From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx0a-001b2d01.pphosted.com (mx0b-001b2d01.pphosted.com [148.163.158.5]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 3zcmDY1qDNzF17y for ; Fri, 9 Feb 2018 04:59:40 +1100 (AEDT) Received: from pps.filterd (m0098416.ppops.net [127.0.0.1]) by mx0b-001b2d01.pphosted.com (8.16.0.22/8.16.0.22) with SMTP id w18Hpih0039345 for ; Thu, 8 Feb 2018 12:59:38 -0500 Received: from e06smtp13.uk.ibm.com (e06smtp13.uk.ibm.com [195.75.94.109]) by mx0b-001b2d01.pphosted.com with ESMTP id 2g0sc8p747-1 (version=TLSv1.2 cipher=AES256-SHA bits=256 verify=NOT) for ; Thu, 08 Feb 2018 12:59:37 -0500 Received: from localhost by e06smtp13.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Thu, 8 Feb 2018 17:59:36 -0000 Date: Thu, 08 Feb 2018 23:29:30 +0530 From: "Naveen N. Rao" Subject: Re: [RFC][PATCH bpf 1/2] bpf: allow 64-bit offsets for bpf function calls To: Alexei Starovoitov , daniel@iogearbox.net, Sandipan Das Cc: linuxppc-dev@lists.ozlabs.org, mpe@ellerman.id.au, netdev@vger.kernel.org References: <20180208120306.2568-1-sandipan@linux.vnet.ibm.com> <4ce54c76-f8d5-e739-d9c2-e3318e398417@fb.com> In-Reply-To: <4ce54c76-f8d5-e739-d9c2-e3318e398417@fb.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Message-Id: <1518112079.b4po0pmm3v.naveen@linux.ibm.com> List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Alexei Starovoitov wrote: > On 2/8/18 4:03 AM, Sandipan Das wrote: >> The imm field of a bpf_insn is a signed 32-bit integer. For >> JIT-ed bpf-to-bpf function calls, it stores the offset from >> __bpf_call_base to the start of the callee function. >> >> For some architectures, such as powerpc64, it was found that >> this offset may be as large as 64 bits because of which this >> cannot be accomodated in the imm field without truncation. >> >> To resolve this, we additionally use the aux data within each >> bpf_prog associated with the caller functions to store the >> addresses of their respective callees. >> >> Signed-off-by: Sandipan Das >> --- >> kernel/bpf/verifier.c | 39 ++++++++++++++++++++++++++++++++++++++- >> 1 file changed, 38 insertions(+), 1 deletion(-) >> >> diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c >> index 5fb69a85d967..52088b4ca02f 100644 >> --- a/kernel/bpf/verifier.c >> +++ b/kernel/bpf/verifier.c >> @@ -5282,6 +5282,19 @@ static int jit_subprogs(struct bpf_verifier_env *= env) >> * run last pass of JIT >> */ >> for (i =3D 0; i <=3D env->subprog_cnt; i++) { >> + u32 flen =3D func[i]->len, callee_cnt =3D 0; >> + struct bpf_prog **callee; >> + >> + /* for now assume that the maximum number of bpf function >> + * calls that can be made by a caller must be at most the >> + * number of bpf instructions in that function >> + */ >> + callee =3D kzalloc(sizeof(func[i]) * flen, GFP_KERNEL); >> + if (!callee) { >> + err =3D -ENOMEM; >> + goto out_free; >> + } >> + >> insn =3D func[i]->insnsi; >> for (j =3D 0; j < func[i]->len; j++, insn++) { >> if (insn->code !=3D (BPF_JMP | BPF_CALL) || >> @@ -5292,6 +5305,26 @@ static int jit_subprogs(struct bpf_verifier_env *= env) >> insn->imm =3D (u64 (*)(u64, u64, u64, u64, u64)) >> func[subprog]->bpf_func - >> __bpf_call_base; >> + >> + /* the offset to the callee from __bpf_call_base >> + * may be larger than what the 32 bit integer imm >> + * can accomodate which will truncate the higher >> + * order bits >> + * >> + * to avoid this, we additionally utilize the aux >> + * data of each caller function for storing the >> + * addresses of every callee associated with it >> + */ >> + callee[callee_cnt++] =3D func[subprog]; >=20 > can you share typical /proc/kallsyms ? > Are you saying that kernel and kernel modules are allocated from > address spaces that are always more than 32-bit apart? Yes. On ppc64, kernel text is linearly mapped from 0xc000000000000000,=20 while vmalloc'ed area starts from 0xd000000000000000 (for radix, this is different, but still beyond a 32-bit offset). > That would mean that all kernel calls into modules are far calls > and the other way around form .ko into kernel? > Performance is probably suffering because every call needs to be built > with full 64-bit offset. No ? Possibly, and I think Michael can give a better perspective, but I think this is due to our ABI. For inter-module calls, we need to setup the TOC pointer (or the address of the function being called with ABIv2), which=20 would require us to load a full address regardless. - Naveen =