Add support for kvm_hypercall4(); PAE wants it.
Signed-off-by: Matias Zabaljauregui <zabaljauregui at gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
replace LHCALL_SET_PMD with LHCALL_SET_PGD hypercall name
(That's really what it is, and the confusion gets worse with PAE support)
Signed-off-by: Matias Zabaljauregui <zabaljauregui@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Reported-by: Jeremy Fitzhardinge <jeremy@goop.org>
Some cleanups and replace direct assignment with native_set_* macros which properly handle 64-bit entries when PAE is activated
Signed-off-by: Matias Zabaljauregui <zabaljauregui@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The downside of the last patch which made restore_flags and irq_enable
check interrupts is that they are now too big to be patched directly
into the callsites, so the C versions are always used.
But the C versions go via PV_CALLEE_SAVE_REGS_THUNK which saves all
the registers. In fact, we don't need any registers in the fast path,
so we can do better than this if we actually code them in assembler.
The results are in the noise, but since it's about the same amount of
code, it's worth applying.
1GB Guest->Host: input(suppressed),output(suppressed)
Before:
Seconds: 0:16.53
Packets: 377268,753673
Interrupts: 22461,24297
Notifications: 1(5245),21303(732370)
Net IRQs triggered: 377023(245),42578(711095)
After:
Seconds: 0:16.48
Packets: 377289,753673
Interrupts: 22281,24465
Notifications: 1(5245),21296(732377)
Net IRQs triggered: 377060(229),42564(711109)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
lguest never checked for pending interrupts when enabling interrupts, and
things still worked. However, it makes a significant difference to TCP
performance, so it's time we fixed it by introducing a pending_irq flag
and checking it on irq_restore and irq_enable.
These two routines are now too big to patch into the 8/10 bytes
patch space, so we drop that code.
Note: The high latency on interrupt delivery had a very curious
effect: once everything else was optimized, networking without GSO was
faster than networking with GSO, since more interrupts were sent and
hence a greater chance of one getting through to the Guest!
Note2: (Almost) Closing the same loophole for iret doesn't have any
measurable effect, so I'm leaving that patch for the moment.
Before:
1GB tcpblast Guest->Host: 30.7 seconds
1GB tcpblast Guest->Host (no GSO): 76.0 seconds
After:
1GB tcpblast Guest->Host: 6.8 seconds
1GB tcpblast Guest->Host (no GSO): 27.8 seconds
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Copy from arch/x86/kernel/irqinit_32.c: we don't use the vectors beyond
LGUEST_IRQS (if any), but we might as well set them all.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Sachin Sant reported these compiler errors:
CC arch/powerpc/kernel/power7-pmu.o
arch/powerpc/kernel/power7-pmu.c:297: error: PERF_COUNT_CPU_CYCLES undeclared here (not in a function)
Which happened because a last-minute rename of symbols crossed with
the Power7 support patch.
Fix this by using the new symbol names.
Reported-by: Sachin Sant <sachinp@in.ibm.com>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: linuxppc-dev@ozlabs.org
LKML-Reference: <1244788494.5554.1.camel@ht.satnam>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Everyone cut and paste this comment from my original one. We now do
it generically, so cut the comments.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Amerigo Wang <amwang@redhat.com>
It's theoretically possible that there are exception table entries
which point into the (freed) init text of modules. These could cause
future problems if other modules get loaded into that memory and cause
an exception as we'd see the wrong fixup. The only case I know of is
kvm-intel.ko (when CONFIG_CC_OPTIMIZE_FOR_SIZE=n).
Amerigo fixed this long-standing FIXME in the x86 version, but this
patch is more general.
This implements trim_init_extable(); most archs are simple since they
use the standard lib/extable.c sort code. Alpha and IA64 use relative
addresses in their fixups, so thier trimming is a slight variation.
Sparc32 is unique; it doesn't seem to define ARCH_HAS_SORT_EXTABLE,
yet it defines its own sort_extable() which overrides the one in lib.
It doesn't sort, so we have to mark deleted entries instead of
actually trimming them.
Inspired-by: Amerigo Wang <amwang@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: linux-alpha@vger.kernel.org
Cc: sparclinux@vger.kernel.org
Cc: linux-ia64@vger.kernel.org
As Christoph Hellwig suggested, module_alloc() actually can be
unified for i386 and x86_64 (of course, also UML).
Signed-off-by: WANG Cong <amwang@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: 'Ingo Molnar' <mingo@elte.hu>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Due to the previous merge, uml needs to be fixed.
Signed-off-by: WANG Cong <amwang@redhat.com>
Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Merge the same functions both in module_32.c and module_64.c into
module.c.
This is the first step to merge both of them finally.
Signed-off-by: WANG Cong <amwang@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The fixed-function performance counters do not work on current Atom
processors. Use the general-purpose ones instead.
Signed-off-by: Yong Wang <yong.y.wang@intel.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <20090612080855.GA2286@ywang-moblin2.bj.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The sparseirq changes (d7e51e66) played poorly with the Blackfin irqchip
implementation as we're still using the old hardirq method. Our bad irq
structure had a NULL kstat_irqs field so when all the common code tries
to increment this field, everything goes big bada boom.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
We have some test code that runs in userspace that exercises the exception
handling of the Blackfin pretty thoroughly. Part of the validation process
is checking the exact exception triggered, so export the last one seen to
userspace via debugfs when debugging is enabled for the test code to check.
Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The kgdb_ebin2mem() was decrementing the count variable to do parsing, but
then later still tries to use it based on its original meaning. So leave
it untouched and use a different variable to walk the memory.
Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The Blackfin kgdb code was all passing back positive errno values when it
really should have been using negative errno values.
Reported-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
There is no need for the L1 attribute to be on the prototype of the
access_ok() function as all consumers of the function do not care where it
lives -- they'll always use pcrel calls to get to it. This prevents
pointless recompiles of most of the system when this config option changes.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The common code already has a prototype for this function and we don't use
it anywhere in the Blackfin code, so punt it from the Blackfin headers.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The core string/clear user functions weren't checking the user pointers
which caused kernel crashes with some bad programs and tests (like LTP).
Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The EVT registers are all contiguous in the memory map, so using a loop to
initialize them all rather than hardcoding the list results in much better
generated code (a hardware loop rather than a whole bunch of individual
loads).
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Make sure the meaning of "lsl" is covered somewhere and it is clear why we
somewhat duplicate the sram alloc/free functions.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The latest Blackfin toolchain has fixed its relocation scheme to match
other ports: always use R_BFIN_ prefix and capitalize everything. This
brings the kernel in line with those fixes.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Rather than having to maintain a hard coded list of Blackfin variants, use
the SIC defines themselves. This fixes build problems on BF51x/BF538 under
some configurations as they were missing from one of the lists.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Make sure the internal core buffers are flushed before telling the DMA
engine to fetch the descriptor structure so that it gets the right values.
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Due to a processor anomaly (05000263 to be exact), most Blackfin parts
cannot keep the embedded filesystem image directly after the kernel in
RAM. Instead, the filesystem needs to be relocated to the end of memory.
As such, we need to tweak the map addr/size during boot for Blackfin
systems.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Commit 6b3087c6 (which introduced Blackfin SMP) broke command line passing
when the DEBUG_DOUBLEFAULT config option was enabled. Switch the code to
using a scratch register and not R7 which holds the command line.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
This restores some L1 reservation logic that was lost during the Blackfin
SMP merge.
Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Now that the sram_init() function exists only to call the bfin_sram_init()
after the punting of the reserve_pda() function, simply merge the two to
avoid pointless overhead.
Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The Per-processor Data Area isn't actually reserved by this function, and
all it ended up doing was issuing a printk(), so punt it.
Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
First we fix the prototypes for functions that return boolean values by
using "int" rather than "uint16_t". Then we introduce a get_gptimer_run()
function for checking the current run status of a timer, and then we add a
disable_gptimers_sync() function which parallels disable_gptimers() with
corresponding normal "_sync" behavior.
Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
People often copy & paste crash messages without surrounding context, so
include common useful information like system/processor stats in the crash
summary. This should smooth over the report/test cycle a bit more.
Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Returning too fast with a bad RETI can trigger false errors.
Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
When displaying a crash dump, make sure accessing the stack is safe so
we don't crash at the same time.
Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Hardware errors on the Blackfin architecture are queued by nature of the
hardware design. Things that could generate a hardware level queue up at
the system interface and might not process until much later, at which
point the system would send a notification back to the core.
As such, it is possible for user space code to do something that would
trigger a hardware error, but have it delay long enough for the process
context to switch. So when the hardware error does signal, we mistakenly
evaluate it as a different process or as kernel context and panic (erp!).
This makes it pretty difficult to find the offending context. But wait,
there is good news somewhere.
By forcing a SSYNC in the interrupt entry, we force all pending queues at
the system level to be processed and all hardware errors to be signaled.
Then we check the current interrupt state to see if the hardware error is
now signaled. If so, we re-queue the current interrupt and return thus
allowing the higher priority hardware error interrupt to process properly.
Since we haven't done any other context processing yet, the right context
will be selected and killed. There is still the possibility that the
exact offending instruction will be unknown, but at least we'll have a
much better idea of where to look.
The downside of course is that this causes system-wide syncs at every
interrupt point which results in significant performance degradation.
Since this situation should not occur in any properly configured system
(as hardware errors are triggered by things like bad pointers), make it a
debug configuration option and disable it by default.
Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
When possible, work around anomaly 05000220 (external memory is write
back cached, but L2 is not cached). If not possible, detect the
conditions at build time and reject any qualifying configurations.
Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Try to keep the naming conventions consistent, so:
SPI_ADC_BF533 -> BFIN_SPI_ADC
TWI_LCD -> BFIN_TWI_LCD
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
This way we properly catch and kill applications that jump to a NULL ptr.
Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
For systems where the core cycles are not a usable tick source (like SMP
or cycles gets updated), enable gptimer0 as an alternative.
Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Add some notes for anomaly 05000120 to make sure we work around it.
Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The two high address lines on the BF51x are not dedicated which means we
need to handle them like any other peripheral pin if we want to access the
upper 2MB of parallel flash.
Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Detect and reject operating conditions for anomaly 05000274 since the
problem cannot be worked around in software.
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Matias Zabaljauregui