From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:45473) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UoFRZ-00008h-HY for qemu-devel@nongnu.org; Sun, 16 Jun 2013 12:04:42 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1UoFRX-0000u7-Rj for qemu-devel@nongnu.org; Sun, 16 Jun 2013 12:04:41 -0400 Received: from e9.ny.us.ibm.com ([32.97.182.139]:45865) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UoFRX-0000u1-Nb for qemu-devel@nongnu.org; Sun, 16 Jun 2013 12:04:39 -0400 Received: from /spool/local by e9.ny.us.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Sun, 16 Jun 2013 12:04:39 -0400 Received: from d01relay05.pok.ibm.com (d01relay05.pok.ibm.com [9.56.227.237]) by d01dlp03.pok.ibm.com (Postfix) with ESMTP id 63C01C90026 for ; Sun, 16 Jun 2013 12:04:37 -0400 (EDT) Received: from d01av02.pok.ibm.com (d01av02.pok.ibm.com [9.56.224.216]) by d01relay05.pok.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id r5GG4cJM345628 for ; Sun, 16 Jun 2013 12:04:38 -0400 Received: from d01av02.pok.ibm.com (loopback [127.0.0.1]) by d01av02.pok.ibm.com (8.14.4/8.13.1/NCO v10.0 AVout) with ESMTP id r5GG4b7O002827 for ; Sun, 16 Jun 2013 13:04:37 -0300 From: Anthony Liguori In-Reply-To: <1371397053-4503-4-git-send-email-lilei@linux.vnet.ibm.com> References: <1371397053-4503-1-git-send-email-lilei@linux.vnet.ibm.com> <1371397053-4503-4-git-send-email-lilei@linux.vnet.ibm.com> Date: Sun, 16 Jun 2013 11:04:34 -0500 Message-ID: <878v2ahwj1.fsf@codemonkey.ws> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: Re: [Qemu-devel] [PATCH 3/7] arch_init: add ram_madvise_free() List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Lei Li , qemu-devel@nongnu.org Cc: lagarcia@br.ibm.com, quintela@redhat.com Lei Li writes: > Signed-off-by: Lei Li > --- > arch_init.c | 13 +++++++++++++ > include/migration/migration.h | 3 +++ > 2 files changed, 16 insertions(+), 0 deletions(-) > > diff --git a/arch_init.c b/arch_init.c > index 872020e..fc66bd2 100644 > --- a/arch_init.c > +++ b/arch_init.c > @@ -524,6 +524,19 @@ uint64_t ram_bytes_total(void) > return total; > } > > +void ram_madvise_free(ram_addr_t size) > +{ > + void *ram; > + RAMBlock *block = NULL; > + > + ram = memory_region_get_ram_ptr(block->mr); > + > + /* XXX. Here just simplely madvise(.., MADV_DONTNEED) the whole ram > + * pages, need more work to keep MADV_DONTNEED ram pages that > + * already sent. */ > + qemu_madvise(ram, size, MADV_DONTNEED); > +} > + I don't think this is right at all. There's no guarantee we have a single linear mapping of all ram. I think you need something a bit more clever than this. > static void migration_end(void) > { > if (migration_bitmap) { > diff --git a/include/migration/migration.h b/include/migration/migration.h > index 8866c3c..9cc5285 100644 > --- a/include/migration/migration.h > +++ b/include/migration/migration.h > @@ -21,6 +21,7 @@ > #include "qapi/error.h" > #include "migration/vmstate.h" > #include "qapi-types.h" > +#include "exec/memory.h" > > enum { > MIG_STATE_ERROR, > @@ -100,6 +101,8 @@ uint64_t ram_bytes_remaining(void); > uint64_t ram_bytes_transferred(void); > uint64_t ram_bytes_total(void); > > +void ram_madvise_free(ram_addr_t size); > + If you introduce new interfaces, please include documentation in the header. Regards, Anthony Liguori > extern SaveVMHandlers savevm_ram_handlers; > > uint64_t dup_mig_bytes_transferred(void); > -- > 1.7.7.6