Mirror of the lttng-dev mailing list
 help / color / mirror / Atom feed
From: greg@kroah.com (Greg KH)
Subject: [lttng-dev] [PATCH 07/20] lttng wrappers
Date: Wed, 30 Nov 2011 15:23:28 +0900	[thread overview]
Message-ID: <20111130062328.GA11610@kroah.com> (raw)
In-Reply-To: <20111129022213.GA24465@Krystal>

On Mon, Nov 28, 2011 at 09:22:13PM -0500, Mathieu Desnoyers wrote:
> * Greg KH (greg at kroah.com) wrote:
> > On Mon, Nov 28, 2011 at 07:42:14AM -0500, Mathieu Desnoyers wrote:
> > > Implement wrappers for compatibility with older kernel versions and
> > > kernels with had the libringbuffer (old) patchset applied.
> > > 
> > > Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers at efficios.com>
> > 
> > Are these going to still be needed anymore, now that this code is in the
> > kernel tree?
> 
> Hrm, this patch changelog could have been a bit more explicit. In
> addition to allow compatibility for older kernel versions, some wrappers
> deal with missing EXPORT_SYMBOL_GPL() from the core kernel.
> 
> The TODO file details what should be done to eliminate those wrappers:
> 
>         2) Cleanup wrappers. The drivers/staging/lttng/wrapper directory
>            contains various wrapper headers that use kallsyms lookups to
>            work around some missing EXPORT_SYMBOL_GPL() in the mainline
>            kernel. Ideally, those few symbols should become exported to
>            modules by the kernel.

Ugh, that's crazy, but I understand.

> So hopefully we can look into adding these missing symbols to the core
> kernel. This would be much nicer, but this does not depend only on the
> lttng driver: it's up to the kernel to export those.
> 
> Another option would be to create a CONFIG bool (y/n) object that wraps
> the missing core kernel symbols. However, this would require this
> support option to be selected at kernel build time in order to allow
> loading the lttng modules, and I doubt it's a good practice for a
> driver.

Yes, that's not good, we can just export the symbols as needed now that
we have an in-kernel user.

> As for those wrappers that deal more specifically with compatibility
> with older kernels, those aren't needed anymore, of course.

Great, please send patches to remove them before I get to it :)

thanks,

greg k-h



  reply	other threads:[~2011-11-30  6:23 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-11-28 12:42 [lttng-dev] LTTng driver submission for staging (2nd attempt) Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 01/20] lttng lib: lttng priority heap Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 02/20] lttng lib: ring buffer Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 03/20] lttng lib: portable bitfield read/write header Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 04/20] lttng: BUILD_RUNTIME_BUG_ON Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 05/20] lttng: offset alignment header Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 06/20] lttng libs: add Makefile Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 07/20] lttng wrappers Mathieu Desnoyers
2011-11-29  1:04   ` Greg KH
2011-11-29  2:22     ` Mathieu Desnoyers
2011-11-30  6:23       ` Greg KH [this message]
2011-11-28 12:42 ` [lttng-dev] [PATCH 08/20] lttng instrumentation: tracepoint events Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 09/20] lttng: syscall instrumentation Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 10/20] lttng: lib ring buffer clients Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 11/20] lttng: tracer control and core structures Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 12/20] lttng: dynamically selectable context information Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 13/20] lttng: timing calibration feature Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 14/20] lttng: debugfs and procfs ABI Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 15/20] lttng: Add documentation and TODO files Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 16/20] lttng: add system call instrumentation probe Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 17/20] lttng: probe callbacks Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 18/20] lttng: toplevel Makefile and Kconfig Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 19/20] staging: add LTTng to build Mathieu Desnoyers
2011-11-28 12:42 ` [lttng-dev] [PATCH 20/20] Add LTTng entry to MAINTAINERS file Mathieu Desnoyers
2011-11-29  1:07 ` [lttng-dev] LTTng driver submission for staging (2nd attempt) Greg KH

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=20111130062328.GA11610@kroah.com \
    --to=greg@kroah.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