From: linas@austin•ibm.com (Linas Vepstas)
To: Nathan Lynch <ntl@pobox•com>
Cc: linuxppc-dev@ozlabs•org, Paul Mackerras <paulus@samba•org>
Subject: Re: [PATCH 1/4] powerpc: Add EEH sysfs blinkenlights
Date: Thu, 24 May 2007 12:34:53 -0500 [thread overview]
Message-ID: <20070524173453.GJ5921@austin.ibm.com> (raw)
In-Reply-To: <20070523232256.GJ29914@localdomain>
On Wed, May 23, 2007 at 06:22:56PM -0500, Nathan Lynch wrote:
> Hi Linas-
>
> Linas Vepstas wrote:
> >
> > Add sysfs blinkenlights for EEH statistics. Shuffle the
> > eeh_add_device_tree() call so that it appears in the correct
> > sequence.
>
> ( blinkenlights? :)
Alles touristen und non-technischen looken peepers! Das machinkontrol is
nicht for gefengerpoken und mittengrabben. Oderwise is easy schnappen
der springenverk, blowenfus, und poppencorken mit spitzensparken. Der
machine is diggen by experten only. Is nicht fur geverken by das
dumpkopfen. Das rubber necken sightseenen keepen das cotton-picken hands
in das pockets. So relaxen, und vatchen das blinkenlights.
> To me this seems a somewhat terse changelog considering that the patch
> introduces a user-visible interface. The changelog does not really
> say what the code is doing or why, or who will use it.
At this time, there are no planned user-space tools that would look
at this. Its intended primarily for sysadmins and service, to determine
what, exactly, is going on in the system. I'd been planning on adding
this for years, but recent email exchanges made it clear that this needs
to get done.
> > +#define EEH_SHOW_ATTR(_name,_memb,_format) \
> > +static ssize_t eeh_show_##_name(struct device *dev, \
> > + struct device_attribute *attr, char *buf) \
>
> I have been frustrated by similar constructions more than once in the
> midst of debugging. I know this has become a common practice with
> sysfs-related code, but using cpp to generate function names
> completely defeats grep etc. when you're trying to track down a
> problem, and I'd not like to see this sort of thing propagated.
I understand the frustration. But I also would hate to have 5 or 6
nearly identidical subroutines, one after the other. Perhaps the
answer is to avoid te ## pste token, and instead do something
like this:
+#define EEH_SHOW_ATTR(_fullname,_memb,_format) \
+static ssize_t _fullname(struct device *dev, \
+ struct device_attribute *attr, char *buf) \
That way, grep will suceed in finding the "full name".
> > +void eeh_sysfs_add_device(struct pci_dev *pdev);
> > +void eeh_sysfs_remove_device(struct pci_dev *pdev);
>
> Don't you need dummy static inline placeholders for CONFIG_EEH=n?
Hmm. I think these are only called fom the dlpar code, and
neither dlpar nor pseries pci will work if eeh is config'ed off...
I don't know how to make CONFIG_PCI depend on CONFIG_EEH, though ...
strange situation.
--linas
next prev parent reply other threads:[~2007-05-24 17:35 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-05-23 17:13 [PATCH 0/4] powerpc: EEH blinkenlights, copyright cleanup Linas Vepstas
2007-05-23 17:16 ` [PATCH 1/4] powerpc: Add EEH sysfs blinkenlights Linas Vepstas
2007-05-23 23:22 ` Nathan Lynch
2007-05-24 17:34 ` Linas Vepstas [this message]
2007-05-23 23:42 ` Stephen Rothwell
2007-05-23 17:20 ` [PATCH 2/4] powerpc: show EEH per-device false positives Linas Vepstas
2007-05-23 17:23 ` [PATCH 3/4] powerpc: remove dead EEH code Linas Vepstas
2007-05-23 23:53 ` Stephen Rothwell
2007-05-24 18:00 ` Linas Vepstas
2007-05-23 17:28 ` [PATCH 4/4] powerpc: tweak EEH copyright info Linas Vepstas
2007-05-23 23:55 ` Stephen Rothwell
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=20070524173453.GJ5921@austin.ibm.com \
--to=linas@austin$(echo .)ibm.com \
--cc=linuxppc-dev@ozlabs$(echo .)org \
--cc=ntl@pobox$(echo .)com \
--cc=paulus@samba$(echo .)org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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