Re: [PATCH v2] arm64/sve: Lower the maximum allocation for the SVE ptrace regset

[Doug Anderson]

Hi,

On Tue, Feb 13, 2024 at 10:24 AM Mark Brown <broonie@xxxxxxxxxx> wrote:
>
> Doug Anderson observed that ChromeOS crashes are being reported which
> include failing allocations of order 7 during core dumps due to ptrace
> allocating storage for regsets:
>
>   chrome: page allocation failure: order:7,
>           mode:0x40dc0(GFP_KERNEL|__GFP_COMP|__GFP_ZERO),
>           nodemask=(null),cpuset=urgent,mems_allowed=0
>    ...
>   regset_get_alloc+0x1c/0x28
>   elf_core_dump+0x3d8/0xd8c
>   do_coredump+0xeb8/0x1378
>
> with further investigation showing that this is:
>
>    [   66.957385] DOUG: Allocating 279584 bytes
>
> which is the maximum size of the SVE regset. As Doug observes it is not
> entirely surprising that such a large allocation of contiguous memory might
> fail on a long running system.
>
> The SVE regset is currently sized to hold SVE registers with a VQ of
> SVE_VQ_MAX which is 512, substantially more than the architectural maximum
> of 16 which we might see even in a system emulating the limits of the
> architecture. Since we don't expose the size we tell the regset core
> externally let's define ARCH_SVE_VQ_MAX with the actual architectural
> maximum and use that for the regset, we'll still overallocate most of the
> time but much less so which will be helpful even if the core is fixed to
> not require contiguous allocations.
>
> Specify ARCH_SVE_VQ_MAX in terms of the maximum value that can be written
> into ZCR_ELx.LEN (where this is set in the hardware). For consistency
> update the maximum SME vector length to be specified in the same style
> while we are at it.
>
> We could also teach the ptrace core about runtime discoverable regset sizes
> but that would be a more invasive change and this is being observed in
> practical systems.
>
> Reported-by: Doug Anderson <dianders@xxxxxxxxxxxx>
> Signed-off-by: Mark Brown <broonie@xxxxxxxxxx>
> ---
> We should probably also use the actual architectural limit for the
> bitmasks we use in the VL enumeration code, though that's both a little
> bit more involved and less immediately a problem.
> ---
> Changes in v2:
> - Specify the value using the size of the bitfield it goes into.
> - Link to v1: https://lore.kernel.org/r/20240203-arm64-sve-ptrace-regset-size-v1-1-2c3ba1386b9e@xxxxxxxxxx
> ---
>  arch/arm64/include/asm/fpsimd.h | 12 ++++++------
>  arch/arm64/kernel/ptrace.c      |  3 ++-
>  2 files changed, 8 insertions(+), 7 deletions(-)

FWIW, v2 still works for me from a black box point of view. :-P I ran
the same test I did from v1. Repeating here:

Confirmed that when I send a "quit" signal to Chrome now that the
allocation I see for "core_note_type" NT_ARM_SVE goes down from
279,584 bytes (n=17474) to just 8,768 bytes (n=548).

Tested-by: Douglas Anderson <dianders@xxxxxxxxxxxx>