linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: Christophe Leroy <christophe.leroy@csgroup.eu>
To: Hari Bathini <hbathini@linux.ibm.com>,
	"bpf@vger.kernel.org" <bpf@vger.kernel.org>,
	linuxppc-dev <linuxppc-dev@lists.ozlabs.org>
Cc: Song Liu <songliubraving@fb.com>,
	Daniel Borkmann <daniel@iogearbox.net>,
	John Fastabend <john.fastabend@gmail.com>,
	Alexei Starovoitov <ast@kernel.org>,
	Andrii Nakryiko <andrii@kernel.org>,
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	Paul Mackerras <paulus@samba.org>,
	"Naveen N. Rao" <naveen.n.rao@linux.ibm.com>,
	Yonghong Song <yhs@fb.com>, KP Singh <kpsingh@kernel.org>,
	Jordan Niethe <jniethe5@gmail.com>,
	Martin KaFai Lau <kafai@fb.com>
Subject: Re: [PATCH v2 4/5] bpf ppc32: add support for BPF_ATOMIC bitwise operations
Date: Sat, 11 Jun 2022 17:14:28 +0000	[thread overview]
Message-ID: <0f41cc76-a214-03c5-8764-808e5001b906@csgroup.eu> (raw)
In-Reply-To: <20220610155552.25892-5-hbathini@linux.ibm.com>



Le 10/06/2022 à 17:55, Hari Bathini a écrit :
> Adding instructions for ppc32 for
> 
> atomic_and
> atomic_or
> atomic_xor
> atomic_fetch_add
> atomic_fetch_and
> atomic_fetch_or
> atomic_fetch_xor
> 
> Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
> ---
> 
> Changes in v2:
> * Used an additional register (BPF_REG_AX)
>      - to avoid clobbering src_reg.
>      - to keep the lwarx reservation as intended.
>      - to avoid the odd switch/goto construct.

Might be a stupid question as I don't know the internals of BPF: Are we 
sure BPF_REG_AX cannot be the src reg or the dst reg ?


> * Zero'ed out the higher 32-bit explicitly when required.
> 
>   arch/powerpc/net/bpf_jit_comp32.c | 53 ++++++++++++++++++++++++-------
>   1 file changed, 41 insertions(+), 12 deletions(-)
> 
> diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c
> index e46ed1e8c6ca..28dc6a1a8f2f 100644
> --- a/arch/powerpc/net/bpf_jit_comp32.c
> +++ b/arch/powerpc/net/bpf_jit_comp32.c
> @@ -294,6 +294,7 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
>   		u32 dst_reg_h = dst_reg - 1;
>   		u32 src_reg = bpf_to_ppc(insn[i].src_reg);
>   		u32 src_reg_h = src_reg - 1;
> +		u32 ax_reg = bpf_to_ppc(BPF_REG_AX);
>   		u32 tmp_reg = bpf_to_ppc(TMP_REG);
>   		u32 size = BPF_SIZE(code);
>   		s16 off = insn[i].off;
> @@ -798,25 +799,53 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *
>   		 * BPF_STX ATOMIC (atomic ops)
>   		 */
>   		case BPF_STX | BPF_ATOMIC | BPF_W:
> -			if (imm != BPF_ADD) {
> -				pr_err_ratelimited("eBPF filter atomic op code %02x (@%d) unsupported\n",
> -						   code, i);
> -				return -ENOTSUPP;
> -			}
> -
> -			/* *(u32 *)(dst + off) += src */
> -
>   			bpf_set_seen_register(ctx, tmp_reg);
> +			bpf_set_seen_register(ctx, ax_reg);
> +
>   			/* Get offset into TMP_REG */
>   			EMIT(PPC_RAW_LI(tmp_reg, off));
> +			tmp_idx = ctx->idx * 4;
>   			/* load value from memory into r0 */
>   			EMIT(PPC_RAW_LWARX(_R0, tmp_reg, dst_reg, 0));
> -			/* add value from src_reg into this */
> -			EMIT(PPC_RAW_ADD(_R0, _R0, src_reg));
> -			/* store result back */
> +
> +			/* Save old value in BPF_REG_AX */
> +			if (imm & BPF_FETCH)
> +				EMIT(PPC_RAW_MR(ax_reg, _R0));
> +
> +			switch (imm) {
> +			case BPF_ADD:
> +			case BPF_ADD | BPF_FETCH:
> +				EMIT(PPC_RAW_ADD(_R0, _R0, src_reg));
> +				break;
> +			case BPF_AND:
> +			case BPF_AND | BPF_FETCH:
> +				EMIT(PPC_RAW_AND(_R0, _R0, src_reg));
> +				break;
> +			case BPF_OR:
> +			case BPF_OR | BPF_FETCH:
> +				EMIT(PPC_RAW_OR(_R0, _R0, src_reg));
> +				break;
> +			case BPF_XOR:
> +			case BPF_XOR | BPF_FETCH:
> +				EMIT(PPC_RAW_XOR(_R0, _R0, src_reg));
> +				break;
> +			default:
> +				pr_err_ratelimited("eBPF filter atomic op code %02x (@%d) unsupported\n",
> +						   code, i);
> +				return -EOPNOTSUPP;
> +			}
> +
> +			/* store new value */
>   			EMIT(PPC_RAW_STWCX(_R0, tmp_reg, dst_reg));
>   			/* we're done if this succeeded */
> -			PPC_BCC_SHORT(COND_NE, (ctx->idx - 3) * 4);
> +			PPC_BCC_SHORT(COND_NE, tmp_idx);
> +
> +			/* For the BPF_FETCH variant, get old data into src_reg */
> +			if (imm & BPF_FETCH) {
> +				EMIT(PPC_RAW_MR(src_reg, ax_reg));
> +				if (!fp->aux->verifier_zext)
> +					EMIT(PPC_RAW_LI(src_reg_h, 0));
> +			}
>   			break;
>   
>   		case BPF_STX | BPF_ATOMIC | BPF_DW: /* *(u64 *)(dst + off) += src */

  reply	other threads:[~2022-06-11 17:15 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-10 15:55 [PATCH v2 0/5] Atomics support for eBPF on powerpc Hari Bathini
2022-06-10 15:55 ` [PATCH v2 1/5] bpf ppc64: add support for BPF_ATOMIC bitwise operations Hari Bathini
2022-06-10 15:55 ` [PATCH v2 2/5] bpf ppc64: add support for atomic fetch operations Hari Bathini
2022-06-10 15:55 ` [PATCH v2 3/5] bpf ppc64: Add instructions for atomic_[cmp]xchg Hari Bathini
2022-06-10 15:55 ` [PATCH v2 4/5] bpf ppc32: add support for BPF_ATOMIC bitwise operations Hari Bathini
2022-06-11 17:14   ` Christophe Leroy [this message]
2022-06-13 19:00     ` Hari Bathini
2022-06-10 15:55 ` [PATCH v2 5/5] bpf ppc32: Add instructions for atomic_[cmp]xchg Hari Bathini
2022-06-11 17:34   ` Christophe Leroy
2022-06-13 19:11     ` Hari Bathini
2022-06-13 19:14       ` Hari Bathini
2022-06-24 10:41         ` Naveen N. Rao
2022-06-24 10:37 ` [PATCH v2 0/5] Atomics support for eBPF on powerpc Naveen N. Rao
2022-07-04 11:33 ` Michael Ellerman

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=0f41cc76-a214-03c5-8764-808e5001b906@csgroup.eu \
    --to=christophe.leroy@csgroup.eu \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=hbathini@linux.ibm.com \
    --cc=jniethe5@gmail.com \
    --cc=john.fastabend@gmail.com \
    --cc=kafai@fb.com \
    --cc=kpsingh@kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=naveen.n.rao@linux.ibm.com \
    --cc=netdev@vger.kernel.org \
    --cc=paulus@samba.org \
    --cc=songliubraving@fb.com \
    --cc=yhs@fb.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).