On Fri, Feb 15, 2019 at 11:06 PM Andy Shevchenko
<andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:
>
> On Fri, Feb 15, 2019 at 12:31:17AM +0800, xiang xiao wrote:
> > On Thu, Feb 14, 2019 at 9:09 PM Andy Shevchenko
> > <andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:
> > > On Thu, Feb 14, 2019 at 02:02:38PM +0800, Xiang Xiao wrote:
>
> > > > This driver allows the remote processor to redirect the output of
> > > > syslog or printf into the kernel log, which is very useful to see
> > > > what happen in the remote side.
> > >
> > > > +struct rpmsg_syslog_header {
> > > > + u32 command;
> > > > + s32 result;
> > > > +} __packed;
> > >
> > > Isn't packed already?
> > >
> >
> > But, I want to make it more explicitly and prepare for struct expansion later.
>
> How? Why it's not in this patch / patch series?
Just for future, not now:).
Since this structure is shared by the different CPU/OS, it's better to
indicate the packed explicitly.
>
> > > > + /* output the message before '\n' to the kernel log */
> > > > + nl = memrchr(msg->data, '\n', msg->count);
> > >
> > > Hmm... To me it sounds somehow fragile.
> > >
> > > If your text contains binary data, how can you guarantee that it would be not
> > > in the middle of two \n:s?
> >
> > This driver is just for log/printf redirection, so we could safely
> > assume the data is pure text.
>
> Then I don't see a point to use memrchr() at all here.
>
> Use strchr or strrchr().
Yes, use strnchr is enough, I will remove memrchr in the next review.
>
> > > OTOH, if it text data, why do you need to take all strings at once?
> >
> > Remote side may decide to buffer more log to reduce the IPC number
> > since IPC is a time consuming operation.
>
> So, you always can do something like
>
> p = msg->data;
> while (...strsep(..., "\n")) {
> pr_info("%s\n", token);
> ...
> }
Can't use strsep here, since log come from remote isn't terminated by '\0'.
>
> >
> > > It might be worse from performance prospective (if you know how and when printk() supplies buffer to the console).
> >
> > Yes, it's very slow if the log send to serial console. But in
> > production environment, printk normally just save in ram and viewed by
> > dmesg which is very fast.
>
> You may not do such assumptions. For someone it would be RAM, for some
> customers it might be a slow channel.
But we need reduce the IPC number, so both fast/slow channel could get
the benefit.
>
>
> > > > + strncpy(priv->buf + priv->next, msg->data + printed, copied);
> > >
> > > Hmm... shouldn't be memcpy()?
> >
> > I use memcpy initially, but found that the unaligned exception happen randomly.
> > To avoid the cache issue, the IPC memory normally map as device memory, but
> > ARM just allow the alignment access to this type of memory.
>
> So, than it's an architecture level issue. With strncpy() here you will get a
> pretty rightful GCC warning.
>
Why GCC warning strncpy here?
> > > > + /* flush the buffered log if need */
> > > > + if (priv->next)
> > > > + pr_info("%.*s\n", priv->next, priv->buf);
> > > > + kfree(priv->buf);
> > >
> > > I don't see how it's serialized. Does rpmsg core take care of this?
> >
> > Yes, the callback come from a dedicated work thread.
>
> Please, add a comment explaining that.
>
Will add in the next review.
> --
> With Best Regards,
> Andy Shevchenko
>
>
