| Age | Commit message (Collapse) | Author |
|---|
|
Add support for bpf_arch_text_poke() and arch_prepare_bpf_trampoline()
for 64-bit powerpc. While the code is generic, BPF trampolines are only
enabled on 64-bit powerpc. 32-bit powerpc will need testing and some
updates.
BPF Trampolines adhere to the existing ftrace ABI utilizing a
two-instruction profiling sequence, as well as the newer ABI utilizing a
three-instruction profiling sequence enabling return with a 'blr'. The
trampoline code itself closely follows x86 implementation.
BPF prog JIT is extended to mimic 64-bit powerpc approach for ftrace
having a single nop at function entry, followed by the function
profiling sequence out-of-line and a separate long branch stub for calls
to trampolines that are out of range. A dummy_tramp is provided to
simplify synchronization similar to arm64.
When attaching a bpf trampoline to a bpf prog, we can patch up to three
things:
- the nop at bpf prog entry to go to the out-of-line stub
- the instruction in the out-of-line stub to either call the bpf trampoline
directly, or to branch to the long_branch stub.
- the trampoline address before the long_branch stub.
We do not need any synchronization here since we always have a valid
branch target regardless of the order in which the above stores are
seen. dummy_tramp ensures that the long_branch stub goes to a valid
destination on other cpus, even when the branch to the long_branch stub
is seen before the updated trampoline address.
However, when detaching a bpf trampoline from a bpf prog, or if changing
the bpf trampoline address, we need synchronization to ensure that other
cpus can no longer branch into the older trampoline so that it can be
safely freed. bpf_tramp_image_put() uses rcu_tasks to ensure all cpus
make forward progress, but we still need to ensure that other cpus
execute isync (or some CSI) so that they don't go back into the
trampoline again. While here, update the stale comment that describes
the redzone usage in ppc64 BPF JIT.
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-18-hbathini@linux.ibm.com
|
|
Add powerpc 32-bit and 64-bit samples for ftrace direct. This serves to
show the sample instruction sequence to be used by ftrace direct calls
to adhere to the ftrace ABI.
On 64-bit powerpc, TOC setup requires some additional work.
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-17-hbathini@linux.ibm.com
|
|
Add support for DYNAMIC_FTRACE_WITH_DIRECT_CALLS similar to the arm64
implementation.
ftrace direct calls allow custom trampolines to be called into directly
from function ftrace call sites, bypassing the ftrace trampoline
completely. This functionality is currently utilized by BPF trampolines
to hook into kernel function entries.
Since we have limited relative branch range, we support ftrace direct
calls through support for DYNAMIC_FTRACE_WITH_CALL_OPS. In this
approach, ftrace trampoline is not entirely bypassed. Rather, it is
re-purposed into a stub that reads direct_call field from the associated
ftrace_ops structure and branches into that, if it is not NULL. For
this, it is sufficient if we can ensure that the ftrace trampoline is
reachable from all traceable functions.
When multiple ftrace_ops are associated with a call site, we utilize a
call back to set pt_regs->orig_gpr3 that can then be tested on the
return path from the ftrace trampoline to branch into the direct caller.
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-16-hbathini@linux.ibm.com
|
|
Implement support for DYNAMIC_FTRACE_WITH_CALL_OPS similar to the
arm64 implementation.
This works by patching-in a pointer to an associated ftrace_ops
structure before each traceable function. If multiple ftrace_ops are
associated with a call site, then a special ftrace_list_ops is used to
enable iterating over all the registered ftrace_ops. If no ftrace_ops
are associated with a call site, then a special ftrace_nop_ops structure
is used to render the ftrace call as a no-op. ftrace trampoline can then
read the associated ftrace_ops for a call site by loading from an offset
from the LR, and branch directly to the associated function.
The primary advantage with this approach is that we don't have to
iterate over all the registered ftrace_ops for call sites that have a
single ftrace_ops registered. This is the equivalent of implementing
support for dynamic ftrace trampolines, which set up a special ftrace
trampoline for each registered ftrace_ops and have individual call sites
branch into those directly.
A secondary advantage is that this gives us a way to add support for
direct ftrace callers without having to resort to using stubs. The
address of the direct call trampoline can be loaded from the ftrace_ops
structure.
To support this, we reserve a nop before each function on 32-bit
powerpc. For 64-bit powerpc, two nops are reserved before each
out-of-line stub. During ftrace activation, we update this location with
the associated ftrace_ops pointer. Then, on ftrace entry, we load from
this location and call into ftrace_ops->func().
For 64-bit powerpc, we ensure that the out-of-line stub area is
doubleword aligned so that ftrace_ops address can be updated atomically.
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-15-hbathini@linux.ibm.com
|
|
We are restricted to a .text size of ~32MB when using out-of-line
function profile sequence. Allow this to be extended up to the previous
limit of ~64MB by reserving space in the middle of .text.
A new config option CONFIG_PPC_FTRACE_OUT_OF_LINE_NUM_RESERVE is
introduced to specify the number of function stubs that are reserved in
.text. On boot, ftrace utilizes stubs from this area first before using
the stub area at the end of .text.
A ppc64le defconfig has ~44k functions that can be traced. A more
conservative value of 32k functions is chosen as the default value of
PPC_FTRACE_OUT_OF_LINE_NUM_RESERVE so that we do not allot more space
than necessary by default. If building a kernel that only has 32k
trace-able functions, we won't allot any more space at the end of .text
during the pass on vmlinux.o. Otherwise, only the remaining functions
get space for stubs at the end of .text. This default value should help
cover a .text size of ~48MB in total (including space reserved at the
end of .text which can cover up to 32MB), which should be sufficient for
most common builds. For a very small kernel build, this can be set to 0.
Or, this can be bumped up to a larger value to support vmlinux .text
size up to ~64MB.
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-14-hbathini@linux.ibm.com
|
|
Function profile sequence on powerpc includes two instructions at the
beginning of each function:
mflr r0
bl ftrace_caller
The call to ftrace_caller() gets nop'ed out during kernel boot and is
patched in when ftrace is enabled.
Given the sequence, we cannot return from ftrace_caller with 'blr' as we
need to keep LR and r0 intact. This results in link stack (return
address predictor) imbalance when ftrace is enabled. To address that, we
would like to use a three instruction sequence:
mflr r0
bl ftrace_caller
mtlr r0
Further more, to support DYNAMIC_FTRACE_WITH_CALL_OPS, we need to
reserve two instruction slots before the function. This results in a
total of five instruction slots to be reserved for ftrace use on each
function that is traced.
Move the function profile sequence out-of-line to minimize its impact.
To do this, we reserve a single nop at function entry using
-fpatchable-function-entry=1 and add a pass on vmlinux.o to determine
the total number of functions that can be traced. This is then used to
generate a .S file reserving the appropriate amount of space for use as
ftrace stubs, which is built and linked into vmlinux.
On bootup, the stub space is split into separate stubs per function and
populated with the proper instruction sequence. A pointer to the
associated stub is maintained in dyn_arch_ftrace.
For modules, space for ftrace stubs is reserved from the generic module
stub space.
This is restricted to and enabled by default only on 64-bit powerpc,
though there are some changes to accommodate 32-bit powerpc. This is
done so that 32-bit powerpc could choose to opt into this based on
further tests and benchmarks.
As an example, after this patch, kernel functions will have a single nop
at function entry:
<kernel_clone>:
addis r2,r12,467
addi r2,r2,-16028
nop
mfocrf r11,8
...
When ftrace is enabled, the nop is converted to an unconditional branch
to the stub associated with that function:
<kernel_clone>:
addis r2,r12,467
addi r2,r2,-16028
b ftrace_ool_stub_text_end+0x11b28
mfocrf r11,8
...
The associated stub:
<ftrace_ool_stub_text_end+0x11b28>:
mflr r0
bl ftrace_caller
mtlr r0
b kernel_clone+0xc
...
This change showed an improvement of ~10% in null_syscall benchmark on a
Power 10 system with ftrace enabled.
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-13-hbathini@linux.ibm.com
|
|
Function tracer on powerpc can only work with vmlinux having a .text
size of up to ~64MB due to powerpc branch instruction having a limited
relative branch range of 32MB. Today, this is only detected on kernel
boot when ftrace is init'ed. Add a post-link script to check the size of
.text so that we can detect this at build time, and break the build if
necessary.
We add a dependency on !COMPILE_TEST for CONFIG_HAVE_FUNCTION_TRACER so
that allyesconfig and other test builds can continue to work without
enabling ftrace.
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-11-hbathini@linux.ibm.com
|
|
bpf_jit_emit_func_call_rel()
Commit 61688a82e047 ("powerpc/bpf: enable kfunc call") enhanced
bpf_jit_emit_func_call_hlp() to handle calls out to module region, where
bpf progs are generated. The only difference now between
bpf_jit_emit_func_call_hlp() and bpf_jit_emit_func_call_rel() is in
handling of the initial pass where target function address is not known.
Fold that logic into bpf_jit_emit_func_call_hlp() and rename it to
bpf_jit_emit_func_call_rel() to simplify bpf function call JIT code.
We don't actually need to load/restore TOC across a call out to a
different kernel helper or to a different bpf program since they all
work with the kernel TOC. We only need to do it if we have to call out
to a module function. So, guard TOC load/restore with appropriate
conditions.
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-10-hbathini@linux.ibm.com
|
|
Move the ftrace stub used to cover inittext before _einittext so that it
is within kernel text, as seen through core_kernel_text(). This is
required for a subsequent change to ftrace.
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-9-hbathini@linux.ibm.com
|
|
To simplify upcoming changes to ftrace, add a check to skip actual
instruction patching if the old and new instructions are the same. We
still validate that the instruction is what we expect, but don't
actually patch the same instruction again.
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-8-hbathini@linux.ibm.com
|
|
Pointer to struct module is only relevant for ftrace records belonging
to kernel modules. Having this field in dyn_arch_ftrace wastes memory
for all ftrace records belonging to the kernel. Remove the same in
favour of looking up the module from the ftrace record address, similar
to other architectures.
Reviewed-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-7-hbathini@linux.ibm.com
|
|
Minor refactor for converting #ifdef to IS_ENABLED().
Reviewed-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-6-hbathini@linux.ibm.com
|
|
On 32-bit powerpc, gcc generates a three instruction sequence for
function profiling:
mflr r0
stw r0, 4(r1)
bl _mcount
On kernel boot, the call to _mcount() is nop-ed out, to be patched back
in when ftrace is actually enabled. The 'stw' instruction therefore is
not necessary unless ftrace is enabled. Nop it out during ftrace init.
When ftrace is enabled, we want the 'stw' so that stack unwinding works
properly. Perform the same within the ftrace handler, similar to 64-bit
powerpc.
Reviewed-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-5-hbathini@linux.ibm.com
|
|
Gcc v5.x emits a 3-instruction sequence for -mprofile-kernel:
mflr r0
std r0, 16(r1)
bl _mcount
Gcc v6.x moved to a simpler 2-instruction sequence by removing the 'std'
instruction. The store saved the return address in the LR save area in
the caller stack frame for stack unwinding. However, with dynamic
ftrace, we no longer have a call to _mcount on kernel boot when ftrace
is not enabled. When ftrace is enabled, that store is performed within
ftrace_caller(). As such, the additional 'std' instruction is redundant.
Nop it out on kernel boot.
With this change, we now use the same 2-instruction profiling sequence
with both -mprofile-kernel, as well as -fpatchable-function-entry on
64-bit powerpc.
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-4-hbathini@linux.ibm.com
|
|
Rather than hard-coding the offset into a function to be used to
determine if a kprobe is at function entry, use ftrace_location() to
determine the ftrace location within the function and categorize all
instructions till that offset to be function entry.
For functions that cannot be traced, we fall back to using a fixed
offset of 8 (two instructions) to categorize a probe as being at
function entry for 64-bit elfv2, unless we are using pcrel.
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-3-hbathini@linux.ibm.com
|
|
So far, we have relied on the fact that gcc supports both
-mprofile-kernel, as well as -fpatchable-function-entry, and clang
supports neither. Our Makefile only checks for CONFIG_MPROFILE_KERNEL to
decide which files to build. Clang has a feature request out [*] to
implement -fpatchable-function-entry, and is unlikely to support
-mprofile-kernel.
Update our Makefile checks so that we pick up the correct files to build
once clang picks up support for -fpatchable-function-entry.
[*] https://github.com/llvm/llvm-project/issues/57031
Signed-off-by: Naveen N Rao <naveen@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241030070850.1361304-2-hbathini@linux.ibm.com
|
|
__pa() is only intended to be used for linear map addresses and using
it for initial_boot_params which is in fixmap for arm64 will give an
incorrect value. Hence save the physical address when it is known at
boot time when calling early_init_dt_scan for arm64 and use it at kexec
time instead of converting the virtual address using __pa().
Note that arm64 doesn't need the FDT region reserved in the DT as the
kernel explicitly reserves the passed in FDT. Therefore, only a debug
warning is fixed with this change.
Reported-by: Breno Leitao <leitao@debian.org>
Suggested-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Usama Arif <usamaarif642@gmail.com>
Fixes: ac10be5cdbfa ("arm64: Use common of_kexec_alloc_and_setup_fdt()")
Link: https://lore.kernel.org/r/20241023171426.452688-1-usamaarif642@gmail.com
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
|
|
The maple platform was added in 2004 [1], to support the "Maple" 970FX
evaluation board.
It was later used for IBM JS20/JS21 machines, as well as the Bimini
machine, aka "Yellow Dog Powerstation".
Sadly all those machines have passed into memory, and there's been no
evidence for years that anyone is still using any of them.
Remove the platform and related code. It can always be reinstated if
there's interest.
Note that this has no impact on support for 970FX based Power Macs.
[1]: https://git.kernel.org/pub/scm/linux/kernel/git/mpe/linux-fullhistory.git/commit/?id=f0d068d65c5e555ffcfbc189de32598f6f00770c
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241013102957.548291-1-mpe@ellerman.id.au
|
|
The help text refers to lilo, but the install script does not run lilo
and never has. The reference to lilo seems to have come originally from
arch/ppc/Makefile, but it was not true there either.
Remove it.
Reported-by: Thorsten Leemhuis <linux@leemhuis.info>
Link: https://fosstodon.org/@kernellogger/113032940928131612
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241009053806.135807-1-mpe@ellerman.id.au
|
|
The dtl_access_lock needs to be a rw_sempahore, a sleeping lock, because
the code calls kmalloc() while holding it, which can sleep:
# echo 1 > /proc/powerpc/vcpudispatch_stats
BUG: sleeping function called from invalid context at include/linux/sched/mm.h:337
in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 199, name: sh
preempt_count: 1, expected: 0
3 locks held by sh/199:
#0: c00000000a0743f8 (sb_writers#3){.+.+}-{0:0}, at: vfs_write+0x324/0x438
#1: c0000000028c7058 (dtl_enable_mutex){+.+.}-{3:3}, at: vcpudispatch_stats_write+0xd4/0x5f4
#2: c0000000028c70b8 (dtl_access_lock){+.+.}-{2:2}, at: vcpudispatch_stats_write+0x220/0x5f4
CPU: 0 PID: 199 Comm: sh Not tainted 6.10.0-rc4 #152
Hardware name: IBM pSeries (emulated by qemu) POWER9 (raw) 0x4e1202 0xf000005 of:SLOF,HEAD hv:linux,kvm pSeries
Call Trace:
dump_stack_lvl+0x130/0x148 (unreliable)
__might_resched+0x174/0x410
kmem_cache_alloc_noprof+0x340/0x3d0
alloc_dtl_buffers+0x124/0x1ac
vcpudispatch_stats_write+0x2a8/0x5f4
proc_reg_write+0xf4/0x150
vfs_write+0xfc/0x438
ksys_write+0x88/0x148
system_call_exception+0x1c4/0x5a0
system_call_common+0xf4/0x258
Fixes: 06220d78f24a ("powerpc/pseries: Introduce rwlock to gatekeep DTLB usage")
Tested-by: Kajol Jain <kjain@linux.ibm.com>
Reviewed-by: Nysal Jan K.A <nysal@linux.ibm.com>
Reviewed-by: Kajol Jain <kjain@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20240819122401.513203-1-mpe@ellerman.id.au
|
|
Drop the include of dma-mapping.h in machdep.h, replace it with forward
declarations of struct device and struct pci_dev, and include time64.h
and page.h which are required for time64_t and pgprot_t respectively.
Add direct includes of some other headers to some files that were
getting them via machdep.h.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241009051826.132805-2-mpe@ellerman.id.au
|
|
Drop the include of seq_file.h in machdep.h, replace it with a forward
declaration of struct seq_file, which is all that's required.
Add direct includes of seq_file.h to some files that were getting
seq_file.h via machdep.h.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241009051826.132805-1-mpe@ellerman.id.au
|
|
The last use of IPI_PRIORITY in asm was removed in commit 37f55d30df2e
("KVM: PPC: Book3S HV: Convert kvmppc_read_intr to a C function").
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241009051701.132282-1-mpe@ellerman.id.au
|
|
Back in the day a lot of logic was implemented inline in dma-mapping.h and
needed various includes. Move of this has long been moved out of line,
so we can drop various includes to improve kernel rebuild times.
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
page_to_phys is duplicated by all architectures, and from some strange
reason placed in <asm/io.h> where it doesn't fit at all.
phys_to_page is only provided by a few architectures despite having a lot
of open coded users.
Provide generic versions in <asm-generic/memory_model.h> to make these
helpers more easily usable.
Note with this patch powerpc loses the CONFIG_DEBUG_VIRTUAL pfn_valid
check. It will be added back in a generic version later.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
Explicitly require "struct page" memory when sharing memory between
guest and host via an Ultravisor. Given the number of pfn_to_page()
calls in the code, it's safe to assume that KVM already requires that the
pfn returned by gfn_to_pfn() is backed by struct page, i.e. this is
likely a bug fix, not a reduction in KVM capabilities.
Switching to gfn_to_page() will eventually allow removing gfn_to_pfn()
and kvm_pfn_to_refcounted_page().
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-79-seanjc@google.com>
|
|
Use kvm_vcpu_map() when patching dcbz in guest memory, as a regular GUP
isn't technically sufficient when writing to data in the target pages.
As per Documentation/core-api/pin_user_pages.rst:
Correct (uses FOLL_PIN calls):
pin_user_pages()
write to the data within the pages
unpin_user_pages()
INCORRECT (uses FOLL_GET calls):
get_user_pages()
write to the data within the pages
put_page()
As a happy bonus, using kvm_vcpu_{,un}map() takes care of creating a
mapping and marking the page dirty.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-75-seanjc@google.com>
|
|
Don't manually do get_page() when patching dcbz, as gfn_to_page() gifts
the caller a reference. I.e. doing get_page() will leak the page due to
not putting all references.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-74-seanjc@google.com>
|
|
Convert Book3S PR to __kvm_faultin_pfn()+kvm_release_faultin_page(), which
are new APIs to consolidate arch code and provide consistent behavior
across all KVM architectures.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-65-seanjc@google.com>
|
|
Mark pages/folios dirty/accessed after installing a PTE, and more
specifically after acquiring mmu_lock and checking for an mmu_notifier
invalidation. Marking a page/folio dirty after it has been written back
can make some filesystems unhappy (backing KVM guests will such filesystem
files is uncommon, and the race is minuscule, hence the lack of complaints).
See the link below for details.
This will also allow converting Book3S to kvm_release_faultin_page(),
which requires that mmu_lock be held (for the aforementioned reason).
Link: https://lore.kernel.org/all/cover.1683044162.git.lstoakes@gmail.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-64-seanjc@google.com>
|
|
Drop @kvm_ro from kvmppc_book3s_instantiate_page() as it is now only
written, and never read.
No functional change intended.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-63-seanjc@google.com>
|
|
Replace Book3s Radix's homebrewed (read: copy+pasted) fault-in logic with
__kvm_faultin_pfn(), which functionally does pretty much the exact same
thing.
Note, when the code was written, KVM indeed didn't do fast GUP without
"!atomic && !async", but that has long since changed (KVM tries fast GUP
for all writable mappings).
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-62-seanjc@google.com>
|
|
Replace Book3s HV's homebrewed fault-in logic with __kvm_faultin_pfn(),
which functionally does pretty much the exact same thing.
Note, when the code was written, KVM indeed didn't do fast GUP without
"!atomic && !async", but that has long since changed (KVM tries fast GUP
for all writable mappings).
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-61-seanjc@google.com>
|
|
Convert PPC e500 to use __kvm_faultin_pfn()+kvm_release_faultin_page(),
and continue the inexorable march towards the demise of
kvm_pfn_to_refcounted_page().
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-55-seanjc@google.com>
|
|
Mark pages accessed before dropping mmu_lock when faulting in guest memory
so that shadow_map() can convert to kvm_release_faultin_page() without
tripping its lockdep assertion on mmu_lock being held. Marking pages
accessed outside of mmu_lock is ok (not great, but safe), but marking
pages _dirty_ outside of mmu_lock can make filesystems unhappy.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-54-seanjc@google.com>
|
|
Mark the underlying page as dirty in kvmppc_e500_ref_setup()'s sole
caller, kvmppc_e500_shadow_map(), which will allow converting e500 to
__kvm_faultin_pfn() + kvm_release_faultin_page() without having to do
a weird dance between ref_setup() and shadow_map().
Opportunistically drop the redundant kvm_set_pfn_accessed(), as
shadow_map() puts the page via kvm_release_pfn_clean().
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-53-seanjc@google.com>
|
|
Drop @hva from __gfn_to_pfn_memslot() now that all callers pass NULL.
No functional change intended.
Tested-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-19-seanjc@google.com>
|
|
Drop @atomic from the myriad "to_pfn" APIs now that all callers pass
"false", and remove a comment blurb about KVM running only the "GUP fast"
part in atomic context.
No functional change intended.
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Tested-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-13-seanjc@google.com>
|
|
Remove KVM_ERR_PTR_BAD_PAGE and instead return NULL, as "bad page" is just
a leftover bit of weirdness from days of old when KVM stuffed a "bad" page
into the guest instead of actually handling missing pages. See commit
cea7bb21280e ("KVM: MMU: Make gfn_to_page() always safe").
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Tested-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20241010182427.1434605-2-seanjc@google.com>
|
|
Cross-merge networking fixes after downstream PR.
No conflicts and no adjacent changes.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
After fixing the HAVE_STACKPROTECTER checks for clang's in-progress
per-task stack protector support [1], the build fails during prepare0
because '-mstack-protector-guard-offset' has not been added to
KBUILD_CFLAGS yet but the other '-mstack-protector-guard' flags have.
clang: error: '-mstack-protector-guard=tls' is used without '-mstack-protector-guard-offset', and there is no default
clang: error: '-mstack-protector-guard=tls' is used without '-mstack-protector-guard-offset', and there is no default
make[4]: *** [scripts/Makefile.build:229: scripts/mod/empty.o] Error 1
make[4]: *** [scripts/Makefile.build:102: scripts/mod/devicetable-offsets.s] Error 1
Mirror other architectures and add all '-mstack-protector-guard' flags
to KBUILD_CFLAGS atomically during stack_protector_prepare, which
resolves the issue and allows clang's implementation to fully work with
the kernel.
Cc: stable@vger.kernel.org # 6.1+
Link: https://github.com/llvm/llvm-project/pull/110928 [1]
Reviewed-by: Keith Packard <keithp@keithp.com>
Tested-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241009-powerpc-fix-stackprotector-test-clang-v2-2-12fb86b31857@kernel.org
|
|
Clang's in-progress per-task stack protector support [1] does not work
with the current Kconfig checks because '-mstack-protector-guard-offset'
is not provided, unlike all other architecture Kconfig checks.
$ fd Kconfig -x rg -l mstack-protector-guard-offset
./arch/arm/Kconfig
./arch/riscv/Kconfig
./arch/arm64/Kconfig
This produces an error from clang, which is interpreted as the flags not
being supported at all when they really are.
$ clang --target=powerpc64-linux-gnu \
-mstack-protector-guard=tls \
-mstack-protector-guard-reg=r13 \
-c -o /dev/null -x c /dev/null
clang: error: '-mstack-protector-guard=tls' is used without '-mstack-protector-guard-offset', and there is no default
This argument will always be provided by the build system, so mirror
other architectures and use '-mstack-protector-guard-offset=0' for
testing support, which fixes the issue for clang and does not regress
support with GCC.
Even with the first problem addressed, the 32-bit test continues to fail
because Kbuild uses the powerpc64le-linux-gnu target for clang and
nothing flips the target to 32-bit, resulting in an error about an
invalid register valid:
$ clang --target=powerpc64le-linux-gnu \
-mstack-protector-guard=tls
-mstack-protector-guard-reg=r2 \
-mstack-protector-guard-offset=0 \
-x c -c -o /dev/null /dev/null
clang: error: invalid value 'r2' in 'mstack-protector-guard-reg=', expected one of: r13
While GCC allows arbitrary registers, the implementation of
'-mstack-protector-guard=tls' in LLVM shares the same code path as the
user space thread local storage implementation, which uses a fixed
register (2 for 32-bit and 13 for 62-bit), so the command line parsing
enforces this limitation.
Use the Kconfig macro '$(m32-flag)', which expands to '-m32' when
supported, in the stack protector support cc-option call to properly
switch the target to a 32-bit one, which matches what happens in Kbuild.
While the 64-bit macro does not strictly need it, add the equivalent
64-bit option for symmetry.
Cc: stable@vger.kernel.org # 6.1+
Link: https://github.com/llvm/llvm-project/pull/110928 [1]
Reviewed-by: Keith Packard <keithp@keithp.com>
Tested-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/20241009-powerpc-fix-stackprotector-test-clang-v2-1-12fb86b31857@kernel.org
|
|
Add hash_supports_debug_pagealloc() helper to detect whether
debug_pagealloc can be supported on hash or not. This checks for both,
whether debug_pagealloc config is enabled and the linear map should
fit within rma_size/4 region size.
This can then be used early during htab_init_page_sizes() to decide
linear map pagesize if hash supports either debug_pagealloc or
kfence.
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/c33c6691b2a2cf619cc74ac100118ca4dbf21a48.1729271995.git.ritesh.list@gmail.com
|
|
Enable kfence on book3s64 hash only when early init is enabled.
This is because, kfence could cause the kernel linear map to be mapped
at PAGE_SIZE level instead of 16M (which I guess we don't want).
Also currently there is no way to -
1. Make multiple page size entries for the SLB used for kernel linear
map.
2. No easy way of getting the hash slot details after the page table
mapping for kernel linear setup. So even if kfence allocate the
pool in late init, we won't be able to get the hash slot details in
kfence linear map.
Thus this patch disables kfence on hash if kfence early init is not
enabled.
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/4a6eea8cfd1cd28fccfae067026bff30cbec1d4b.1729271995.git.ritesh.list@gmail.com
|
|
Both radix and hash on book3s requires to detect if kfence
early init is enabled or not. Hash needs to disable kfence
if early init is not enabled because with kfence the linear map is
mapped using PAGE_SIZE rather than 16M mapping.
We don't support multiple page sizes for slb entry used for kernel
linear map in book3s64.
This patch refactors out the common functions required to detect kfence
early init is enabled or not.
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/f4a787224fbe5bb787158ace579780c0257f6602.1729271995.git.ritesh.list@gmail.com
|
|
Now that linear map functionality of debug_pagealloc is made generic,
enable kfence to use this generic infrastructure.
1. Define kfence related linear map variables.
- u8 *linear_map_kf_hash_slots;
- unsigned long linear_map_kf_hash_count;
- DEFINE_RAW_SPINLOCK(linear_map_kf_hash_lock);
2. The linear map size allocated in RMA region is quite small
(KFENCE_POOL_SIZE >> PAGE_SHIFT) which is 512 bytes by default.
3. kfence pool memory is reserved using memblock_phys_alloc() which has
can come from anywhere.
(default 255 objects => ((1+255) * 2) << PAGE_SHIFT = 32MB)
4. The hash slot information for kfence memory gets added in linear map
in hash_linear_map_add_slot() (which also adds for debug_pagealloc).
Reported-by: Pavithra Prakash <pavrampu@linux.vnet.ibm.com>
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/5c2b61941b344077a2b8654dab46efa0322af3af.1729271995.git.ritesh.list@gmail.com
|
|
Make size of the linear map to be allocated in RMA region to be of
ppc64_rma_size / 4. If debug_pagealloc requires more memory than that
then do not allocate any memory and disable debug_pagealloc.
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/e1ef66f32a1fe63bcbb89d5c11d86c65beef5ded.1729271995.git.ritesh.list@gmail.com
|
|
Currently kernel_map_linear_page() function assumes to be working on
linear_map_hash_slots array. But since in later patches we need a
separate linear map array for kfence, hence make
kernel_map_linear_page() take a linear map array and lock in it's
function argument.
This is needed to separate out kfence from debug_pagealloc
infrastructure.
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/5b67df7b29e68d7c78d6fc1f42d41137299bac6b.1729271995.git.ritesh.list@gmail.com
|
|
This refactors hash__kernel_map_pages() function to call
hash_debug_pagealloc_map_pages(). This will come useful when we will add
kfence support.
No functionality changes in this patch.
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/0cb8ddcccdcf61ea06ab4d92aacd770c16cc0f2c.1729271995.git.ritesh.list@gmail.com
|
|
This adds hash_debug_pagealloc_alloc_slots() function instead of open
coding that in htab_initialize(). This is required since we will be
separating the kfence functionality to not depend upon debug_pagealloc.
Now that everything required for debug_pagealloc is under a #ifdef
config. Bring in linear_map_hash_slots and linear_map_hash_count
variables under the same config too.
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://patch.msgid.link/d1d5aabe1e4c693a983e59ccf3de08e3c28c5161.1729271995.git.ritesh.list@gmail.com
|
