Re: [RESEND PATCH bpf-next v2 1/4] bpf: Introduce global percpu data

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Subject: Re: [RESEND PATCH bpf-next v2 1/4] bpf: Introduce global percpu data
From: Alexei Starovoitov <alexei.starovoitov@xxxxxxxxx>
Date: Tue, 18 Feb 2025 17:47:05 -0800
Cc: bpf <bpf@xxxxxxxxxxxxxxx>, Alexei Starovoitov <ast@xxxxxxxxxx>, Daniel Borkmann <daniel@xxxxxxxxxxxxx>, Andrii Nakryiko <andrii@xxxxxxxxxx>, Yonghong Song <yonghong.song@xxxxxxxxx>, Song Liu <song@xxxxxxxxxx>, Eddy Z <eddyz87@xxxxxxxxx>, Quentin Monnet <qmo@xxxxxxxxxx>, Daniel Xu <dxu@xxxxxxxxx>, kernel-patches-bot@xxxxxx
In-reply-to: <20250213161931.46399-2-leon.hwang@linux.dev>

On Thu, Feb 13, 2025 at 8:19 AM Leon Hwang <leon.hwang@xxxxxxxxx> wrote:
>
>
> +#ifdef CONFIG_SMP

What happens on !SMP ?
I think the code assumes that the verifier will adjust ld_imm64
with percpu insn.
On !SMP ld_imm64 will be pointing where?

References:
- [RESEND PATCH bpf-next v2 0/4] bpf: Introduce global percpu data
  - From: Leon Hwang
- [RESEND PATCH bpf-next v2 1/4] bpf: Introduce global percpu data
  - From: Leon Hwang

Prev by Date: Re: [PATCH v3 bpf 0/2] bpf: skip non exist keys in generic_map_lookup_batch
Next by Date: Re: [PATCH bpf-next v2 1/3] tcp: add TCP_RTO_MAX_MIN_SEC definition
Previous by thread: [RESEND PATCH bpf-next v2 1/4] bpf: Introduce global percpu data
Next by thread: [PATCH v3 0/6] scripts/sorttable: ftrace: Remove place holders for weak functions in available_filter_functions
Index(es):
- Date
- Thread

[Index of Archives] [Linux Samsung SoC] [Linux Rockchip SoC] [Linux Actions SoC] [Linux for Synopsys ARC Processors] [Linux NFS] [Linux NILFS] [Linux USB Devel] [Video for Linux] [Linux Audio Users] [Yosemite News] [Linux Kernel] [Linux SCSI]

Powered by Linux