From: Andrii Nakryiko <andrii.nakryiko@gmail.com>
To: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: bpf@vger.kernel.org, daniel@iogearbox.net, andrii@kernel.org,
torvalds@linux-foundation.org, brho@google.com,
hannes@cmpxchg.org, akpm@linux-foundation.org, urezki@gmail.com,
hch@infradead.org, linux-mm@kvack.org, kernel-team@fb.com
Subject: Re: [PATCH v3 bpf-next 03/14] bpf: Add x86-64 JIT support for PROBE_MEM32 pseudo instructions.
Date: Mon, 11 Mar 2024 15:05:59 -0700 [thread overview]
Message-ID: <CAEf4BzYuou1A-3_1Q5qoecLsE42880MYvDCv6shOUVYZKHQb_Q@mail.gmail.com> (raw)
In-Reply-To: <20240308010812.89848-4-alexei.starovoitov@gmail.com>
On Thu, Mar 7, 2024 at 5:08 PM Alexei Starovoitov
<alexei.starovoitov@gmail.com> wrote:
>
> From: Alexei Starovoitov <ast@kernel.org>
>
> Add support for [LDX | STX | ST], PROBE_MEM32, [B | H | W | DW] instructions.
> They are similar to PROBE_MEM instructions with the following differences:
> - PROBE_MEM has to check that the address is in the kernel range with
> src_reg + insn->off >= TASK_SIZE_MAX + PAGE_SIZE check
> - PROBE_MEM doesn't support store
> - PROBE_MEM32 relies on the verifier to clear upper 32-bit in the register
> - PROBE_MEM32 adds 64-bit kern_vm_start address (which is stored in %r12 in the prologue)
> Due to bpf_arena constructions such %r12 + %reg + off16 access is guaranteed
> to be within arena virtual range, so no address check at run-time.
> - PROBE_MEM32 allows STX and ST. If they fault the store is a nop.
> When LDX faults the destination register is zeroed.
>
> Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
> ---
> arch/x86/net/bpf_jit_comp.c | 191 +++++++++++++++++++++++++++++++++++-
> include/linux/bpf.h | 1 +
> include/linux/filter.h | 3 +
> 3 files changed, 194 insertions(+), 1 deletion(-)
>
[...]
> +static u8 add_3mod(u8 byte, u32 r1, u32 r2, u32 index)
> +{
> + if (is_ereg(r1))
> + byte |= 1;
> + if (is_ereg(index))
> + byte |= 2;
> + if (is_ereg(r2))
> + byte |= 4;
> + return byte;
> +}
> +
> /* Encode 'dst_reg' register into x86-64 opcode 'byte' */
> static u8 add_1reg(u8 byte, u32 dst_reg)
> {
> @@ -645,6 +659,8 @@ static void emit_bpf_tail_call_indirect(struct bpf_prog *bpf_prog,
> pop_r12(&prog);
> } else {
> pop_callee_regs(&prog, callee_regs_used);
> + if (bpf_arena_get_kern_vm_start(bpf_prog->aux->arena))
ah, I guess this is where NULL is expected?.. But isn't `if
(bpf_prog->aux->arena)` equivalent and more straightforward check?
> + pop_r12(&prog);
> }
>
> EMIT1(0x58); /* pop rax */
> @@ -704,6 +720,8 @@ static void emit_bpf_tail_call_direct(struct bpf_prog *bpf_prog,
> pop_r12(&prog);
> } else {
> pop_callee_regs(&prog, callee_regs_used);
> + if (bpf_arena_get_kern_vm_start(bpf_prog->aux->arena))
> + pop_r12(&prog);
> }
>
> EMIT1(0x58); /* pop rax */
[...]
> @@ -1147,11 +1276,14 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image, u8 *rw_image
> bool tail_call_seen = false;
> bool seen_exit = false;
> u8 temp[BPF_MAX_INSN_SIZE + BPF_INSN_SAFETY];
> + u64 arena_vm_start;
> int i, excnt = 0;
> int ilen, proglen = 0;
> u8 *prog = temp;
> int err;
>
> + arena_vm_start = bpf_arena_get_kern_vm_start(bpf_prog->aux->arena);
and I'm guessing here you didn't want that check... I'd probably go
with explicit pointer checks, but ok, it's fine
> +
> detect_reg_usage(insn, insn_cnt, callee_regs_used,
> &tail_call_seen);
>
[...]
next prev parent reply other threads:[~2024-03-11 22:06 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-03-08 1:07 [PATCH v3 bpf-next 00/14] bpf: Introduce BPF arena Alexei Starovoitov
2024-03-08 1:07 ` [PATCH v3 bpf-next 01/14] bpf: Introduce bpf_arena Alexei Starovoitov
2024-03-11 22:01 ` Andrii Nakryiko
2024-03-11 22:41 ` Alexei Starovoitov
2024-03-11 22:59 ` Andrii Nakryiko
2024-03-12 0:47 ` Alexei Starovoitov
2024-03-08 1:08 ` [PATCH v3 bpf-next 02/14] bpf: Disasm support for addr_space_cast instruction Alexei Starovoitov
2024-03-08 1:08 ` [PATCH v3 bpf-next 03/14] bpf: Add x86-64 JIT support for PROBE_MEM32 pseudo instructions Alexei Starovoitov
2024-03-11 22:05 ` Andrii Nakryiko [this message]
2024-03-11 22:44 ` Alexei Starovoitov
2024-03-08 1:08 ` [PATCH v3 bpf-next 04/14] bpf: Add x86-64 JIT support for bpf_addr_space_cast instruction Alexei Starovoitov
2024-03-08 1:08 ` [PATCH v3 bpf-next 05/14] bpf: Recognize addr_space_cast instruction in the verifier Alexei Starovoitov
2024-03-08 1:08 ` [PATCH v3 bpf-next 06/14] bpf: Recognize btf_decl_tag("arg:arena") as PTR_TO_ARENA Alexei Starovoitov
2024-03-08 1:08 ` [PATCH v3 bpf-next 07/14] libbpf: Add __arg_arena to bpf_helpers.h Alexei Starovoitov
2024-03-08 1:08 ` [PATCH v3 bpf-next 08/14] libbpf: Add support for bpf_arena Alexei Starovoitov
2024-03-08 1:08 ` [PATCH v3 bpf-next 09/14] bpftool: Recognize arena map type Alexei Starovoitov
2024-03-11 17:08 ` Quentin Monnet
2024-03-08 1:08 ` [PATCH v3 bpf-next 10/14] libbpf: Recognize __arena global varaibles Alexei Starovoitov
2024-03-11 17:09 ` Quentin Monnet
2024-03-08 1:08 ` [PATCH v3 bpf-next 11/14] bpf: Add helper macro bpf_addr_space_cast() Alexei Starovoitov
2024-03-08 1:08 ` [PATCH v3 bpf-next 12/14] selftests/bpf: Add unit tests for bpf_arena_alloc/free_pages Alexei Starovoitov
2024-03-08 1:08 ` [PATCH v3 bpf-next 13/14] selftests/bpf: Add bpf_arena_list test Alexei Starovoitov
2024-03-08 1:08 ` [PATCH v3 bpf-next 14/14] selftests/bpf: Add bpf_arena_htab test Alexei Starovoitov
2024-03-11 22:45 ` [PATCH v3 bpf-next 00/14] bpf: Introduce BPF arena Andrii Nakryiko
2024-03-11 23:02 ` Alexei Starovoitov
2024-03-11 22:50 ` patchwork-bot+netdevbpf
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAEf4BzYuou1A-3_1Q5qoecLsE42880MYvDCv6shOUVYZKHQb_Q@mail.gmail.com \
--to=andrii.nakryiko@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=alexei.starovoitov@gmail.com \
--cc=andrii@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=brho@google.com \
--cc=daniel@iogearbox.net \
--cc=hannes@cmpxchg.org \
--cc=hch@infradead.org \
--cc=kernel-team@fb.com \
--cc=linux-mm@kvack.org \
--cc=torvalds@linux-foundation.org \
--cc=urezki@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).