powerpc: Make cpu_spec __ro_after_init

The cpu_spec is a struct holding various information about the CPU the
kernel is executing on. It's populated early in boot and must not change
after that.

In

powerpc: Make cpu_spec __ro_after_init

The cpu_spec is a struct holding various information about the CPU the
kernel is executing on. It's populated early in boot and must not change
after that.

In particular the cpu_features and mmu_features hold the set of
discovered CPU/MMU features and are used to set static keys for each
feature, and do binary patching of assembly. So any change to the
cpu_features/mmu_features later in boot will not be reflected in
the state of the static keys or patched code.

There is already logic to check that cpu_features/mmu_features don't
change, see check_features() in feature-fixups.c.

But as another layer of protection the entire cpu_spec should be read
only after init, annotate it as such.

Signed-off-by: Michael Ellerman <[email protected]>
Link: https://msgid.link/[email protected]

show more ...

powerpc/kuap: Use MMU_FTR_KUAP on all and refactor disabling kuap

All but book3s/64 use a static branch key for disabling kuap.
book3s/64 uses an mmu feature.

Refactor all targets to use MMU_FTR_KU

powerpc/kuap: Use MMU_FTR_KUAP on all and refactor disabling kuap

All but book3s/64 use a static branch key for disabling kuap.
book3s/64 uses an mmu feature.

Refactor all targets to use MMU_FTR_KUAP like book3s/64.

For PPC32 that implies updating mmu features fixups once KUAP
has been initialised.

Signed-off-by: Christophe Leroy <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://msgid.link/6b3d7c977bad73378ea368bc6818e9c94ea95ab0.1689091022.git.christophe.leroy@csgroup.eu

show more ...

powerpc/cputable: Split cpu_specs[] out of cputable.h

cpu_specs[] is full of #ifdefs depending on the different
types of CPU.

CPUs are mutually exclusive, it is therefore possible to split
cpu_spec

powerpc/cputable: Split cpu_specs[] out of cputable.h

cpu_specs[] is full of #ifdefs depending on the different
types of CPU.

CPUs are mutually exclusive, it is therefore possible to split
cpu_specs[] into smaller more readable pieces.

Create cpu_specs_XXX.h that will each be dedicated on one
of the following mutually exclusive families:
- 40x
- 44x
- 47x
- 8xx
- e500
- book3s/32
- book3s/64

In book3s/32, the block for 603 has been moved in front in order
to not have two 604 blocks.

Signed-off-by: Christophe Leroy <[email protected]>
[mpe: Fix CONFIG_47x to be CONFIG_PPC_47x, tweak some formatting]
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/a44b865e0318286155273b10cdf524ab697928c1.1663606875.git.christophe.leroy@csgroup.eu

show more ...

powerpc/cputable: Move __cpu_setup() prototypes out of cputable.h

Move all prototypes out of cputable.h

For that rename cpu_setup_power.h to cpu_setup.h and move all
prototypes in it.

Signed-off-b

powerpc/cputable: Move __cpu_setup() prototypes out of cputable.h

Move all prototypes out of cputable.h

For that rename cpu_setup_power.h to cpu_setup.h and move all
prototypes in it.

Signed-off-by: Christophe Leroy <[email protected]>
[mpe: Standardise cpu_spec *spec formatting]
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/f45118489ee450db654db8bbcdfd8f5907337c22.1663606875.git.christophe.leroy@csgroup.eu

show more ...

powerpc/cputable: Remove __machine_check_early_realmode_p{7/8/9} prototypes

__machine_check_early_realmode_p{7/8/9} are already in mce.h
which is included. Remove them from cputable.c

Signed-off-by

powerpc/cputable: Remove __machine_check_early_realmode_p{7/8/9} prototypes

__machine_check_early_realmode_p{7/8/9} are already in mce.h
which is included. Remove them from cputable.c

Signed-off-by: Christophe Leroy <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/b77fc0f90e3a9c065324cbff549b718ccf0809f8.1663606875.git.christophe.leroy@csgroup.eu

show more ...

powerpc/64e: Tie PPC_BOOK3E_64 to PPC_E500MC

The only 64-bit Book3E CPUs we support require the selection
of CONFIG_PPC_E500MC.

However our Kconfig allows configurating a kernel that has 64-bit
Boo

powerpc/64e: Tie PPC_BOOK3E_64 to PPC_E500MC

The only 64-bit Book3E CPUs we support require the selection
of CONFIG_PPC_E500MC.

However our Kconfig allows configurating a kernel that has 64-bit
Book3E support, but without CONFIG_PPC_E500MC enabled. Such a kernel
would never boot, it doesn't know about any CPUs.

To fix this, force CONFIG_PPC_E500MC to be selected whenever we are
building a 64-bit Book3E kernel.

And add a test to detect future situations where cpu_specs is empty.

Signed-off-by: Christophe Leroy <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/ae5d8b8b3ccc346e61d2ec729767f92766273f0b.1663606875.git.christophe.leroy@csgroup.eu

show more ...

powerpc: Remove remaining parts of oprofile

Commit 9850b6c69356 ("arch: powerpc: Remove oprofile") removed
oprofile.

Remove all remaining parts of it.

Signed-off-by: Christophe Leroy <christophe.l

powerpc: Remove remaining parts of oprofile

Commit 9850b6c69356 ("arch: powerpc: Remove oprofile") removed
oprofile.

Remove all remaining parts of it.

Signed-off-by: Christophe Leroy <[email protected]>
Acked-by: Viresh Kumar <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/298432fe1a14c0a415760011d72c3f0999efd5e2.1657204631.git.christophe.leroy@csgroup.eu

show more ...

powerpc: Enable the DAWR on POWER9 DD2.3 and above

The hardware bug in POWER9 preventing use of the DAWR was fixed in
DD2.3. Set the CPU_FTR_DAWR feature bit on these newer systems to start
using it

powerpc: Enable the DAWR on POWER9 DD2.3 and above

The hardware bug in POWER9 preventing use of the DAWR was fixed in
DD2.3. Set the CPU_FTR_DAWR feature bit on these newer systems to start
using it again, and update the documentation accordingly.

The CPU features for DD2.3 are currently determined by "DD2.2 or later"
logic. In adding DD2.3 as a discrete case for the first time here, I'm
carrying the quirks of DD2.2 forward to keep all behavior outside of
this DAWR change the same. This leaves the assessment and potential
removal of those quirks on DD2.3 for later.

Signed-off-by: Reza Arbab <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

powerpc: Export mmu_feature_keys[] as non-GPL

When the mmu_feature_keys[] was introduced in the commit c12e6f24d413
("powerpc: Add option to use jump label for mmu_has_feature()"),
it is unlikely th

powerpc: Export mmu_feature_keys[] as non-GPL

When the mmu_feature_keys[] was introduced in the commit c12e6f24d413
("powerpc: Add option to use jump label for mmu_has_feature()"),
it is unlikely that it would be used either directly or indirectly in
the out of tree modules. So we exported it as GPL only.

But with the evolution of the codes, especially the PPC_KUAP support, it
may be indirectly referenced by some primitive macro or inline functions
such as get_user() or __copy_from_user_inatomic(), this will make it
impossible to build many non GPL modules (such as ZFS) on ppc
architecture. Fix this by exposing the mmu_feature_keys[] to the non-GPL
modules too.

Fixes: 7613f5a66bec ("powerpc/64s/kuap: Use mmu_has_feature()")
Reported-by: Nathaniel Filardo <[email protected]>
Signed-off-by: Kevin Hao <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

powerpc: Add missing headers

Don't inherit headers "by chances" from asm/prom.h, asm/mpc52xx.h,
asm/pci.h etc...

Include the needed headers, and remove asm/prom.h when it was
needed exclusively for

powerpc: Add missing headers

Don't inherit headers "by chances" from asm/prom.h, asm/mpc52xx.h,
asm/pci.h etc...

Include the needed headers, and remove asm/prom.h when it was
needed exclusively for pulling necessary headers.

Signed-off-by: Christophe Leroy <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/be8bdc934d152a7d8ee8d1a840d5596e2f7d85e0.1646767214.git.christophe.leroy@csgroup.eu

show more ...

powerpc: fix typos in comments

Various spelling mistakes in comments.
Detected with the help of Coccinelle.

Signed-off-by: Julia Lawall <[email protected]>
Reviewed-by: Joel Stanley <[email protected]

powerpc: fix typos in comments

Various spelling mistakes in comments.
Detected with the help of Coccinelle.

Signed-off-by: Julia Lawall <[email protected]>
Reviewed-by: Joel Stanley <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

arch: powerpc: Stop building and using oprofile

The "oprofile" user-space tools don't use the kernel OPROFILE support
any more, and haven't in a long time. User-space has been converted to
the perf

arch: powerpc: Stop building and using oprofile

The "oprofile" user-space tools don't use the kernel OPROFILE support
any more, and haven't in a long time. User-space has been converted to
the perf interfaces.

This commits stops building oprofile for powerpc and removes any
reference to it from directories in arch/powerpc/ apart from
arch/powerpc/oprofile, which will be removed in the next commit (this is
broken into two commits as the size of the commit became very big, ~5k
lines).

Note that the member "oprofile_cpu_type" in "struct cpu_spec" isn't
removed as it was also used by other parts of the code.

Suggested-by: Christoph Hellwig <[email protected]>
Suggested-by: Linus Torvalds <[email protected]>
Signed-off-by: Viresh Kumar <[email protected]>
Acked-by: Robert Richter <[email protected]>
Acked-by: William Cohen <[email protected]>
Acked-by: Al Viro <[email protected]>
Acked-by: Thomas Gleixner <[email protected]>

show more ...

powerpc/32s: Make support for 603 and 604+ selectable

book3s/32 has two main families:
- CPU with 603 cores that don't have HASH PTE table and
perform SW TLB loading.
- Other CPUs based on 604+ core

powerpc/32s: Make support for 603 and 604+ selectable

book3s/32 has two main families:
- CPU with 603 cores that don't have HASH PTE table and
perform SW TLB loading.
- Other CPUs based on 604+ cores that have HASH PTE table.

This leads to some complex logic and additionnal code to
support both. This makes sense for distribution kernels
that aim at running on any CPU, but when you are fine
tuning a kernel for an embedded 603 based board you
don't need all the HASH logic.

Allow selection of support for each family, in order to opt
out unneeded parts of code. At least one must be selected.

Note that some of the CPU supporting HASH also support SW TLB
loading, however it is not supported by Linux kernel at the
time being, because they do not have alternate registers in
the TLB miss exception handlers.

Signed-off-by: Christophe Leroy <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/8dde0cdb629a71abc29b0d85a52a86e920376cb6.1603348103.git.christophe.leroy@csgroup.eu

show more ...

powerpc/32s: Regroup 603 based CPUs in cputable

In order to selectively build the kernel for 603 SW TLB handling,
regroup all 603 based CPUs together.

Signed-off-by: Christophe Leroy <christophe.le

powerpc/32s: Regroup 603 based CPUs in cputable

In order to selectively build the kernel for 603 SW TLB handling,
regroup all 603 based CPUs together.

Signed-off-by: Christophe Leroy <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/45065263fdb9f5cc2a2d210ec2a762ac8bf5b2bc.1603348103.git.christophe.leroy@csgroup.eu

show more ...

powerpc/32s: Remove CONFIG_PPC_BOOK3S_6xx

As 601 is gone, CONFIG_PPC_BOO3S_6xx and CONFIG_PPC_BOOK3S_32
are dedundant.

Remove CONFIG_PPC_BOOK3S_6xx.

Signed-off-by: Christophe Leroy <christophe.ler

powerpc/32s: Remove CONFIG_PPC_BOOK3S_6xx

As 601 is gone, CONFIG_PPC_BOO3S_6xx and CONFIG_PPC_BOOK3S_32
are dedundant.

Remove CONFIG_PPC_BOOK3S_6xx.

Signed-off-by: Christophe Leroy <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/f18c16af37f6f77b577bed8d9e12831b695617ae.1603348103.git.christophe.leroy@csgroup.eu

show more ...

powerpc: Retire e200 core (mpc555x processor)

There is no defconfig selecting CONFIG_E200, and no platform.

e200 is an earlier version of booke, a predecessor of e500,
with some particularities lik

powerpc: Retire e200 core (mpc555x processor)

There is no defconfig selecting CONFIG_E200, and no platform.

e200 is an earlier version of booke, a predecessor of e500,
with some particularities like an unified cache instead of both an
instruction cache and a data cache.

Remove it.

Signed-off-by: Christophe Leroy <[email protected]>
Acked-by: Scott Wood <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/34ebc3ba2c768d97f363bd5f2deea2356e9ae127.1605589460.git.christophe.leroy@csgroup.eu

show more ...

powerpc/44x: Don't support 440 when CONFIG_PPC_47x is set

As stated in platform/44x/Kconfig, CONFIG_PPC_47x is not
compatible with 440 and 460 variants.

This is confirmed in asm/cache.h as L1_CACHE

powerpc/44x: Don't support 440 when CONFIG_PPC_47x is set

As stated in platform/44x/Kconfig, CONFIG_PPC_47x is not
compatible with 440 and 460 variants.

This is confirmed in asm/cache.h as L1_CACHE_SHIFT is different
for 47x, meaning a kernel built for 47x will not run correctly
on a 440.

In cputable, opt out all 440 and 460 variants when CONFIG_PPC_47x
is set. Also add a default match dedicated to 470.

Signed-off-by: Christophe Leroy <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/822833ce3dc10634339818f7d1ab616edf63b0c6.1603041883.git.christophe.leroy@csgroup.eu

show more ...

powerpc/64s: Convert some cpu_setup() and cpu_restore() functions to C

The only thing keeping the cpu_setup() and cpu_restore() functions
used in the cputable entries for Power7, Power8, Power9 and

powerpc/64s: Convert some cpu_setup() and cpu_restore() functions to C

The only thing keeping the cpu_setup() and cpu_restore() functions
used in the cputable entries for Power7, Power8, Power9 and Power10 in
assembly was cpu_restore() being called before there was a stack in
generic_secondary_smp_init(). Commit ("powerpc/64: Set up a kernel
stack for secondaries before cpu_restore()") means that it is now
possible to use C.

Rewrite the functions in C so they are a little bit easier to read.
This is not changing their functionality.

Signed-off-by: Jordan Niethe <[email protected]>
[mpe: Tweak copyright and authorship notes]
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

powerpc/64s: Remove TM from Power10 features

ISA v3.1 removes transactional memory and hence it should not be present
in cpu_features or cpu_user_features2. Remove CPU_FTR_TM_COMP from
CPU_FTRS_POWE

powerpc/64s: Remove TM from Power10 features

ISA v3.1 removes transactional memory and hence it should not be present
in cpu_features or cpu_user_features2. Remove CPU_FTR_TM_COMP from
CPU_FTRS_POWER10. Remove PPC_FEATURE2_HTM_COMP and
PPC_FEATURE2_HTM_NOSC_COMP from COMMON_USER2_POWER10.

Fixes: a3ea40d5c736 ("powerpc: Add POWER10 architected mode")
Signed-off-by: Jordan Niethe <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

powerpc: Remove PowerPC 601

Powerpc 601 is 25 years old.

It is not selected by any defconfig.

It requires a lot of special handling as it deviates from the
standard 6xx.

Retire it.

Signed-off-by

powerpc: Remove PowerPC 601

Powerpc 601 is 25 years old.

It is not selected by any defconfig.

It requires a lot of special handling as it deviates from the
standard 6xx.

Retire it.

Signed-off-by: Christophe Leroy <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/00a6948d659e017f8ca63437d1384222c3aede57.1601362098.git.christophe.leroy@csgroup.eu

show more ...

powerpc: untangle cputable mce include

Having cputable.h include mce.h means it pulls in a bunch of low level
headers (e.g., synch.h) which then can't use CPU_FTR_ definitions.

Signed-off-by: Nicho

powerpc: untangle cputable mce include

Having cputable.h include mce.h means it pulls in a bunch of low level
headers (e.g., synch.h) which then can't use CPU_FTR_ definitions.

Signed-off-by: Nicholas Piggin <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

powerpc/kernel: Cleanup machine check function declarations

__machine_check_early_realmode_p*() are currently declared as extern
in cputable.c and because of this when compiled with "C=1" (which
ena

powerpc/kernel: Cleanup machine check function declarations

__machine_check_early_realmode_p*() are currently declared as extern
in cputable.c and because of this when compiled with "C=1" (which
enables semantic checker) produces these warnings.

CHECK arch/powerpc/kernel/mce_power.c
arch/powerpc/kernel/mce_power.c:709:6: warning: symbol '__machine_check_early_realmode_p7' was not declared. Should it be static?
arch/powerpc/kernel/mce_power.c:717:6: warning: symbol '__machine_check_early_realmode_p8' was not declared. Should it be static?
arch/powerpc/kernel/mce_power.c:722:6: warning: symbol '__machine_check_early_realmode_p9' was not declared. Should it be static?
arch/powerpc/kernel/mce_power.c:740:6: warning: symbol '__machine_check_early_realmode_p10' was not declared. Should it be static?

Patch here moves the declaration to asm/mce.h and includes the same in
cputable.c

Fixes: ae744f3432d3 ("powerpc/book3s: Flush SLB/TLBs if we get SLB/TLB machine check errors on power8")
Fixes: 7b9f71f974a1 ("powerpc/64s: POWER9 machine check handler")
Signed-off-by: Madhavan Srinivasan <[email protected]>
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

powerpc: Add POWER10 raw mode cputable entry

Add a raw mode cputable entry for POWER10. Copies most of the fields
from commit a3ea40d5c736 ("powerpc: Add POWER10 architected mode")
except for oprofi

powerpc: Add POWER10 raw mode cputable entry

Add a raw mode cputable entry for POWER10. Copies most of the fields
from commit a3ea40d5c736 ("powerpc: Add POWER10 architected mode")
except for oprofile_cpu_type, machine_check_early, pvr_mask and
pvr_mask fields. On bare metal systems we use DT CPU features, which
doesn't need a cputable entry. But in VMs we still rely on the raw
cputable entry to set the correct values for the PMU related fields.

Fixes: a3ea40d5c736 ("powerpc: Add POWER10 architected mode")
Signed-off-by: Madhavan Srinivasan <[email protected]>
[mpe: Reorder vs cleanup patch and add Fixes tag]
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

powerpc/64s: system call support for scv/rfscv instructions

Add support for the scv instruction on POWER9 and later CPUs.

For now this implements the zeroth scv vector 'scv 0', as identical to
'sc'

powerpc/64s: system call support for scv/rfscv instructions

Add support for the scv instruction on POWER9 and later CPUs.

For now this implements the zeroth scv vector 'scv 0', as identical to
'sc' system calls, with the exception that LR is not preserved, nor
are volatile CR registers, and error is not indicated with CR0[SO],
but by returning a negative errno.

rfscv is implemented to return from scv type system calls. It can not
be used to return from sc system calls because those are defined to
preserve LR.

getpid syscall throughput on POWER9 is improved by 26% (428 to 318
cycles), largely due to reducing mtmsr and mtspr.

Signed-off-by: Nicholas Piggin <[email protected]>
[mpe: Fix ppc64e build]
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

powerpc: Add POWER10 architected mode

PVR value of 0x0F000006 means we are arch v3.1 compliant (i.e.
POWER10). This is used by phyp and kvm when booting as a pseries guest
to detect the presence of

powerpc: Add POWER10 architected mode

PVR value of 0x0F000006 means we are arch v3.1 compliant (i.e.
POWER10). This is used by phyp and kvm when booting as a pseries guest
to detect the presence of new P10 features and to enable the
appropriate hwcap and facility bits.

Signed-off-by: Alistair Popple <[email protected]>
Signed-off-by: Cédric Le Goater <[email protected]>
[mpe: Fall through to __init_FSCR rather than duplicating it, drop
hack to set current->thread.fscr now that is handled elsewhere.]
Signed-off-by: Michael Ellerman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...