IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
This threshold is no longer used now that all invalidates issue a single
ATSD to each active NPU.
Signed-off-by: Mark Hairgrove <mhairgrove@nvidia.com>
Reviewed-by: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Prior to this change only two types of ATSDs were issued to the NPU:
invalidates targeting a single page and invalidates targeting the whole
address space. The crossover point happened at the configurable
atsd_threshold which defaulted to 2M. Invalidates that size or smaller
would issue per-page invalidates for the whole range.
The NPU supports more invalidation sizes however: 64K, 2M, 1G, and all.
These invalidates target addresses aligned to their size. 2M is a common
invalidation size for GPU-enabled applications because that is a GPU
page size, so reducing the number of invalidates by 32x in that case is a
clear improvement.
ATSD latency is high in general so now we always issue a single invalidate
rather than multiple. This will over-invalidate in some cases, but for any
invalidation size over 2M it matches or improves the prior behavior.
There's also an improvement for single-page invalidates since the prior
version issued two invalidates for that case instead of one.
With this change all issued ATSDs now perform a flush, so the flush
parameter has been removed from all the helpers.
To show the benefit here are some performance numbers from a
microbenchmark which creates a 1G allocation then uses mprotect with
PROT_NONE to trigger invalidates in strides across the allocation.
One NPU (1 GPU):
mprotect rate (GB/s)
Stride Before After Speedup
64K 5.3 5.6 5%
1M 39.3 57.4 46%
2M 49.7 82.6 66%
4M 286.6 285.7 0%
Two NPUs (6 GPUs):
mprotect rate (GB/s)
Stride Before After Speedup
64K 6.5 7.4 13%
1M 33.4 67.9 103%
2M 38.7 93.1 141%
4M 356.7 354.6 -1%
Anything over 2M is roughly the same as before since both cases issue a
single ATSD.
Signed-off-by: Mark Hairgrove <mhairgrove@nvidia.com>
Reviewed-By: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
There are two types of ATSDs issued to the NPU: invalidates targeting a
specific virtual address and invalidates targeting the whole address
space. In both cases prior to this change, the sequence was:
for each NPU
- Write the target address to the XTS_ATSD_AVA register
- EIEIO
- Write the launch value to issue the ATSD
First, a target address is not required when invalidating the whole
address space, so that write and the EIEIO have been removed. The AP
(size) field in the launch is not needed either.
Second, for per-address invalidates the above sequence is inefficient in
the common case of multiple NPUs because an EIEIO is issued per NPU. This
unnecessarily forces the launches of later ATSDs to be ordered with the
launches of earlier ones. The new sequence only issues a single EIEIO:
for each NPU
- Write the target address to the XTS_ATSD_AVA register
EIEIO
for each NPU
- Write the launch value to issue the ATSD
Performance results were gathered using a microbenchmark which creates a
1G allocation then uses mprotect with PROT_NONE to trigger invalidates in
strides across the allocation.
With only a single NPU active (one GPU) the difference is in the noise for
both types of invalidates (+/-1%).
With two NPUs active (on a 6-GPU system) the effect is more noticeable:
mprotect rate (GB/s)
Stride Before After Speedup
64K 5.9 6.5 10%
1M 31.2 33.4 7%
2M 36.3 38.7 7%
4M 322.6 356.7 11%
Signed-off-by: Mark Hairgrove <mhairgrove@nvidia.com>
Reviewed-by: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The code in machine_check_exception excludes 64s hvmode when
incrementing the MCE counter only to call opal_machine_check to
increment it specifically for this case.
Remove the exclusion and special case.
Fixes: a43c159042 ("powerpc/pseries: Flush SLB contents on SLB MCE
errors.")
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
There is a mismatch between function pnv_platform_error_reboot() definition
and declaration regarding function modifiers. In the declaration part, it
contains the function attribute __noreturn, while function definition
itself lacks it.
This was reported by sparse tool as an error:
arch/powerpc/platforms/powernv/opal.c:538:6: error: symbol 'pnv_platform_error_reboot' redeclared with different type (originally declared at arch/powerpc/platforms/powernv/powernv.h:11) - different modifiers
I checked and the function is already being considered as being 'noreturn'
by the compiler, thus, I understand this patch does not change any code
being generated.
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Ever since fast reboot is enabled by default in opal,
opal_cec_reboot() will use fast-reset instead of full IPL to perform
system reboot. This leaves the user with no direct way to force a full
IPL reboot except changing an nvram setting that persistently disables
fast-reset for all subsequent reboots.
This patch provides a more direct way for the user to force a one-shot
full IPL reboot by passing the command line argument 'full' to the
reboot command. So the user will be able to tweak the reboot behavior
via:
$ sudo reboot full # Force a full ipl reboot skipping fast-reset
or
$ sudo reboot # default reboot path (usually fast-reset)
The reboot command passes the un-parsed command argument to the kernel
via the 'Reboot' syscall which is then passed on to the arch function
pnv_restart(). The patch updates pnv_restart() to handle this cmd-arg
and issues opal_cec_reboot2 with OPAL_REBOOT_FULL_IPL to force a full
IPL reset.
Signed-off-by: Vaibhav Jain <vaibhav@linux.ibm.com>
Acked-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
mpe: This was fixed originally in commit d3d4ffaae4
("powerpc/powernv/ioda2: Reduce upper limit for DMA window size"), but
contrary to what the merge commit says was inadvertently lost by me in
commit ce57c6610c ("Merge branch 'topic/ppc-kvm' into next") which
brought in changes that moved the code to a new file. So reapply it to
the new file.
Original commit message follows:
We use PHB in mode1 which uses bit 59 to select a correct DMA window.
However there is mode2 which uses bits 59:55 and allows up to 32 DMA
windows per a PE.
Even though documentation does not clearly specify that, it seems that
the actual hardware does not support bits 59:55 even in mode1, in
other words we can create a window as big as 1<<58 but DMA simply
won't work.
This reduces the upper limit from 59 to 55 bits to let the userspace
know about the hardware limits.
Fixes: ce57c6610c ("Merge branch 'topic/ppc-kvm' into next")
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
On pseries, as of today system crashes if we get a machine check
exceptions due to SLB errors. These are soft errors and can be fixed
by flushing the SLBs so the kernel can continue to function instead of
system crash. We do this in real mode before turning on MMU. Otherwise
we would run into nested machine checks. This patch now fetches the
rtas error log in real mode and flushes the SLBs on SLB/ERAT errors.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Michal Suchanek <msuchanek@suse.com>
Reviewed-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
When hot-removing memory release_mem_region_adjustable() splits iomem
resources if they are not the exact size of the memory being
hot-deleted. Adding this memory back to the kernel adds a new resource.
Eg a node has memory 0x0 - 0xfffffffff. Hot-removing 1GB from
0xf40000000 results in the single resource 0x0-0xfffffffff being split
into two resources: 0x0-0xf3fffffff and 0xf80000000-0xfffffffff.
When we hot-add the memory back we now have three resources:
0x0-0xf3fffffff, 0xf40000000-0xf7fffffff, and 0xf80000000-0xfffffffff.
This is an issue if we try to remove some memory that overlaps
resources. Eg when trying to remove 2GB at address 0xf40000000,
release_mem_region_adjustable() fails as it expects the chunk of memory
to be within the boundaries of a single resource. We then get the
warning: "Unable to release resource" and attempting to use memtrace
again gives us this error: "bash: echo: write error: Resource
temporarily unavailable"
This patch makes memtrace remove memory in chunks that are always the
same size from an address that is always equal to end_of_memory -
n*size, for some n. So hotremoving and hotadding memory of different
sizes will now not attempt to remove memory that spans multiple
resources.
Signed-off-by: Rashmica Gupta <rashmica.g@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Deciding wich govenors should be built into the kernel can be left to
users to configure.
Fixes: 81f359027a ("cpufreq: powernv: Select CPUFreq related Kconfig options for powernv")
Signed-off-by: Joel Stanley <joel@jms.id.au>
[mpe: Update powernv/ppc64 defconfigs to enable them by default]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Pull IDA updates from Matthew Wilcox:
"A better IDA API:
id = ida_alloc(ida, GFP_xxx);
ida_free(ida, id);
rather than the cumbersome ida_simple_get(), ida_simple_remove().
The new IDA API is similar to ida_simple_get() but better named. The
internal restructuring of the IDA code removes the bitmap
preallocation nonsense.
I hope the net -200 lines of code is convincing"
* 'ida-4.19' of git://git.infradead.org/users/willy/linux-dax: (29 commits)
ida: Change ida_get_new_above to return the id
ida: Remove old API
test_ida: check_ida_destroy and check_ida_alloc
test_ida: Convert check_ida_conv to new API
test_ida: Move ida_check_max
test_ida: Move ida_check_leaf
idr-test: Convert ida_check_nomem to new API
ida: Start new test_ida module
target/iscsi: Allocate session IDs from an IDA
iscsi target: fix session creation failure handling
drm/vmwgfx: Convert to new IDA API
dmaengine: Convert to new IDA API
ppc: Convert vas ID allocation to new IDA API
media: Convert entity ID allocation to new IDA API
ppc: Convert mmu context allocation to new IDA API
Convert net_namespace to new IDA API
cb710: Convert to new IDA API
rsxx: Convert to new IDA API
osd: Convert to new IDA API
sd: Convert to new IDA API
...
Pull powerpc fixes from Michael Ellerman:
- An implementation for the newly added hv_ops->flush() for the OPAL
hvc console driver backends, I forgot to apply this after merging the
hvc driver changes before the merge window.
- Enable all PCI bridges at boot on powernv, to avoid races when
multiple children of a bridge try to enable it simultaneously. This
is a workaround until the PCI core can be enhanced to fix the races.
- A fix to query PowerVM for the correct system topology at boot before
initialising sched domains, seen in some configurations to cause
broken scheduling etc.
- A fix for pte_access_permitted() on "nohash" platforms.
- Two commits to fix SIGBUS when using remap_pfn_range() seen on Power9
due to a workaround when using the nest MMU (GPUs, accelerators).
- Another fix to the VFIO code used by KVM, the previous fix had some
bugs which caused guests to not start in some configurations.
- A handful of other minor fixes.
Thanks to: Aneesh Kumar K.V, Benjamin Herrenschmidt, Christophe Leroy,
Hari Bathini, Luke Dashjr, Mahesh Salgaonkar, Nicholas Piggin, Paul
Mackerras, Srikar Dronamraju.
* tag 'powerpc-4.19-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/mce: Fix SLB rebolting during MCE recovery path.
KVM: PPC: Book3S: Fix guest DMA when guest partially backed by THP pages
powerpc/mm/radix: Only need the Nest MMU workaround for R -> RW transition
powerpc/mm/books3s: Add new pte bit to mark pte temporarily invalid.
powerpc/nohash: fix pte_access_permitted()
powerpc/topology: Get topology for shared processors at boot
powerpc64/ftrace: Include ftrace.h needed for enable/disable calls
powerpc/powernv/pci: Work around races in PCI bridge enabling
powerpc/fadump: cleanup crash memory ranges support
powerpc/powernv: provide a console flush operation for opal hvc driver
powerpc/traps: Avoid rate limit messages from show unhandled signals
powerpc/64s: Fix PACA_IRQ_HARD_DIS accounting in idle_power4()
Removes a custom spinlock and simplifies the code. Also fix an
error where we could allocate one ID too many.
Signed-off-by: Matthew Wilcox <willy@infradead.org>
The generic code is racy when multiple children of a PCI bridge try to
enable it simultaneously.
This leads to drivers trying to access a device through a
not-yet-enabled bridge, and this EEH errors under various
circumstances when using parallel driver probing.
There is work going on to fix that properly in the PCI core but it
will take some time.
x86 gets away with it because (outside of hotplug), the BIOS enables
all the bridges at boot time.
This patch does the same thing on powernv by enabling all bridges that
have child devices at boot time, thus avoiding subsequent races. It's
suitable for backporting to stable and distros, while the proper PCI
fix will probably be significantly more invasive.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: stable@vger.kernel.org
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Pull powerpc updates from Michael Ellerman:
"Notable changes:
- A fix for a bug in our page table fragment allocator, where a page
table page could be freed and reallocated for something else while
still in use, leading to memory corruption etc. The fix reuses
pt_mm in struct page (x86 only) for a powerpc only refcount.
- Fixes to our pkey support. Several are user-visible changes, but
bring us in to line with x86 behaviour and/or fix outright bugs.
Thanks to Florian Weimer for reporting many of these.
- A series to improve the hvc driver & related OPAL console code,
which have been seen to cause hardlockups at times. The hvc driver
changes in particular have been in linux-next for ~month.
- Increase our MAX_PHYSMEM_BITS to 128TB when SPARSEMEM_VMEMMAP=y.
- Remove Power8 DD1 and Power9 DD1 support, neither chip should be in
use anywhere other than as a paper weight.
- An optimised memcmp implementation using Power7-or-later VMX
instructions
- Support for barrier_nospec on some NXP CPUs.
- Support for flushing the count cache on context switch on some IBM
CPUs (controlled by firmware), as a Spectre v2 mitigation.
- A series to enhance the information we print on unhandled signals
to bring it into line with other arches, including showing the
offending VMA and dumping the instructions around the fault.
Thanks to: Aaro Koskinen, Akshay Adiga, Alastair D'Silva, Alexey
Kardashevskiy, Alexey Spirkov, Alistair Popple, Andrew Donnellan,
Aneesh Kumar K.V, Anju T Sudhakar, Arnd Bergmann, Bartosz Golaszewski,
Benjamin Herrenschmidt, Bharat Bhushan, Bjoern Noetel, Boqun Feng,
Breno Leitao, Bryant G. Ly, Camelia Groza, Christophe Leroy, Christoph
Hellwig, Cyril Bur, Dan Carpenter, Daniel Klamt, Darren Stevens, Dave
Young, David Gibson, Diana Craciun, Finn Thain, Florian Weimer,
Frederic Barrat, Gautham R. Shenoy, Geert Uytterhoeven, Geoff Levand,
Guenter Roeck, Gustavo Romero, Haren Myneni, Hari Bathini, Joel
Stanley, Jonathan Neuschäfer, Kees Cook, Madhavan Srinivasan, Mahesh
Salgaonkar, Markus Elfring, Mathieu Malaterre, Mauro S. M. Rodrigues,
Michael Hanselmann, Michael Neuling, Michael Schmitz, Mukesh Ojha,
Murilo Opsfelder Araujo, Nicholas Piggin, Parth Y Shah, Paul
Mackerras, Paul Menzel, Ram Pai, Randy Dunlap, Rashmica Gupta, Reza
Arbab, Rodrigo R. Galvao, Russell Currey, Sam Bobroff, Scott Wood,
Shilpasri G Bhat, Simon Guo, Souptick Joarder, Stan Johnson, Thiago
Jung Bauermann, Tyrel Datwyler, Vaibhav Jain, Vasant Hegde, Venkat
Rao, zhong jiang"
* tag 'powerpc-4.19-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux: (234 commits)
powerpc/mm/book3s/radix: Add mapping statistics
powerpc/uaccess: Enable get_user(u64, *p) on 32-bit
powerpc/mm/hash: Remove unnecessary do { } while(0) loop
powerpc/64s: move machine check SLB flushing to mm/slb.c
powerpc/powernv/idle: Fix build error
powerpc/mm/tlbflush: update the mmu_gather page size while iterating address range
powerpc/mm: remove warning about ‘type’ being set
powerpc/32: Include setup.h header file to fix warnings
powerpc: Move `path` variable inside DEBUG_PROM
powerpc/powermac: Make some functions static
powerpc/powermac: Remove variable x that's never read
cxl: remove a dead branch
powerpc/powermac: Add missing include of header pmac.h
powerpc/kexec: Use common error handling code in setup_new_fdt()
powerpc/xmon: Add address lookup for percpu symbols
powerpc/mm: remove huge_pte_offset_and_shift() prototype
powerpc/lib: Use patch_site to patch copy_32 functions once cache is enabled
powerpc/pseries: Fix endianness while restoring of r3 in MCE handler.
powerpc/fadump: merge adjacent memory ranges to reduce PT_LOAD segements
powerpc/fadump: handle crash memory ranges array index overflow
...
Fix the below build error using strlcpy instead of strncpy
In function 'pnv_parse_cpuidle_dt',
inlined from 'pnv_init_idle_states' at arch/powerpc/platforms/powernv/idle.c:840:7,
inlined from '__machine_initcall_powernv_pnv_init_idle_states' at arch/powerpc/platforms/powernv/idle.c:870:1:
arch/powerpc/platforms/powernv/idle.c:820:3: error: 'strncpy' specified bound 16 equals destination size [-Werror=stringop-truncation]
strncpy(pnv_idle_states[i].name, temp_string[i],
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
PNV_IDLE_NAME_LEN);
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This patch allows the memory removed by memtrace to be readded to the
kernel. So now you don't have to reboot your system to add the memory
back to the kernel or to have a different amount of memory removed.
Signed-off-by: Rashmica Gupta <rashmica.g@gmail.com>
Tested-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
For (bad) historical reasons, OPAL used to create a non-standard pair
of properties "opal-interrupts" and "opal-interrupts-names" for
representing the list of interrupts it wants Linux to request on its
behalf.
Among other issues, the opal-interrupts doesn't have a way to carry
the type of interrupts, and they were assumed to be all level
sensitive.
This is wrong on some recent systems where some of them are edge
sensitive causing warnings in the XIVE code and possible misbehaviours
if they need to be retriggered (typically the NPU2 TCE error
interrupts).
This makes Linux switch to using the standard "interrupts" and
"interrupt-names" properties instead when they are available, using
standard of_irq helpers, which can carry all the desired type
information.
Newer versions of OPAL will generate those properties in addition to
the legacy ones.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
[mpe: Fixup prefix logic to check strlen(r->name). Reinstate setting
of start = 0 in opal_event_shutdown() to avoid double free warnings]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
NX increments readOffset by FIFO size in receive FIFO control register
when CRB is read. But the index in RxFIFO has to match with the
corresponding entry in FIFO maintained by VAS in kernel. Otherwise NX
may be processing incorrect CRBs and can cause CRB timeout.
VAS FIFO offset is 0 when the receive window is opened during
initialization. When the module is reloaded or in kexec boot, readOffset
in FIFO control register may not match with VAS entry. This patch adds
nx_coproc_init OPAL call to reset readOffset and queued entries in FIFO
control register for both high and normal FIFOs.
Signed-off-by: Haren Myneni <haren@us.ibm.com>
[mpe: Fixup uninitialized variable warning]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Export opal_check_token symbol for modules to check the availability
of OPAL calls before using them.
Signed-off-by: Haren Myneni <haren@us.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Look for fw-features properties to determine the appropriate settings
for the count cache flush, and then call the generic powerpc code to
set it up based on the security feature flags.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Currently we require platform code to call setup_barrier_nospec(). But
if we add an empty definition for the !CONFIG_PPC_BARRIER_NOSPEC case
then we can call it in setup_arch().
Signed-off-by: Diana Craciun <diana.craciun@nxp.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
It's identical to xive_teardown_cpu() so just use the latter
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
We've encountered a performance issue when multiple processors stress
{get,put}_mmio_atsd_reg(). These functions contend for
mmio_atsd_usage, an unsigned long used as a bitmask.
The accesses to mmio_atsd_usage are done using test_and_set_bit_lock()
and clear_bit_unlock(). As implemented, both of these will require
a (successful) stwcx to that same cache line.
What we end up with is thread A, attempting to unlock, being slowed by
other threads repeatedly attempting to lock. A's stwcx instructions
fail and retry because the memory reservation is lost every time a
different thread beats it to the punch.
There may be a long-term way to fix this at a larger scale, but for
now resolve the immediate problem by gating our call to
test_and_set_bit_lock() with one to test_bit(), which is obviously
implemented without using a store.
Fixes: 1ab66d1fba ("powerpc/powernv: Introduce address translation services for Nvlink2")
Signed-off-by: Reza Arbab <arbab@linux.ibm.com>
Acked-by: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Commit 9c7b185ab2 ("powernv/cpuidle: Parse dt idle properties into
global structure") parses dt idle states into structs, but never marks
them valid. This results in all idle states being lost.
Fixes: 9c7b185ab2 ("powernv/cpuidle: Parse dt idle properties into global structure")
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Akshay Adiga <akshay.adiga@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
When a PCI device is detected, pdev->is_added is set to 1 and proc and
sysfs entries are created.
When the device is removed, pdev->is_added is checked for one and then
device is detached with clearing of proc and sys entries and at end,
pdev->is_added is set to 0.
is_added and is_busmaster are bit fields in pci_dev structure sharing same
memory location.
A strange issue was observed with multiple removal and rescan of a PCIe
NVMe device using sysfs commands where is_added flag was observed as zero
instead of one while removing device and proc,sys entries are not cleared.
This causes issue in later device addition with warning message
"proc_dir_entry" already registered.
Debugging revealed a race condition between the PCI core setting the
is_added bit in pci_bus_add_device() and the NVMe driver reset work-queue
setting the is_busmaster bit in pci_set_master(). As these fields are not
handled atomically, that clears the is_added bit.
Move the is_added bit to a separate private flag variable and use atomic
functions to set and retrieve the device addition state. This avoids the
race because is_added no longer shares a memory location with is_busmaster.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=200283
Signed-off-by: Hari Vyas <hari.vyas@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Lukas Wunner <lukas@wunner.de>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Adds support to enable/disable a sensor group at runtime. This
can be used to select the sensor groups that needs to be copied to
main memory by OCC. Sensor groups like power, temperature, current,
voltage, frequency, utilization can be enabled/disabled at runtime.
Signed-off-by: Shilpasri G Bhat <shilpa.bhat@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Device-tree parsing happens twice, once while deciding idle state to be
used for hotplug and once during cpuidle init. Hence, parsing the device
tree and caching it will reduce code duplication. Parsing code has been
moved to pnv_parse_cpuidle_dt() from pnv_probe_idle_states(). In addition
to the properties in the device tree the number of available states is
also required.
Signed-off-by: Akshay Adiga <akshay.adiga@linux.vnet.ibm.com>
Reviewed-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
files not using feature fixup don't need asm/feature-fixups.h
files using feature fixup need asm/feature-fixups.h
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Only include linux/stringify.h is files using __stringify()
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This patch moves ASM_CONST() and stringify_in_c() into
dedicated asm-const.h, then cleans all related inclusions.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
[mpe: asm-compat.h should include asm-const.h]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The RAW console does not need writes to be atomic, so relax
opal_put_chars to be able to do partial writes, and implement an
_atomic variant which does not take a spinlock. This API is used
in xmon, so the less locking that is used, the better chance there
is that a crash can be debugged.
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
OPAL console writes do not have to synchronously flush firmware /
hardware buffers unless they are going through the udbg path.
Remove the unconditional flushing from opal_put_chars. Flush if
there was no space in the buffer as an optimisation (callers loop
waiting for success in that case). udbg flushing is moved to
udbg_opal_putc.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
opal_put_chars deals with partial writes because in OPALv1,
opal_console_write_buffer_space did not work correctly. That firmware
is not supported.
This reworks the opal_put_chars code to no longer deal with partial
writes by turning them into full writes. Partial write handling is still
supported in terms of what gets returned to the caller, but it may not
go to the console atomically. A warning message is printed in this
case.
This allows console flushing to be moved out of the opal_write_lock
spinlock. That could cause the lock to be held for long periods if the
console is busy (especially if it was being spammed by firmware),
which is dangerous because the lock is taken by xmon to debug the
system. Flushing outside the lock improves the situation a bit.
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
A new console flushing firmware API was introduced to replace event
polling loops, and implemented in opal-kmsg with affddff69c
("powerpc/powernv: Add a kmsg_dumper that flushes console output on
panic"), to flush the console in the panic path.
The OPAL console driver has other situations where interrupts are off
and it needs to flush the console synchronously. These still use a
polling loop.
So move the opal-kmsg flush code to opal_flush_console, and use the
new function in opal-kmsg and opal_put_chars.
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Reviewed-by: Russell Currey <ruscur@russell.cc>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Use the more refined and tested event polling loop from opal_put_chars
as the fallback console flush in the opal-kmsg path. This loop is used
by the console driver today, whereas the opal-kmsg fallback is not
likely to have been used for years.
Use WARN_ONCE rather than a printk when the fallback is invoked to
prepare for moving the console flush into a common function.
Reviewed-by: Russell Currey <ruscur@russell.cc>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
OPAL_CONSOLE_FLUSH is documented as being able to return OPAL_BUSY,
so implement the standard OPAL_BUSY handling for it.
Reviewed-by: Russell Currey <ruscur@russell.cc>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The OPAL console driver does not delay in case it gets OPAL_BUSY or
OPAL_BUSY_EVENT from firmware.
It can't yet be made to sleep because it is called under spinlock,
but it can be changed to the standard OPAL_BUSY loop form, and a
delay added to keep it from hitting the firmware too frequently.
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The intention here is to consume and discard the remaining buffer
upon error. This works if there has not been a previous partial write.
If there has been, then total_len is no longer total number of bytes
to copy. total_len is always "bytes left to copy", so it should be
added to written bytes.
This code may not be exercised any more if partial writes will not be
hit, but this is a small bugfix before a larger change.
Reviewed-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Fixes: 8034f715f ("powernv/opal-dump: Convert to irq domain")
Converts all the return explicit number to a more proper IRQ_HANDLED,
which looks proper incase of interrupt handler returning case.
Here, It also removes error message like "nobody cared" which was
getting unveiled while returning -1 or 0 from handler.
Signed-off-by: Mukesh Ojha <mukesh02@linux.vnet.ibm.com>
Reviewed-by: Vasant Hegde <hegdevasant@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Moves the return value check of 'opal_dump_info' to a proper place which
was previously unnecessarily filling all the dump info even on failure.
Signed-off-by: Mukesh Ojha <mukesh02@linux.vnet.ibm.com>
Acked-by: Stewart Smith <stewart@linux.vnet.ibm.com>
Acked-by: Jeremy Kerr <jk@ozlabs.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The threshold at which it becomes more efficient to coalesce a range
of ATSDs into a single per-PID ATSD is currently not well understood
due to a lack of real-world work loads. This patch adds a debugfs
parameter allowing the threshold to be altered at runtime in order to
aid future development and refinement of the value.
Signed-off-by: Alistair Popple <alistair@popple.id.au>
Acked-by: Balbir Singh <bsingharora@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Merge in some commits we're sharing with the KVM tree.
I manually propagated the change from commit d3d4ffaae4
("powerpc/powernv/ioda2: Reduce upper limit for DMA window size") into
pci-ioda-tce.c.
Conflicts:
arch/powerpc/include/asm/cputable.h
arch/powerpc/platforms/powernv/pci-ioda.c
arch/powerpc/platforms/powernv/pci.h
At the moment we allocate the entire TCE table, twice (hardware part and
userspace translation cache). This normally works as we normally have
contigous memory and the guest will map entire RAM for 64bit DMA.
However if we have sparse RAM (one example is a memory device), then
we will allocate TCEs which will never be used as the guest only maps
actual memory for DMA. If it is a single level TCE table, there is nothing
we can really do but if it a multilevel table, we can skip allocating
TCEs we know we won't need.
This adds ability to allocate only first level, saving memory.
This changes iommu_table::free() to avoid allocating of an extra level;
iommu_table::set() will do this when needed.
This adds @alloc parameter to iommu_table::exchange() to tell the callback
if it can allocate an extra level; the flag is set to "false" for
the realmode KVM handlers of H_PUT_TCE hcalls and the callback returns
H_TOO_HARD.
This still requires the entire table to be counted in mm::locked_vm.
To be conservative, this only does on-demand allocation when
the usespace cache table is requested which is the case of VFIO.
The example math for a system replicating a powernv setup with NVLink2
in a guest:
16GB RAM mapped at 0x0
128GB GPU RAM window (16GB of actual RAM) mapped at 0x244000000000
the table to cover that all with 64K pages takes:
(((0x244000000000 + 0x2000000000) >> 16)*8)>>20 = 4556MB
If we allocate only necessary TCE levels, we will only need:
(((0x400000000 + 0x400000000) >> 16)*8)>>20 = 4MB (plus some for indirect
levels).
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This moves actual pages allocation to a separate function which is going
to be reused later in on-demand TCE allocation.
While we are at it, remove unnecessary level size round up as the caller
does this already.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
We want to support sparse memory and therefore huge chunks of DMA windows
do not need to be mapped. If a DMA window big enough to require 2 or more
indirect levels, and a DMA window is used to map all RAM (which is
a default case for 64bit window), we can actually save some memory by
not allocation TCE for regions which we are not going to map anyway.
The hardware tables alreary support indirect levels but we also keep
host-physical-to-userspace translation array which is allocated by
vmalloc() and is a flat array which might use quite some memory.
This converts it_userspace from vmalloc'ed array to a multi level table.
As the format becomes platform dependend, this replaces the direct access
to it_usespace with a iommu_table_ops::useraddrptr hook which returns
a pointer to the userspace copy of a TCE; future extension will return
NULL if the level was not allocated.
This should not change non-KVM handling of TCE tables and it_userspace
will not be allocated for non-KVM tables.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Right now we have allocation code in pci-ioda.c and traversing code in
pci.c, let's keep them toghether. However both files are big enough
already so let's move this business to a new file.
While we at it, move the code which links IOMMU table groups to
IOMMU tables as it is not specific to any PNV PHB model.
These puts exported symbols from the new file together.
This fixes several warnings from checkpatch.pl like this:
"WARNING: Prefer 'unsigned int' to bare use of 'unsigned'".
As this is almost cut-n-paste, there should be no behavioral change.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
