Provide a kernel API and a sysfs entry which allow a user to specify
that when a card is PERSTed, it's image will stay the same, allowing
it to participate in EEH.
cxl_reset is used to reflash the card. In that case, we cannot safely
assert that the image will not change. Therefore, disallow cxl_reset
if the flag is set.
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
If the driver doesn't participate in EEH, the AFUs will be removed
by cxl_remove, which will be invoked by EEH.
If the driver does particpate in EEH, the vPHB needs to stick around
so that the it can particpate.
In both cases, we shouldn't remove the AFU/vPHB.
Reviewed-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
As with an adapter, some aspects of initialisation are done only once
in the lifetime of an AFU: for example, allocating memory, or setting
up sysfs/debugfs files.
However, we may want to be able to do some parts of the initialisation
multiple times: for example, in error recovery we want to be able to
tear down and then re-map IO memory and IRQs.
Therefore, refactor AFU init/teardown as follows.
- Create two new functions: 'cxl_configure_afu', and its pair
'cxl_deconfigure_afu'. As with the adapter functions,
these (de)configure resources that do not need to last the entire
lifetime of the AFU.
- Allocating and releasing memory remain the task of 'cxl_alloc_afu'
and 'cxl_release_afu'.
- Once-only functions that do not involve allocating/releasing memory
stay in the overarching 'cxl_init_afu'/'cxl_remove_afu' pair.
However, the task of picking an AFU mode and activating it has been
broken out.
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Some aspects of initialisation are done only once in the lifetime of
an adapter: for example, allocating memory for the adapter,
allocating the adapter number, or setting up sysfs/debugfs files.
However, we may want to be able to do some parts of the
initialisation multiple times: for example, in error recovery we
want to be able to tear down and then re-map IO memory and IRQs.
Therefore, refactor CXL init/teardown as follows.
- Keep the overarching functions 'cxl_init_adapter' and its pair,
'cxl_remove_adapter'.
- Move all 'once only' allocation/freeing steps to the existing
'cxl_alloc_adapter' function, and its pair 'cxl_release_adapter'
(This involves moving allocation of the adapter number out of
cxl_init_adapter.)
- Create two new functions: 'cxl_configure_adapter', and its pair
'cxl_deconfigure_adapter'. These two functions 'wire up' the
hardware --- they (de)configure resources that do not need to
last the entire lifetime of the adapter
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
- MMIO pointer unmapping is guarded by a null pointer check.
However, iounmap doesn't null the pointer, just invalidate it.
Therefore, explicitly null the pointer after unmapping.
- afu_desc_mmio also needs to be unmapped.
- PCI regions are allocated in cxl_map_adapter_regs.
Therefore they should be released in unmap, not elsewhere.
Acked-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Check if an IRQ is mapped before releasing it.
This will simplify future EEH code by allowing unconditional unmapping
of IRQs.
Acked-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Previously the SPA was allocated and freed upon entering and leaving
AFU-directed mode. This causes some issues for error recovery - contexts
hold a pointer inside the SPA, and they may persist after the AFU has
been detached.
We would ideally like to allocate the SPA when the AFU is allocated, and
release it until the AFU is released. However, we don't know how big the
SPA needs to be until we read the AFU descriptor.
Therefore, restructure the code:
- Allocate the SPA only once, on the first attach.
- Release the SPA only when the entire AFU is being released (not
detached). Guard the release with a NULL check, so we don't free
if it was never allocated (e.g. dedicated mode)
Acked-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
If the PCI channel has gone down, don't attempt to poke the hardware.
We need to guard every time cxl_whatever_(read|write) is called. This
is because a call to those functions will dereference an offset into an
mmio register, and the mmio mappings get invalidated in the EEH
teardown.
Check in the read/write functions in the header.
We give them the same semantics as usual PCI operations:
- a write to a channel that is down is ignored.
- a read from a channel that is down returns all fs.
Also, we try to access the MMIO space of a vPHB device as part of the
PCI disable path. Because that's a read that bypasses most of our usual
checks, we handle it explicitly.
As far as user visible warnings go:
- Check link state in file ops, return -EIO if down.
- Be reasonably quiet if there's an error in a teardown path,
or when we already know the hardware is going down.
- Throw a big WARN if someone tries to start a CXL operation
while the card is down. This gives a useful stacktrace for
debugging whatever is doing that.
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
We're about to make these more complex, so make them functions
first.
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
In the complete hotplug case, EEH PEs are supposed to be released
and set to NULL. Normally, this is done by eeh_remove_device(),
which is called from pcibios_release_device().
However, if something is holding a kref to the device, it will not
be released, and the PE will remain. eeh_add_device_late() has
a check for this which will explictly destroy the PE in this case.
This check in eeh_add_device_late() occurs after a call to
eeh_ops->probe(). On PowerNV, probe is a pointer to pnv_eeh_probe(),
which will exit without probing if there is an existing PE.
This means that on PowerNV, devices with outstanding krefs will not
be rediscovered by EEH correctly after a complete hotplug. This is
affecting CXL (CAPI) devices in the field.
Put the probe after the kref check so that the PE is destroyed
and affected devices are correctly rediscovered by EEH.
Fixes: d91dafc02f ("powerpc/eeh: Delay probing EEH device during hotplug")
Cc: stable@vger.kernel.org
Cc: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Daniel Axtens <dja@axtens.net>
Acked-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Section 3.7 of Version 1.2 of the Power8 Processor User's Manual
prescribes that updates to HID0 be preceded by a SYNC instruction and
followed by an ISYNC instruction (Page 91).
Create an inline function name update_power8_hid0() which follows this
recipe and invoke it from the static split core path.
Signed-off-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>
Reviewed-by: Sam Bobroff <sam.bobroff@au1.ibm.com>
Tested-by: Sam Bobroff <sam.bobroff@au1.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Replace hard coded values with existing DRCONF flags while procesing
detected LMBs from the device tree. Does not change any functionality.
Signed-off-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The value of 'valid' is always zero when 'esid' is zero, and if 'esid'
is non-zero then the value of 'valid' is irrelevant because we are using
logical or in the if expression.
In fact 'valid' can be dropped completely from dump_segments() by
simply doing the check with SLB_ESID_V directly in the if.
Signed-off-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
[mpe: Rewrite change log]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The code to fetch the SLB size from the device tree wants to first look
for "slb-size" and then if that's not found "ibm,slb-size".
We can simplify the code by looking for the properties and then if we
find one of them we set mmu_slb_size.
We also change the function name from check_cpu_slb_size() to
init_mmu_slb_size() as the function doesn't check anything, it only
initialises mmu_slb_size.
Signed-off-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
[mpe: Rewrite change log]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This patch adds some documentation to patch_slb_encoding() explaining
how it works.
Signed-off-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
[mpe: Update change log and mention the signedness of the immediate]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The SLB code uses 'slot' and 'entry' interchangeably, change it to always
use 'entry'.
Signed-off-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
[mpe: Rewrite change log]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This patch just removes one redundant entry for one extern variable
'slb_compare_rr_to_size' from the scope. This patch does not change
any functionality.
Signed-off-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
An IO address, tagged with __iomem, is passed to debugfs_create_file
as private data. This requires that it be cast to void *. The cast
drops the __iomem annotation and so creates a sparse warning:
drivers/misc/cxl/debugfs.c:51:57: warning: cast removes address space of expression
The address space marker is added back in the file operations
(fops_io_u64).
Silence the warning with __force.
Signed-off-by: Daniel Axtens <dja@axtens.net>
Acked-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
A few declarations were identified by sparse as needing to be static:
drivers/misc/cxl/irq.c:408:6: warning: symbol 'afu_irq_name_free' was not declared. Should it be static?
drivers/misc/cxl/irq.c:467:6: warning: symbol 'afu_register_hwirqs' was not declared. Should it be static?
drivers/misc/cxl/file.c:254:6: warning: symbol 'afu_compat_ioctl' was not declared. Should it be static?
drivers/misc/cxl/file.c:399:30: warning: symbol 'afu_master_fops' was not declared. Should it be static?
Make them static.
Signed-off-by: Daniel Axtens <dja@axtens.net>
Acked-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
It's a good idea, and it brings us in line with the rest of arch/powerpc.
Signed-off-by: Daniel Axtens <dja@axtens.net>
Acked-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Add a new powerpc-specific trace clock using the timebase register,
similar to x86-tsc. This gives us
- a fast, monotonic, hardware clock source for trace entries, and
- a clock that can be used to correlate events across cpus as well as across
hypervisor and guests.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
In case of error, the functions platform_get_resource() and kmalloc()
returns NULL not ERR_PTR(). The IS_ERR() test in the return value check
should be replaced with NULL test.
Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
wf_find_control(), wf_find_sensor(), and wf_is_overtemp() are exported
but unused. Remove these three functions.
Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
wf_critical_overtemp() is exported. But nothing uses that export.
That's unsurprising because there's no header that defines it. Stop
exporting that function and make it static.
Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
wf_unregister_client() increments the client count when a client
unregisters. That is obviously incorrect. Decrement that client count
instead.
Fixes: 75722d3992 ("[PATCH] ppc64: Thermal control for SMU based machines")
Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Use %pR to simplify the debug code. This also make the debug info more
readable.
Signed-off-by: Kevin Hao <haokexin@gmail.com>
[mpe: Unsplit multi-line printk strings]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Currently when attaching a context in dedicated mode, we ignore the
result of add_process_element(), which could potentially fail.
If add_process_element() returns an error, pass it back to the caller.
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Invoke new opal_cec_reboot2() call with reboot type
OPAL_REBOOT_PLATFORM_ERROR (for unrecoverable HMI interrupts) to inform
BMC/OCC about this error, so that BMC can collect relevant data for error
analysis and decide what component to de-configure before rebooting.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
On non-recoverable MCE errors in kernel space, Linux kernel panics
and system reboots. On BMC based system opal-prd runs as a daemon
in the host. Hence, kernel crash may prevent opal-prd to detect and
analyze this MCE error. This may land us in a situation where the faulty
memory never gets de-configured and Linux would keep hitting same MCE error
again and again. If this happens in early stage of kernel initialization,
then Linux will keep crashing and rebooting in a loop.
This patch fixes this issue by invoking new opal_cec_reboot2() call with
reboot type OPAL_REBOOT_PLATFORM_ERROR to inform BMC/OCC about this
error, so that BMC can collect relevant data for error analysis and
decide what component to de-configure before rebooting.
This patch is dependent on OPAL patchset posted on skiboot mailing list
at https://lists.ozlabs.org/pipermail/skiboot/2015-July/001771.html that
introduces opal_cec_reboot2() opal call.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
In the event of unrecovered HMI the existing code panics as soon as
it receives the first unrecovered HMI event. This makes host to report
partial information about HMIs before panic. There may be more errors
which would have caused the HMI and hence more HMI event would have been
generated waiting to be pulled by host. This patch implements a logic to
pull and display all the HMI event before going down panic path.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The V2 version of HMI event now carries additional information for
Malfunction Alert. It now contains error information about CORE and NX
checkstop. This patch checks and displays the check stop reason before
panic.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Acked-by: Stewart Smith <stewart@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Wire up the syscall number and regs so the tests work on powerpc.
With the powerpc kernel support just merged, all tests pass on ppc64,
ppc64 (compat), ppc64le, ppc, ppc64e and ppc64e (compat).
Acked-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The seccomp_bpf test uses BPF_LD|BPF_W|BPF_ABS to load 32-bit values
from seccomp_data->args. On big endian machines this will load the high
word of the argument, which is not what the test wants.
Borrow a hack from samples/seccomp/bpf-helper.h which changes the offset
on big endian to account for this.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Acked-by: Kees Cook <keescook@chromium.org>
This commit enables seccomp filter on powerpc, now that we have all the
necessary pieces in place.
To support seccomp's desire to modify the syscall return value under
some circumstances, we use a different ABI to the ptrace ABI. That is we
use r3 as the syscall return value, and orig_gpr3 is the first syscall
parameter.
This means the seccomp code, or a ptracer via SECCOMP_RET_TRACE, will
see -ENOSYS preloaded in r3. This is identical to the behaviour on x86,
and allows seccomp or the ptracer to either leave the -ENOSYS or change
it to something else, as well as rejecting or not the syscall by
modifying r0.
If seccomp does not reject the syscall, we restore the register state to
match what ptrace and audit expect, ie. r3 is the first syscall
parameter again. We do this restore using orig_gpr3, which may have been
modified by seccomp, which allows seccomp to modify the first syscall
paramater and allow the syscall to proceed.
We need to #ifdef the the additional handling of r3 for seccomp, so move
it all out of line.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Kees Cook <keescook@chromium.org>
SIG_SYS was added in commit a0727e8ce5 "signal, x86: add SIGSYS info
and make it synchronous."
Because we use the asm-generic struct siginfo, we got support for
SIG_SYS for free as part of that commit.
However there was no compat handling added for powerpc. That means we've
been advertising the existence of signfo._sifields._sigsys to compat
tasks, but not actually filling in the fields correctly.
Luckily it looks like no one has noticed, presumably because the only
user of SIGSYS in the kernel is seccomp filter, which we don't support
yet.
So before we enable seccomp filter, add compat handling for SIGSYS.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Kees Cook <keescook@chromium.org>
The documentation for syscall_get_nr() in asm-generic says:
Note this returns int even on 64-bit machines. Only 32 bits of
system call number can be meaningful. If the actual arch value
is 64 bits, this truncates to 32 bits so 0xffffffff means -1.
However our implementation was never updated to reflect this.
Generally it's not important, but there is once case where it matters.
For seccomp filter with SECCOMP_RET_TRACE, the tracer will set
regs->gpr[0] to -1 to reject the syscall. When the task is a compat
task, this means we end up with 0xffffffff in r0 because ptrace will
zero extend the 32-bit value.
If syscall_get_nr() returns an unsigned long, then a 64-bit kernel will
see a positive value in r0 and will incorrectly allow the syscall
through seccomp.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Kees Cook <keescook@chromium.org>
Currently syscall_get_arguments() is used by syscall tracepoints, and
collect_syscall() which is used in some debugging as well as
/proc/pid/syscall.
The current implementation just copies regs->gpr[3 .. 5] out, which is
fine for all the current use cases.
When we enable seccomp filter, that will also start using
syscall_get_arguments(). However for seccomp filter we want to use r3
as the return value of the syscall, and orig_gpr3 as the first
parameter. This will allow seccomp to modify the return value in r3.
To support this we need to modify syscall_get_arguments() to return
orig_gpr3 instead of r3. This is safe for all uses because orig_gpr3
always contains the r3 value that was passed to the syscall. We store it
in the syscall entry path and never modify it.
Update syscall_set_arguments() while we're here, even though it's never
used.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Kees Cook <keescook@chromium.org>
Currently syscall_get_arguments() has two loops, one for compat and one
for regular tasks. In prepartion for the next patch, which changes which
registers we use, switch it to only have one loop, so we only have one
place to update.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Kees Cook <keescook@chromium.org>
Currently the only caller of syscall_set_return_value() is seccomp
filter, which is not enabled on powerpc.
This means we have not noticed that our implementation of
syscall_set_return_value() negates error, even though the value passed
in is already negative.
So remove the negation in syscall_set_return_value(), and expect the
caller to do it like all other implementations do.
Also add a comment about the ccr handling.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Kees Cook <keescook@chromium.org>
syscall_get_error() is unused, and never has been.
It's also probably wrong, as it negates r3 before returning it, but that
depends on what the caller is expecting.
It also doesn't deal with compat, and doesn't deal with TIF_NOERROR.
Although we could fix those, until it has a caller and it's clear what
semantics the caller wants it's just untested code. So drop it.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Kees Cook <keescook@chromium.org>
The API for calling do_syscall_trace_enter() is currently sensible
enough, it just returns the (modified) syscall number.
However once we enable seccomp filter it will get more complicated. When
seccomp filter runs, the seccomp kernel code (via SECCOMP_RET_ERRNO), or
a ptracer (via SECCOMP_RET_TRACE), may reject the syscall and *may* or may
*not* set a return value in r3.
That means the assembler that calls do_syscall_trace_enter() can not
blindly return ENOSYS, it needs to only return ENOSYS if a return value
has not already been set.
There is no way to implement that logic with the current API. So change
the do_syscall_trace_enter() API to make it deal with the return code
juggling, and the assembler can then just return whatever return code it
is given.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Kees Cook <keescook@chromium.org>
Currently on powerpc we have our own #define for the highest (negative)
errno value, called _LAST_ERRNO. This is defined to be 516, for reasons
which are not clear.
The generic code, and x86, use MAX_ERRNO, which is defined to be 4095.
In particular seccomp uses MAX_ERRNO to restrict the value that a
seccomp filter can return.
Currently with the mismatch between _LAST_ERRNO and MAX_ERRNO, a seccomp
tracer wanting to return 600, expecting it to be seen as an error, would
instead find on powerpc that userspace sees a successful syscall with a
return value of 600.
To avoid this inconsistency, switch powerpc to use MAX_ERRNO.
We are somewhat confident that generic syscalls that can return a
non-error value above negative MAX_ERRNO have already been updated to
use force_successful_syscall_return().
I have also checked all the powerpc specific syscalls, and believe that
none of them expect to return a non-error value between -MAX_ERRNO and
-516. So this change should be safe ...
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Kees Cook <keescook@chromium.org>
This patch just changes data type of bhrb_users variable from
int to unsigned int because it never contains a negative value.
Reported-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Simplify code that extracts a 24x7 counter from the HCALL's result buffer.
Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Fix parameter alignment to be consistent with coding style.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The hardware RNG on POWER8 and POWER7+ can be relatively slow, since
it can only supply one 64-bit value per microsecond. Currently we
read it in arch_get_random_long(), but that slows down reading from
/dev/urandom since the code in random.c calls arch_get_random_long()
for every longword read from /dev/urandom.
Since the hardware RNG supplies high-quality entropy on every read, it
matches the semantics of arch_get_random_seed_long() better than those
of arch_get_random_long(). Therefore this commit makes the code use
the POWER8/7+ hardware RNG only for arch_get_random_seed_{long,int}
and not for arch_get_random_{long,int}.
This won't affect any other PowerPC-based platforms because none of
them currently support a hardware RNG. To make it clear that the
ppc_md function pointer is used for arch_get_random_seed_*, we rename
it from get_random_long to get_random_seed.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The EPOW interrupt handler uses rtas_get_sensor(), which in turn
uses rtas_busy_delay() to wait for RTAS becoming ready in case it
is necessary. But rtas_busy_delay() is annotated with might_sleep()
and thus may not be used by interrupts handlers like the EPOW handler!
This leads to the following BUG when CONFIG_DEBUG_ATOMIC_SLEEP is
enabled:
BUG: sleeping function called from invalid context at arch/powerpc/kernel/rtas.c:496
in_atomic(): 1, irqs_disabled(): 1, pid: 0, name: swapper/1
CPU: 1 PID: 0 Comm: swapper/1 Not tainted 4.2.0-rc2-thuth #6
Call Trace:
[c00000007ffe7b90] [c000000000807670] dump_stack+0xa0/0xdc (unreliable)
[c00000007ffe7bc0] [c0000000000e1f14] ___might_sleep+0x134/0x180
[c00000007ffe7c20] [c00000000002aec0] rtas_busy_delay+0x30/0xd0
[c00000007ffe7c50] [c00000000002bde4] rtas_get_sensor+0x74/0xe0
[c00000007ffe7ce0] [c000000000083264] ras_epow_interrupt+0x44/0x450
[c00000007ffe7d90] [c000000000120260] handle_irq_event_percpu+0xa0/0x300
[c00000007ffe7e70] [c000000000120524] handle_irq_event+0x64/0xc0
[c00000007ffe7eb0] [c000000000124dbc] handle_fasteoi_irq+0xec/0x260
[c00000007ffe7ef0] [c00000000011f4f0] generic_handle_irq+0x50/0x80
[c00000007ffe7f20] [c000000000010f3c] __do_irq+0x8c/0x200
[c00000007ffe7f90] [c0000000000236cc] call_do_irq+0x14/0x24
[c00000007e6f39e0] [c000000000011144] do_IRQ+0x94/0x110
[c00000007e6f3a30] [c000000000002594] hardware_interrupt_common+0x114/0x180
Fix this issue by introducing a new rtas_get_sensor_fast() function
that does not use rtas_busy_delay() - and thus can only be used for
sensors that do not cause a BUSY condition - known as "fast" sensors.
The EPOW sensor is defined to be "fast" in sPAPR - mpe.
Fixes: 587f83e8dd ("powerpc/pseries: Use rtas_get_sensor in RAS code")
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
rtas.h already has some nice #defines for RTAS return status
codes - let's use them instead of hard-coded "magic" values!
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
When detecting EEH error on non-existing PE, including the reserved
one, the PE is simply unfrozen without dumping the PHB diag-data,
which is useful for locating the root cause of the EEH error. The
patch dumps the PHB diag-data when non-existing PE reports error.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Daniel Axtens