linux-trace-devel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Steven Rostedt <rostedt@goodmis.org>
To: Tzvetomir Stoyanov <tz.stoyanov@gmail.com>
Cc: linux-trace-devel@vger.kernel.org
Subject: Re: [PATCH v20 13/15] trace-cmd: Basic infrastructure for host - guest timestamp synchronization
Date: Mon, 2 Mar 2020 08:43:56 -0500	[thread overview]
Message-ID: <20200302084356.4e021c6c@oasis.local.home> (raw)
In-Reply-To: <CAPpZLN67B53wpkZATXnXViaTYW=t9XsX0hVQXu==EtLFzNhotA@mail.gmail.com>

On Mon, 2 Mar 2020 11:43:52 +0200
Tzvetomir Stoyanov <tz.stoyanov@gmail.com> wrote:

> On Sat, Feb 29, 2020 at 2:28 AM Steven Rostedt <rostedt@goodmis.org> wrote:
> >
> >
> > Now that I'm playing with patch 14, I took more interest in this code.
> >
> > On Thu, 27 Feb 2020 16:19:59 +0200
> > "Tzvetomir Stoyanov (VMware)" <tz.stoyanov@gmail.com> wrote:
> >  
> > > +
> > > +#define PROTO_MASK_SIZE (sizeof(char))  
> >
> > Hmm, this is the size in bytes of the mask, not bits. You may need both.
> >
> > #define PROTO_MASK_BITS (PROTO_MASK_SIZE * 8)
> >
> > Because we can have up to 8 protocols per mask size (8 bits in a byte).
> >  
> > > +/**
> > > + * tracecmd_tsync_proto_select - Select time sync protocol, to be used for
> > > + *           timestamp synchronization with a peer
> > > + *
> > > + * @proto_mask: bitmask array of time sync protocols, supported by the peer
> > > + * @length: size of the @protos array
> > > + *
> > > + * Retuns Id of a time sync protocol, that can be used with the peer, or 0
> > > + *     in case there is no match with supported protocols
> > > + */
> > > +unsigned int tracecmd_tsync_proto_select(char *proto_mask, int length)
> > > +{
> > > +     struct tsync_proto *selected = NULL;
> > > +     struct tsync_proto *proto;
> > > +     int word;
> > > +     int id;
> > > +
> > > +     for (word = 0; word < length; word++) {
> > > +             for (proto = tsync_proto_list; proto; proto = proto->next) {
> > > +                     if (proto->proto_id < word * PROTO_MASK_SIZE)
> > > +                             continue;  
> >
> > The above should be: proto->proto_id < word * PROTO_MASK_BITS
> >
> > Because what you have currently is:
> >
> >    proto->proto_id < word * 1
> >
> >  
> > > +
> > > +                     id = proto->proto_id - word * PROTO_MASK_SIZE;  
> >
> > And here you want PROTO_MASK_BITS, otherwise if we have a proto_id of 2
> > (which would fit as a bit in a char), this would become:
> >
> >         id = 2 - 0 * 1 = 1
> >  
> > > +                     if (id >= PROTO_MASK_SIZE)  
> >
> > Then this is: 2 >= 1 which would skip it.
> >
> > Hmm, maybe you don't even need PROTO_MASK_SIZE and only need
> > PROTO_MASK_BITS.
> >  
> The PROTO_MASK_SIZE is the size of 1 word of the mask array, in bytes.
> In the first implementation the array was of integers, and the define
> was 4. It was very easy to switch to char, that's why I prefer to keep
> the implementation as is - even though that "proto->proto_id < word *
> 1".
> The confusion is the name of the variable "id", which is actually the
> index of the bit inside the given bitmask's word. In the example of
> proto_id 2:
>   - the word is 0, this proto_id is in the first word from the bitmask array.
>   - the index of the bit in this word is 2 ( id = 2 - 0 * 1 = 2 )

Yes, 2 (that's what I meant), but then my point still is valid.

>   - then we can check if the bit, representing the proto_id 2 is up:
>         if ((1 << id) & proto_mask[word]) ...

I'm talking about this:

+			id = proto->proto_id - word * PROTO_MASK_SIZE;
+			if (id >= PROTO_MASK_SIZE)
+				continue;


Where id = 2, then if (2 >= 1) continue, means we skip it.

You need: if (id >= PROTO_MASK_SIZE_BITS)

-- Steve

  reply	other threads:[~2020-03-02 13:43 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-27 14:19 [PATCH v20 00/15]Timestamp synchronization of host - guest tracing session Tzvetomir Stoyanov (VMware)
2020-02-27 14:19 ` [PATCH v20 01/15] trace-cmd: Add support for negative time offsets in trace.dat file Tzvetomir Stoyanov (VMware)
2020-02-27 14:19 ` [PATCH v20 02/15] trace-cmd: Add new library API for local CPU count Tzvetomir Stoyanov (VMware)
2020-02-28 23:42   ` Steven Rostedt
2020-02-27 14:19 ` [PATCH v20 03/15] trace-cmd: Find and store pids of tasks, which run virtual CPUs of given VM Tzvetomir Stoyanov (VMware)
2020-02-27 17:23   ` Steven Rostedt
2020-03-02  9:29     ` Tzvetomir Stoyanov
2020-02-27 14:19 ` [PATCH v20 04/15] trace-cmd: Implement new API tracecmd_add_option_v() Tzvetomir Stoyanov (VMware)
2020-02-27 14:19 ` [PATCH v20 05/15] trace-cmd: Add new API to generate a unique ID of the tracing session Tzvetomir Stoyanov (VMware)
2020-02-27 14:19 ` [PATCH v20 06/15] trace-cmd: Store the session tracing ID in the trace.dat file Tzvetomir Stoyanov (VMware)
2020-02-27 14:19 ` [PATCH v20 07/15] trace-cmd: Add definitions of htonll() and ntohll() Tzvetomir Stoyanov (VMware)
2020-02-27 14:19 ` [PATCH v20 08/15] trace-cmd: Exchange tracing IDs between host and guest Tzvetomir Stoyanov (VMware)
2020-02-27 14:19 ` [PATCH v20 09/15] trace-cmd: Implement new option in trace.dat file: TRACECMD_OPTION_TIME_SHIFT Tzvetomir Stoyanov (VMware)
2020-02-27 17:28   ` Steven Rostedt
2020-02-27 14:19 ` [PATCH v20 10/15] trace-cmd: Add guest information in host's trace.dat file Tzvetomir Stoyanov (VMware)
2020-02-27 14:19 ` [PATCH v20 11/15] trace-cmd: Add host trace clock as guest trace argument Tzvetomir Stoyanov (VMware)
2020-02-27 14:19 ` [PATCH v20 12/15] trace-cmd: Refactor few trace-cmd internal functions Tzvetomir Stoyanov (VMware)
2020-02-27 14:19 ` [PATCH v20 13/15] trace-cmd: Basic infrastructure for host - guest timestamp synchronization Tzvetomir Stoyanov (VMware)
2020-02-28 17:41   ` Steven Rostedt
2020-02-29  0:28   ` Steven Rostedt
2020-03-02  9:43     ` Tzvetomir Stoyanov
2020-03-02 13:43       ` Steven Rostedt [this message]
2020-02-27 14:20 ` [PATCH v20 14/15] trace-cmd: [POC] PTP-like algorithm " Tzvetomir Stoyanov (VMware)
2020-02-27 14:20 ` [PATCH v20 15/15] trace-cmd: Debug scripts for " Tzvetomir Stoyanov (VMware)

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=20200302084356.4e021c6c@oasis.local.home \
    --to=rostedt@goodmis.org \
    --cc=linux-trace-devel@vger.kernel.org \
    --cc=tz.stoyanov@gmail.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).