netdev - Re: [PATCH net v4 1/2] net: sched: tbf: fix calculation of max

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]

Date:	Wed, 4 Dec 2013 09:53:02 +0800
From:	Yang Yingliang <yangyingliang@...wei.com>
To:	Eric Dumazet <eric.dumazet@...il.com>
CC:	<davem@...emloft.net>, <netdev@...r.kernel.org>,
	<brouer@...hat.com>, <jpirko@...hat.com>, <jbrouer@...hat.com>
Subject: Re: [PATCH net v4 1/2] net: sched: tbf: fix calculation of max_size

On 2013/12/3 22:51, Eric Dumazet wrote:
> On Tue, 2013-12-03 at 19:32 +0800, Yang Yingliang wrote:
> 
>>>
>>>>
>>>> TSO packet of 64KB -> about 45 frames if MSS=1448, 45*1514 = 68130 bytes
>>
>> Maybe MAX_PKT_LEN should be much bigger. Hmm, I'm uncertain how big is the proper value.
>>
> 
> Thats the thing : When user is able to compute the appropriate burst and
> give precise instructions to the kernel, why tbf would reduce it to
> whatever fixed threshold ?
> 
> If I set 200000 as a burst, I do not want tbf use 65536 or even less.
> 
> If tbf rounds my 200000 to 199800, its fine.
> 
> You had problems to set burst to appropriate values, but most other
> users did that fine.
> 
> tbf_segment() is an attempt to help for very low rates, to not overcome
> the small bursts (as low as 2000 bytes) programmed on atbf qdisc, but
> for high rates, we _definitely_ want to avoid segmentation as hell.
> 

Thanks for your opinions!
>From your opinions, how about calculating burst directly with q->buffer and psched_ns_t2l().

I did it in my v1 patch:

/* Time to Length, convert time in ns to length in bytes
 * to determinate how many bytes can be sent in given time.
 */
static inline u64 psched_ns_t2l(const struct psched_ratecfg *r,
				u64 time_in_ns)
{
	u64 len = time_in_ns;
	u8 shift = r->shift;
	bool is_div = false;

	/* The formula is :
	 * len = (time_in_ns << shift) / mult
	 * when time_in_ns does shift, it would overflow.
	 * If overflow happens first time, do division.
	 * Then do shift. If it happens again,
	 * set lenth to ~0ULL.
	 */
	while (shift) {
		if (len & (1ULL << 63)) {
			if (!is_div) {
				len = div64_u64(len, r->mult);
				is_div = true;
			} else {
				/* overflow happens */
				len = ~0ULL;
				is_div = true;
				break;
			}
		}
		len <<= 1;
		shift--;
	}
	if (!is_div)
		len = div64_u64(len, r->mult);

	if (unlikely(r->linklayer == TC_LINKLAYER_ATM))
		len = (len / 53) * 48;

	if (len > r->overhead)
		len -= r->overhead;
	else
		len = 0;

	return len;
}
max_size = min_t(u64, psched_ns_t2l(&q->rate, q->buffer), ~0);

Regards,
Yang


--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html