lkml.org 
[lkml]   [2006]   [Apr]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [RFC PATCH 1/2] Makefile: export-symbol usage report generator.
From
Date
On Thu, 2006-04-20 at 01:41 +0200, Adrian Bunk wrote:
> On Wed, Apr 19, 2006 at 04:25:58PM -0700, Ram Pai wrote:
> > On Tue, 2006-04-18 at 16:09 +0200, Adrian Bunk wrote:
> > > On Thu, Apr 13, 2006 at 05:38:26AM -0700, Ram Pai wrote:
> > >
> > > > I ran a report to extract export-symbol usage by kernel modules. The results
> > > > are at http://www.sudhaa.com/~ram/misc/export_report.txt
> > > >
> > > > The report lists:
> > > > 1. All the exported symbols and their usage counts by in-kernel modules.
> > > > 2. for each in-kernel module, lists the modules and the exported symbols
> > > > from those modules, that it depends on.
> > > >
> > > > Highlights:
> > > > On x86 architecture
> > > > (1) 880 exported symbols not used by any in-kernel modules.
> > > > (2) 1792 exported symbols used only once.
> > > >
> > > > I hope this report/tool shall help all inkernel modules to revisit their usage
> > > > of kernel interfaces.
> > > >
> > > > This patch integrates the report-generator into the kernel build process. After
> > > > applying this patch, invoke 'make export_report' and it creates the report in
> > > > Documentation/export_report.txt
> > > >...
> > >
> > > I like your patch, but I have observed two issues:
> > > - please don't force an allmodconfig, simply use the currently compiled
> > > kernel
> >
> > Looking for ideas. Its hard to extract out the imported symbols unless
> > they are compiled as modules. Running the report on a currently compiled
> > kernel will miss most of the subsystems that are not compile as modules.
> > Hence the report wont be complete. Any suggestions?
> >...
>
> - it's a tool for kernel hackers who know what they are doing
> - it's a tool to help you finding unused exports, but each one still
> requires manual verification

I am not sure I understand you. Are you saying:

A) the tool should report the export symbol usage with whatever
set of modules the user has configured in his .config ?

or

B) the tool should report the export symbol usage for all the
modules irrespective of what has been configured in the
users .config ?
RP
>
> Besides this, e.g. CONFIG_SMP=n brings you more modules and therefore a
> better coverage than allmodconfig.
>
> cu
> Adrian
>

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2006-04-20 02:56    [W:1.968 / U:0.012 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site