From nobody Mon May 6 23:34:08 2024 Delivered-To: importer@patchew.org Received-SPF: pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) client-ip=208.118.235.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Authentication-Results: mx.zohomail.com; spf=pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org Return-Path: Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) by mx.zohomail.com with SMTPS id 1502285754364894.3379606276397; Wed, 9 Aug 2017 06:35:54 -0700 (PDT) Received: from localhost ([::1]:47559 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfR9M-0001Pz-4o for importer@patchew.org; Wed, 09 Aug 2017 09:35:52 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59103) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfR7U-0008T8-2S for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:33:58 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dfR7R-0003Q6-MQ for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:33:56 -0400 Received: from mx1.redhat.com ([209.132.183.28]:57750) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dfR7R-0003PG-Db for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:33:53 -0400 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.13]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 5B33AA48E1; Wed, 9 Aug 2017 13:33:52 +0000 (UTC) Received: from t460s.redhat.com (ovpn-117-198.ams2.redhat.com [10.36.117.198]) by smtp.corp.redhat.com (Postfix) with ESMTP id 3926B60A98; Wed, 9 Aug 2017 13:33:50 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 5B33AA48E1 Authentication-Results: ext-mx06.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx06.extmail.prod.ext.phx2.redhat.com; spf=fail smtp.mailfrom=david@redhat.com From: David Hildenbrand To: qemu-devel@nongnu.org Date: Wed, 9 Aug 2017 15:33:41 +0200 Message-Id: <20170809133346.30271-2-david@redhat.com> In-Reply-To: <20170809133346.30271-1-david@redhat.com> References: <20170809133346.30271-1-david@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.13 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.30]); Wed, 09 Aug 2017 13:33:52 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] [fuzzy] X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PATCH RFC 1/6] kvm: require JOIN_MEMORY_REGIONS_WORKS X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Paolo Bonzini , kvm@vger.kernel.org, david@redhat.com Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail: RSF_0 Z_629925259 SPT_0 Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" We already require DESTROY_MEMORY_REGION_WORKS, JOIN_MEMORY_REGIONS_WORKS was added just half a year later. In addition, with flatview overlapping memory regions are first removed before adding the changed one. So we can't really detect joining memory regions this way. Let's just get rid of this special handling. Signed-off-by: David Hildenbrand --- accel/kvm/kvm-all.c | 42 +----------------------------------------- 1 file changed, 1 insertion(+), 41 deletions(-) diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index 46ce479..eae61bb 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -79,7 +79,6 @@ struct KVMState int coalesced_mmio; struct kvm_coalesced_mmio_ring *coalesced_mmio_ring; bool coalesced_flush_in_progress; - int broken_set_mem_region; int vcpu_events; int robust_singlestep; int debugregs; @@ -127,6 +126,7 @@ static bool kvm_immediate_exit; static const KVMCapabilityInfo kvm_required_capabilites[] =3D { KVM_CAP_INFO(USER_MEMORY), KVM_CAP_INFO(DESTROY_MEMORY_REGION_WORKS), + KVM_CAP_INFO(JOIN_MEMORY_REGIONS_WORKS), KVM_CAP_LAST_INFO }; =20 @@ -696,7 +696,6 @@ kvm_check_extension_list(KVMState *s, const KVMCapabili= tyInfo *list) static void kvm_set_phys_mem(KVMMemoryListener *kml, MemoryRegionSection *section, bool add) { - KVMState *s =3D kvm_state; KVMSlot *mem, old; int err; MemoryRegion *mr =3D section->mr; @@ -763,35 +762,6 @@ static void kvm_set_phys_mem(KVMMemoryListener *kml, abort(); } =20 - /* Workaround for older KVM versions: we can't join slots, even no= t by - * unregistering the previous ones and then registering the larger - * slot. We have to maintain the existing fragmentation. Sigh. - * - * This workaround assumes that the new slot starts at the same - * address as the first existing one. If not or if some overlapping - * slot comes around later, we will fail (not seen in practice so = far) - * - and actually require a recent KVM version. */ - if (s->broken_set_mem_region && - old.start_addr =3D=3D start_addr && old.memory_size < size && = add) { - mem =3D kvm_alloc_slot(kml); - mem->memory_size =3D old.memory_size; - mem->start_addr =3D old.start_addr; - mem->ram =3D old.ram; - mem->flags =3D kvm_mem_flags(mr); - - err =3D kvm_set_user_memory_region(kml, mem); - if (err) { - fprintf(stderr, "%s: error updating slot: %s\n", __func__, - strerror(-err)); - abort(); - } - - start_addr +=3D old.memory_size; - ram +=3D old.memory_size; - size -=3D old.memory_size; - continue; - } - /* register prefix slot */ if (old.start_addr < start_addr) { mem =3D kvm_alloc_slot(kml); @@ -833,10 +803,6 @@ static void kvm_set_phys_mem(KVMMemoryListener *kml, } } =20 - /* in case the KVM bug workaround already "consumed" the new slot */ - if (!size) { - return; - } if (!add) { return; } @@ -1692,12 +1658,6 @@ static int kvm_init(MachineState *ms) =20 s->coalesced_mmio =3D kvm_check_extension(s, KVM_CAP_COALESCED_MMIO); =20 - s->broken_set_mem_region =3D 1; - ret =3D kvm_check_extension(s, KVM_CAP_JOIN_MEMORY_REGIONS_WORKS); - if (ret > 0) { - s->broken_set_mem_region =3D 0; - } - #ifdef KVM_CAP_VCPU_EVENTS s->vcpu_events =3D kvm_check_extension(s, KVM_CAP_VCPU_EVENTS); #endif --=20 2.9.4 From nobody Mon May 6 23:34:08 2024 Delivered-To: importer@patchew.org Received-SPF: pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) client-ip=208.118.235.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Authentication-Results: mx.zohomail.com; spf=pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org Return-Path: Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) by mx.zohomail.com with SMTPS id 1502285754435885.8565167438142; Wed, 9 Aug 2017 06:35:54 -0700 (PDT) Received: from localhost ([::1]:47558 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfR9L-0001Pc-H0 for importer@patchew.org; Wed, 09 Aug 2017 09:35:51 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59110) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfR7U-0008TB-Cp for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:33:58 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dfR7T-0003RO-9v for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:33:56 -0400 Received: from mx1.redhat.com ([209.132.183.28]:59414) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dfR7T-0003Qs-0z for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:33:55 -0400 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.13]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id F25146880E; Wed, 9 Aug 2017 13:33:53 +0000 (UTC) Received: from t460s.redhat.com (ovpn-117-198.ams2.redhat.com [10.36.117.198]) by smtp.corp.redhat.com (Postfix) with ESMTP id AD86F60A98; Wed, 9 Aug 2017 13:33:52 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com F25146880E Authentication-Results: ext-mx04.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx04.extmail.prod.ext.phx2.redhat.com; spf=fail smtp.mailfrom=david@redhat.com From: David Hildenbrand To: qemu-devel@nongnu.org Date: Wed, 9 Aug 2017 15:33:42 +0200 Message-Id: <20170809133346.30271-3-david@redhat.com> In-Reply-To: <20170809133346.30271-1-david@redhat.com> References: <20170809133346.30271-1-david@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.13 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.28]); Wed, 09 Aug 2017 13:33:54 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] [fuzzy] X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PATCH RFC 2/6] kvm: factor out alignment of memory section X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Paolo Bonzini , kvm@vger.kernel.org, david@redhat.com Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail: RSF_0 Z_629925259 SPT_0 Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Factor it out, so we can reuse it later. Signed-off-by: David Hildenbrand --- accel/kvm/kvm-all.c | 59 +++++++++++++++++++++++++++++++++++--------------= ---- 1 file changed, 39 insertions(+), 20 deletions(-) diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index eae61bb..f6ceeba 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -190,6 +190,36 @@ static KVMSlot *kvm_lookup_matching_slot(KVMMemoryList= ener *kml, } =20 /* + * Calculate and align the start address and the size of the section. + * Return the size. If the size is 0, the aligned section is empty. + */ +static hwaddr kvm_align_section(MemoryRegionSection *section, + hwaddr *start) +{ + hwaddr size =3D int128_get64(section->size); + hwaddr delta; + + *start =3D section->offset_within_address_space; + + /* kvm works in page size chunks, but the function may be called + with sub-page size and unaligned start address. Pad the start + address to next and truncate size to previous page boundary. */ + delta =3D qemu_real_host_page_size - (*start & ~qemu_real_host_page_ma= sk); + delta &=3D ~qemu_real_host_page_mask; + *start +=3D delta; + if (delta > size) { + return 0; + } + size -=3D delta; + size &=3D qemu_real_host_page_mask; + if (*start & ~qemu_real_host_page_mask) { + return 0; + } + + return size; +} + +/* * Find overlapping slot with lowest start address */ static KVMSlot *kvm_lookup_overlapping_slot(KVMMemoryListener *kml, @@ -700,25 +730,8 @@ static void kvm_set_phys_mem(KVMMemoryListener *kml, int err; MemoryRegion *mr =3D section->mr; bool writeable =3D !mr->readonly && !mr->rom_device; - hwaddr start_addr =3D section->offset_within_address_space; - ram_addr_t size =3D int128_get64(section->size); - void *ram =3D NULL; - unsigned delta; - - /* kvm works in page size chunks, but the function may be called - with sub-page size and unaligned start address. Pad the start - address to next and truncate size to previous page boundary. */ - delta =3D qemu_real_host_page_size - (start_addr & ~qemu_real_host_pag= e_mask); - delta &=3D ~qemu_real_host_page_mask; - if (delta > size) { - return; - } - start_addr +=3D delta; - size -=3D delta; - size &=3D qemu_real_host_page_mask; - if (!size || (start_addr & ~qemu_real_host_page_mask)) { - return; - } + hwaddr start_addr, size; + void *ram; =20 if (!memory_region_is_ram(mr)) { if (writeable || !kvm_readonly_mem_allowed) { @@ -730,7 +743,13 @@ static void kvm_set_phys_mem(KVMMemoryListener *kml, } } =20 - ram =3D memory_region_get_ram_ptr(mr) + section->offset_within_region = + delta; + size =3D kvm_align_section(section, &start_addr); + if (!size) { + return; + } + + ram =3D memory_region_get_ram_ptr(mr) + section->offset_within_region + + (section->offset_within_address_space - start_addr); =20 while (1) { mem =3D kvm_lookup_overlapping_slot(kml, start_addr, start_addr + = size); --=20 2.9.4 From nobody Mon May 6 23:34:08 2024 Delivered-To: importer@patchew.org Received-SPF: pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) client-ip=208.118.235.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Authentication-Results: mx.zohomail.com; spf=pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org Return-Path: Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) by mx.zohomail.com with SMTPS id 150228596019797.87417700570575; Wed, 9 Aug 2017 06:39:20 -0700 (PDT) Received: from localhost ([::1]:47569 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfRCh-0004EE-0d for importer@patchew.org; Wed, 09 Aug 2017 09:39:19 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59138) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfR7W-0008TC-D2 for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:34:03 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dfR7U-0003T0-Ou for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:33:57 -0400 Received: from mx1.redhat.com ([209.132.183.28]:33132) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dfR7U-0003S6-IY for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:33:56 -0400 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.13]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 88893191697; Wed, 9 Aug 2017 13:33:55 +0000 (UTC) Received: from t460s.redhat.com (ovpn-117-198.ams2.redhat.com [10.36.117.198]) by smtp.corp.redhat.com (Postfix) with ESMTP id 4BDD45C684; Wed, 9 Aug 2017 13:33:54 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 88893191697 Authentication-Results: ext-mx10.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx10.extmail.prod.ext.phx2.redhat.com; spf=fail smtp.mailfrom=david@redhat.com From: David Hildenbrand To: qemu-devel@nongnu.org Date: Wed, 9 Aug 2017 15:33:43 +0200 Message-Id: <20170809133346.30271-4-david@redhat.com> In-Reply-To: <20170809133346.30271-1-david@redhat.com> References: <20170809133346.30271-1-david@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.13 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.39]); Wed, 09 Aug 2017 13:33:55 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] [fuzzy] X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PATCH RFC 3/6] kvm: use start + size for memory ranges X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Paolo Bonzini , kvm@vger.kernel.org, david@redhat.com Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail: RSF_0 Z_629925259 SPT_0 Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Convert kvm_lookup_matching_slot(). Signed-off-by: David Hildenbrand --- accel/kvm/kvm-all.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index f6ceeba..bec157e 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -172,7 +172,7 @@ static KVMSlot *kvm_alloc_slot(KVMMemoryListener *kml) =20 static KVMSlot *kvm_lookup_matching_slot(KVMMemoryListener *kml, hwaddr start_addr, - hwaddr end_addr) + hwaddr size) { KVMState *s =3D kvm_state; int i; @@ -180,8 +180,7 @@ static KVMSlot *kvm_lookup_matching_slot(KVMMemoryListe= ner *kml, for (i =3D 0; i < s->nr_slots; i++) { KVMSlot *mem =3D &kml->slots[i]; =20 - if (start_addr =3D=3D mem->start_addr && - end_addr =3D=3D mem->start_addr + mem->memory_size) { + if (start_addr =3D=3D mem->start_addr && size =3D=3D mem->memory_s= ize) { return mem; } } @@ -414,7 +413,7 @@ static int kvm_section_update_flags(KVMMemoryListener *= kml, { hwaddr phys_addr =3D section->offset_within_address_space; ram_addr_t size =3D int128_get64(section->size); - KVMSlot *mem =3D kvm_lookup_matching_slot(kml, phys_addr, phys_addr + = size); + KVMSlot *mem =3D kvm_lookup_matching_slot(kml, phys_addr, size); =20 if (mem =3D=3D NULL) { return 0; --=20 2.9.4 From nobody Mon May 6 23:34:08 2024 Delivered-To: importer@patchew.org Received-SPF: pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) client-ip=208.118.235.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Authentication-Results: mx.zohomail.com; spf=pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org Return-Path: Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) by mx.zohomail.com with SMTPS id 1502285871202438.74119590712064; Wed, 9 Aug 2017 06:37:51 -0700 (PDT) Received: from localhost ([::1]:47563 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfRBG-0003Ah-0X for importer@patchew.org; Wed, 09 Aug 2017 09:37:50 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59161) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfR7Y-0008UX-3m for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:34:01 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dfR7W-0003V2-O8 for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:34:00 -0400 Received: from mx1.redhat.com ([209.132.183.28]:55888) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dfR7W-0003U7-F8 for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:33:58 -0400 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.13]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 6C55094538; Wed, 9 Aug 2017 13:33:57 +0000 (UTC) Received: from t460s.redhat.com (ovpn-117-198.ams2.redhat.com [10.36.117.198]) by smtp.corp.redhat.com (Postfix) with ESMTP id D693E5C684; Wed, 9 Aug 2017 13:33:55 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 6C55094538 Authentication-Results: ext-mx02.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx02.extmail.prod.ext.phx2.redhat.com; spf=fail smtp.mailfrom=david@redhat.com From: David Hildenbrand To: qemu-devel@nongnu.org Date: Wed, 9 Aug 2017 15:33:44 +0200 Message-Id: <20170809133346.30271-5-david@redhat.com> In-Reply-To: <20170809133346.30271-1-david@redhat.com> References: <20170809133346.30271-1-david@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.13 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.26]); Wed, 09 Aug 2017 13:33:57 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] [fuzzy] X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PATCH RFC 4/6] kvm: we never have overlapping slots in kvm_set_phys_mem() X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Paolo Bonzini , kvm@vger.kernel.org, david@redhat.com Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail: RSF_0 Z_629925259 SPT_0 Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Due to how flatviews are handled, we will never have overlapping memory sections in kvm. address_space_update_topology_pass() will make sure that we will only get called for a) an existing memory section for which we only update parameters (log_start, log_stop). b) an existing memory section we want to delete (region_del) c) a brand new memory section we want to add (region_add) We cannot have overlapping memory sections in kvm as we will first remove the overlapping sections and then add the ones without conflicts. Therefore we can remove the complexity for handling prefix and suffix slots. Signed-off-by: David Hildenbrand --- accel/kvm/kvm-all.c | 69 +++++++++----------------------------------------= ---- 1 file changed, 11 insertions(+), 58 deletions(-) diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index bec157e..7b45506 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -725,7 +725,7 @@ kvm_check_extension_list(KVMState *s, const KVMCapabili= tyInfo *list) static void kvm_set_phys_mem(KVMMemoryListener *kml, MemoryRegionSection *section, bool add) { - KVMSlot *mem, old; + KVMSlot *mem; int err; MemoryRegion *mr =3D section->mr; bool writeable =3D !mr->readonly && !mr->rom_device; @@ -750,28 +750,17 @@ static void kvm_set_phys_mem(KVMMemoryListener *kml, ram =3D memory_region_get_ram_ptr(mr) + section->offset_within_region + (section->offset_within_address_space - start_addr); =20 - while (1) { - mem =3D kvm_lookup_overlapping_slot(kml, start_addr, start_addr + = size); + mem =3D kvm_lookup_matching_slot(kml, start_addr, size); + if (!add) { if (!mem) { - break; - } - - if (add && start_addr >=3D mem->start_addr && - (start_addr + size <=3D mem->start_addr + mem->memory_size) && - (ram - start_addr =3D=3D mem->ram - mem->start_addr)) { - /* The new slot fits into the existing one and comes with - * identical parameters - update flags and done. */ - kvm_slot_update_flags(kml, mem, mr); + g_assert(!memory_region_is_ram(mr) && !writeable && !mr->romd_= mode); return; } - - old =3D *mem; - if (mem->flags & KVM_MEM_LOG_DIRTY_PAGES) { kvm_physical_sync_dirty_bitmap(kml, section); } =20 - /* unregister the overlapping slot */ + /* unregister the slot */ mem->memory_size =3D 0; err =3D kvm_set_user_memory_region(kml, mem); if (err) { @@ -779,57 +768,21 @@ static void kvm_set_phys_mem(KVMMemoryListener *kml, __func__, strerror(-err)); abort(); } - - /* register prefix slot */ - if (old.start_addr < start_addr) { - mem =3D kvm_alloc_slot(kml); - mem->memory_size =3D start_addr - old.start_addr; - mem->start_addr =3D old.start_addr; - mem->ram =3D old.ram; - mem->flags =3D kvm_mem_flags(mr); - - err =3D kvm_set_user_memory_region(kml, mem); - if (err) { - fprintf(stderr, "%s: error registering prefix slot: %s\n", - __func__, strerror(-err)); -#ifdef TARGET_PPC - fprintf(stderr, "%s: This is probably because your kernel'= s " \ - "PAGE_SIZE is too big. Please try to use 4= k " \ - "PAGE_SIZE!\n", __func__); -#endif - abort(); - } - } - - /* register suffix slot */ - if (old.start_addr + old.memory_size > start_addr + size) { - ram_addr_t size_delta; - - mem =3D kvm_alloc_slot(kml); - mem->start_addr =3D start_addr + size; - size_delta =3D mem->start_addr - old.start_addr; - mem->memory_size =3D old.memory_size - size_delta; - mem->ram =3D old.ram + size_delta; - mem->flags =3D kvm_mem_flags(mr); - - err =3D kvm_set_user_memory_region(kml, mem); - if (err) { - fprintf(stderr, "%s: error registering suffix slot: %s\n", - __func__, strerror(-err)); - abort(); - } - } + return; } =20 - if (!add) { + if (mem) { + /* update the slot */ + kvm_slot_update_flags(kml, mem, mr); return; } + + /* register the new slot */ mem =3D kvm_alloc_slot(kml); mem->memory_size =3D size; mem->start_addr =3D start_addr; mem->ram =3D ram; mem->flags =3D kvm_mem_flags(mr); - err =3D kvm_set_user_memory_region(kml, mem); if (err) { fprintf(stderr, "%s: error registering slot: %s\n", __func__, --=20 2.9.4 From nobody Mon May 6 23:34:08 2024 Delivered-To: importer@patchew.org Received-SPF: pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) client-ip=208.118.235.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Authentication-Results: mx.zohomail.com; spf=pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org Return-Path: Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) by mx.zohomail.com with SMTPS id 1502285871417949.0894418567613; Wed, 9 Aug 2017 06:37:51 -0700 (PDT) Received: from localhost ([::1]:47564 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfRBG-0003B5-80 for importer@patchew.org; Wed, 09 Aug 2017 09:37:50 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59174) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfR7Z-0008VN-4p for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:34:02 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dfR7Y-0003WB-4Y for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:34:01 -0400 Received: from mx1.redhat.com ([209.132.183.28]:37842) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dfR7X-0003Va-Us for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:34:00 -0400 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.13]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id E9AE43C7A99; Wed, 9 Aug 2017 13:33:58 +0000 (UTC) Received: from t460s.redhat.com (ovpn-117-198.ams2.redhat.com [10.36.117.198]) by smtp.corp.redhat.com (Postfix) with ESMTP id BBA8960A98; Wed, 9 Aug 2017 13:33:57 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com E9AE43C7A99 Authentication-Results: ext-mx03.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx03.extmail.prod.ext.phx2.redhat.com; spf=fail smtp.mailfrom=david@redhat.com From: David Hildenbrand To: qemu-devel@nongnu.org Date: Wed, 9 Aug 2017 15:33:45 +0200 Message-Id: <20170809133346.30271-6-david@redhat.com> In-Reply-To: <20170809133346.30271-1-david@redhat.com> References: <20170809133346.30271-1-david@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.13 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.27]); Wed, 09 Aug 2017 13:33:59 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] [fuzzy] X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PATCH RFC 5/6] kvm: kvm_log_start/stop are only called with known sections X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Paolo Bonzini , kvm@vger.kernel.org, david@redhat.com Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail: RSF_0 Z_629925259 SPT_0 Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Let's properly align the sections first and bail out if we would ever be called with a memory region we don't know yet. Signed-off-by: David Hildenbrand --- accel/kvm/kvm-all.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index 7b45506..78a7f01 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -411,15 +411,21 @@ static int kvm_slot_update_flags(KVMMemoryListener *k= ml, KVMSlot *mem, static int kvm_section_update_flags(KVMMemoryListener *kml, MemoryRegionSection *section) { - hwaddr phys_addr =3D section->offset_within_address_space; - ram_addr_t size =3D int128_get64(section->size); - KVMSlot *mem =3D kvm_lookup_matching_slot(kml, phys_addr, size); + hwaddr start_addr, size; + KVMSlot *mem; =20 - if (mem =3D=3D NULL) { + size =3D kvm_align_section(section, &start_addr); + if (!size) { return 0; - } else { - return kvm_slot_update_flags(kml, mem, section->mr); } + + mem =3D kvm_lookup_matching_slot(kml, start_addr, size); + if (!mem) { + fprintf(stderr, "%s: error finding slot\n", __func__); + abort(); + } + + return kvm_slot_update_flags(kml, mem, section->mr); } =20 static void kvm_log_start(MemoryListener *listener, --=20 2.9.4 From nobody Mon May 6 23:34:08 2024 Delivered-To: importer@patchew.org Received-SPF: pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) client-ip=208.118.235.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Authentication-Results: mx.zohomail.com; spf=pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org Return-Path: Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) by mx.zohomail.com with SMTPS id 1502285766071220.805289625608; Wed, 9 Aug 2017 06:36:06 -0700 (PDT) Received: from localhost ([::1]:47560 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfR9X-0001Ya-NZ for importer@patchew.org; Wed, 09 Aug 2017 09:36:03 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59196) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dfR7e-00007U-02 for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:34:10 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dfR7Z-0003Xc-Nu for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:34:06 -0400 Received: from mx1.redhat.com ([209.132.183.28]:58034) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dfR7Z-0003Wi-F2 for qemu-devel@nongnu.org; Wed, 09 Aug 2017 09:34:01 -0400 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.13]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 76CABA48F7; Wed, 9 Aug 2017 13:34:00 +0000 (UTC) Received: from t460s.redhat.com (ovpn-117-198.ams2.redhat.com [10.36.117.198]) by smtp.corp.redhat.com (Postfix) with ESMTP id 4561E60A98; Wed, 9 Aug 2017 13:33:59 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 76CABA48F7 Authentication-Results: ext-mx06.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx06.extmail.prod.ext.phx2.redhat.com; spf=fail smtp.mailfrom=david@redhat.com From: David Hildenbrand To: qemu-devel@nongnu.org Date: Wed, 9 Aug 2017 15:33:46 +0200 Message-Id: <20170809133346.30271-7-david@redhat.com> In-Reply-To: <20170809133346.30271-1-david@redhat.com> References: <20170809133346.30271-1-david@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.13 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.30]); Wed, 09 Aug 2017 13:34:00 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] [fuzzy] X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PATCH RFC 6/6] kvm: kvm_log_sync() is only called with known memory sections X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Paolo Bonzini , kvm@vger.kernel.org, david@redhat.com Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail: RSF_0 Z_629925259 SPT_0 Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Flatview will make sure that we can only end up in this function with memory sections that correspond to exactly one slot. So we don't have to iterate multiple times. There won't be overlapping slots but only matching slots. Properly align the section and look up the corresponding slot. This heavily simplifies this function. We can now get rid of kvm_lookup_overlapping_slot(). Signed-off-by: David Hildenbrand --- accel/kvm/kvm-all.c | 101 +++++++++++++++++-------------------------------= ---- 1 file changed, 33 insertions(+), 68 deletions(-) diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index 78a7f01..5f1463d 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -218,34 +218,6 @@ static hwaddr kvm_align_section(MemoryRegionSection *s= ection, return size; } =20 -/* - * Find overlapping slot with lowest start address - */ -static KVMSlot *kvm_lookup_overlapping_slot(KVMMemoryListener *kml, - hwaddr start_addr, - hwaddr end_addr) -{ - KVMState *s =3D kvm_state; - KVMSlot *found =3D NULL; - int i; - - for (i =3D 0; i < s->nr_slots; i++) { - KVMSlot *mem =3D &kml->slots[i]; - - if (mem->memory_size =3D=3D 0 || - (found && found->start_addr < mem->start_addr)) { - continue; - } - - if (end_addr > mem->start_addr && - start_addr < mem->start_addr + mem->memory_size) { - found =3D mem; - } - } - - return found; -} - int kvm_physical_memory_addr_from_host(KVMState *s, void *ram, hwaddr *phys_addr) { @@ -489,55 +461,48 @@ static int kvm_physical_sync_dirty_bitmap(KVMMemoryLi= stener *kml, MemoryRegionSection *section) { KVMState *s =3D kvm_state; - unsigned long size, allocated_size =3D 0; struct kvm_dirty_log d =3D {}; KVMSlot *mem; - int ret =3D 0; - hwaddr start_addr =3D section->offset_within_address_space; - hwaddr end_addr =3D start_addr + int128_get64(section->size); + hwaddr start_addr, size; =20 - d.dirty_bitmap =3D NULL; - while (start_addr < end_addr) { - mem =3D kvm_lookup_overlapping_slot(kml, start_addr, end_addr); - if (mem =3D=3D NULL) { - break; - } + size =3D kvm_align_section(section, &start_addr); + if (!size) { + return 0; + } =20 - /* XXX bad kernel interface alert - * For dirty bitmap, kernel allocates array of size aligned to - * bits-per-long. But for case when the kernel is 64bits and - * the userspace is 32bits, userspace can't align to the same - * bits-per-long, since sizeof(long) is different between kernel - * and user space. This way, userspace will provide buffer which - * may be 4 bytes less than the kernel will use, resulting in - * userspace memory corruption (which is not detectable by valgrind - * too, in most cases). - * So for now, let's align to 64 instead of HOST_LONG_BITS here, in - * a hope that sizeof(long) won't become >8 any time soon. - */ - size =3D ALIGN(((mem->memory_size) >> TARGET_PAGE_BITS), - /*HOST_LONG_BITS*/ 64) / 8; - if (!d.dirty_bitmap) { - d.dirty_bitmap =3D g_malloc(size); - } else if (size > allocated_size) { - d.dirty_bitmap =3D g_realloc(d.dirty_bitmap, size); - } - allocated_size =3D size; - memset(d.dirty_bitmap, 0, allocated_size); + mem =3D kvm_lookup_matching_slot(kml, start_addr, size); + if (!mem) { + fprintf(stderr, "%s: error finding slot\n", __func__); + abort(); + } =20 - d.slot =3D mem->slot | (kml->as_id << 16); - if (kvm_vm_ioctl(s, KVM_GET_DIRTY_LOG, &d) =3D=3D -1) { - DPRINTF("ioctl failed %d\n", errno); - ret =3D -1; - break; - } + /* XXX bad kernel interface alert + * For dirty bitmap, kernel allocates array of size aligned to + * bits-per-long. But for case when the kernel is 64bits and + * the userspace is 32bits, userspace can't align to the same + * bits-per-long, since sizeof(long) is different between kernel + * and user space. This way, userspace will provide buffer which + * may be 4 bytes less than the kernel will use, resulting in + * userspace memory corruption (which is not detectable by valgrind + * too, in most cases). + * So for now, let's align to 64 instead of HOST_LONG_BITS here, in + * a hope that sizeof(long) won't become >8 any time soon. + */ + size =3D ALIGN(((mem->memory_size) >> TARGET_PAGE_BITS), + /*HOST_LONG_BITS*/ 64) / 8; + d.dirty_bitmap =3D g_malloc0(size); =20 - kvm_get_dirty_pages_log_range(section, d.dirty_bitmap); - start_addr =3D mem->start_addr + mem->memory_size; + d.slot =3D mem->slot | (kml->as_id << 16); + if (kvm_vm_ioctl(s, KVM_GET_DIRTY_LOG, &d) =3D=3D -1) { + DPRINTF("ioctl failed %d\n", errno); + g_free(d.dirty_bitmap); + return -1; } + + kvm_get_dirty_pages_log_range(section, d.dirty_bitmap); g_free(d.dirty_bitmap); =20 - return ret; + return 0; } =20 static void kvm_coalesce_mmio_region(MemoryListener *listener, --=20 2.9.4