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: <20160624104619.GA4378@localhost.localdomain>
Date:	Fri, 24 Jun 2016 06:46:20 -0400
From:	Konrad Rzeszutek Wilk <konrad@...nok.org>
To:	Jisheng Zhang <jszhang@...vell.com>
Cc:	Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
	Catalin Marinas <catalin.marinas@....com>,
	Manjeet Pawar <manjeet.p@...sung.com>,
	linux-doc@...r.kernel.org, corbet@....net, will.deacon@....com,
	linux-kernel@...r.kernel.org, ajeet.y@...sung.com,
	akhilesh.k@...sung.com, r.thapliyal@...sung.com,
	linux-arm-kernel@...ts.infradead.org, pankaj.m@...sung.com
Subject: Re: [PATCH] arm64:swiotlb:Enable only when Input size through
 command line

On Fri, Jun 24, 2016 at 10:57:29AM +0800, Jisheng Zhang wrote:
> Dear Konrad,
> 
> On Thu, 23 Jun 2016 12:06:10 -0400 Konrad Rzeszutek Wilk wrote:
> 
> > On June 23, 2016 10:30:34 AM EDT, Catalin Marinas <catalin.marinas@....com> wrote:
> > >On Thu, Jun 23, 2016 at 05:43:40PM +0530, Manjeet Pawar wrote:  
> > >> From: Rohit Thapliyal <r.thapliyal@...sung.com>
> > >> 
> > >> swiotlb default size of 64M is too big as
> > >> default value therefore it is made configurable
> > >> through command line through swiotlb_size parameter.
> > >> swiotlb allocation shall be done only when the
> > >> swiotlb size is given through command line.
> > >> Otherwise no swiotlb is allocated.  
> > >
> > >I already queued this patch:
> > >
> > >http://lkml.kernel.org/g/1465372426-4077-1-git-send-email-jszhang@marvell.com
> > >
> > >If you have any objections to it, please reply there.  
> > 
> > 
> > I do (sorry about duplicate email, the other got rejected by mailing lists).
> > 
> > Why not expand the swiotlb= parameter instead of introducing a new one?
> 
> Do you mean pass "swiotlb=" for those platforms(most probably, arm64 with less
> than 4GB DDR) which don't need swiotlb? I'm afraid this is not convenient, and

Why not just have a function that checks the amount of memory? x86 has
that - if it finds that the machine has less than 4GB it will not setup
SWIOTLB?

> users even don't notice swiotlb parameter. From another side, pass "swiotlb=0"
> will make the swiotlb reserve 64MB instead, so how can we achieve zero reserved
> memory for swiotlb through "swiotlb=" parameter?

Obviously make the function understand that 0 is to turn it off.
> 
> PS: my patch didn't introduce new boot parameter.

swiotlb_sz ?

> 
> I'm not sure I got your meaning, so could you please comment my patch
> directly?
> 
> Thanks,
> Jisheng
> 
> > 
> > Also, why not use the swiotlb by itself? That does the job as well?
> > 
> > 
> > _______________________________________________
> > linux-arm-kernel mailing list
> > linux-arm-kernel@...ts.infradead.org
> > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
> 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ