netdev - Re: [PATCH bpf 2/2] bpf, mips: limit to 33 tail calls

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 linux-cve-announce 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:   Wed, 18 Dec 2019 12:32:53 +0300
From:   Alexander Lobakin <alobakin@...nk.ru>
To:     Paul Burton <paulburton@...nel.org>
Cc:     Paul Chaignon <paul.chaignon@...nge.com>,
        Björn Töpel 
        <bjorn.topel@...il.com>, Mahshid Khezri <khezri.mahshid@...il.com>,
        paul.chaignon@...il.com, bpf@...r.kernel.org,
        netdev@...r.kernel.org, Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Martin KaFai Lau <kafai@...com>,
        Song Liu <songliubraving@...com>, Yonghong Song <yhs@...com>,
        Andrii Nakryiko <andriin@...com>
Subject: Re: [PATCH bpf 2/2] bpf, mips: limit to 33 tail calls

Paul Burton wrote 11.12.2019 02:23:
> Hi Paul,
> 
> On Mon, Dec 09, 2019 at 07:52:52PM +0100, Paul Chaignon wrote:
>> All BPF JIT compilers except RISC-V's and MIPS' enforce a 33-tail 
>> calls
>> limit at runtime.  In addition, a test was recently added, in 
>> tailcalls2,
>> to check this limit.
>> 
>> This patch updates the tail call limit in MIPS' JIT compiler to allow
>> 33 tail calls.

Hi Paul,

You've restored MIPS cBPF in mips-fixes tree, doesn't it require any
changes to limit tail calls to 33? This series includes only eBPF as
there was no MIPS cBPF at the moment of writing.

>> Fixes: b6bd53f9c4e8 ("MIPS: Add missing file for eBPF JIT.")
>> Reported-by: Mahshid Khezri <khezri.mahshid@...il.com>
>> Signed-off-by: Paul Chaignon <paul.chaignon@...nge.com>
> 
> I'd be happy to take this through mips-fixes, but equally happy for it
> to go through the BPF/net trees in which case:
> 
>   Acked-by: Paul Burton <paulburton@...nel.org>
> 
> Thanks,
>     Paul
> 
>> ---
>>  arch/mips/net/ebpf_jit.c | 9 +++++----
>>  1 file changed, 5 insertions(+), 4 deletions(-)
>> 
>> diff --git a/arch/mips/net/ebpf_jit.c b/arch/mips/net/ebpf_jit.c
>> index 46b76751f3a5..3ec69d9cbe88 100644
>> --- a/arch/mips/net/ebpf_jit.c
>> +++ b/arch/mips/net/ebpf_jit.c
>> @@ -604,6 +604,7 @@ static void emit_const_to_reg(struct jit_ctx *ctx, 
>> int dst, u64 value)
>>  static int emit_bpf_tail_call(struct jit_ctx *ctx, int this_idx)
>>  {
>>  	int off, b_off;
>> +	int tcc_reg;
>> 
>>  	ctx->flags |= EBPF_SEEN_TC;
>>  	/*
>> @@ -616,14 +617,14 @@ static int emit_bpf_tail_call(struct jit_ctx 
>> *ctx, int this_idx)
>>  	b_off = b_imm(this_idx + 1, ctx);
>>  	emit_instr(ctx, bne, MIPS_R_AT, MIPS_R_ZERO, b_off);
>>  	/*
>> -	 * if (--TCC < 0)
>> +	 * if (TCC-- < 0)
>>  	 *     goto out;
>>  	 */
>>  	/* Delay slot */
>> -	emit_instr(ctx, daddiu, MIPS_R_T5,
>> -		   (ctx->flags & EBPF_TCC_IN_V1) ? MIPS_R_V1 : MIPS_R_S4, -1);
>> +	tcc_reg = (ctx->flags & EBPF_TCC_IN_V1) ? MIPS_R_V1 : MIPS_R_S4;
>> +	emit_instr(ctx, daddiu, MIPS_R_T5, tcc_reg, -1);
>>  	b_off = b_imm(this_idx + 1, ctx);
>> -	emit_instr(ctx, bltz, MIPS_R_T5, b_off);
>> +	emit_instr(ctx, bltz, tcc_reg, b_off);
>>  	/*
>>  	 * prog = array->ptrs[index];
>>  	 * if (prog == NULL)
>> --
>> 2.17.1
>> 

Regards,
ᚷ ᛖ ᚢ ᚦ ᚠ ᚱ