| 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
| ||
|
Message-ID: <AANLkTingurFim5kWw_uLh0xJNiweYkissvf-L9Rv5ypL@mail.gmail.com>
Date: Thu, 20 May 2010 19:53:03 +0200
From: Dhaval Giani <dhaval.giani@...il.com>
To: Rafael Tinoco <rafael.tinoco@...aweb.com.br>
Cc: balbir <balbir@...ux.vnet.ibm.com>,
LKML <linux-kernel@...r.kernel.org>
Subject: Re: [Libcg-devel] Some changes to taskstats.
Hi Rafael
I am cc'ing the LKML and dropping libcg-devel, since the LKML is the
right place to discuss this.
On Thu, May 20, 2010 at 3:38 AM, Rafael Tinoco
<rafael.tinoco@...aweb.com.br> wrote:
> Hello Balbir,
>
> Im sure this isn't the right place to post it, but I would like some
> help to change a few things inside taskstats struct on main kernel.
> Since you have its copyright maybe its a little bit easier.
>
> Ive changed:
>
> /usr/src/linux-2.6.33.2/include/linux/task_io_accounting_ops.h
> /usr/src/linux-2.6.33.2/include/linux/task_io_accounting.h
> /usr/src/linux-2.6.33.2/include/linux/taskstats.h
> /usr/src/linux-2.6.33.2/net/socket.c
>
> Extending the TASKSTATS_HAS_IO_ACCOUNTING (at the end)
>
> with
>
> 167 __u64 read_net_bytes; /* bytes of network read I/O */
> 168 __u64 write_net_bytes; /* bytes of network write I/O */
>
> And the taskstats_ops.h
>
> with
>
> 56 static inline void task_io_account_read_net(size_t bytes)
> 57 {
> 58 current->ioac.read_net_bytes += bytes;
> 59 }
> 60
> 61 static inline void task_io_account_write_net(size_t bytes)
> 62 {
> 63 current->ioac.write_net_bytes += bytes;
> 64 }
>
> And socket.c
>
> with
>
> 673 static inline int __sock_recvmsg_nosec(struct kiocb *iocb, struct
> socket *sock,
> 674 struct msghdr *msg, size_t
> size, int flags)
> 675 {
> 676 int ret = 0;
> 677 struct sock_iocb *si = kiocb_to_siocb(iocb);
> 678
> 679 si->sock = sock;
> 680 si->scm = NULL;
> 681 si->msg = msg;
> 682 si->size = size;
> 683 si->flags = flags;
> 684
> 685 ret = sock->ops->recvmsg(iocb, sock, msg, size, flags);
> 686
> 687 if(ret > 0)
> 688 task_io_account_read_net(ret);
> 689
> 690 return ret;
> 691 }
>
> And taskstats.h (at the end)
>
> with
>
> 167 __u64 read_net_bytes; /* bytes of network read I/O */
> 168 __u64 write_net_bytes; /* bytes of network write I/O */
>
>
> So i could have read and written bytes per pid inside taskstats. This
> is needed for my "ustats" program Ill be releasing soon.
> Ustats is responsible for UID accounting of everything happening about
> procs (using cn_msg and taskstats).
> It maintains statistics of a short timequantum (less than 1 sec) and a
> long timequantum (1 minute) of EVERYTHING any user is actually doing
> or has done. It maintains lists of active and inactive processes..
> couting cpu, mem, disks and network.
> In some benchmark tests, with 1500 users Im using less than 0.3% of
> cpu each 1 or 2 secs for all this accounting. The results are sent to
> a queue for mongodb distribution.
>
> Take a look:
>
> { "host" : "hm2306", "uid" : "11494", "name" : "revistaharvard",
> "time" : "2010:05:12:13:36", "ucpu" : "2424000", "pucpu" : "17.24",
> "scpu" : "652000", "pscpu" : "4.64", "mem" : "413.41" }
> { "host" : "hm2306", "uid" : "11562", "name" :
> "anunciecarrogratis1", "time" : "2010:05:12:13:36", "ucpu" :
> "96000", "pucpu" : "0.68", "scpu" : "44000", "pscpu" : "0.31",
> "mem" : "17.03" }
> { "host" : "hm2306", "uid" : "11568", "name" : "altodoengenho",
> "time" : "2010:05:12:13:36", "ucpu" : "612000", "pucpu" : "4.42",
> "scpu" : "192000", "pscpu" : "1.39", "mem" : "106.77" }
> { "host" : "hm2306", "uid" : "11421", "name" : "callplusnet",
> "time" : "2010:05:12:13:36", "ucpu" : "100000", "pucpu" : "0.72",
> "scpu" : "16000", "pscpu" : "0.12", "mem" : "15.05" }
> { "host" : "hm2306", "uid" : "11190", "name" : "okimagem", "time" :
> "2010:05:12:13:36", "ucpu" : "472000", "pucpu" : "3.48", "scpu" :
> "48000", "pscpu" : "0.35", "mem" : "41.63" }
> { "host" : "hm2306", "uid" : "11444", "name" : "ammp1", "time" :
> "2010:05:12:13:36", "ucpu" : "300000", "pucpu" : "2.24", "scpu" :
> "36000", "pscpu" : "0.27", "mem" : "33.57" }
> { "host" : "hm2306", "uid" : "11498", "name" : "atleticomonteazul",
> "time" : "2010:05:12:13:36", "ucpu" : "296000", "pucpu" : "2.34",
> "scpu" : "72000", "pscpu" : "0.57", "mem" : "45.86" }
> { "host" : "hm2306", "uid" : "11586", "name" : "ou", "time" :
> "2010:05:12:13:36", "ucpu" : "248000", "pucpu" : "2.01", "scpu" :
> "96000", "pscpu" : "0.78", "mem" : "28.69" }
> { "host" : "hm2306", "uid" : "11479", "name" : "firefriend1",
> "time" : "2010:05:12:13:36", "ucpu" : "88000", "pucpu" : "0.71",
> "scpu" : "28000", "pscpu" : "0.23", "mem" : "14.89" }
> { "host" : "hm2306", "uid" : "11469", "name" : "enotel2", "time" :
> "2010:05:12:13:36", "ucpu" : "152000", "pucpu" : "1.24", "scpu" :
> "40000", "pscpu" : "0.33", "mem" : "29.29" }
> { "host" : "hm2306", "uid" : "11551", "name" : "abrhrj1", "time" :
> "2010:05:12:13:36", "ucpu" : "760000", "pucpu" : "6.55", "scpu" :
> "156000", "pscpu" : "1.34", "mem" : "93.41" }
> { "host" : "hm2306", "uid" : "11538", "name" : "k6s2", "time" :
> "2010:05:12:13:36", "ucpu" : "2576000", "pucpu" : "22.23", "scpu" :
> "416000", "pscpu" : "3.59", "mem" : "256.04" }
> { "host" : "hm2306", "uid" : "11537", "name" : "indacarros", "time"
> : "2010:05:12:13:36", "ucpu" : "304000", "pucpu" : "2.67", "scpu" :
> "68000", "pscpu" : "0.60", "mem" : "26.57" }
>
> So, basically pucpu and pscpu is a percentage of 0% to ncpu * 100%
> based on my timequantum. The error rates are in 0.0001 order.
> Im loosing 0.07% of statistics (very very very short lived processes),
> only on extreme high load conditions.
>
> Do you think its possible changes like this in the taskstats struct
> and socket.c ?
>
> With my changes:
>
> [inaddy][inerddao][ustats]$ netcat -l -p 8080
> hehehehehe
>
> Gave me exactly:
> proc_event_exit: pid 1000 uid 1564 (ucpu: 0) (scpu: 0) (teste: 11)
>
> teste is read bytes of this process (10 + \0).
>
> Thank you
>
> Regards
>
> Rafael Tinoco
>
> ------------------------------------------------------------------------------
>
> _______________________________________________
> Libcg-devel mailing list
> Libcg-devel@...ts.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/libcg-devel
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists