- The MP code no longer knows anything specific about an MP Table.
Instead, the local APIC code adds CPUs via the cpu_add() function when
a local APIC is enumerated by an APIC enumerator.
- Don't divide the argument to mp_bootaddress() by 1024 just so that we
can turn around and mulitply it by 1024 again.
- We no longer panic if SMP is enabled but we are booted on a UP machine.
- init_secondary(), the asm code between init_secondary() and ap_init()
in mpboot.s and ap_init() have all been merged together in C into
init_secondary().
- We now use the cpuid feature bits to determine if we should enable
PSE, PGE, or VME on each AP.
- Due to the change in the implementation of critical sections, acquire
the SMP TLB mutex around a slightly larger chunk of code for TLB
shootdowns.
- Remove some of the debug code from the original SMP implementation
that is no longer used or no longer applies to the new APIC code.
- Use a temporary hack to disable the ACPI module until the SMP code has
been further reorganized to allow ACPI to work as a module again.
- Add a DDB command to dump the interesting contents of the IDT.
devices claiming resources that they don't actually use. The PIC drivers
only register valid interrupt sources, so we don't need to rely on these
drivers to claim invalid IRQs to prevent their use by other drivers.
APIC Descriptor Table to enumerate both I/O APICs and local APICs. ACPI
does not embed PCI interrupt routing information in the MADT like the MP
Table does. Instead, ACPI stores the PCI interrupt routing information
in the _PRT object under each PCI bus device. The MADT table simply
provides hints about which interrupt vectors map to which I/O APICs. Thus
when using ACPI, the existing ACPI PCI bridge drivers are sufficient to
route PCI interrupts.
- The apic interrupt entry points have been rewritten so that each entry
point can serve 32 different vectors. When the entry is executed, it
uses one of the 32-bit ISR registers to determine which vector in its
assigned range was triggered. Thus, the apic code can support 159
different interrupt vectors with only 5 entry points.
- We now always to disable the local APIC to work around an errata in
certain PPros and then re-enable it again if we decide to use the APICs
to route interrupts.
- We no longer map IO APICs or local APICs using special page table
entries. Instead, we just use pmap_mapdev(). We also no longer
export the virtual address of the local APIC as a global symbol to
the rest of the system, but only in local_apic.c. To aid this, the
APIC ID of each CPU is exported as a per-CPU variable.
- Interrupt sources are provided for each intpin on each IO APIC.
Currently, each source is given a unique interrupt vector meaning that
PCI interrupts are not shared on most machines with an I/O APIC.
That mapping for interrupt sources to interrupt vectors is up to the
APIC enumerator driver however.
- We no longer probe to see if we need to use mixed mode to route IRQ 0,
instead we always use mixed mode to route IRQ 0 for now. This can be
disabled via the 'NO_MIXED_MODE' kernel option.
- The npx(4) driver now always probes to see if a built-in FPU is present
since this test can now be performed with the new APIC code. However,
an SMP kernel will panic if there is more than one CPU and a built-in
FPU is not found.
- PCI interrupts are now properly routed when using APICs to route
interrupts, so remove the hack to psuedo-route interrupts when the
intpin register was read.
- The apic.h header was moved to apicreg.h and a new apicvar.h header
that declares the APIs used by the new APIC code was added.
default we provide 16 interrupt sources for IRQs 0 through 15. However,
if the I/O APIC driver has already registered sources for any of those IRQs
then we will silently fail to register our own source for that IRQ.
Note that i386/isa/icu.h is now specific to the 8259A and no longer
contains any info relevant to APICs. Also note that fast interrupts no
longer use a separate entry point. Instead, both fast and threaded
interrupts share the same entry point which merely looks up the appropriate
source and passes control to intr_execute_handlers().
that provides methods via a PIC driver to do things like mask a source,
unmask a source, enable it when the first interrupt handler is added, etc.
The interrupt code provides a table of interrupt sources indexed by IRQ
numbers, or vectors. These vectors are what new-bus uses for its IRQ
resources and for bus_setup_intr()/bus_teardown_intr(). The interrupt
code then maps that vector a given interrupt source object. When an
interrupt comes in, the low-level interrupt code looks up the interrupt
source for the source that triggered the interrupt and hands it off to
this code to execute the appropriate handlers.
By having an interrupt source abstraction, this allows us to have different
types of interrupt source providers within the shared IRQ address space.
For example, IRQ 0 may map to pin 0 of the master 8259A PIC, IRQs 1
through 60 may map to pins on various I/O APICs, and IRQs 120 through
128 may map to MSI interrupts for various PCI devices.
setting. Make va_copy be an alias if __ISO_C_VISIBLE >= 1999.
Why? more than a few ports have an autoconf that looks for __va_copy
because it is available on glibc. It is critical that we use it if
at all possible on amd64. It generally isn't a problem for i386 and its
ilk because autoconf driven code tends to fall back to an assignment.
implement i386 compat numbers where it makes sense. This would save a
syscall translation layer. Yes, this breaks the abi slightly again, but
fortunately its just a recompile rather than tweaking the source. I will
be fixing the libc stubs while I'm here.
Xcpustop(). %es is used in at least the call to savectx() when savectx()
calls bcopy(), so not loading it was fatal if a stop IPI interrupts
user mode.
This reduces bugs starting and stopping CPUs for debuggers. CPUs are
stopped mainly in kdb_trap() and cpu_reset(). At reset time there is
a good chance that all the CPUs are in the kernel, so the bug was
probably harmless then.
sigreturn() ABI and the signal context on the stack.
Make the trapframe (and its shadows in the ucontext and sigframe etc)
8 bytes larger in order to preserve 16 byte stack alignment for the
following C code calls. I could have done some padding after the
trapframe was saved, but some of the C code still expects an argument of
'struct trapframe'. Anyway, this gives me a spare field that can be used
to store things like 'partial trapframe' status or something else in
the future.
The runtime impact is fairly small, *except* for threaded apps and things
that decode contexts and the signal stack (eg: cvsup binary). Signal
delivery isn't too badly affected because the kernel generates the
sigframe that sigreturn uses after the handler has been called.
The size of mcontext_t and struct sigframe hasn't changed. Only
the last few fields (sc_eip etc) got moved a little and I eliminated
a spare field. mc_len/sc_len did change location though so the
sanity checks there will still trap it.
A small helper function pmap_is_prefaultable() is added. This function
encapsulate the few lines of pmap_prefault() that actually vary from
machine to machine. Note: pmap_is_prefaultable() and pmap_mincore() have
much in common. Going forward, it's worth considering their merger.
avoid problems with some Pentium 4 cpus and some older PPro/Pentium2
cpus. There are several problems, some documented in Intel errata.
This patch:
1) moves the kernel to the second page in the PSE case. There is an
errata that says that you Must Not point a 4MB page at physical
address zero on older cpus. We avoided bugs here due to sheer luck.
2) sets up PSE page tables right from the start in locore, rather than
trying to switch from 4K to 4M (or 2M) pages part way through the boot
sequence at the same time that we're messing with PG_G.
For some reason, the pmap work over the last 18 months seems to tickle
the problems, and the PAE infrastructure changes disturb the cpu
bugs even more.
A couple of people have reported a problem with APM bios calls during
boot. I'll work with people to get this resolved.
Obtained from: bmilekic
Reimplement pmap_release() such that it uses the page table rather than
the pte object to locate the page table directory pages. (Temporarily,
retain an assertion on the emptiness of the pte object.)
systems where the data/stack/etc limits are too big for a 32 bit process.
Move the 5 or so identical instances of ELF_RTLD_ADDR() into imgact_elf.c.
Supply an ia32_fixlimits function. Export the clip/default values to
sysctl under the compat.ia32 heirarchy.
Have mmap(0, ...) respect the current p->p_limits[RLIMIT_DATA].rlim_max
value rather than the sysctl tweakable variable. This allows mmap to
place mappings at sensible locations when limits have been reduced.
Have the imgact_elf.c ld-elf.so.1 placement algorithm use the same
method as mmap(0, ...) now does.
Note that we cannot remove all references to the sysctl tweakable
maxdsiz etc variables because /etc/login.conf specifies a datasize
of 'unlimited'. And that causes exec etc to fail since it can no
longer find space to mmap things.
32 bit binary stuff. 32 bit binaries do not like it much when the kernel
tries hard to put things above the 8GB mark.
I have a work-in-progress to fix this properly, but I didn't want to burn
anybody with this yet.
initialize a TSC timecounter until we know if it is broke or not.
XXX I think there is a bug in the i386 code here. init_TSC_tc() comes
after:
if (statclock_disable)
return;
ie: if you turn off the statclock interrupt, you dont get the TSC either.
for breakpoint and trace traps from usermode. Although all the setidt
entries are interrupt gates on amd64, all but the trace and bpt trap
entry handlers reenable interrupts after the swapgs instruction in order
to simulate the trap/interrupt gate distinction. In other words, the
amd64 code behaves the same way that i386 does here.
for ddb input in some atkbd-based console drivers. ddb must not use any
normal locks but DELAY() normally calls getit() which needs clock_lock.
This also removes the need for recursion on clock_lock.
of "dumb" PCI-based serial/parallel boards get a hint how to enable
them.
I wasn't sure about the ia64, pc98, powerpc, and sparc64 archs whether
they'd support puc(4) or not.
pmap_remove_pte(), passed NULL instead of the required page table
page to pmap_unuse_pt(). Compute the necessary page table page
in pmap_remove_pte(). Also, remove some unreachable code from
pmap_remove_pte().
on data structures on the kernel stack which are guaranteed to be 16 byte
aligned by gcc, the amd64 ABI and __aligned(16).
Ensire the tss_rsp0 initial stack pointer is 16 byte aligned in case
sizeof(pcb) becomes odd at some point. This is convenient for the
interrupt handler case because the ring crossing pushes cause the
required odd alignment before the call to the C code.
Have fast_syscall add an additional 8 bytes to ensure that the trapframe
has the correct odd alignment for the call to C code. Note that there are
no checks to make sure that the trapframe size is appropriate for this.
This makes get/setfpcontext work properly (finally). You get a GPF in
kernel mode if any of this is botched without the alignment fixup code
that is apparently needed on i386.
atomically extracts and holds the physical page that is associated with the
given pmap and virtual address. Such a function is needed to make the
memory mapping optimizations used by, for example, pipes and raw disk I/O
MP-safe.
Reviewed by: tegge
rl(4) driver and put it in a new re(4) driver. The re(4) driver shares
the if_rlreg.h file with rl(4) but is a separate module. (Ultimately
I may change this. For now, it's convenient.)
rl(4) has been modified so that it will never attach to an 8139C+
chip, leaving it to re(4) instead. Only re(4) has the PCI IDs to
match the 8169/8169S/8110S gigE chips. if_re.c contains the same
basic code that was originally bolted onto if_rl.c, with the
following updates:
- Added support for jumbo frames. Currently, there seems to be
a limit of approximately 6200 bytes for jumbo frames on transmit.
(This was determined via experimentation.) The 8169S/8110S chips
apparently are limited to 7.5K frames on transmit. This may require
some more work, though the framework to handle jumbo frames on RX
is in place: the re_rxeof() routine will gather up frames than span
multiple 2K clusters into a single mbuf list.
- Fixed bug in re_txeof(): if we reap some of the TX buffers,
but there are still some pending, re-arm the timer before exiting
re_txeof() so that another timeout interrupt will be generated, just
in case re_start() doesn't do it for us.
- Handle the 'link state changed' interrupt
- Fix a detach bug. If re(4) is loaded as a module, and you do
tcpdump -i re0, then you do 'kldunload if_re,' the system will
panic after a few seconds. This happens because ether_ifdetach()
ends up calling the BPF detach code, which notices the interface
is in promiscuous mode and tries to switch promisc mode off while
detaching the BPF listner. This ultimately results in a call
to re_ioctl() (due to SIOCSIFFLAGS), which in turn calls re_init()
to handle the IFF_PROMISC flag change. Unfortunately, calling re_init()
here turns the chip back on and restarts the 1-second timeout loop
that drives re_tick(). By the time the timeout fires, if_re.ko
has been unloaded, which results in a call to invalid code and
blows up the system.
To fix this, I cleared the IFF_UP flag before calling ether_ifdetach(),
which stops the ioctl routine from trying to reset the chip.
- Modified comments in re_rxeof() relating to the difference in
RX descriptor status bit layout between the 8139C+ and the gigE
chips. The layout is different because the frame length field
was expanded from 12 bits to 13, and they got rid of one of the
status bits to make room.
- Add diagnostic code (re_diag()) to test for the case where a user
has installed a broken 32-bit 8169 PCI NIC in a 64-bit slot. Some
NICs have the REQ64# and ACK64# lines connected even though the
board is 32-bit only (in this case, they should be pulled high).
This fools the chip into doing 64-bit DMA transfers even though
there is no 64-bit data path. To detect this, re_diag() puts the
chip into digital loopback mode and sets the receiver to promiscuous
mode, then initiates a single 64-byte packet transmission. The
frame is echoed back to the host, and if the frame contents are
intact, we know DMA is working correctly, otherwise we complain
loudly on the console and abort the device attach. (At the moment,
I don't know of any way to work around the problem other than
physically modifying the board, so until/unless I can think of a
software workaround, this will have do to.)
- Created re(4) man page
- Modified rlphy.c to allow re(4) to attach as well as rl(4).
Note that this code works for the sample 8169/Marvell 88E1000 NIC
that I have, but probably won't work for the 8169S/8110S chips.
RealTek has sent me some sample NICs, but they haven't arrived yet.
I will probably need to add an rlgphy driver to handle the on-board
PHY in the 8169S/8110S (it needs special DSP initialization).
sockets into machine-dependent files. The rationale for this
migration is illustrated by the modified amd64 allocator. It uses the
amd64's direct map to avoid emphemeral mappings in the kernel's
address space. On an SMP, the emphemeral mappings result in an IPI
for TLB shootdown for each transmitted page. Yuck.
Maintainers of other 64-bit platforms with direct maps should be able
to use the amd64 allocator as a reference implementation.
- Add a new PCIM_HDRTYPE constant for the field in PCIR_HDRTYPE that holds
the header type.
- Replace several magic numbers with appropriate constants for the header
type register and a couple of PCI_FUNCMAX.
- Merge to amd64 the fix to the i386 bridge code to skip devices with
unknown header types.
Requested by: imp (1, 2)
from the ia32 specific stuff. Some of this still needs to move to the MI
freebsd32 area, and some needs to move to the MD area. This is still
work-in-progress.
prototypes of cpu_halt(), cpu_reset() and swi_vm() from md_var.h to
cpu.h. This affects db_command.c and kern_shutdown.c.
ia64: move all MD prototypes from cpu.h to md_var.h. This affects
madt.c, interrupt.c and mp_machdep.c. Remove is_physical_memory().
It's not used (vm_machdep.c).
alpha: the MD prototypes have been left in cpu.h with a comment
that they should be there. Moving them is left for later. It was
expected that the impact would be significant enough to be done in
a seperate commit.
powerpc: MD prototypes left in cpu.h. Comment added.
Suggested by: bde
Tested with: make universe (pc98 incomplete)
- Add a macro for the logical shift needed to extract an APIC ID from
either from the local APIC ICR Hi register or the APIC ID registers of
the local and IO APICs.
parameter. The new name better reflects what the function does and
how it is used. The last parameter was always FALSE.
Note: In theory, gcc would perform constant propagation and dead code
elimination to achieve the same effect as removing the last parameter,
which is always FALSE. In practice, recent versions do not. So, there
is little point in letting unused code pessimize execution.
they haven't been counted before. This test was ommitted when bus_dmamap_load()
was merged into this function, and results in the pagesneeded field growing
without bounds when multiple deferrals happen.
Thanks to Paul Saab for beating his head against this for a few hours =-)
set in cpu_critical_fork_exit() anymore.
- As far as I can tell, cpu_thread_link() has never been used, not even
when it was originally added, so remove it.
created not only with UMA_ZONE_VM but also with UMA_ZONE_NOFREE. In
the i386 case in particular, the pmap code would hook a special
page allocation routine that allocated from kernel_map and not kmem_map,
and so when/if the pageout daemon drained the zones, it could actually
push out slabs from the PV ENTRY zone but call UMA's default page_free,
which resulted in pages allocated from kernel_map being freed to
kmem_map; bad. kmem_free() ignores the return value of the
vm_map_delete and just returns. I'm not sure what the exact
repercussions could be, but it doesn't look good.
In the PAE case on i386, we also set-up a zone in pmap, so be
conservative for now and make that zone also ZONE_NOFREE and
ZONE_VM. Do this for the pmap zones for the other archs too,
although in some cases it may not be entirely necessarily. We'd
rather be safe than sorry at this point.
Perhaps all UMA_ZONE_VM zones should by default be also
UMA_ZONE_NOFREE?
May fix some of silby's crashes on the PV ENTRY zone.
memory in bus_dmamem_alloc(). This is possible now that
contigmalloc() supports the M_ZERO flag.
- Remove the locking of Giant around calls to contigmalloc() since
contigmalloc() now grabs Giant itself.
written as a template that when inlined is specialized for the caller
through constant value propagation and dead code elimination. Thus,
the specialized code that is generated for pmap_clear_reference() et
al. avoids several conditional branches inside of a loop.
fields in the low 32 bits of the local APIC ICR register. Use this macro
in place of APIC_RESV2_MASK when masking off existing bits from the ICR
when writing to it to send an IPI.
Tested by: scottl
AMD64, gcc (and the ABI) expects the x87 unit to be running in 80/64
mode (not 64/53) so that it can use it for 'long double' operations. It
takes the expected precision differences into account when generating
code.
the SSE mxcsr register as well. Since gcc will intermix SSE2 and x87
FP code, the fpsetround() etc mode had better be the same.
There are hooks to enable these inlines to be instantiated inside libc
for non-gcc or C++ callers. (g++ doesn't like the inlines that tried
to extract an integer and convert it to an enum).
on the implied sign extension. The single unified VADDR() macro was
not able to avoid sign extending the VM_MAXUSER_ADDRESS/USRSTACK values.
Be explicit about UVADDR() (positive address space) and KVADDR()
(kernel negative address space) to make mistakes show up more
spectacularly.
Increase user VM space from 1/2TB (512GB) to 128TB.
corresponding release code. This was preventing the use of more than
1/2TB of user VM. I also spent a week staring at this code only to
eventually find that I'd mistakenly typed a P as an R.
rather than a non-existing pte. There is code elsewhere in i386/amd64
pmap that neglects to handle the large page cases because it knows that
it will see PG_PS in the returned "pte".
order to avoid the overhead of later page faults. In general, it
implements two cases: one for vnode-backed objects and one for
device-backed objects. Only the device-backed case is really
machine-dependent, belonging in the pmap.
This commit moves the vnode-backed case into the (relatively) new
function vm_map_pmap_enter(). On amd64 and i386, this commit only
amounts to code rearrangement. On alpha and ia64, the new machine
independent (MI) implementation of the vnode case is smaller and more
efficient than their pmap-based implementations. (The MI
implementation takes advantage of the fact that objects in -CURRENT
are ordered collections of pages.) On sparc64, pmap_object_init_pt()
hadn't (yet) been implemented.
Add two new arguments to bus_dma_tag_create(): lockfunc and lockfuncarg.
Lockfunc allows a driver to provide a function for managing its locking
semantics while using busdma. At the moment, this is used for the
asynchronous busdma_swi and callback mechanism. Two lockfunc implementations
are provided: busdma_lock_mutex() performs standard mutex operations on the
mutex that is specified from lockfuncarg. dftl_lock() is a panic
implementation and is defaulted to when NULL, NULL are passed to
bus_dma_tag_create(). The only time that NULL, NULL should ever be used is
when the driver ensures that bus_dmamap_load() will not be deferred.
Drivers that do not provide their own locking can pass
busdma_lock_mutex,&Giant args in order to preserve the former behaviour.
sparc64 and powerpc do not provide real busdma_swi functions, so this is
largely a noop on those platforms. The busdma_swi on is64 is not properly
locked yet, so warnings will be emitted on this platform when busdma
callback deferrals happen.
If anyone gets panics or warnings from dflt_lock() being called, please
let me know right away.
Reviewed by: tmm, gibbs
implementation of a largely MI pmap_object_init_pt() for vnode-backed
objects. pmap_enter_quick() is implemented via pmap_enter() on sparc64
and powerpc.
- Correct a mismatch between pmap_object_init_pt()'s prototype and its
various implementations. (I plan to keep pmap_object_init_pt() as
the MD hook for device-backed objects on i386 and amd64.)
- Correct an error in ia64's pmap_enter_quick() and adjust its interface
to match the other versions. Discussed with: marcel
being requested is outside of the range of the direct map region. eg:
for pci windows. While here, increase the minimum size of the direct
map region to be 4GB instead of 1GB.
to the machine-independent parts of the VM. At the same time, this
introduces vm object locking for the non-i386 platforms.
Two details:
1. KSTACK_GUARD has been removed in favor of KSTACK_GUARD_PAGES. The
different machine-dependent implementations used various combinations
of KSTACK_GUARD and KSTACK_GUARD_PAGES. To disable guard page, set
KSTACK_GUARD_PAGES to 0.
2. Remove the (unnecessary) clearing of PG_ZERO in vm_thread_new. In
5.x, (but not 4.x,) PG_ZERO can only be set if VM_ALLOC_ZERO is passed
to vm_page_alloc() or vm_page_grab().
