Dave Olson [Fri, 3 Apr 2015 04:28:45 +0000 (21:28 -0700)]
powerpc: Fix missing L2 cache size in /sys/devices/system/cpu
This problem appears to have been introduced in 2.6.29 by commit 93197a36a9c1 "Rewrite sysfs processor cache info code".
This caused lscpu to error out on at least e500v2 devices, eg:
error: cannot open /sys/devices/system/cpu/cpu0/cache/index2/size: No such file or directory
Some embedded powerpc systems use cache-size in DTS for the unified L2
cache size, not d-cache-size, so we need to allow for both DTS names.
Added a new CACHE_TYPE_UNIFIED_D cache_type_info structure to handle
this.
Fixes: 93197a36a9c1 ("powerpc: Rewrite sysfs processor cache info code") Signed-off-by: Dave Olson <olson@cumulusnetworks.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Anton Blanchard [Thu, 9 Apr 2015 02:52:56 +0000 (12:52 +1000)]
powerpc: Add ppc64 hard lockup detector support
The hard lockup detector uses a PMU event as a periodic NMI to
detect if we are stuck (where stuck means no timer interrupts have
occurred).
Ben's rework of the ppc64 soft disable code has made ppc64 PMU
exceptions a partial NMI. They can get disabled if an external
interrupt comes in, but otherwise PMU interrupts will fire in
interrupt disabled regions.
We disable the hard lockup detector by default for a few reasons:
- It breaks userspace event based branches on POWER8.
- It is likely to produce false positives on KVM guests.
- Since PMCs can only count to 2^31, counting cycles means we might
take multiple PMU exceptions per second per hardware thread even
if our hard lockup timeout is 10 seconds.
It can be enabled via a boot option, or via procfs.
Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Anton Blanchard [Thu, 9 Apr 2015 02:52:55 +0000 (12:52 +1000)]
oprofile: Disable oprofile NMI timer on ppc64
We want to enable the hard lockup detector on ppc64, but right now
that enables the oprofile NMI timer too.
We'd prefer not to enable the oprofile NMI timer, it adds another
element to our PMU testing and it requires us to increase our
exported symbols (eg cpu_khz).
Modify the config entry for OPROFILE_NMI_TIMER to disable it on PPC64.
Signed-off-by: Anton Blanchard <anton@samba.org> Acked-by: Robert Richter <rric@kernel.org> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/perf/hv-24x7: Break up single_24x7_request
Break up the function single_24x7_request() into smaller functions.
This would later enable us to "prepare" a multi-event request
buffer and then submit a single hcall for several events.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/perf/hv-24x7: Use pr_devel() to log message
Use pr_devel_ratelimited() to log error message when the 24x7 HCALL
fails. Since users specify events by their sysfs name, the HCALL should
succeed. Any errors reported by the HCALL would be of interest to the
developer, rather than the user/administrator.
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/perf/hv-24x7: Modify definition of request and result buffers
The parameters to the 24x7 HCALL have variable number of elements in them.
Set the minimum number of such elements to 1 rather than 0 and eliminate
the temporary structures.
This would enable us to submit multiple counter requests and process
multiple results from a single HCALL (in a follow on patch).
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/powernv: Add interfaces for flash device access
This change adds the OPAL interface definitions to allow Linux to read,
write and erase from system flash devices. We register platform devices
for the flash devices exported by firmware.
We clash with the existing opal_flash_init function, which is really for
the FSP flash update functionality, so we rename that initcall to
opal_flash_update_init().
A future change will add an mtd driver that uses this interface.
Changes from Joel Stanley and Jeremy Kerr.
Signed-off-by: Cyril Bur <cyrilbur@gmail.com> Signed-off-by: Jeremy Kerr <jk@ozlabs.org> Signed-off-by: Joel Stanley <joel@jms.id.au> Acked-by: Stewart Smith <stewart@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Sam bobroff [Fri, 10 Apr 2015 04:16:49 +0000 (14:16 +1000)]
selftests/powerpc: Add transactional syscall test
Check that a syscall made during an active transaction will fail with
the correct failure code and that one made during a suspended
transaction will succeed.
Signed-off-by: Sam Bobroff <sam.bobroff@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Sam bobroff [Fri, 10 Apr 2015 04:16:47 +0000 (14:16 +1000)]
powerpc/tm: Abort syscalls in active transactions
This patch changes the syscall handler to doom (tabort) active
transactions when a syscall is made and return immediately without
performing the syscall.
Currently, the system call instruction automatically suspends an
active transaction which causes side effects to persist when an active
transaction fails.
This does change the kernel's behaviour, but in a way that was
documented as unsupported. It doesn't reduce functionality because
syscalls will still be performed after tsuspend. It also provides a
consistent interface and makes the behaviour of user code
substantially the same across powerpc and platforms that do not
support suspended transactions (e.g. x86 and s390).
Performance measurements using
http://ozlabs.org/~anton/junkcode/null_syscall.c
indicate the cost of a system call increases by about 0.5%.
Signed-off-by: Sam Bobroff <sam.bobroff@au1.ibm.com> Acked-By: Michael Neuling <mikey@neuling.org> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Daniel Axtens [Fri, 10 Apr 2015 03:15:47 +0000 (13:15 +1000)]
powerpc: fsl_pci, swiotlb: Move controller ops from ppc_md to controller_ops
Move the installation of DMA operations out of swiotlb's subsys
initcall, and into the generic PCI controller operations struct.
These ops are installed conditionally, based on the ppc_swiotlb_enable
global. The global can be set in two places:
- swiotlb_detect_4g, which is always called at the arch initcall level
- setup_pci_atmu, which is called as part of the fsl_add_bridge and
fsl_pci_syscore_do_resume.
fsl_pci_syscore_do_resume is called late enough that any changes as a
result of that call will have no effect.
As such, if we test the global and set the operations as part of
fsl_add_bridge, after the call to setup_pci_atmu, we can be confident
that it will cover all the PCI implementations affected by the changes
to dma-swiotlb.c.
Signed-off-by: Daniel Axtens <dja@axtens.net> Acked-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Daniel Axtens [Tue, 31 Mar 2015 05:00:48 +0000 (16:00 +1100)]
powerpc: dart_iommu: optionally populate controller_ops on init
If a pci_controller_ops struct is provided to iommu_init_early_dart,
populate that with the DMA setup ops, rather than ppc_md. If NULL is
provided, populate ppc_md as before.
This also patches the call sites for Maple and Power Mac to pass
NULL, so existing behaviour is preserved.
The benefit of making this optional is that it means we don't have
to change dart, Maple and Power Mac over to the controller_ops
system in one fell swoop.
Signed-off-by: Daniel Axtens <dja@axtens.net> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Daniel Axtens [Tue, 31 Mar 2015 05:00:44 +0000 (16:00 +1100)]
powerpc: Create pci_controller_ops.probe_mode and shim
Add pci_controller_ops.probe_mode, shadowing ppc_md.pci_probe_mode.
Add a shim, and changes the callsites to use the shim.
We also need to move the probe mode defines to pci-bridge.h from pci.h.
They are required by the shim in order to return a sensible default.
Previously, the were defined in pci.h, but pci.h includes pci-bridge.h
before the relevant #defines. This means the definitions are absent
if pci.h is included before pci-bridge.h. This occurs in some drivers.
So, move the definitons now, and move them back when we remove the shim.
Anything that wants the defines would have had to include pci.h, and
since pci.h includes pci-bridge.h, nothing will lose access to the
defines.
Signed-off-by: Daniel Axtens <dja@axtens.net> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Daniel Axtens [Tue, 31 Mar 2015 05:00:42 +0000 (16:00 +1100)]
powerpc: Create pci_controller_ops.dma_dev_setup and shim
Introduces the pci_controller_ops structure.
Add pci_controller_ops.dma_dev_setup, shadowing ppc_md.pci_dma_dev_setup.
Add a shim, and change the callsites to use the shim.
Signed-off-by: Daniel Axtens <dja@axtens.net> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Daniel Axtens [Tue, 31 Mar 2015 05:00:41 +0000 (16:00 +1100)]
powerpc: pcibios_enable_device_hook: return bool rather than int
pcibios_enable_device_hook returned an int. Every implementation
returned either -EINVAL or 0. The return value wasn't propagated by
the caller: any non-zero return value caused pcibios_enable_device
to return -EINVAL itself. Therefore, make the hook return a bool.
Signed-off-by: Daniel Axtens <dja@axtens.net> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Daniel Axtens [Tue, 31 Mar 2015 05:00:39 +0000 (16:00 +1100)]
powerpc: move find_and_init_phbs() to pSeries specific code
Previously, find_and_init_phbs() was used in both PowerNV and pSeries
setup. However, since RTAS support has been dropped from PowerNV, we
can move it into a platform-specific file.
Signed-off-by: Daniel Axtens <dja@axtens.net> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
smp_ops->probe() is currently supposed to return the number of cpus in
the system.
The last actual usage of the value was removed in May 2007 in e147ec8f1808
"[POWERPC] Simplify smp_space_timers". We still passed the value around
until June 2010 when even that was finally removed in c1aa687d499a
"powerpc: Clean up obsolete code relating to decrementer and timebase".
So drop that requirement, probe() now returns void, and update all
implementations.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/cell: Fix cell iommu after it_page_shift changes
The patch to add it_page_shift incorrectly changed the increment of
uaddr to use it_page_shift, rather then (1 << it_page_shift).
This broke booting on at least some Cell blades, as the iommu was
basically non-functional.
Fixes: 3a553170d35d ("powerpc/iommu: Add it_page_shift field to determine iommu page size") Signed-off-by: Michael Ellerman <michael@ellerman.id.au> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Michael Ellerman [Mon, 30 Mar 2015 06:38:09 +0000 (17:38 +1100)]
powerpc: Reword the "returning from prom_init" message
We get way too many bug reports that say "the kernel is hung in
prom_init", which stems from the fact that the last piece of output
people see is "returning from prom_init".
The kernel is almost never hung in prom_init(), it's just that it's
crashed somewhere after prom_init() but prior to the console coming up.
The existing message should give a clue to that, ie. "returning from"
indicates that prom_init() has finished, but it doesn't seem to work.
Let's try something different.
This prints:
Quiescing Open Firmware ...
Booting Linux via __start() ...
Which hopefully makes it clear that prom_init() is not the problem, and
although __start() probably isn't either, it's at least the right place
to begin looking.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Wistfully-Acked-by: Jeremy Kerr <jk@ozlabs.org>
Michael Ellerman [Mon, 30 Mar 2015 03:10:37 +0000 (14:10 +1100)]
powerpc: Replace mem_init_done with slab_is_available()
We have a powerpc specific global called mem_init_done which is "set on
boot once kmalloc can be called".
But that's not *quite* true. We set it at the bottom of mem_init(), and
rely on the fact that mm_init() calls kmem_cache_init() immediately
after that, and nothing is running in parallel.
So replace it with the generic and 100% correct slab_is_available().
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Michael Ellerman [Wed, 25 Mar 2015 09:11:55 +0000 (20:11 +1100)]
powerpc/mm: Change setbat() to take a pgprot_t rather than flags
The callers of setbat() are actually passing a pgprot_t for the flags
parameter. This doesn't matter unless STRICT_MM_TYPECHECKS is enabled.
So we can turn that on without breaking the build, change setbat() to
take a pgprot_t and have it convert it to an unsigned long internally.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Michael Ellerman [Thu, 19 Mar 2015 04:15:20 +0000 (15:15 +1100)]
powerpc: Remove the celleb support
The celleb code has seen no actual development for ~7 years.
We (maintainers) have no access to test hardware, and it is highly
likely the code has bit-rotted.
As far as we're aware the hardware was never widely available, and is
certainly no longer available, and no one on the list has shown any
interest in it over the years.
So remove it. If anyone has one and cares please speak up.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Acked-by: Jeremy Kerr <jk@ozlabs.org>
Michael Ellerman [Thu, 12 Mar 2015 06:27:11 +0000 (17:27 +1100)]
powerpc/powernv: Remove powernv RTAS support
The powernv code has some conditional support for running on bare metal
machines that have no OPAL firmware, but provide RTAS.
No released machines ever supported that, and even in the lab it was
just a transitional hack in the days when OPAL was still being
developed.
So remove the code.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Acked-by: Stewart Smith <stewart@linux.vnet.ibm.com>
Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/scottwood/linux into next
Freescale updates from Scott:
"Highlights include BMan device tree nodes, an MSI erratum workaround, a
couple minor performance improvements, config updates, and misc
fixes/cleanup."
After previous discussions regarding the subject [1][2], there's no clear
explanation or reason why the call was needed in the first place. The sensible
argument is some sort of synchronization between the CPU and the MPIC, which
hasn't been pointed out precisely and is no longer required (at least on BookE
platforms).
The benefit of this change is saving a MMIO trap per interrupt when running in a
KVM guest.
Yanjiang Jin [Mon, 2 Mar 2015 08:35:35 +0000 (16:35 +0800)]
powerpc/mpc85xx: call k(un)map_atomic rather than k(un)map
The k(un)map function may be called in atomic context in the
function map_and_flush(), so use k(un)map_atomic to replace it,
else we would get the below warning during kdump:
Kevin Hao [Tue, 10 Mar 2015 12:41:31 +0000 (20:41 +0800)]
powerpc: book3e_64: fix the align size for paca_struct
All the cache line size of the current book3e 64bit SoCs are 64 bytes.
So we should use this size to align the member of paca_struct.
This only change the paca_struct's members which are private to book3e
CPUs, and should not have any effect to book3s ones. With this, we save
192 bytes. Also change it to __aligned(size) since it is preferred over
__attribute__((aligned(size))).
WARNING: vmlinux.o(.text+0x213b6): Section mismatch in reference from the function chrp_init_early() to the variable .init.data:boot_command_line
The function chrp_init_early() references
the variable __initdata boot_command_line.
This is often because chrp_init_early lacks a __initdata
annotation or the annotation of boot_command_line is wrong.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/powernv: handle OPAL_SUCCESS return in opal_sensor_read
Currently, when a sensor value is read, the kernel calls OPAL, which in
turn builds a message for the FSP, and waits for a message back.
The new device tree for OPAL sensors [1] adds new sensors that can be
read synchronously (core temperatures for instance) and that don't need
to wait for a response.
This patch modifies the opal call to accept an OPAL_SUCCESS return value
and cover the case above.
powerpc/powernv: convert codes returned by OPAL calls
OPAL has its own list of return codes. The patch provides a translation
of such codes in errnos for the opal_sensor_read call, and possibly
others if needed.
Joe Perches [Mon, 30 Mar 2015 23:46:04 +0000 (16:46 -0700)]
powerpc: Use bool function return values of true/false not 1/0
Use the normal return values for bool functions
Signed-off-by: Joe Perches <joe@perches.com> Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Gavin Shan [Fri, 27 Mar 2015 00:29:00 +0000 (11:29 +1100)]
powerpc/powernv: Don't map M64 segments using M32DT
If M64 has been supported, the prefetchable 64-bits memory resources
shouldn't be mapped to the corresponding PE# via M32DT. Unfortunately,
we're doing that in pnv_ioda_setup_pe_seg() wrongly. The issue was
introduced by commit 262af55 ("powerpc/powernv: Enable M64 aperatus
for PHB3"). The patch fixes the issue by simply skipping M64 resources
when updating to M32DT.
Gavin Shan [Fri, 27 Mar 2015 00:22:17 +0000 (11:22 +1100)]
powerpc/eeh: Fix PE#0 check in eeh_add_to_parent_pe()
The function eeh_add_parent_pe() is used to create a PE or add one
edev to its parent PE. Current code checks if PE#0 is valid for the
later case. Actually, we should validate PE#0 for both cases when
EEH core regards PE#0 as invalid one (without flag EEH_VALID_PE_ZERO).
Otherwise, not all EEH devices can be added to its parent PE#0 for
EEH on P7IOC.
The patch fixes the issue by validating PE#0 for the two cases. So far,
we don't have PE#0 for EEH on P7IOC, but it will show up when we enable
M64 for P7IOC. The patch also makes the error message more meaningful.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:59 +0000 (16:23 +0800)]
powerpc/powernv: Group VF PE when IOV BAR is big on PHB3
When IOV BAR is big, each is covered by 4 M64 windows. This leads to
several VF PE sits in one PE in terms of M64.
Group VF PEs according to the M64 allocation.
[bhelgaas: use dev_printk() when possible] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:58 +0000 (16:23 +0800)]
powerpc/powernv: Reserve additional space for IOV BAR, with m64_per_iov supported
M64 aperture size is limited on PHB3. When the IOV BAR is too big, this
will exceed the limitation and failed to be assigned.
Introduce a different mechanism based on the IOV BAR size:
- if IOV BAR size is smaller than 64MB, expand to total_pe
- if IOV BAR size is bigger than 64MB, roundup power2
[bhelgaas: make dev_printk() output more consistent, use PCI_SRIOV_NUM_BARS] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:57 +0000 (16:23 +0800)]
powerpc/powernv: Shift VF resource with an offset
On PowerNV platform, resource position in M64 BAR implies the PE# the
resource belongs to. In some cases, adjustment of a resource is necessary
to locate it to a correct position in M64 BAR .
This patch adds pnv_pci_vf_resource_shift() to shift the 'real' PF IOV BAR
address according to an offset.
Note:
After doing so, there would be a "hole" in the /proc/iomem when offset
is a positive value. It looks like the device return some mmio back to
the system, which actually no one could use it.
[bhelgaas: rework loops, rework overlap check, index resource[]
conventionally, remove pci_regs.h include, squashed with next patch] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:56 +0000 (16:23 +0800)]
powerpc/powernv: Implement pcibios_iov_resource_alignment() on powernv
Implement pcibios_iov_resource_alignment() on powernv platform.
On PowerNV platform, there are 3 cases for the IOV BAR:
1. initial state, the IOV BAR size is multiple times of VF BAR size
2. after expanded, the IOV BAR size is expanded to meet the M64 segment size
3. sizing stage, the IOV BAR is truncated to 0
pnv_pci_iov_resource_alignment() handle these three cases respectively.
[bhelgaas: adjust to drop "align" parameter, return pci_iov_resource_size()
if no ppc_md machdep_call version] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:55 +0000 (16:23 +0800)]
powerpc/powernv: Reserve additional space for IOV BAR according to the number of total_pe
On PHB3, PF IOV BAR will be covered by M64 BAR to have better PE isolation.
M64 BAR is a type of hardware resource in PHB3, which could map a range of
MMIO to PE numbers on powernv platform. And this range is divided equally
by the number of total_pe with each divided range mapping to a PE number.
Also, the M64 BAR must map a MMIO range with power-of-two size.
The total_pe number is usually different from total_VFs, which can lead to
a conflict between MMIO space and the PE number.
For example, if total_VFs is 128 and total_pe is 256, the second half of
M64 BAR will be part of other PCI device, which may already belong to other
PEs.
This patch prevents the conflict by reserving additional space for the PF
IOV BAR, which is total_pe number of VF's BAR size.
[bhelgaas: make dev_printk() output more consistent, index resource[]
conventionally] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Previously the iommu_table had the same lifetime as a struct pnv_ioda_pe
and was embedded in it. The pnv_ioda_pe was assigned to a PE on the bootup
stage. Since PEs are based on the hardware layout which is static in the
system, they will never get released. This means the iommu_table in the
pnv_ioda_pe will never get released either.
This no longer works for VF PE. VF PEs are created and released dynamically
when VFs are created and released. So we need to assign pnv_ioda_pe to VF
PEs respectively when VFs are enabled and clean up those resources for VF
PE when VFs are disabled. And iommu_table is one of the resources we need
to handle dynamically.
Current iommu_table is a static field in pnv_ioda_pe, which will face a
problem when freeing it. During the disabling of a VF,
pnv_pci_ioda2_release_dma_pe will call iommu_free_table to release the
iommu_table for this PE. A static iommu_table will fail in
iommu_free_table.
According to these requirement, this patch allocates iommu_table
dynamically.
Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:53 +0000 (16:23 +0800)]
powerpc/pci: Don't unset PCI resources for VFs
Flag PCI_REASSIGN_ALL_RSRC is used to ignore resources information setup by
firmware, so that kernel would re-assign all resources of pci devices.
On powerpc arch, this happens in a header fixup function
pcibios_fixup_resources(), which will clean up the resources if this flag
is set. This works fine for PFs, since after clean up, kernel will
re-assign the resources in pcibios_resource_survey().
Below is a simple call flow on how it works:
pcibios_init
pcibios_scan_phb
pci_scan_child_bus
...
pci_device_add
pci_fixup_device(pci_fixup_header)
pcibios_fixup_resources # header fixup
for (i = 0; i < DEVICE_COUNT_RESOURCE; i++)
dev->resource[i].start = 0
pcibios_resource_survey # re-assign
pcibios_allocate_resources
However, the VF resources won't be re-assigned, since the VF resources are
completely determined by the PF resources, and the PF resources have
already been reassigned. This means we need to leave VF's resources
un-cleared in pcibios_fixup_resources().
In this patch, we skip the resource unset process in
pcibios_fixup_resources(), if the pci_dev is a VF.
Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Gavin Shan [Wed, 25 Mar 2015 08:23:52 +0000 (16:23 +0800)]
powerpc/pci: Create pci_dn for VFs
pci_dn is the extension of PCI device node and is created from device node.
Unfortunately, VFs are enabled dynamically by PF's driver and they don't
have corresponding device nodes and pci_dn, which is required to access
VFs' config spaces.
The patch creates pci_dn for VFs in pcibios_sriov_enable() on their PF,
and removes pci_dn for VFs in pcibios_sriov_disable() on their PF. When
VF's pci_dn is created, it's put to the child list of the pci_dn of PF's
upstream bridge. The pci_dn is linked to pci_dev during early fixup time
to setup the fast path.
[bhelgaas: add ifdef around add_one_dev_pci_info(), use dev_printk()] Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:51 +0000 (16:23 +0800)]
PCI: Consider additional PF's IOV BAR alignment in sizing and assigning
When sizing and assigning resources, we divide the resources into two
lists: the requested list and the additional list. We don't consider the
alignment of additional VF(n) BAR space.
This is because the alignment required for the VF(n) BAR space is the size
of an individual VF BAR, not the size of the space for *all* VFs. But we
want additional alignment to support partitioning on PowerNV.
Consider the additional IOV BAR alignment when sizing and assigning
resources. When there is not enough system MMIO space to accomodate both
the requested list and the additional list, the PF's IOV BAR alignment will
not contribute to the bridge. When there is enough system MMIO space for
both lists, the additional alignment will contribute to the bridge.
The additional alignment is stored in the min_align of pci_dev_resource,
which is stored in the additional list by add_to_list() at the end of
pbus_size_mem(). The additional alignment is calculated in
pci_resource_alignment(). For an IOV BAR, we have arch dependent function
to get the alignment for different arch.
[bhelgaas: changelog, printk cast] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Per the SR-IOV spec r1.1, sec 3.3.14, the required alignment of a PF's IOV
BAR is the size of an individual VF BAR, and the size consumed is the
individual VF BAR size times NumVFs.
The PowerNV platform has additional alignment requirements to help support
its Partitionable Endpoint device isolation feature (see
Documentation/powerpc/pci_iov_resource_on_powernv.txt).
Add a pcibios_iov_resource_alignment() interface to allow platforms to
request additional alignment.
[bhelgaas: changelog, adapt to reworked pci_sriov_resource_alignment(),
drop "align" parameter] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:48 +0000 (16:23 +0800)]
PCI: Export pci_iov_virtfn_bus() and pci_iov_virtfn_devfn()
On PowerNV, some resource reservation is needed for SR-IOV VFs that don't
exist at the bootup stage. To do the match between resources and VFs, the
code need to get the VF's BDF in advance.
Rename virtfn_bus() and virtfn_devfn() to pci_iov_virtfn_bus() and
pci_iov_virtfn_devfn() and export them.
[bhelgaas: changelog, make "busnr" int] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:47 +0000 (16:23 +0800)]
PCI: Calculate maximum number of buses required for VFs
An SR-IOV device can change its First VF Offset and VF Stride based on the
values of ARI Capable Hierarchy and NumVFs. The number of buses required
for all VFs is determined by NumVFs, First VF Offset, and VF Stride (see
SR-IOV spec r1.1, sec 2.1.2).
Previously pci_iov_bus_range() computed how many buses would be required by
TotalVFs, but this was based on a single NumVFs value and may not have been
the maximum for all NumVFs configurations.
Iterate over all valid NumVFs and calculate the maximum number of bus
numbers that could ever be required for VFs of this device.
[bhelgaas: changelog, compute busnr of NumVFs, not TotalVFs, remove
kerenl-doc comment marker] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:46 +0000 (16:23 +0800)]
PCI: Refresh First VF Offset and VF Stride when updating NumVFs
The First VF Offset and VF Stride fields depend on the NumVFs setting, so
refresh the cached fields in struct pci_sriov when updating NumVFs. See
the SR-IOV spec r1.1, sec 3.3.9 and 3.3.10.
[bhelgaas: changelog, remove kernel-doc comment marker] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:44 +0000 (16:23 +0800)]
PCI: Keep individual VF BAR size in struct pci_sriov
Currently we don't store the individual VF BAR size. We calculate it when
needed by dividing the PF's IOV resource size (which contains space for
*all* the VFs) by total_VFs or by reading the BAR in the SR-IOV capability
again.
Keep the individual VF BAR size in struct pci_sriov.barsz[], add
pci_iov_resource_size() to retrieve it, and use that instead of doing the
division or reading the SR-IOV capability BAR.
[bhelgaas: rename to "barsz[]", simplify barsz[] index computation, remove
SR-IOV capability BAR sizing] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Wei Yang [Wed, 25 Mar 2015 08:23:43 +0000 (16:23 +0800)]
PCI: Print PF SR-IOV resource that contains all VF(n) BAR space
When we size VF BAR0, VF BAR1, etc., from the SR-IOV Capability of a PF, we
learn the alignment requirement and amount of space consumed by a single
VF. But when VFs are enabled, *each* of the NumVFs consumes that amount of
space, so the total size of the PF resource is "VF BAR size * NumVFs".
Add a printk of the total space consumed by the VFs corresponding to what
we already do for normal non-IOV BARs.
No functional change; new message only.
[bhelgaas: split out into its own patch] Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Michael Ellerman [Sat, 28 Mar 2015 10:35:17 +0000 (21:35 +1100)]
selftests/powerpc: Add a test of the switch_endian() syscall
This adds a test of the switch_endian() syscall we added in the previous
commit.
We test it by calling the endian switch syscall, and then executing some
code in the other endian to check everything went as expected. That code
checks registers we expect to be maintained are. If the endian switch
failed to happen that code sequence will be illegal and cause the test
to abort.
We then switch back to the original endian, do the same checks and
finally write a success message and exit(0).
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Michael Ellerman [Sat, 28 Mar 2015 10:35:16 +0000 (21:35 +1100)]
powerpc: Add a proper syscall for switching endianness
We currently have a "special" syscall for switching endianness. This is
syscall number 0x1ebe, which is handled explicitly in the 64-bit syscall
exception entry.
That has a few problems, firstly the syscall number is outside of the
usual range, which confuses various tools. For example strace doesn't
recognise the syscall at all.
Secondly it's handled explicitly as a special case in the syscall
exception entry, which is complicated enough without it.
As a first step toward removing the special syscall, we need to add a
regular syscall that implements the same functionality.
The logic is simple, it simply toggles the MSR_LE bit in the userspace
MSR. This is the same as the special syscall, with the caveat that the
special syscall clobbers fewer registers.
This version clobbers r9-r12, XER, CTR, and CR0-1,5-7.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Tyrel Datwyler [Fri, 27 Mar 2015 19:47:25 +0000 (12:47 -0700)]
powerpc/pseries: Simplify check for suspendability during suspend/migration
During suspend/migration operation we must wait for the VASI state reported
by the hypervisor to become Suspending prior to making the ibm,suspend-me
RTAS call. Calling routines to rtas_ibm_supend_me() pass a vasi_state variable
that exposes the VASI state to the caller. This is unnecessary as the caller
only really cares about the following three conditions; if there is an error
we should bailout, success indicating we have suspended and woken back up so
proceed to device tree update, or we are not suspendable yet so try calling
rtas_ibm_suspend_me again shortly.
This patch removes the extraneous vasi_state variable and simply uses the
return code to communicate how to proceed. We either succeed, fail, or get
-EAGAIN in which case we sleep for a second before trying to call
rtas_ibm_suspend_me again. The behaviour of ppc_rtas() remains the same,
but migrate_store() now returns the propogated error code on failure.
Previously -1 was returned from migrate_store() in the failure case which
equates to -EPERM and was clearly wrong.
Jan Stancek [Tue, 24 Mar 2015 12:33:22 +0000 (08:33 -0400)]
powerpc/perf: add missing put_cpu_var in power_pmu_event_init
One path in power_pmu_event_init() calls get_cpu_var(), but is
missing matching call to put_cpu_var(), which causes preemption
imbalance and crash in user-space:
Yanjiang Jin [Fri, 27 Feb 2015 05:30:34 +0000 (13:30 +0800)]
powerpc/mm: Free string after creating kmem cache
kmem_cache_create()->kmem_cache_create_memcg()->kstrdup() allocates new
space and copys name's content, so it is safe to free name memory after
calling kmem_cache_create(). Else kmemleak will report the below
warning:
unreferenced object 0xc0000000f9002160 (size 16):
comm "swapper/0", pid 0, jiffies 4294892296 (age 1386.640s)
hex dump (first 16 bytes):
70 67 74 61 62 6c 65 2d 32 5e 39 00 de ad be ef pgtable-2^9.....
backtrace:
[<c0000000004e03ec>] .kvasprintf+0x5c/0xa0
[<c0000000004e045c>] .kasprintf+0x2c/0x50
[<c00000000002e36c>] .pgtable_cache_add+0xac/0x100
[<c00000000002e3e4>] .pgtable_cache_init+0x24/0x80
[<c000000000c6c67c>] .start_kernel+0x228/0x4c8
[<c000000000000594>] .start_here_common+0x24/0x90
Signed-off-by: Yanjiang Jin <yanjiang.jin@windriver.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
projects / linux-beck.git / log