RFR(S): 8016696: PPC64 (part 4): add relocation for trampoline stubs

[Lindenmaier, Goetz]

Hi Albert,

I will look into it, once I am done preparing part 5
according to Chris' proposals.

I will check the distribution of actual offsets
to be able to judge the effect of a smaller offset 
field.  I guess x86 is the bottleneck here, as they
have offset_unit = 1 and need format_width > 0.
On Power there is offset_unit = 4, so offsets
reach far enough anyways. 

I'm not sure whether I can judge how flags
could be combined, as I don't know who else
wants to use the last entry.

Best regards,
  Goetz.


-----Original Message-----
From: ppc-aix-port-dev-bounces at openjdk.java.net [mailto:ppc-aix-port-dev-bounces at openjdk.java.net] On Behalf Of Albert Noll
Sent: Donnerstag, 20. Juni 2013 09:30
To: ppc-aix-port-dev at openjdk.java.net
Subject: Re: RFR(S): 8016696: PPC64 (part 4): add relocation for trampoline stubs

Hi,

is it absolutely necessary to use the (until now) unused slot or is 
there the chance that we can recycle an existing one? If not, we need 
someone who makes the necessary changes. Any thoughts?

Best,
Albert

On 18.06.2013 21:40, John Rose wrote:
> cords are stored as an array of unsigned shorts (u2).  This slightly 
> simplifies random acce