linux-kernel - Re: RFC: remove sn2, hpsim and ia64 machvecs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20190807134401.GA14792@lst.de>
Date:   Wed, 7 Aug 2019 15:44:01 +0200
From:   Christoph Hellwig <hch@....de>
To:     John Paul Adrian Glaubitz <glaubitz@...sik.fu-berlin.de>
Cc:     Christoph Hellwig <hch@....de>, Tony Luck <tony.luck@...el.com>,
        Fenghua Yu <fenghua.yu@...el.com>,
        Arnd Bergmann <arnd@...db.de>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        linux-ia64@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: RFC: remove sn2, hpsim and ia64 machvecs

On Wed, Aug 07, 2019 at 03:35:37PM +0200, John Paul Adrian Glaubitz wrote:
> Hi Christoph!
> 
> On 8/7/19 3:30 PM, Christoph Hellwig wrote:
> > let me know what you think of this series.  This drops the pretty much
> > dead sn2 and hpsim support, which then allows us to build a single ia64
> > kernel image that supports all remaining systems without extra indirections
> > in the fast path.
> 
> Interesting. Does that mean Debian no longer needs to maintain two different
> kernels for ia64, currently named "itanium" and "mckinley"?

I don't think so.  Assumeing one of them sets CONFIG_ITANIUM and the
other just CONFIG_MCKINLEY, the first one already covers everything that
the second can.  But assuming both set CONFIG_IA64_GENERIC they should
both get a little faster as a lot of indirect calls there just exist for
SN2 are gone with this.  And if my memory serves me right indirect calls
have always been rather expensive on ia64 vs other architectures (at least
before spectre mitigations made others slower as well).

> 
> Adrian
> 
> -- 
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaubitz@...ian.org
> `. `'   Freie Universitaet Berlin - glaubitz@...sik.fu-berlin.de
>   `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
---end quoted text---