
now part of RS's initialization process.
