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]
Message-ID: <aWcvE-WN09pvpS4g@rkannoth-OptiPlex-7090>
Date: Wed, 14 Jan 2026 11:22:19 +0530
From: Ratheesh Kannoth <rkannoth@...vell.com>
To: Jakub Kicinski <kuba@...nel.org>
CC: <netdev@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <sgoutham@...vell.com>, <davem@...emloft.net>, <edumazet@...gle.com>,
        <pabeni@...hat.com>, <andrew+netdev@...n.ch>
Subject: Re: [PATCH net-next v3 01/13] octeontx2-af: npc: cn20k: Index
 management

On 2026-01-13 at 20:37:32, Jakub Kicinski (kuba@...nel.org) wrote:
> On Tue, 13 Jan 2026 15:59:02 +0530 Ratheesh Kannoth wrote:
> > On 2026-01-11 at 04:28:42, Jakub Kicinski (kuba@...nel.org) wrote:
> > > On Fri, 9 Jan 2026 11:18:16 +0530 Ratheesh Kannoth wrote:
> > > > +static int
> > > > +npc_subbank_srch_order_parse_n_fill(struct rvu *rvu, char *options,
> > > > +				    int num_subbanks)
> > >
> > > Please avoid writable debugfs files,
> > I explored devlink option, but the input string is too big.
> >
> > > do you really need them?
> > Customer can change subbank search order by modifying the search order
> > thru this devlink.
>
> Unclear what you're trying to say.
TCAM memory is dividied into 32 horizontal chunks. Each chunk is called
a subbank. When a request to alloc a free TCAM slot is requested by PF,
these 32 subbank are searched in a specific order. Since bottom subbank rules
have higher priority than top subbanks, customer may need to alter the
search order to control the distribution of allocation to different subbanks.

Example search order format to debugfs entry is as below

"[0]=[8],[1]=7,[2]=30,[3]=0,[4]=1,[5]=2,[6]=3,[4]=4,[5]=5,[6]=6,[7]=9,[9]=10,[10]=11,[11]=12,[12]=13,[13]=14......[31]=0"

This input "string" is too long for devlink ?

union devlink_param_value {
        u8 vu8;
        u16 vu16;
        u32 vu32;
        char vstr[__DEVLINK_PARAM_MAX_STRING_VALUE];
        bool vbool;
};

>Hope you documented this much better
> in v4,
My bad—already sent V4. I will make sure to document this in V5

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ