Re: [PATCH] kprobe: increase kprobe_hash_table size

[Masami Hiramatsu <mhiramat@redhat.com>]

Andrew Morton wrote:
>> I agree that there may be many opinions about what is the best suited size.
>> Why I chose 512 was that I thought the table (byte) size was less than or
>> equal 4096 even on 64-bit arch.
> 
> Well...
> 
>    text    data     bss     dec     hex filename
>    7036     744    9380   17160    4308 kernel/kprobes.o
>    7048     744   73892   81684   13f14 kernel/kprobes.o
> 
> That's 64 kbytes more memory.  It will be kretprobe_table_locks[] which
> is hurting here, due to the ____cacheline_aligned.

Oops! It's really bad.


> I expected CONFIG_X86_VSMP=y to make this far worse, but fortunately
> that only affects ____cacheline_internodealigned_in_smp.
> 
> btw, that array wastes a ton of memory on uniprocessor builds.  Using
> ____cacheline_aligned_in_smp should fix that.
> 
> Please always check these thigns with /usr/bin/size.

I see. I'll check that and try to find the best way...

> btw2, could/should kprobe_table[] and kretprobe_inst_table[] be
> aggregated into kretprobe_table_locks[]?  That would save some memory
> and might save some cache misses as well?

Indeed, thank you for good idea.

> Anyway, enough pos-facto code review.  Is this change which you're
> proposing worth increasing kernel memory usage by 64k?

Not really. Hmm, I have to investigate more on this problem.

Thanks a lot.

-- 
Masami Hiramatsu

Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division

e-mail: mhiramat@redhat.com