From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ozlabs.org (ozlabs.org [IPv6:2401:3900:2:1::2]) (using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 3w6WLQ250KzDq5b for ; Tue, 18 Apr 2017 13:47:58 +1000 (AEST) From: Michael Ellerman To: Anshuman Khandual , Michal Suchanek , linuxppc-dev@lists.ozlabs.org Cc: "'Kirill A. Shutemov'" Subject: Re: [PATCH] selftests/powerpc/mm: Add a test for virtual address mapping In-Reply-To: References: <20170412094112.13181-1-khandual@linux.vnet.ibm.com> <20170414221857.58171fce@neko> Date: Tue, 18 Apr 2017 13:47:54 +1000 Message-ID: <87y3uy74wl.fsf@concordia.ellerman.id.au> MIME-Version: 1.0 Content-Type: text/plain List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Anshuman Khandual writes: > On 04/15/2017 01:48 AM, Michal Suchanek wrote: >> Hello, >> >> On Wed, 12 Apr 2017 15:11:12 +0530 >> Anshuman Khandual wrote: >> >>> This verifies virtual address mapping below and above the >>> 128TB range and makes sure that address returned are within >>> the expected range depending upon the hint passed from the >>> user space. >> >> This description does not mention anything PPC specific. Can this be >> generalized to work with Intel L5 enabled kernel as well? > > Hmm, I am not completely aware about the proposed mmap interface > for 128TB and beyond till 512TB by Kirill on x86. If both powerpc > and x86 will have the same semantics, then this test is generic > enough to cover both the architectures and will move it into > selftests/vm/ instead of current selftests/powerpc/mm. I think it can be made generic, maybe it needs a few #ifdefs, but that's fine for a selftest. So yeah if you can move it to selftests/vm I think that would be better. cheers