Re: [PATCH V2 2/2] module: Fix performance regression on modules with large symbol tables

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

On Sun, Nov 20, 2011 at 9:06 PM, Rusty Russell <rusty@xxxxxxxxxxxxxxx> wrote:
> On Thu, 17 Nov 2011 19:15:02 -0800, Kevin Cernekee <cernekee@xxxxxxxxx> wrote:
> I'm deeply tempted.  It's very simple, 46 lines shorter, preserves exact
> matches, and doesn't have any strange slowdowns on corner cases.

Unfortunately, the last patch I posted still makes duplicate copies of
the exact matches.  And the copy of nvidia.ko I'm looking at right now
does reuse strtab entries for the duplicated symbols.

FWIW, preserving exact matches would save about 35KB on a kernel
module whose .text + .data + .rodata sections add up to ~9.5MB.

I'm sure there are more clever ways to do it, but a fast brute force
approach could involve storing the core_strtab offsets in a temporary
array indexed by the original strtab st_name values.
--
To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Linux USB Devel]     [Linux Media]     [Video for Linux]     [Linux Audio Users]     [Photo]     [Yosemite News]    [Yosemite Photos]    [Free Online Dating]     [Linux Kernel]     [Linux SCSI]     [XFree86]

Add to Google Powered by Linux