netdev - Re: [PATCH bpf-next 07/13] bpf, s390x: remove obsolete exception handling from div/mod

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] [day] [month] [year] [list]

Message-ID: <e2f2f242-0181-1aed-af07-1a64f3a838cb@iogearbox.net>
Date:   Mon, 29 Jan 2018 16:52:59 +0100
From:   Daniel Borkmann <daniel@...earbox.net>
To:     Michael Holzheu <holzheu@...ux.vnet.ibm.com>
Cc:     ast@...nel.org, netdev@...r.kernel.org
Subject: Re: [PATCH bpf-next 07/13] bpf, s390x: remove obsolete exception
 handling from div/mod

On 01/29/2018 03:33 PM, Michael Holzheu wrote:
> Am Fri, 26 Jan 2018 23:33:42 +0100
> schrieb Daniel Borkmann <daniel@...earbox.net>:
> 
>> Since we've changed div/mod exception handling for src_reg in
>> eBPF verifier itself, 
> 
> Maybe add the commit that introduced that to the patch description?

I couldn't add it here since it was all part of the same series and
thus didn't have a stable commit id for future reference yet. Maybe
I should have just put the commit subject in this case; will do next
time.

>> remove the leftovers from s390x JIT.
>>
>> Signed-off-by: Daniel Borkmann <daniel@...earbox.net>
>> Cc: Michael Holzheu <holzheu@...ux.vnet.ibm.com>
>> ---
>>  arch/s390/net/bpf_jit_comp.c | 10 ----------
>>  1 file changed, 10 deletions(-)
>>
>> diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
>> index e501887..78a19c9 100644
>> --- a/arch/s390/net/bpf_jit_comp.c
>> +++ b/arch/s390/net/bpf_jit_comp.c
>> @@ -610,11 +610,6 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i
>>  	{
>>  		int rc_reg = BPF_OP(insn->code) == BPF_DIV ? REG_W1 : REG_W0;
>>
>> -		jit->seen |= SEEN_RET0;
>> -		/* ltr %src,%src (if src == 0 goto fail) */
>> -		EMIT2(0x1200, src_reg, src_reg);
>> -		/* jz <ret0> */
>> -		EMIT4_PCREL(0xa7840000, jit->ret0_ip - jit->prg);
>>  		/* lhi %w0,0 */
>>  		EMIT4_IMM(0xa7080000, REG_W0, 0);
>>  		/* lr %w1,%dst */
>> @@ -630,11 +625,6 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i
>>  	{
>>  		int rc_reg = BPF_OP(insn->code) == BPF_DIV ? REG_W1 : REG_W0;
>>
>> -		jit->seen |= SEEN_RET0;
>> -		/* ltgr %src,%src (if src == 0 goto fail) */
>> -		EMIT4(0xb9020000, src_reg, src_reg);
>> -		/* jz <ret0> */
>> -		EMIT4_PCREL(0xa7840000, jit->ret0_ip - jit->prg);
>>  		/* lghi %w0,0 */
>>  		EMIT4_IMM(0xa7090000, REG_W0, 0);
>>  		/* lgr %w1,%dst */
> 
> If the check is done in the verifier now, this looks good to me.
> 
> Reviewed-by: Michael Holzheu <holzheu@...ux.vnet.ibm.com>

Thanks for the review, Michael!