xfs
[Top] [All Lists]

Re: [PATCH] fix dir2 shortform structures on ARM old ABI

To: "Josef 'Jeff' Sipek" <jeffpc@xxxxxxxxxxxxxx>
Subject: Re: [PATCH] fix dir2 shortform structures on ARM old ABI
From: Eric Sandeen <sandeen@xxxxxxxxxxx>
Date: Mon, 17 Mar 2008 15:04:13 -0500
Cc: xfs-oss <xfs@xxxxxxxxxxx>
In-reply-to: <20080317195313.GB16500@xxxxxxxxxxxxxx>
References: <47DB4181.7040603@xxxxxxxxxxx> <20080315041722.GA25621@xxxxxxxxxxxxxx> <47DB4F4F.8030407@xxxxxxxxxxx> <20080315042703.GA28242@xxxxxxxxxxxxxx> <47DB51A3.70200@xxxxxxxxxxx> <20080315045147.GB28242@xxxxxxxxxxxxxx> <47DEB930.7020108@xxxxxxxxxxx> <20080317195313.GB16500@xxxxxxxxxxxxxx>
Sender: xfs-bounce@xxxxxxxxxxx
User-agent: Thunderbird 2.0.0.12 (Macintosh/20080213)
Josef 'Jeff' Sipek wrote:
> On Mon, Mar 17, 2008 at 01:32:16PM -0500, Eric Sandeen wrote:
>> Josef 'Jeff' Sipek wrote:
>>
>>> Josef 'Jeff' Sipek, wondering exactly how passionate one can get about
>>> structure member alignment :)
>> Very.  ;)
>>
>> Tossing packed at all the ondisk stuctures bloats things badly on ia64.
>>
>> cvs/linux-2.6-xfs> wc -l before.dis
>> 166688 before.dis
>> cvs/linux-2.6-xfs> wc -l after.dis
>> 182294 after.dis
>>
>> That's +15606 lines.
>  
> I'm not done yet! :-P
> 
> First of all, the patch I showed you actually breaks a few things that I
> still need to fix.

Oh, I wasn't trying to blame you or our patch specifically, just wanted
to highlight what I consider to be the bad idea of giving gcc a bunch of
directives that IMHO we don't need.

> Second, I need to find out whether all the affected structures are always
> aligned on some boundary (probably 4 or 8 byte). If there indeed is some
> alignment, there might be a way to reduce those 15k extra lines to something
> a whole lot less - I hope.

To what end?  What are you trying to fix?  If it's not reduced to 0 then
your change is introducing regressions, IMHO.

Respectfully, ;)
-Eric


<Prev in Thread] Current Thread [Next in Thread>