From: "Michael S. Tsirkin" <mst@redhat.com>
To: qemu-devel@nongnu.org
Cc: Peter Maydell <peter.maydell@linaro.org>,
Richard Henderson <rth@twiddle.net>,
Paolo Bonzini <pbonzini@redhat.com>,
Anthony Liguori <aliguori@amazon.com>,
Igor Mammedov <imammedo@redhat.com>
Subject: [Qemu-devel] [PULL 06/15] pc: align DIMM's address/size by backend's alignment value
Date: Sun, 23 Nov 2014 13:17:57 +0200 [thread overview]
Message-ID: <1416741307-32307-7-git-send-email-mst@redhat.com> (raw)
In-Reply-To: <1416741307-32307-1-git-send-email-mst@redhat.com>
From: Igor Mammedov <imammedo@redhat.com>
Performance wise it's better to align GVA by the backend's
page size.
Also do not allow to create DIMM device with suboptimal
size (i.e. not aligned to backends page size) to aviod
memory loss.
Do above only for 2.2 and newer machine types to avoid
breaking working configs with 2.1 machine type.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
include/hw/i386/pc.h | 4 ++++
hw/i386/pc.c | 17 +++++++++++++++++
hw/i386/pc_piix.c | 2 ++
hw/i386/pc_q35.c | 3 +++
4 files changed, 26 insertions(+)
diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index 7c3731f..9d85b89 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -24,6 +24,8 @@
* address space begins.
* @hotplug_memory: hotplug memory addess space container
* @acpi_dev: link to ACPI PM device that performs ACPI hotplug handling
+ * @enforce_aligned_dimm: check that DIMM's address/size is aligned by
+ * backend's alignment value if provided
*/
struct PCMachineState {
/*< private >*/
@@ -38,12 +40,14 @@ struct PCMachineState {
uint64_t max_ram_below_4g;
bool vmport;
+ bool enforce_aligned_dimm;
};
#define PC_MACHINE_ACPI_DEVICE_PROP "acpi-device"
#define PC_MACHINE_MEMHP_REGION_SIZE "hotplug-memory-region-size"
#define PC_MACHINE_MAX_RAM_BELOW_4G "max-ram-below-4g"
#define PC_MACHINE_VMPORT "vmport"
+#define PC_MACHINE_ENFORCE_ALIGNED_DIMM "enforce-aligned-dimm"
/**
* PCMachineClass:
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index 33928b9..021ec44 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -1564,6 +1564,10 @@ static void pc_dimm_plug(HotplugHandler *hotplug_dev,
goto out;
}
+ if (memory_region_get_alignment(mr) && pcms->enforce_aligned_dimm) {
+ align = memory_region_get_alignment(mr);
+ }
+
addr = pc_dimm_get_free_addr(pcms->hotplug_memory_base,
memory_region_size(&pcms->hotplug_memory),
!addr ? NULL : &addr, align,
@@ -1732,6 +1736,13 @@ static void pc_machine_set_vmport(Object *obj, bool value, Error **errp)
pcms->vmport = value;
}
+static bool pc_machine_get_aligned_dimm(Object *obj, Error **errp)
+{
+ PCMachineState *pcms = PC_MACHINE(obj);
+
+ return pcms->enforce_aligned_dimm;
+}
+
static void pc_machine_initfn(Object *obj)
{
PCMachineState *pcms = PC_MACHINE(obj);
@@ -1744,11 +1755,17 @@ static void pc_machine_initfn(Object *obj)
pc_machine_get_max_ram_below_4g,
pc_machine_set_max_ram_below_4g,
NULL, NULL, NULL);
+
pcms->vmport = !xen_enabled();
object_property_add_bool(obj, PC_MACHINE_VMPORT,
pc_machine_get_vmport,
pc_machine_set_vmport,
NULL);
+
+ pcms->enforce_aligned_dimm = true;
+ object_property_add_bool(obj, PC_MACHINE_ENFORCE_ALIGNED_DIMM,
+ pc_machine_get_aligned_dimm,
+ NULL, NULL);
}
static void pc_machine_class_init(ObjectClass *oc, void *data)
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index 7bb97a4..741dffd 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix.c
@@ -305,10 +305,12 @@ static void pc_init_pci(MachineState *machine)
static void pc_compat_2_1(MachineState *machine)
{
+ PCMachineState *pcms = PC_MACHINE(machine);
smbios_uuid_encoded = false;
x86_cpu_compat_set_features("coreduo", FEAT_1_ECX, CPUID_EXT_VMX, 0);
x86_cpu_compat_set_features("core2duo", FEAT_1_ECX, CPUID_EXT_VMX, 0);
x86_cpu_compat_kvm_no_autodisable(FEAT_8000_0001_ECX, CPUID_EXT3_SVM);
+ pcms->enforce_aligned_dimm = false;
}
static void pc_compat_2_0(MachineState *machine)
diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c
index 598e679..e9ba1a2 100644
--- a/hw/i386/pc_q35.c
+++ b/hw/i386/pc_q35.c
@@ -284,6 +284,9 @@ static void pc_q35_init(MachineState *machine)
static void pc_compat_2_1(MachineState *machine)
{
+ PCMachineState *pcms = PC_MACHINE(machine);
+
+ pcms->enforce_aligned_dimm = false;
smbios_uuid_encoded = false;
x86_cpu_compat_set_features("coreduo", FEAT_1_ECX, CPUID_EXT_VMX, 0);
x86_cpu_compat_set_features("core2duo", FEAT_1_ECX, CPUID_EXT_VMX, 0);
--
MST
next prev parent reply other threads:[~2014-11-23 11:18 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-23 11:17 [Qemu-devel] [PULL 00/15] pc, pci, misc bugfixes Michael S. Tsirkin
2014-11-23 11:17 ` [Qemu-devel] [PULL 01/15] qemu-char: fix tcp_get_fds Michael S. Tsirkin
2014-11-23 11:17 ` [Qemu-devel] [PULL 02/15] pc: kvm: check if KVM has free memory slots to avoid abort() Michael S. Tsirkin
2014-11-23 11:17 ` [Qemu-devel] [PULL 03/15] pc: make pc_dimm_plug() more readble Michael S. Tsirkin
2014-11-23 11:17 ` [Qemu-devel] [PULL 04/15] pc: limit DIMM address and size to page aligned values Michael S. Tsirkin
2014-11-23 11:17 ` [Qemu-devel] [PULL 05/15] memory: expose alignment used for allocating RAM as MemoryRegion API Michael S. Tsirkin
2014-11-23 11:17 ` Michael S. Tsirkin [this message]
2014-11-23 11:18 ` [Qemu-devel] [PULL 07/15] pc: pc-dimm: use backend alignment during address auto allocation Michael S. Tsirkin
2014-11-23 11:18 ` [Qemu-devel] [PULL 08/15] pc: explicitly check maxmem limit when adding DIMM Michael S. Tsirkin
2014-11-23 11:18 ` [Qemu-devel] [PULL 09/15] pc: count in 1Gb hugepage alignment when sizing hotplug-memory container Michael S. Tsirkin
2014-11-23 11:18 ` [Qemu-devel] [PULL 10/15] hw/pci: fix crash on shpc error flow Michael S. Tsirkin
2014-11-23 11:18 ` [Qemu-devel] [PULL 11/15] acpi-build: mark RAM dirty on table update Michael S. Tsirkin
2014-11-23 11:18 ` [Qemu-devel] [PULL 12/15] target-i386: move generic memory hotplug methods to DSDTs Michael S. Tsirkin
2014-11-23 11:18 ` [Qemu-devel] [PULL 13/15] pcie: fix typo in pcie_cap_deverr_init() Michael S. Tsirkin
2014-11-23 11:18 ` [Qemu-devel] [PULL 14/15] pcie: fix improper use of negative value Michael S. Tsirkin
2014-11-23 11:18 ` [Qemu-devel] [PULL 15/15] pc: acpi: mark all possible CPUs as enabled in SRAT Michael S. Tsirkin
2014-11-24 13:07 ` [Qemu-devel] [PULL 00/15] pc, pci, misc bugfixes Peter Maydell
2014-11-24 14:16 ` [Qemu-devel] [PATCH 08/15] fixup! pc: explicitly check maxmem limit when adding DIMM Igor Mammedov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1416741307-32307-7-git-send-email-mst@redhat.com \
--to=mst@redhat.com \
--cc=aliguori@amazon.com \
--cc=imammedo@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).