Seems like needless clutter having a bunch of #if defined(CONFIG_$ARCH) in
include/linux/cache.h. Move the per architecture section definition to
asm/cache.h, and keep the if-not-defined dummy case in linux/cache.h to
catch architectures which don't implement the section.
Verified that symbols still go in .data.read_mostly on parisc,
and the compile doesn't break.
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
1) huge_pte_offset() did not check the page table hierarchy
elements as being empty correctly, resulting in an OOPS
2) Need platform specific hugetlb_get_unmapped_area() to handle
the top-down vs. bottom-up address space allocation strategies.
Signed-off-by: David S. Miller <davem@davemloft.net>
We only need to write an invalid tag every 16 bytes,
so taking advantage of this can save many instructions
compared to the simple memset() call we make now.
A prefetching implementation is implemented for sun4u
and a block-init store version if implemented for Niagara.
The next trick is to be able to perform an init and
a copy_tsb() in parallel when growing a TSB table.
Signed-off-by: David S. Miller <davem@davemloft.net>
Put it one page below the top of the 32-bit address space.
This gives us ~16MB more address space to work with.
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently allocations are very constrained for 32-bit processes.
It grows down-up from 0x70000000 to 0xf0000000 which gives about
2GB of stack + dynamic mmap() space.
So support the top-down method, and we need to override the
generic helper function in order to deal with D-cache coloring.
With these changes I was able to squeeze out a mmap() just over
3.6GB in size in a 32-bit process.
Signed-off-by: David S. Miller <davem@davemloft.net>
This is good for up to %50 performance improvement of some test cases.
The problem has been the race conditions, and hopefully I've plugged
them all up here.
1) There was a serious race in switch_mm() wrt. lazy TLB
switching to and from kernel threads.
We could erroneously skip a tsb_context_switch() and thus
use a stale TSB across a TSB grow event.
There is a big comment now in that function describing
exactly how it can happen.
2) All code paths that do something with the TSB need to be
guarded with the mm->context.lock spinlock. This makes
page table flushing paths properly synchronize with both
TSB growing and TLB context changes.
3) TSB growing events are moved to the end of successful fault
processing. Previously it was in update_mmu_cache() but
that is deadlock prone. At the end of do_sparc64_fault()
we hold no spinlocks that could deadlock the TSB grow
sequence. We also have dropped the address space semaphore.
While we're here, add prefetching to the copy_tsb() routine
and put it in assembler into the tsb.S file. This piece of
code is quite time critical.
There are some small negative side effects to this code which
can be improved upon. In particular we grab the mm->context.lock
even for the tsb insert done by update_mmu_cache() now and that's
a bit excessive. We can get rid of that locking, and the same
lock taking in flush_tsb_user(), by disabling PSTATE_IE around
the whole operation including the capturing of the tsb pointer
and tsb_nentries value. That would work because anyone growing
the TSB won't free up the old TSB until all cpus respond to the
TSB change cross call.
I'm not quite so confident in that optimization to put it in
right now, but eventually we might be able to and the description
is here for reference.
This code seems very solid now. It passes several parallel GCC
bootstrap builds, and our favorite "nut cruncher" stress test which is
a full "make -j8192" build of a "make allmodconfig" kernel. That puts
about 256 processes on each cpu's run queue, makes lots of process cpu
migrations occur, causes lots of page table and TLB flushing activity,
incurs many context version number changes, and it swaps the machine
real far out to disk even though there is 16GB of ram on this test
system. :-)
Signed-off-by: David S. Miller <davem@davemloft.net>
Report 'sun4v' when appropriate in /proc/cpuinfo
Remove all the verifications of the OBP version string. Just
make sure it's there, and report it raw in the bootup logs and
via /proc/cpuinfo.
Signed-off-by: David S. Miller <davem@davemloft.net>
The mapping is a simple "(cpuid >> 2) == core" for now.
Later we'll add more sophisticated code that will walk
the sun4v machine description and figure this out from
there.
We should also add core mappings for jaguar and panther
processors.
Signed-off-by: David S. Miller <davem@davemloft.net>
This has been pending for a long time, and the fact
that we waste a ton of ram on some configurations
kind of pushed things over the edge.
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't piggy back the SMP receive signal code to do the
context version change handling.
Instead allocate another fixed PIL number for this
asynchronous cross-call. We can't use smp_call_function()
because this thing is invoked with interrupts disabled
and a few spinlocks held.
Also, fix smp_call_function_mask() to count "cpus" correctly.
There is no guarentee that the local cpu is in the mask
yet that is exactly what this code was assuming.
Signed-off-by: David S. Miller <davem@davemloft.net>
1) Always spin_lock_init() in init_context(). The caller essentially
clears it out, or copies the mm info from the parent. In both
cases we need to explicitly initialize the spinlock.
2) Always do explicit IRQ disabling while taking mm->context.lock
and ctx_alloc_lock.
Signed-off-by: David S. Miller <davem@davemloft.net>
The UltraSPARC T1 manual recommends this because the chip
could instruction prefetch into the VA hole, and this would
also make decoding certain kinds of memory access traps
more difficult (because the chip sign extends certain pieces
of trap state).
Signed-off-by: David S. Miller <davem@davemloft.net>
There were several bugs in the SUN4V cpu mondo dispatch code.
In fact, if we ever got a EWOULDBLOCK or other error from
the hypervisor call, we'd potentially send a cpu mondo multiple
times to the same cpu and even worse we could loop until the
timeout resending the same mondo over and over to such cpus.
So let's bulletproof this thing as follows:
1) Implement cpu_mondo_send() and cpu_state() hypervisor calls
in arch/sparc64/kernel/entry.S, add prototypes to asm/hypervisor.h
2) Don't build and update the cpulist using inline functions, this
was causing the cpu mask to not get updated in the caller.
3) Disable interrupts during the entire mondo send, otherwise our
cpu list and/or mondo block could get overwritten if we take
an interrupt and do a cpu mondo send on the current cpu.
4) Check for all possible error return types from the cpu_mondo_send()
hypervisor call. In particular:
HV_EOK) Our work is done, all cpus have received the mondo.
HV_CPUERROR) One or more of the cpus in the cpu list we passed
to the hypervisor are in error state. Use cpu_state()
calls over the entries in the cpu list to see which
ones. Record them in "error_mask" and report this
after we are done sending the mondo to cpus which are
not in error state.
HV_EWOULDBLOCK) We need to keep trying.
Any other error we consider fatal, we report the event and exit
immediately.
5) We only timeout if forward progress is not made. Forward progress
is defined as having at least one cpu get the mondo successfully
in a given cpu_mondo_send() call. Otherwise we bump a counter
and delay a little. If the counter hits a limit, we signal an
error and report the event.
Also, smp_call_function_mask() error handling reports the number
of cpus incorrectly.
Signed-off-by: David S. Miller <davem@davemloft.net>
The context allocation scheme we use depends upon there being a 1<-->1
mapping from cpu to physical TLB for correctness. Chips like Niagara
break this assumption.
So what we do is notify all cpus with a cross call when the context
version number changes, and if necessary this makes them allocate
a valid context for the address space they are running at the time.
Stress tested with make -j1024, make -j2048, and make -j4096 kernel
builds on a 32-strand, 8 core, T2000 with 16GB of ram.
Signed-off-by: David S. Miller <davem@davemloft.net>
Niagara helps us find a ancient bug in the sparc64 port :-)
The ASI_* values are plain constant defines, thus signed 32-bit
on sparc64. To put shift this into the regs->tstate value we were
doing or'ing "(ASI_PNF << 24)" into there.
ASI_PNF is 0x82 and shifted left by 24 makes that topmost bit the
sign bit in a 32-bit value. This would get sign extended to 64-bits
and thus corrupt the top-half of the reg->tstate value.
This never caused problems in pre-Niagara cpus because the only thing
up there were the condition code values. But Niagara has the global
register level field, and this all 1's value is illegal there so
Niagara gives an illegal instruction trap due to this bug.
I'm pretty sure this bug is about as old as the sparc64 port itself.
This also points out that we weren't setting ASI_PNF for 32-bit tasks.
We should, so fix that while we're here.
Signed-off-by: David S. Miller <davem@davemloft.net>
It can map all of the linear kernel mappings with zero TSB hash
conflicts for systems with 16GB or less ram. In such cases, on
SUN4V, once we load up this TSB the first time with all the
mappings, we never take a linear kernel mapping TLB miss ever
again, the hypervisor handles them all.
Signed-off-by: David S. Miller <davem@davemloft.net>
Set, but never used.
We used to use this for dynamic IRQ retargetting, but that
code died a long time ago.
Signed-off-by: David S. Miller <davem@davemloft.net>
The SUN4V convention with non-shared TSBs is that the context
bit of the TAG is clear. So we have to choose an "invalid"
bit and initialize new TSBs appropriately. Otherwise a zero
TAG looks "valid".
Make sure, for the window fixup cases, that we use the right
global registers and that we don't potentially trample on
the live global registers in etrap/rtrap handling (%g2 and
%g6) and that we put the missing virtual address properly
in %g5.
Signed-off-by: David S. Miller <davem@davemloft.net>
This gives more consistent bogomips and delay() semantics,
especially on sun4v. It gives weird looking values though...
Signed-off-by: David S. Miller <davem@davemloft.net>
We need to use the real hardware processor ID when
targetting interrupts, not the "define to 0" thing
the uniprocessor build gives us.
Also, fill in the Node-ID and Agent-ID fields properly
on sun4u/Safari.
Signed-off-by: David S. Miller <davem@davemloft.net>
The sibling cpu bringup is extremely fragile. We can only
perform the most basic calls until we take over the trap
table from the firmware/hypervisor on the new cpu.
This means no accesses to %g4, %g5, %g6 since those can't be
TLB translated without our trap handlers.
In order to achieve this:
1) Change sun4v_init_mondo_queues() so that it can operate in
several modes.
It can allocate the queues, or install them in the current
processor, or both.
The boot cpu does both in it's call early on.
Later, the boot cpu allocates the sibling cpu queue, starts
the sibling cpu, then the sibling cpu loads them in.
2) init_cur_cpu_trap() is changed to take the current_thread_info()
as an argument instead of reading %g6 directly on the current
cpu.
3) Create a trampoline stack for the sibling cpus. We do our basic
kernel calls using this stack, which is locked into the kernel
image, then go to our proper thread stack after taking over the
trap table.
4) While we are in this delicate startup state, we put 0xdeadbeef
into %g4/%g5/%g6 in order to catch accidental accesses.
5) On the final prom_set_trap_table*() call, we put &init_thread_union
into %g6. This is a hack to make prom_world(0) work. All that
wants to do is restore the %asi register using
get_thread_current_ds().
Longer term we should just do the OBP calls to set the trap table by
hand just like we do for everything else. This would avoid that silly
prom_world(0) issue, then we can remove the init_thread_union hack.
Signed-off-by: David S. Miller <davem@davemloft.net>
Use prom_startcpu_cpuid() on SUN4V instead of prom_startcpu().
We should really test for "SUNW,start-cpu-by-cpuid" presence
and use it if present even on SUN4U.
Signed-off-by: David S. Miller <davem@davemloft.net>
It doesn't like const variables being passed into
"i" constraing asm operations. It's a bug, but
there is nothing we can really do but work around
it.
Based upon a report from Andrew Morton.
Signed-off-by: David S. Miller <davem@davemloft.net>
Yes, you heard it right, they changed the PTE layout for
SUN4V. Ho hum...
This is the simple and inefficient way to support this.
It'll get optimized, don't worry.
Signed-off-by: David S. Miller <davem@davemloft.net>
Code patching did not sign extend negative branch
offsets correctly.
Kernel TLB miss path needs patching and %g4 register
preservation in order to handle SUN4V correctly.
Signed-off-by: David S. Miller <davem@davemloft.net>
This is where the virtual address of the fault status
area belongs.
To set it up we don't make a hypervisor call, instead
we call OBP's SUNW,set-trap-table with the real address
of the fault status area as the second argument. And
right before that call we write the virtual address into
ASI_SCRATCHPAD vaddr 0x0.
Signed-off-by: David S. Miller <davem@davemloft.net>
Add assembler file for PCI hypervisor calls.
Setup basic skeleton of SUN4V PCI controller driver.
Add 32-bit devhandle to PBM struct, as this is needed for
hypervisor calls.
Signed-off-by: David S. Miller <davem@davemloft.net>
Abstract out IOMMU operations so that we can have a different
set of calls on sun4v, which needs to do things through
hypervisor calls.
Signed-off-by: David S. Miller <davem@davemloft.net>
When we register a TSB with the hypervisor, so that it or hardware can
handle TLB misses and do the TSB walk for us, the hypervisor traps
down to these trap when it incurs a TSB miss.
Processing is simple, we load the missing virtual address and context,
and do a full page table walk.
Signed-off-by: David S. Miller <davem@davemloft.net>
We look for "SUNW,sun4v" in the 'compatible' property
of the root OBP device tree node.
Protect every %ver register access, to make sure it is
not touched on sun4v, as %ver is hyperprivileged there.
Lock kernel TLB entries using hypervisor calls instead of
calls into OBP.
Signed-off-by: David S. Miller <davem@davemloft.net>
Technically the hypervisor call supports sending in a list
of all cpus to get the cross-call, but I only pass in one
cpu at a time for now.
The multi-cpu support is there, just ifdef'd out so it's easy to
enable or delete it later.
Signed-off-by: David S. Miller <davem@davemloft.net>
Sun4v has 4 interrupt queues: cpu, device, resumable errors,
and non-resumable errors. A set of head/tail offset pointers
help maintain a work queue in physical memory. The entries
are 64-bytes in size.
Each queue is allocated then registered with the hypervisor
as we bring cpus up.
The two error queues each get a kernel side buffer that we
use to quickly empty the main interrupt queue before we
call up to C code to log the event and possibly take evasive
action.
Signed-off-by: David S. Miller <davem@davemloft.net>
Things are a little tricky because, unlike sun4u, we have
to:
1) do a hypervisor trap to do the TLB load.
2) do the TSB lookup calculations by hand
Signed-off-by: David S. Miller <davem@davemloft.net>
If we're just switching between different alternate global
sets, nop it out on sun4v. Also, get rid of all of the
alternate global save/restore in the OBP CIF trampoline code.
Signed-off-by: David S. Miller <davem@davemloft.net>
They are totally unnecessary because:
1) Interrupts are already disabled when switch_to()
runs.
2) We don't use hard-coded alternate globals any longer.
This found a case in rtrap, which still assumed alternate
global %g6 was current_thread_info(), and that is fixed
by this changeset as well.
Signed-off-by: David S. Miller <davem@davemloft.net>
When saving and restoing trap state, do the window spill/fill
handling inline so that we never trap deeper than 2 trap levels.
This is important for chips like Niagara.
The window fixup code is massively simplified, and many more
improvements are now possible.
Signed-off-by: David S. Miller <davem@davemloft.net>
On uniprocessor, it's always zero for optimize that.
On SMP, the jmpl to the stub kills the return address stack in the cpu
branch prediction logic, so expand the code sequence inline and use a
code patching section to fix things up. This also always better and
explicit register selection, which will be taken advantage of in a
future changeset.
The hard_smp_processor_id() function is big, so do not inline it.
Fix up tests for Jalapeno to also test for Serrano chips too. These
tests want "jbus Ultra-IIIi" cases to match, so that is what we should
test for.
Signed-off-by: David S. Miller <davem@davemloft.net>
The are distrupting, which by the sparc v9 definition means they
can only occur when interrupts are enabled in the %pstate register.
This never occurs in any of the trap handling code running at
trap levels > 0.
So just mark it as an unexpected trap.
This allows us to kill off the cee_stuff member of struct thread_info.
Signed-off-by: David S. Miller <davem@davemloft.net>
This way we don't need to lock the TSB into the TLB.
The trick is that every TSB load/store is registered into
a special instruction patch section. The default uses
virtual addresses, and the patch instructions use physical
address load/stores.
We can't do this on all chips because only cheetah+ and later
have the physical variant of the atomic quad load.
Signed-off-by: David S. Miller <davem@davemloft.net>
Since we use %g5 itself as a temporary, it can get clobbered
if we take an interrupt mid-stream and thus cause end up with
the final %g5 value too early as a result of rtrap processing.
Set %g5 at the very end, atomically, to avoid this problem.
Signed-off-by: David S. Miller <davem@davemloft.net>
The TSB_LOCK_BIT define is actually a special
value shifted down by 32-bits for the assembler
code macros.
In C code, this isn't what we want.
Signed-off-by: David S. Miller <davem@davemloft.net>
As the RSS grows, grow the TSB in order to reduce the likelyhood
of hash collisions and thus poor hit rates in the TSB.
This definitely needs some serious tuning.
Signed-off-by: David S. Miller <davem@davemloft.net>
This also cleans up tsb_context_switch(). The assembler
routine is now __tsb_context_switch() and the former is
an inline function that picks out the bits from the mm_struct
and passes it into the assembler code as arguments.
setup_tsb_parms() computes the locked TLB entry to map the
TSB. Later when we support using the physical address quad
load instructions of Cheetah+ and later, we'll simply use
the physical address for the TSB register value and set
the map virtual and PTE both to zero.
Signed-off-by: David S. Miller <davem@davemloft.net>
Move {init_new,destroy}_context() out of line.
Do not put huge pages into the TSB, only base page size translations.
There are some clever things we could do here, but for now let's be
correct instead of fancy.
Signed-off-by: David S. Miller <davem@davemloft.net>
UltraSPARC has special sets of global registers which are switched to
for certain trap types. There is one set for MMU related traps, one
set of Interrupt Vector processing, and another set (called the
Alternate globals) for all other trap types.
For what seems like forever we've hard coded the values in some of
these trap registers. Some examples include:
1) Interrupt Vector global %g6 holds current processors interrupt
work struct where received interrupts are managed for IRQ handler
dispatch.
2) MMU global %g7 holds the base of the page tables of the currently
active address space.
3) Alternate global %g6 held the current_thread_info() value.
Such hardcoding has resulted in some serious issues in many areas.
There are some code sequences where having another register available
would help clean up the implementation. Taking traps such as
cross-calls from the OBP firmware requires some trick code sequences
wherein we have to save away and restore all of the special sets of
global registers when we enter/exit OBP.
We were also using the IMMU TSB register on SMP to hold the per-cpu
area base address, which doesn't work any longer now that we actually
use the TSB facility of the cpu.
The implementation is pretty straight forward. One tricky bit is
getting the current processor ID as that is different on different cpu
variants. We use a stub with a fancy calling convention which we
patch at boot time. The calling convention is that the stub is
branched to and the (PC - 4) to return to is in register %g1. The cpu
number is left in %g6. This stub can be invoked by using the
__GET_CPUID macro.
We use an array of per-cpu trap state to store the current thread and
physical address of the current address space's page tables. The
TRAP_LOAD_THREAD_REG loads %g6 with the current thread from this
table, it uses __GET_CPUID and also clobbers %g1.
TRAP_LOAD_IRQ_WORK is used by the interrupt vector processing to load
the current processor's IRQ software state into %g6. It also uses
__GET_CPUID and clobbers %g1.
Finally, TRAP_LOAD_PGD_PHYS loads the physical address base of the
current address space's page tables into %g7, it clobbers %g1 and uses
__GET_CPUID.
Many refinements are possible, as well as some tuning, with this stuff
in place.
Signed-off-by: David S. Miller <davem@davemloft.net>
Taking a nod from the powerpc port.
With the per-cpu caching of both the page allocator and SLAB, the
pgtable quicklist scheme becomes relatively silly and primitive.
Signed-off-by: David S. Miller <davem@davemloft.net>
We now use the TSB hardware assist features of the UltraSPARC
MMUs.
SMP is currently knowingly broken, we need to find another place
to store the per-cpu base pointers. We hid them away in the TSB
base register, and that obviously will not work any more :-)
Another known broken case is non-8KB base page size.
Also noticed that flush_tlb_all() is not referenced anywhere, only
the internal __flush_tlb_all() (local cpu only) is used by the
sparc64 port, so we can get rid of flush_tlb_all().
The kernel gets it's own 8KB TSB (swapper_tsb) and each address space
gets it's own private 8K TSB. Later we can add code to dynamically
increase the size of per-process TSB as the RSS grows. An 8KB TSB is
good enough for up to about a 4MB RSS, after which the TSB starts to
incur many capacity and conflict misses.
We even accumulate OBP translations into the kernel TSB.
Another area for refinement is large page size support. We could use
a secondary address space TSB to handle those.
Signed-off-by: David S. Miller <davem@davemloft.net>
We must use the "a" (allocate) attribute every time we
emit an entry into the __ex_table section.
For consistency, use "a" instead of #alloc which is some
Solaris compat cruft GNU as provides on Sparc.
Signed-off-by: David S. Miller <davem@davemloft.net>
The change to kernel/sched.c's init code to use for_each_cpu()
requires that the cpu_possible_map be setup much earlier.
Set it up via setup_arch(), constrained to NR_CPUS, and later
constrain it to max_cpus in smp_prepare_cpus().
This fixes SMP booting on sparc64.
Signed-off-by: David S. Miller <davem@davemloft.net>
Make new MADV_REMOVE, MADV_DONTFORK, MADV_DOFORK consistent across all
arches. The idea is to make it possible to use them portably even before
distros include them in libc headers.
Move common flags to asm-generic/mman.h
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Cc: Roland Dreier <rolandd@cisco.com>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Currently, copy-on-write may change the physical address of a page even if the
user requested that the page is pinned in memory (either by mlock or by
get_user_pages). This happens if the process forks meanwhile, and the parent
writes to that page. As a result, the page is orphaned: in case of
get_user_pages, the application will never see any data hardware DMA's into
this page after the COW. In case of mlock'd memory, the parent is not getting
the realtime/security benefits of mlock.
In particular, this affects the Infiniband modules which do DMA from and into
user pages all the time.
This patch adds madvise options to control whether memory range is inherited
across fork. Useful e.g. for when hardware is doing DMA from/into these
pages. Could also be useful to an application wanting to speed up its forks
by cutting large areas out of consideration.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Acked-by: Hugh Dickins <hugh@veritas.com>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Kyle McMartin