DPDK patches and discussions
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: Matan Azrad <matan@mellanox.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>,
	Stephen Hemminger <sthemmin@microsoft.com>
Subject: Re: [dpdk-dev] [PATCH v5 1/4] examples/multi_process/client_server_mp: check port validity
Date: Tue, 6 Aug 2019 08:39:55 -0700	[thread overview]
Message-ID: <20190806083955.59124799@hermes.lan> (raw)
In-Reply-To: <AM0PR0502MB4019394E04168A1337E7A516D2D50@AM0PR0502MB4019.eurprd05.prod.outlook.com>

On Tue, 6 Aug 2019 08:19:01 +0000
Matan Azrad <matan@mellanox.com> wrote:

> From: Stephen Hemminger
> > On Sun, 4 Aug 2019 08:31:54 +0000
> > Matan Azrad <matan@mellanox.com> wrote:
> >   
> > > > > >  	/* convert parameter to a number and verify */
> > > > > >  	pm = strtoul(portmask, &end, 16);
> > > > > > -	if (end == NULL || *end != '\0' || pm == 0)
> > > > > > +	if (end == NULL || *end != '\0' || pm > UINT16_MAX || pm  
> > == 0)  
> > > > >
> > > > > Why pm > UINT16_MAX ? should be something like > (1 <<  
> > > > RTE_MAX_ETHPORTS) - 1.  
> > > > > And need to be sure pm type can hold RTE_MAX_ETHPORTS bits,  
> > > > otherwise port 0 may unlikely be all the time visible in the loop below.  
> > > > >  
> > > >
> > > > The DPDK assumes a lot of places that unsigned long will hold a port  
> > mask.  
> > >
> > > So, all are bugs, no?  
> > 
> > I don't think 32 bit build is that well tested. But yes a mask needs to hold 64
> > ports.  
> 
> What if someone changes RTE_MAX_ETHPORTS to be bigger than 64 in config file?
> 
> Assume the user changes RTE_MAX_ETHPORTS to 128, and there is a valid port in range [64, 127].
> Then,  assume the failsafe sub device owns port ID 0.
> 
> Because the mask bits are not enough to handle the above range, you will get port 0 as valid port - bug.
> 
> I think you need one more check to the RTE_MAX_ETHPORTS > 64 case. 

Not really needed.

The DPDK has lots of hard coded assumptions of all ports fitting in 64 bits.
Examples include testpmd/parameters.c etc.

The original concept of a small set of assigned values for portid is not going
to scale. It really should have been more like ifindex; something that is not
used by common API's much larger range; and assigned purely sequentially.

The API's should all be using names, but the DPDK port naming is also a mess...

  reply	other threads:[~2019-08-06 15:40 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-07-09 15:09 [dpdk-dev] [PATCH v3 0/2] examples/client_server_mp: fix port issues Stephen Hemminger
2019-07-09 15:09 ` [dpdk-dev] [PATCH v3 1/2] examples/multi_process/client_server_mp: check port validity Stephen Hemminger
2019-07-09 15:09 ` [dpdk-dev] [PATCH v3 2/2] examples/multi_process - fix crash in mp_client with sparse ports Stephen Hemminger
2019-07-26 16:50 ` [dpdk-dev] [PATCH v4 0/4] examples/client_server_mp: fix port issues Stephen Hemminger
2019-07-26 16:50   ` [dpdk-dev] [PATCH v4 1/4] examples/multi_process/client_server_mp: check port validity Stephen Hemminger
2019-07-30  9:21     ` Matan Azrad
2019-07-30 15:56       ` Stephen Hemminger
2019-07-30 16:39         ` Matan Azrad
2019-07-30 16:50           ` Stephen Hemminger
2019-07-26 16:50   ` [dpdk-dev] [PATCH v4 2/4] examples/multi_process/client_server_mp - fix crash in mp_client with sparse ports Stephen Hemminger
2019-07-26 16:50   ` [dpdk-dev] [PATCH v4 3/4] examples/multi_process/client_server_mp/mp_server: fix style Stephen Hemminger
2019-07-26 16:50   ` [dpdk-dev] [PATCH v4 4/4] examples/multi_process/client_server_mp/mp_server: use ether format address Stephen Hemminger
2019-08-02  2:58   ` [dpdk-dev] [PATCH v5 0/4] examples/client_server_mp: port id fixes Stephen Hemminger
2019-08-02  2:58     ` [dpdk-dev] [PATCH v5 1/4] examples/multi_process/client_server_mp: check port validity Stephen Hemminger
2019-08-02  5:33       ` Matan Azrad
2019-08-02 15:53         ` Stephen Hemminger
2019-08-04  8:31           ` Matan Azrad
2019-08-05 16:00             ` Stephen Hemminger
2019-08-06  8:19               ` Matan Azrad
2019-08-06 15:39                 ` Stephen Hemminger [this message]
2019-08-06 20:03                   ` Matan Azrad
2019-08-06 23:09                     ` Stephen Hemminger
2019-08-07  5:38                       ` Matan Azrad
2019-08-07  5:53                         ` Stephen Hemminger
2019-08-07  7:02                           ` Matan Azrad
2019-08-07 15:15                             ` Stephen Hemminger
2019-08-02  2:58     ` [dpdk-dev] [PATCH v5 2/4] examples/multi_process/client_server_mp - fix crash in mp_client with sparse ports Stephen Hemminger
2019-08-02  2:58     ` [dpdk-dev] [PATCH v5 3/4] examples/multi_process/client_server_mp/mp_server: fix style Stephen Hemminger
2019-08-02 11:09       ` David Marchand
2019-08-02  2:58     ` [dpdk-dev] [PATCH v5 4/4] examples/multi_process/client_server_mp/mp_server: use ether format address Stephen Hemminger
2019-08-02 23:52   ` [dpdk-dev] [PATCH v6 0/2] examples/client_server_mp: port id (fixes only) Stephen Hemminger
2019-08-02 23:52     ` [dpdk-dev] [PATCH v6 1/2] examples/multi_process/client_server_mp: check port validity Stephen Hemminger
2019-08-02 23:52     ` [dpdk-dev] [PATCH v6 2/2] examples/multi_process/client_server_mp - fix crash in mp_client with sparse ports Stephen Hemminger
2019-08-05 16:38   ` [dpdk-dev] [PATCH v7 0/2] examples/client_server_mp: port id (fixes only) Stephen Hemminger
2019-08-05 16:38     ` [dpdk-dev] [PATCH v7 1/2] examples/multi_process/client_server_mp: check port validity Stephen Hemminger
2019-08-06 12:07       ` Matan Azrad
2019-11-13 18:53         ` Stephen Hemminger
2019-08-05 16:38     ` [dpdk-dev] [PATCH v7 2/2] examples/multi_process/client_server_mp - fix crash in mp_client with sparse ports Stephen Hemminger
2019-08-07  5:40     ` [dpdk-dev] [PATCH v7 0/2] examples/client_server_mp: port id (fixes only) Matan Azrad
2019-11-25 22:51       ` Thomas Monjalon

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190806083955.59124799@hermes.lan \
    --to=stephen@networkplumber.org \
    --cc=dev@dpdk.org \
    --cc=matan@mellanox.com \
    --cc=sthemmin@microsoft.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).