netdev - Re: [PATCH bpf-next v7 4/4] bpf, arm64: bpf trampoline for arm64

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 11 Jul 2022 22:16:00 +0800
From:   Xu Kuohai <xukuohai@...wei.com>
To:     Jean-Philippe Brucker <jean-philippe@...aro.org>
CC:     <bpf@...r.kernel.org>, <linux-arm-kernel@...ts.infradead.org>,
        <linux-kernel@...r.kernel.org>, <netdev@...r.kernel.org>,
        Will Deacon <will@...nel.org>, KP Singh <kpsingh@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        Catalin Marinas <catalin.marinas@....com>,
        Daniel Borkmann <daniel@...earbox.net>,
        Alexei Starovoitov <ast@...nel.org>,
        Andrii Nakryiko <andrii@...nel.org>,
        Zi Shen Lim <zlim.lnx@...il.com>,
        Martin KaFai Lau <kafai@...com>,
        Song Liu <songliubraving@...com>, Yonghong Song <yhs@...com>,
        John Fastabend <john.fastabend@...il.com>,
        "David S . Miller" <davem@...emloft.net>,
        Hideaki YOSHIFUJI <yoshfuji@...ux-ipv6.org>,
        David Ahern <dsahern@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        Dave Hansen <dave.hansen@...ux.intel.com>, <x86@...nel.org>,
        "H . Peter Anvin" <hpa@...or.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Jesper Dangaard Brouer <hawk@...nel.org>,
        Russell King <rmk+kernel@...linux.org.uk>,
        James Morse <james.morse@....com>,
        Hou Tao <houtao1@...wei.com>,
        Jason Wang <wangborong@...rlc.com>
Subject: Re: [PATCH bpf-next v7 4/4] bpf, arm64: bpf trampoline for arm64

On 7/11/2022 7:57 PM, Jean-Philippe Brucker wrote:
> On Fri, Jul 08, 2022 at 05:30:32AM -0400, Xu Kuohai wrote:
>> +static void invoke_bpf_prog(struct jit_ctx *ctx, struct bpf_tramp_link *l,
>> +			    int args_off, int retval_off, int run_ctx_off,
>> +			    bool save_ret)
>> +{
>> +	u32 *branch;
>> +	u64 enter_prog;
>> +	u64 exit_prog;
>> +	u8 r0 = bpf2a64[BPF_REG_0];
>> +	struct bpf_prog *p = l->link.prog;
>> +	int cookie_off = offsetof(struct bpf_tramp_run_ctx, bpf_cookie);
>> +
>> +	if (p->aux->sleepable) {
>> +		enter_prog = (u64)__bpf_prog_enter_sleepable;
>> +		exit_prog = (u64)__bpf_prog_exit_sleepable;
>> +	} else {
>> +		enter_prog = (u64)__bpf_prog_enter;
>> +		exit_prog = (u64)__bpf_prog_exit;
>> +	}
>> +
>> +	if (l->cookie == 0) {
>> +		/* if cookie is zero, one instruction is enough to store it */
>> +		emit(A64_STR64I(A64_ZR, A64_SP, run_ctx_off + cookie_off), ctx);
>> +	} else {
>> +		emit_a64_mov_i64(A64_R(10), l->cookie, ctx);
>> +		emit(A64_STR64I(A64_R(10), A64_SP, run_ctx_off + cookie_off),
>> +		     ctx);
>> +	}
>> +
>> +	/* save p to callee saved register x19 to avoid loading p with mov_i64
>> +	 * each time.
>> +	 */
>> +	emit_addr_mov_i64(A64_R(19), (const u64)p, ctx);
>> +
>> +	/* arg1: prog */
>> +	emit(A64_MOV(1, A64_R(0), A64_R(19)), ctx);
>> +	/* arg2: &run_ctx */
>> +	emit(A64_ADD_I(1, A64_R(1), A64_SP, run_ctx_off), ctx);
>> +
>> +	emit_call(enter_prog, ctx);
>> +
>> +	/* if (__bpf_prog_enter(prog) == 0)
>> +	 *         goto skip_exec_of_prog;
>> +	 */
>> +	branch = ctx->image + ctx->idx;
>> +	emit(A64_NOP, ctx);
>> +
>> +	/* save return value to callee saved register x20 */
>> +	emit(A64_MOV(1, A64_R(20), A64_R(0)), ctx);
>> +
>> +	emit(A64_ADD_I(1, A64_R(0), A64_SP, args_off), ctx);
>> +	if (!p->jited)
>> +		emit_addr_mov_i64(A64_R(1), (const u64)p->insnsi, ctx);
>> +
>> +	emit_call((const u64)p->bpf_func, ctx);
>> +
>> +	/* store return value, which is held in r0 for JIT and in x0
>> +	 * for interpreter.
>> +	 */
>> +	if (save_ret)
>> +		emit(A64_STR64I(p->jited ? r0 : A64_R(0), A64_SP, retval_off),
>> +		     ctx);
> 
> This should be only A64_R(0), not r0. r0 happens to equal A64_R(0) when
> jitted due to the way build_epilogue() builds the function at the moment,
> but we shouldn't rely on that.
> 

looks like I misunderstood something, will change it to:

/* store return value, which is held in x0 for interpreter and in
 * bpf register r0 for JIT, but r0 happens to equal x0 due to the
 * way build_epilogue() builds the JIT image.
 */
if (save_ret)
        emit(A64_STR64I(A64_R(0), A64_SP, retval_off), ctx);

> Apart from that, for the series
> 
> Reviewed-by: Jean-Philippe Brucker <jean-philippe@...aro.org>
> 
> .