To the issue: some point during 2.6.20 development, Paul Mackerras
introduced the "lazy IRQ disabling" patch (very cool work, BTW).
In that patch, the performance monitor unit exception was marked as
"maskable", in the sense that if interrupts were soft-disabled, that
exception could be ignored. This broke my PowerPC profiling code.
The symptom that I see is that a varying number of interrupts
(from 0 to $n$, typically closer to 0) get delivered, when, in
reality, it should always be very close to $n$.
The issue stems from the way masking is being done. Masking in
this fashion seems to work well with the decrementer and external
interrupts, because they are raised again until "really" handled.
For the PMU, however, this does not apply (at least on my Xserver
machine with a 970FX processor). If the PMU exception is not handled,
it will _not_ be re-raised (at least on my machine). The documentation
states that the PMXE bit in MMCR0 is set to 0 when the PMU exception
is raised. However, software must re-set the bit to re-enable PMU
exceptions. If the exception is ignored (as currently) not only is
that interrupt lost, but because software does not re-set PMXE, the
PMU registers are "frozen" forever.
[This patch means that performance monitor exceptions are taken and
handled even if irqs are off, as long as some other interrupt hasn't
come along and caused interrupts to be hard-disabled. In this sense
the PMU exception becomes like an NMI. The oprofile code for most
powerpc processors does nothing that is unsafe in an NMI context, but
the Cell oprofile code does a spin_lock_irqsave. However, that turns
out to be OK because Cell doesn't actually use the performance
monitor exception; performance monitor interrupts come in as a
regular interrupt on Cell, so will be disabled when irqs are off.
-- paulus.]
Signed-off-by: Paul Mackerras <paulus@samba.org>
Commit 3ccfc65c50 missed the same fixes for
legacy iSeries specific code, so make some more symbols no longer global.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Since iSeries is merged to MULTIPLATFORM, there is no way to build a 64bit
kernel without MULTIPLATFORM, so PPC_MULTIPLATFORM can be removed in
64bit-only files.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
It turns out that the linker warnings on 64-bit powerpc about "section
blah exceeds stub group size" were being triggered by conditional
branches in head_64.S branching to global symbols, whether in
head_64.S or in other files. This eliminates the warnings by making
some global symbols in head_64.S no longer global, and by rearranging
some branches.
Signed-off-by: Paul Mackerras <paulus@samba.org>
[ Yee-haa. Maybe I'll notice newly introduced real warnings now - Linus ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There are currently two versions of the functions for applying the
feature fixups, one for CPU features and one for firmware features. In
addition, they are both in assembly and with separate implementations
for 32 and 64 bits. identify_cpu() is also implemented in assembly and
separately for 32 and 64 bits.
This patch replaces them with a pair of C functions. The call sites are
slightly moved on ppc64 as well to be called from C instead of from
assembly, though it's a very small change, and thus shouldn't cause any
problem.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The lazy IRQ disable patch missed a couple of places where the
interrupt enable flags need to be restored correctly. First, we
weren't restoring the paca->hard_enabled flag on interrupt exit.
Instead of saving it on entry, we compute it from the MSR_EE bit
in the MSR we are restoring at exit. Secondly, the MMU hash miss
code was clearing both paca->soft_enabled and paca->hard_enabled
but not restoring them in the case where hash_page was able to
resolve the miss from the Linux page tables.
Signed-off-by: Paul Mackerras <paulus@samba.org>
This implements a lazy strategy for disabling interrupts. This means
that local_irq_disable() et al. just clear the 'interrupts are
enabled' flag in the paca. If an interrupt comes along, the interrupt
entry code notices that interrupts are supposed to be disabled, and
clears the EE bit in SRR1, clears the 'interrupts are hard-enabled'
flag in the paca, and returns. This means that interrupts only
actually get disabled in the processor when an interrupt comes along.
When interrupts are enabled by local_irq_enable() et al., the code
sets the interrupts-enabled flag in the paca, and then checks whether
interrupts got hard-disabled. If so, it also sets the EE bit in the
MSR to hard-enable the interrupts.
This has the potential to improve performance, and also makes it
easier to make a kernel that can boot on iSeries and on other 64-bit
machines, since this lazy-disable strategy is very similar to the
soft-disable strategy that iSeries already uses.
This version renames paca->proc_enabled to paca->soft_enabled, and
changes a couple of soft-disables in the kexec code to hard-disables,
which should fix the crash that Michael Ellerman saw. This doesn't
yet use a reserved CR field for the soft_enabled and hard_enabled
flags. This applies on top of Stephen Rothwell's patches to make it
possible to build a combined iSeries/other kernel.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Reduce default cacheline size on 64-bit powerpc from 128 bytes to 64.
This is the architected minimum. In most cases we'll still end up using
cache line information from the device tree, but defaults are used during
early boot and doing a few dcbst/icbi's too many there won't do any harm.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Cleanup CPU inits a bit more, Geoff Levand already did some earlier.
* Move CPU state save to cpu_setup, since cpu_setup is only ever done
on cpu 0 on 64-bit and save is never done more than once.
* Rename __restore_cpu_setup to __restore_cpu_ppc970 and add
function pointers to the cputable to use instead. Powermac always
has 970 so no need to check there.
* Rename __970_cpu_preinit to __cpu_preinit_ppc970 and check PVR before
calling it instead of in it, it's too early to use cputable.
* Rename pSeries_secondary_smp_init to generic_secondary_smp_init since
everyone but powermac and iSeries use it.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The firmware of POWER4 and JS20 systems does not switch the cpu to 64bit
mode when the registered system_reset and machine_check handlers get called.
If a 32bit process runs on that cpu at the time of the event, the cpu
remains in 32bit mode. xmon and kdump can not deal with it, the result is
an error like 'Bad kernel stack pointer fff2aad0 at 3200'.
xmon just loses some register info, but booting the kdump kernel usually fails.
Both handlers are not hot paths. Duplicate the EXCEPTION_PROLOG_PSERIES macro
and add two instructions to switch to 64bit:
li r11,5;
rldimi r10,r11,61,0;
Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
There's a small period early in boot where we don't know which cpu we're
running on. That's ok, except that it means we have no paca, or more
correctly that our paca pointer points somewhere random.
So that we can safely call things like smp_processor_id(), we need a paca,
so just assume we're on cpu 0. No code should _write_ to the paca before
we've set the correct one up.
We setup the proper paca after we've scanned the flat device tree in
early_setup(), so there's no need to do it again in start_here_common.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This patch allows the kernel to recognized that it was loaded at zero
and the copy down of the image is unnecessary. This is useful for
Simulators and kexec models.
On a typical 3.8 MiB vmlinux.strip this saves about 2.3 million instructions.
Signed-off-by: Paul Mackerras <paulus@samba.org>
There's no need to set the boot cpu paca in asm, so do it in C so us
mere mortals can understand it.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
None of this seems to be necessary, so let's see if can remove it and not
break anything. Booted on iSeries & pSeries here.
NB. we don't remove the hvReleaseData, we just move it down so that the file
reads more clearly.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This is a first version of support for the Cell BE "Reliability,
Availability and Serviceability" features.
It doesn't yet handle some of the RAS interrupts (the ones described in
iic_is/iic_irr), I'm still working on a proper way to expose these. They
are essentially a cascaded controller by themselves (sic !) though I may
just handle them locally to the iic driver. I need also to sync with
David Erb on the way he hooked in the performance monitor interrupt.
So that's all for 2.6.17 and I'll do more work on that with my rework of
the powerpc interrupt layer that I'm hacking on at the moment.
Signed-off-by: Arnd Bergmann <arnd.bergmann@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
We weren't using the recommended sequence for putting the CPU into
nap mode. When I changed the idle loop, for some reason 7447A cpus
started hanging when we put them into nap mode. Changing to the
recommended sequence fixes that.
The complexity here is that the recommended sequence is a loop that
keeps putting the cpu back into nap mode. Clearly we need some way
to break out of the loop when an interrupt (external interrupt,
decrementer, performance monitor) occurs. Here we use a bit in
the thread_info struct to indicate that we need this, and the exception
entry code notices this and arranges for the exception to return
to the value in the link register, thus breaking out of the loop.
We use a new `local_flags' field in the thread_info which we can
alter without needing to use an atomic update sequence.
The PPC970 has the same recommended sequence, so we do the same thing
there too.
This also fixes a bug in the kernel stack overflow handling code on
32-bit, since it was causing a value that we needed in a register to
get trashed.
Signed-off-by: Paul Mackerras <paulus@samba.org>
We currently have a hack to flip the boot cpu and its secondary thread
to logical cpuid 0 and 1. This means the logical - physical mapping will
differ depending on which cpu is boot cpu. This is most apparent on
kexec, where we might kexec on any cpu and therefore change the mapping
from boot to boot.
The patch below does a first pass early on to work out the logical cpuid
of the boot thread. We then fix up some paca structures to match.
Ive also removed the boot_cpuid_phys variable for ppc64, to be
consistent we use get_hard_smp_processor_id(boot_cpuid) everywhere.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The address of variable val in prom_init_stdout is passed to prom_getprop.
prom_getprop casts the pointer to u32 and passes it to call_prom in the hope
that OpenFirmware stores something there.
But the pointer is truncated in the lower bits and the expected value is
stored somewhere else.
In my testing I had a stackpointer of 0x0023e6b4. val was at offset 120,
wich has address 0x0023e72c. But the value passed to OF was 0x0023e728.
c00000000040b710: 3b 01 00 78 addi r24,r1,120
...
c00000000040b754: 57 08 00 38 rlwinm r8,r24,0,0,28
...
c00000000040b784: 80 01 00 78 lwz r0,120(r1)
...
c00000000040b798: 90 1b 00 0c stw r0,12(r27)
...
The stackpointer came from 32bit code.
The chain was yaboot -> zImage -> vmlinux
PowerMac OpenFirmware does appearently not handle the ELF sections
correctly. If yaboot was compiled in
/usr/src/packages/BUILD/lilo-10.1.1/yaboot, then the stackpointer is
unaligned. But the stackpointer is correct if yaboot is compiled in
/tmp/yaboot.
This bug triggered since 2.6.15, now prom_getprop is an inline
function. gcc clears the lower bits, instead of just clearing the
upper 32 bits.
Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
yaboot is scrogged and calls us with an invalid stack alignment,
it seems.
Thanks to David Woodhouse to pointing me to the problem.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This implements accurate task and cpu time accounting for 64-bit
powerpc kernels. Instead of accounting a whole jiffy of time to a
task on a timer interrupt because that task happened to be running at
the time, we now account time in units of timebase ticks according to
the actual time spent by the task in user mode and kernel mode. We
also count the time spent processing hardware and software interrupts
accurately. This is conditional on CONFIG_VIRT_CPU_ACCOUNTING. If
that is not set, we do tick-based approximate accounting as before.
To get this accurate information, we read either the PURR (processor
utilization of resources register) on POWER5 machines, or the timebase
on other machines on
* each entry to the kernel from usermode
* each exit to usermode
* transitions between process context, hard irq context and soft irq
context in kernel mode
* context switches.
On POWER5 systems with shared-processor logical partitioning we also
read both the PURR and the timebase at each timer interrupt and
context switch in order to determine how much time has been taken by
the hypervisor to run other partitions ("steal" time). Unfortunately,
since we need values of the PURR on both threads at the same time to
accurately calculate the steal time, and since we can only calculate
steal time on a per-core basis, the apportioning of the steal time
between idle time (time which we ceded to the hypervisor in the idle
loop) and actual stolen time is somewhat approximate at the moment.
This is all based quite heavily on what s390 does, and it uses the
generic interfaces that were added by the s390 developers,
i.e. account_system_time(), account_user_time(), etc.
This patch doesn't add any new interfaces between the kernel and
userspace, and doesn't change the units in which time is reported to
userspace by things such as /proc/stat, /proc/<pid>/stat, getrusage(),
times(), etc. Internally the various task and cpu times are stored in
timebase units, but they are converted to USER_HZ units (1/100th of a
second) when reported to userspace. Some precision is therefore lost
but there should not be any accumulating error, since the internal
accumulation is at full precision.
Signed-off-by: Paul Mackerras <paulus@samba.org>
HMT support is currently broken and needs to be reworked to play nicely
with the SMT scheduler. Remove the bit rotten bits for the time being.
I also updated an incorrect comment, we enter __secondary_hold with the
physical cpu id in r3.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The runlatch SPR can take a lot of time to write. My original runlatch
code would set it on every exception entry even though most of the time
this was not required. It would also continually set it in the idle
loop, which is an issue on an SMT capable processor.
Now we cache the runlatch value in a threadinfo bit, and only check for
it in decrementer and hardware interrupt exceptions as well as the idle
loop. Boot on POWER3, POWER5 and iseries, and compile tested on pmac32.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
For UP to SMP kexec to work we need to jump into pSeries_secondary_smp_init
event on a UP + KEXEC kernel. The secondary cpus will not find their hw_cpu_id
in the paca and so they'll jump into kexec_wait, ready for a kexec.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This patch removes all self references and fixes references to files
in the now defunct arch/ppc64 tree. I think this accomplises
everything wanted, though there might be a few references I missed.
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Since 404849bbd2 we've been using
LOAD_REG_ADDRBASE, which uses the toc pointer, in decrementer_iSeries_masked.
This can explode if we take the decrementer interrupt while we're in a module,
because the toc pointer in r2 will be the module's toc pointer.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
At present the lppaca - the structure shared with the iSeries
hypervisor and phyp - is contained within the PACA, our own low-level
per-cpu structure. This doesn't have to be so, the patch below
removes it, making a separate array of lppaca structures.
This saves approximately 500*NR_CPUS bytes of image size and kernel
memory, because we don't need aligning gap between the Linux and
hypervisor portions of every PACA. On the other hand it means an
extra level of dereference in many accesses to the lppaca.
The patch also gets rid of several places where we assign the paca
address to a local variable for no particular reason.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This patch consolidates the variety of macros used for loading 32 or
64-bit constants in assembler (LOADADDR, LOADBASE, SET_REG_TO_*). The
idea is to make the set of macros consistent across 32 and 64 bit and
to make it more obvious which is the appropriate one to use in a given
situation. The new macros and their semantics are described in the
comments in ppc_asm.h.
In the process, we change several places that were unnecessarily using
immediate loads on ppc64 to use the GOT/TOC. Likewise we cleanup a
couple of places where we were clumsily subtracting PAGE_OFFSET with
asm instructions to use assemble-time arithmetic or the toreal() macro
instead.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
We don't set CONFIG_PPC_MULTIPLATFORM on iSeries (yet). Avoid
compiling in the prom_init stuff on iSeries.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Remove a comment in head.S which is no longer relevant.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
There's a few places where we need to fix things up for the kernel to work
if it's linked at 32MB:
- platforms/powermac/smp.c
To start secondary cpus on pmac we patch the reset vector, which is fine.
Except if we're above 32MB we don't have enough bits for an absolute branch,
it needs to relative.
- kernel/head_64.s
- A few branches in the cpu hold code need to load the full target address
and do a bctr.
- after_prom_start needs to load PHYSICAL_START as the dest address, not 0.
- The exception prolog needs to load the low word of the target adddress,
not just the low halfword.
- Fixup handling of the initial stab address.
- kernel/setup_64.c
smp_release_cpus() needs to write 1 to the spinloop flag near 0, not 32 MB.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The fwnmi vectors can be anywhere < 32 MB, so we need to use a trampoline
for them. The kdump kernel will register the trampoline addresses, which will
then jump up to the real code above 32 MB.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This patch removes several unnecessary fields from the paca:
- next_jiffy_update_tb was simply unused. Remove trivially.
- The exdsi exception save area was not used. There were plans to use
it, but they never seem to have gone anywhere. If they ever do, we
can put it back. Remove from the paca, and from asm-offsets.c
- The default_decr field was used from asm, but was only ever assigned
the value of tb_ticks_per_jiffy. Just access tb_ticks_per_jiffy from
asm directly instead.
Built and booted on POWER5 LPAR and iSeries RS64.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This patch merges platform codes. systemcfg->platform is no longer used,
systemcfg use in general is deprecated as much as possible (and renamed
_systemcfg before it gets completely moved elsewhere in a future patch),
_machine is now used on ppc64 along as ppc32. Platform codes aren't gone
yet but we are getting a step closer. A bunch of asm code in head[_64].S
is also turned into C code.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Adds a new CONFIG_PPC_64K_PAGES which, when enabled, changes the kernel
base page size to 64K. The resulting kernel still boots on any
hardware. On current machines with 4K pages support only, the kernel
will maintain 16 "subpages" for each 64K page transparently.
Note that while real 64K capable HW has been tested, the current patch
will not enable it yet as such hardware is not released yet, and I'm
still verifying with the firmware architects the proper to get the
information from the newer hypervisors.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There's no reason for smp_release_cpus() to be asm, and most people can make
more sense of C code. Add an extern declaration to smp.h and remove the custom
one in machine_kexec.c
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
This brings in a lot of changes from arch/ppc64/kernel/pmac_*.c to
arch/powerpc/platforms/powermac/*.c and makes various minor tweaks
elsewhere. On the powermac we now initialize ppc_md by copying
the whole pmac_md structure into it, which required some changes in
the ordering of initializations of individual fields of it.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Merge ppc32 and ppc64 versions of thread_info.h. They were pretty
similar already, the chief changes are:
- Instead of inline asm to implement current_thread_info(),
which needs to be different for ppc32 and ppc64, we use C with an
asm("r1") register variable. gcc turns it into the same asm as we
used to have for both platforms.
- We replace ppc32's 'local_flags' with the ppc64
'syscall_noerror' field. The noerror flag was in fact the only thing
in the local_flags field anyway, so the ppc64 approach is simpler, and
means we only need a load-immediate/store instead of load/mask/store
when clearing the flag.
- In readiness for 64k pages, when THREAD_SIZE will be less
than a page, ppc64 used kmalloc() rather than get_free_pages() to
allocate the kernel stack. With this patch we do the same for ppc32,
since there's no strong reason not to.
- For ppc64, we no longer export THREAD_SHIFT and THREAD_SIZE
via asm-offsets, thread_info.h can now be safely included in asm, as
on ppc32.
Built and booted on G4 Powerbook (ARCH=ppc and ARCH=powerpc) and
Power5 (ARCH=ppc64 and ARCH=powerpc).
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This is a bunch of mostly small fixes that are needed to get
ARCH=powerpc to compile for 64-bit. This adds setup_64.c from
arch/ppc64/kernel/setup.c and locks.c from arch/ppc64/lib/locks.c.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Although both ppc32 and ppc64 have a reloc_offset function, the ppc64
one produced the opposite sign to the ppc32 one. This standardizes
on the ppc32 sign and fixes the merged 64-bit code to account for that.
Signed-off-by: Paul Mackerras <paulus@samba.org>
This changes symbols like HID0, SPRG3, SRR0, SRR1 etc. that refer to
special purpose registers to SPRN_HID0, SPRN_SPRG3, etc. Using the
SPRN_ symbols clutters the namespace less, and the forthcoming merge
of asm/processor.h and asm/reg.h is going to remove the non-SPRN_
versions.
Signed-off-by: Paul Mackerras <paulus@samba.org>
These macros help in writing assembly code that works for both ppc32
and ppc64. With this we now have a common fpu.S. This takes out
load_up_fpu from head_64.S.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Livio Soares