From mboxrd@z Thu Jan 1 00:00:00 1970 From: Julien Grall Subject: Re: [PATCH v4 08/11] xen/arm: Handle remove foreign mapping Date: Mon, 16 Dec 2013 17:06:20 +0000 Message-ID: <52AF330C.3050305@linaro.org> References: <1386963461-6520-1-git-send-email-julien.grall@linaro.org> <1386963461-6520-9-git-send-email-julien.grall@linaro.org> <20131216115131.GB35881@deinos.phlegethon.org> <52AF1D6E.7080402@linaro.org> <1387208403.21086.12.camel@kazak.uk.xensource.com> <52AF29C3.1030700@linaro.org> <1387211584.21086.27.camel@kazak.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: Received: from mail6.bemta5.messagelabs.com ([195.245.231.135]) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1Vsbce-0008IQ-JS for xen-devel@lists.xenproject.org; Mon, 16 Dec 2013 17:06:24 +0000 Received: by mail-ee0-f45.google.com with SMTP id d49so2379065eek.4 for ; Mon, 16 Dec 2013 09:06:22 -0800 (PST) In-Reply-To: <1387211584.21086.27.camel@kazak.uk.xensource.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Ian Campbell Cc: Keir Fraser , patches@linaro.org, Tim Deegan , stefano.stabellini@citrix.com, Jan Beulich , xen-devel@lists.xenproject.org List-Id: xen-devel@lists.xenproject.org On 12/16/2013 04:33 PM, Ian Campbell wrote: > On Mon, 2013-12-16 at 16:26 +0000, Julien Grall wrote: >> I have reworked this patch. I get a simpler patch: >> >> commit aab2e5d2ae7d0fa87c74cae2f22044f87be33f70 >> Author: Julien Grall >> Date: Fri Dec 13 16:51:03 2013 +0000 >> >> xen/arm: Handle remove foreign mapping >> >> Modify get_page_from_gfn to take reference on foreign mapping. This will avoid >> specific handling in the common code. >> >> Signed-off-by: Julien Grall >> >> --- >> Changes in v5: >> - Remove specific p2m handling in common code >> - Handle foreign mapping in get_page_from_gfn >> Changes in v4: >> - Split patch #6 from dom0 pvh series v6.2 to retrieve only common >> code. >> - Rework commit title >> - Rename xen_rem_foreign_from_p2m in p2m_remove_foreign >> - Get the mfn from the pte. We are not sure that maddr given in >> parameters is valid >> Changes in v3: >> - Move put_page in create_p2m_entries >> - Move xenmem_rem_foreign_from_p2m in arch/arm/p2m.c >> Changes in v2: >> - Introduce the patch >> >> diff --git a/xen/arch/arm/p2m.c b/xen/arch/arm/p2m.c >> index 39d8a03..f7bd7e2 100644 >> --- a/xen/arch/arm/p2m.c >> +++ b/xen/arch/arm/p2m.c >> @@ -317,10 +317,21 @@ static int create_p2m_entries(struct domain *d, >> break; >> case REMOVE: >> { >> - lpae_t pte; >> + lpae_t pte = third[third_table_offset(addr)]; >> + unsigned long mfn; >> + >> + maddr = (pte.bits & PADDR_MASK & PAGE_MASK); > > I thought we had a macro for this, but apparently not. While looking for > it I spotted that x86 has pte_to_mfn, which sounds like a useful > innovation... (not essential as part of this series though). > >> + mfn = paddr_to_pfn(maddr); >> + >> + /* TODO: Handle other p2m type */ >> + if ( pte.p2m.valid && p2m_is_foreign(pte.p2m.type) ) >> + { >> + ASSERT(mfn_valid(mfn)); > > Something somewhere is making sure we don't put foreign MMIO regions > into the p2m, right? I misread this part. And the answer is still yes because in this case MMIO won't belong to a domain (there is no reference on it), so get_page will return NULL when the foreign mapping is created in xenmem_add_to_physmap_one. >> + put_page(mfn_to_page(mfn)); >> + } >> + >> memset(&pte, 0x00, sizeof(pte)); >> write_pte(&third[third_table_offset(addr)], pte); >> - maddr += PAGE_SIZE; >> } >> break; >> } >> diff --git a/xen/include/asm-arm/p2m.h b/xen/include/asm-arm/p2m.h >> index 0eb07a8..e0b58da 100644 >> --- a/xen/include/asm-arm/p2m.h >> +++ b/xen/include/asm-arm/p2m.h >> @@ -122,9 +122,21 @@ static inline struct page_info *get_page_from_gfn( >> if ( !mfn_valid(mfn) ) >> return NULL; >> page = mfn_to_page(mfn); >> - if ( !get_page(page, d) ) >> - return NULL; >> - return page; >> + >> + if ( get_page(page, d) ) > > This isn't noisy (even at debug level) on failure, I thought so? > > Might be safer (and TBH more logical) to move it after the foreign > special case. > >> + return page; >> + >> + /* get_page won't work on foreign mapping because the page doesn't >> + * belong to the current domain. >> + */ >> + if ( p2mt == p2m_map_foreign ) >> + { >> + struct domain *fdom = page_get_owner_and_reference(page); >> + ASSERT(fdom != NULL); > > ASSERT(fdom != d) > ? > >> + return page; >> + } >> + >> + return NULL; >> } >> >> int get_page_type(struct page_info *page, unsigned long type); >> > > -- Julien Grall