Re: [PATCH bpf-next v3 11/11] bpf: Fall back to nospec for sanitization-failures

2025-05-14 Thread Kumar Kartikeya Dwivedi
On Wed, 14 May 2025 at 13:30, Luis Gerhorst wrote: > > Kumar Kartikeya Dwivedi writes: > > (including relevant part from other message) > > > On Thu, 1 May 2025 at 04:00, Luis Gerhorst wrote: > > > >> +static bool error_recoverable_with_nospec(int err) > >> +{ > >> + /* Should only return

Re: [PATCH bpf-next v3 11/11] bpf: Fall back to nospec for sanitization-failures

2025-05-14 Thread Luis Gerhorst
Kumar Kartikeya Dwivedi writes: (including relevant part from other message) > On Thu, 1 May 2025 at 04:00, Luis Gerhorst wrote: > >> +static bool error_recoverable_with_nospec(int err) >> +{ >> + /* Should only return true for non-fatal errors that are allowed to >> +* occur dur

Re: [PATCH bpf-next v3 11/11] bpf: Fall back to nospec for sanitization-failures

2025-05-13 Thread Kumar Kartikeya Dwivedi
On Thu, 1 May 2025 at 04:10, Luis Gerhorst wrote: > > ALU sanitization was introduced to ensure that a subsequent ptr access > can never go OOB, even under speculation. This is required because we > currently allow speculative scalar confusion. Spec. scalar confusion is > possible because Spectre

[PATCH bpf-next v3 11/11] bpf: Fall back to nospec for sanitization-failures

2025-05-01 Thread Luis Gerhorst
ALU sanitization was introduced to ensure that a subsequent ptr access can never go OOB, even under speculation. This is required because we currently allow speculative scalar confusion. Spec. scalar confusion is possible because Spectre v4 sanitization only adds a nospec after critical stores (e.g