lists.openwall.net  lists / announce owlusers owldev johnusers johndev passwdqcusers yescrypt popa3dusers / osssecurity kernelhardening musl sabotage tlsify passwords / cryptdev xvendor / Bugtraq FullDisclosure linuxkernel linuxnetdev linuxext4 linuxhardening PHC  
Open Source and information security mailing list archives
 

Date: Tue, 8 Dec 2020 23:19:40 +0200 From: Vladimir Oltean <olteanv@...il.com> To: DENG Qingfang <dqfext@...il.com> Cc: netdev@...r.kernel.org, linuxmediatek@...ts.infradead.org, Sean Wang <sean.wang@...iatek.com>, Landen Chao <Landen.Chao@...iatek.com>, Andrew Lunn <andrew@...n.ch>, Vivien Didelot <vivien.didelot@...il.com>, Florian Fainelli <f.fainelli@...il.com>, "David S . Miller" <davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>, Matthias Brugger <matthias.bgg@...il.com>, Russell King <linux@...linux.org.uk>, Frank Wunderlich <frankw@...licfiles.de> Subject: Re: [PATCH netnext] net: dsa: mt7530: support setting ageing time On Tue, Dec 08, 2020 at 10:56:21PM +0200, Vladimir Oltean wrote: > On Tue, Dec 08, 2020 at 03:00:28PM +0800, DENG Qingfang wrote: > > MT7530 has a global address age control register, so use it to set > > ageing time. > > > > The applied timer is (AGE_CNT + 1) * (AGE_UNIT + 1) seconds > > > > Signedoffby: DENG Qingfang <dqfext@...il.com> > >  > > drivers/net/dsa/mt7530.c  41 ++++++++++++++++++++++++++++++++++++++++ > > drivers/net/dsa/mt7530.h  13 +++++++++++++ > > 2 files changed, 54 insertions(+) > > > > diff git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c > > index 6408402a44f5..99bf8fed6536 100644 > >  a/drivers/net/dsa/mt7530.c > > +++ b/drivers/net/dsa/mt7530.c > > @@ 870,6 +870,46 @@ mt7530_get_sset_count(struct dsa_switch *ds, int port, int sset) > > return ARRAY_SIZE(mt7530_mib); > > } > > > > +static int > > +mt7530_set_ageing_time(struct dsa_switch *ds, unsigned int msecs) > > +{ > > + struct mt7530_priv *priv = ds>priv; > > + unsigned int secs = msecs / 1000; > > + unsigned int tmp_age_count; > > + unsigned int error = 1; > > + unsigned int age_count; > > + unsigned int age_unit; > > + > > + /* Applied timer is (AGE_CNT + 1) * (AGE_UNIT + 1) seconds */ > > + if (secs < 1  secs > (AGE_CNT_MAX + 1) * (AGE_UNIT_MAX + 1)) > > + return ERANGE; > > + > > + /* iterate through all possible age_count to find the closest pair */ > > + for (tmp_age_count = 0; tmp_age_count <= AGE_CNT_MAX; ++tmp_age_count) { > > + unsigned int tmp_age_unit = secs / (tmp_age_count + 1)  1; > > + > > + if (tmp_age_unit <= AGE_UNIT_MAX) { > > + unsigned int tmp_error = secs  > > + (tmp_age_count + 1) * (tmp_age_unit + 1); > > + > > + /* found a closer pair */ > > + if (error > tmp_error) { > > + error = tmp_error; > > + age_count = tmp_age_count; > > + age_unit = tmp_age_unit; > > + } > > I feel that the error calculation is just snake oil. This would be enough: > > if (tmp_age_unit <= AGE_UNIT_MAX) > break; > > Explanation: > > You are given a number X, and must find A and B such that the error > E = (A x B)  X should be minimum, with > 1 <= A <= A_max > 1 <= B <= B_max > > It is logical to try with A=1 first. If X / A <= B_max, then of course, > B = X / 1, and the error E is 0. > > If that doesn't work out, and B > B_max, then you go to A=2. That gives > you another B = X / 2, and the error E is 1. You check again if B <= > B_max. If it is, that's your answer. B = X / 2, with an error E of 1. > > You get my point. Iterating ascendingly through A, and calculating B as > X / A, already gives you the smallest error as soon as it satisfies the > B <= B_max requirement. > > > + > > + /* found the exact match, so break the loop */ > > + if (!error) > > + break; > > + } > > + } > > + > > + mt7530_write(priv, MT7530_AAC, AGE_CNT(age_count)  AGE_UNIT(age_unit)); > > + > > + return 0; > > +} Thinking more about it, it's not snake oil but is actually correct. Where I was wrong was the division giving you a certain error, but it actually only gives a maximum error. Other, larger, divisors can still give you an error of 0. If the number you're searching for, X, is, say, a multiple of 3 but not of 2, and is: B_max * 1 < X < B_max * 2 then exiting the loop at A=2 would give you an error E=1. But if you continued the loop, then A=3 would have yielded an error E=0, because X is a multiple of 3 but not of 2. I should probably go and refresh my basic arithmetic. Reviewedby: Vladimir Oltean <olteanv@...il.com>
Powered by blists  more mailing lists