On Fri, 2023-05-19 at 15:14 +1000, Michael Ellerman wrote:
> Andrew Morton writes:
> > On Fri, 5 May 2023 16:02:17 + David Laight
> > wrote:
> >
> > > From: Michael Ellerman
> > > > Sent: 05 May 2023 04:51
> > > >
> > > > Since commit 1ba3cbf3ec3b ("mm: kfence: improve the performance
> > >
Le 18/05/2023 à 00:20, Andrew Morton a écrit :
> On Fri, 5 May 2023 16:02:17 + David Laight
> wrote:
>
>> From: Michael Ellerman
>>> Sent: 05 May 2023 04:51
>>>
>>> Since commit 1ba3cbf3ec3b ("mm: kfence: improve the performance of
>>> __kfence_alloc() and __kfence_free()"), kfence reports
The functions here use struct thread_struct fields, so need to import
the full definition from . The header
that defines current only forward declares struct thread_struct.
Failing to include this header leads to a compilation
error when a translation unit does not also include
indirectly.
Sig
Andrew Morton writes:
> On Fri, 5 May 2023 16:02:17 + David Laight
> wrote:
>
>> From: Michael Ellerman
>> > Sent: 05 May 2023 04:51
>> >
>> > Since commit 1ba3cbf3ec3b ("mm: kfence: improve the performance of
>> > __kfence_alloc() and __kfence_free()"), kfence reports failures in
>> > rand
Test the kernel DEXCR[NPHIE] interface and hashchk exception handling.
Introduces with it a DEXCR utils library for common DEXCR operations.
Volatile is used to prevent the compiler optimising away the signal
tests.
Signed-off-by: Benjamin Gray
---
v1: * Clean up dexcr makefile
* I
Make the DEXCR value configurable at config time. Intentionally don't
limit possible values to support future aspects without needing kernel
updates.
The default config value enables hashst/hashchk in problem state.
This should be safe, as generally software needs to request these
instructions be
The HASHKEYR register contains a secret per-process key to enable unique
hashes per process. In general it should not be exposed to userspace
at all and a regular process has no need to know its key.
However, checkpoint restore in userspace (CRIU) functionality requires
that a process be able to s
ptrace-decl.h uses user_regset_get2_fn (among other things) from
regset.h. While all current users of ptrace-decl.h include regset.h
before it anyway, it adds an implicit ordering dependency and breaks
source tooling that tries to inspect ptrace-decl.h by itself.
Signed-off-by: Benjamin Gray
Revi
The DEXCR register is of interest when ptracing processes. Currently it
is static, but eventually will be dynamically controllable by a process.
If a process can control its own, then it is useful for it to be
ptrace-able to (e.g., for checkpoint-restore functionality).
It is also relevant to core
Adds _MSG assertion variants to provide more context behind why a
failure occurred. Also include unistd.h for _exit() and stdio.h for
fprintf(), and move ARRAY_SIZE macro to utils.h.
The _MSG variants and ARRAY_SIZE will be used by the following
DEXCR selftests.
Signed-off-by: Benjamin Gray
Revi
Add a utility 'lsdexcr' to print the current DEXCR status. Useful for
quickly checking the status such as when debugging test failures or
verifying the new default DEXCR does what you want (for userspace at
least). Example output:
# ./lsdexcr
uDEXCR: 0400 (NPHIE)
HDEXCR:
The ISA 3.1B hashst and hashchk instructions use a per-cpu SPR HASHKEYR
to hold a key used in the hash calculation. This key should be different
for each process to make it harder for a malicious process to recreate
valid hash values for a victim process.
Add support for storing a per-thread hash
Describe the DEXCR and document how to configure it.
Signed-off-by: Benjamin Gray
Reviewed-by: Russell Currey
---
v3: * Add ruscur reviewed-by
v2: * Document coredump & ptrace support
v1: * Remove the dynamic control docs, describe the static config
option
This documenta
The DEXCR is a SPR that allows control over various execution
'aspects', such as indirect branch prediction and enabling the
hashst/hashchk instructions. Further details are in ISA 3.1B
Book 3 chapter 9.
This series adds static (compile time) support for initialising
the DEXCR, and basic HASHKEYR
Recognise and pass the appropriate signal to the user program when a
hashchk instruction triggers. This is independent of allowing
configuration of DEXCR[NPHIE], as a hypervisor can enforce this aspect
regardless of the kernel.
The signal mirrors how ARM reports their similar check failure. For
ex
ISA 3.1B introduces the Dynamic Execution Control Register (DEXCR). It
is a per-cpu register that allows control over various CPU behaviours
including branch hint usage, indirect branch speculation, and
hashst/hashchk support.
Add some definitions and basic support for the DEXCR in the kernel.
Rig
On Tue, May 09, 2023 at 09:18:35AM +0200, Pali Rohár wrote:
On Tuesday 09 May 2023 17:14:48 Michael Ellerman wrote:
Randy Dunlap writes:
> Hi--
>
> Just a heads up. This patch can cause build errors.
> I sent a patch for these on 2023-APR-28:
>
https://lore.kernel.org/linuxppc-dev/2023042904
On Mon, May 01, 2023 at 12:27:55PM -0700, Vishal Moola (Oracle) wrote:
> The MM subsystem is trying to shrink struct page. This patchset
> introduces a memory descriptor for page table tracking - struct ptdesc.
>
> This patchset introduces ptdesc, splits ptdesc from struct page, and
> converts man
On 16/05/2023 01:00, Jason Gunthorpe wrote:
> This is on github:
> https://github.com/jgunthorpe/linux/commits/iommu_all_defdom
Tested-by: Steven Price
Works fine on my Firefly-RK3288.
Thanks,
Steve
From: Anshuman Khandual
The entire scheme of deferred TLB flush in reclaim path rests on the
fact that the cost to refill TLB entries is less than flushing out
individual entries by sending IPI to remote CPUs. But architecture
can have different ways to evaluate that. Hence apart from checking
TT
From: Yicong Yang
Though ARM64 has the hardware to do tlb shootdown, the hardware
broadcasting is not free.
A simplest micro benchmark shows even on snapdragon 888 with only
8 cores, the overhead for ptep_clear_flush is huge even for paging
out one page mapped by only one process:
5.36% a.out
From: Barry Song
on x86, batched and deferred tlb shootdown has lead to 90%
performance increase on tlb shootdown. on arm64, HW can do
tlb shootdown without software IPI. But sync tlbi is still
quite expensive.
Even running a simplest program which requires swapout can
prove this is true,
#incl
22 matches
Mail list logo
