SeaBIOS July 2012

seabios@seabios.org

37 participants
119 discussions

Re: [SeaBIOS] [Qemu-devel] [QEMU RFC PATCH 5/7] pc: write lapic info (apic IDs) to fw_cfg so seabios can use it
by Eduardo Habkost 13 Jul '12

13 Jul '12

On Thu, Jul 12, 2012 at 07:29:02PM +0000, Blue Swirl wrote: [...] > CODING_STYLE requires CamelCase, also below. > [...] > > + struct lapic_info_table *lapic_table = g_malloc0(lapic_info_size);; > > ;-- > > > + lapic_table->count = max_cpus; > > + for (i = 0; i < max_cpus; i++) { > > + lapic_table->entries[i].apic_id = apic_id_for_cpu(i); > > + } > > + fw_cfg_add_bytes(fw_cfg, FW_CFG_LAPIC_INFO, (uint8_t *)lapic_table, > > You are passing a structure with host endianness. Please convert the > fields with cpu_to_le32(). Thanks. I will fix the issues if I send a new version of this patch. But I believe we'll find a way to make FW_CFG_LAPIC_INFO unnecessary. -- Eduardo

1 0

Re: [SeaBIOS] [Qemu-devel] [RFC PATCH v2 00/21] ACPI memory hotplug
by Vasilis Liaskovitis 13 Jul '12

13 Jul '12

On Thu, Jul 12, 2012 at 08:04:56PM +0000, Blue Swirl wrote: > On Wed, Jul 11, 2012 at 10:31 AM, Vasilis Liaskovitis > <vasilis.liaskovitis(a)profitbricks.com> wrote: > > This is v2 of the ACPI memory hotplug prototype for x86_64 target. > > I think the concept of DIMMs (what about SIMMs? SODIMMs? I liked > memslot) would be useful for most targets, but hotplugging may be > limited to x86 only. It would be nice to keep these two separate or as > loosely coupled as possible. agreed. what specific usecases besides hotplugging are you thinking about? Also are there non-acpi hotplug platforms? I am trying to keep generic dimm manipulation functions (e.g. population / depopulation and searching) in hw/dimm[.ch]. Currently the x86-acpi_piix4 "backend" registers a callback for hot-add / hot-remove. In theory other hotplug backends can hook in. btw I don't mind using "-memslot" (I think someone during v1 mentioned -dimm), we just need some consensus on the naming. > > > > > Changes v1->v2 > > > > - memory map is automatically calculated for hotplug dimms. Dimms are added from > > top-of-memory skipping the pci hole at [PCI_HOLE_START, 4G). > > - Renamed from "-memslot" to "-dimm". Commands changed to "dimm_add", "dimm_del". > > - Seabios ejection array reduced to a byte. Use extraction macros for dimm ssdt. > > - additional SRAT paravirt info does not break previous SRAT fw_cfg layout. > > - Documentation of new acpi_piix4 registers and paravirt data. > > - add ACPI _OST support for _OST enabled guests. This allows qemu to receive > > notification for success / failure of memory hot-add and hot-remove operations. > > Guest needs to support _OST (https://lkml.org/lkml/2012/6/25/321) > > - add monitor info command to report total guest memory (initial + hot-added) > > - add command line options and monitor commands for batch dimm creation/population > > > > Overview: > > > > Dimm devices are modeled with a new qemu command line > > > > "-dimm id=name,size=sz,node=pxm,populated=on|off" > > > > As already mentioned, the starting physical address for all dimms is calculated > > automatically from top of memory, skipping the pci hole at [PCI_HOLE_START, 4G). > > Node is defining numa proximity for this dimm. When not defined it defaults > > to zero. > > "-dimm id=dimm0,size=512M,node=0,populated=off" > > will define a 512M memory slot belonging to numa node 0. > > > > Dimms are added or removed with a new hmp command "dimm_add/dimm_del": > > Hot-add syntax: "dimm_add id" > > Hot-remove syntax: "dimm_del id" > > > > Issues: > > > > - Live migration works as long as populated field is changed to "on" for > > hotplugged dimms at the destination qemu command line (patch 12/21 lifts > > this requirement). The DimmState structure does not yet define a > > VMStateDescription, but i assume this is the preferred way to pass state > > for migration. > > > > - Dimms are abstracted as qdevices attached to the main system bus. However, > > memory hotplugging has its own side channel ignoring main_system_bus's hotplug > > incapability. A cleaner integration is still needed, probably attaching memory > > devices as children-links of an acpi-capable device (in the pc case acpi_piix4) > > instead of the system bus (TBD). Then device_add/device_del instead of new > > commands can hopefully be used. > > > > Comments/review welcome. > > > > series is based on uq/master for qemu-kvm, and master for seabios. Can be found > > also at: > > http://github.com/vliaskov/qemu-kvm/commits/memhp-v2 > > http://github.com/vliaskov/seabios/commits/memhp-v2 > > > > Vasilis Liaskovitis (14): > > dimm: Implement memory device abstraction > > acpi_piix4: Implement memory device hotplug registers > > pc: calculate dimm physical addresses and adjust memory map > > pc: Add dimm paravirt SRAT info > > Implement "-dimm" command line option > > Implement dimm_add and dimm_del commands for hmp and qmp > > fix live-migration when "populated=on" is missing > > Implement memory hotplug notification lists > > acpi_piix4: _OST dimm support > > acpi_piix4: Update dimm state on VM reboot > > acpi_piix4: Update dimm bitmap state on hot-remove fail > > Implement "info memtotal" and "query-memtotal" > > Implement -dimms, -dimmspop command line options > > Implement mem_increase, mem_decrease hmp/qmp commands > > > > arch_init.c | 23 ++- > > docs/specs/acpi_hotplug.txt | 46 +++++ > > docs/specs/fwcfg.txt | 28 +++ > > hmp-commands.hx | 67 +++++++ > > hmp.c | 24 +++ > > hmp.h | 2 + > > hw/Makefile.objs | 2 +- > > hw/acpi_piix4.c | 131 ++++++++++++- > > hw/dimm.c | 449 +++++++++++++++++++++++++++++++++++++++++++ > > hw/dimm.h | 72 +++++++ > > hw/pc.c | 94 +++++++++- > > hw/pc.h | 6 + > > hw/pc_piix.c | 18 ++- > > monitor.c | 35 ++++ > > monitor.h | 5 + > > qapi-schema.json | 38 ++++ > > qemu-config.c | 70 +++++++ > > qemu-options.hx | 15 ++ > > qmp-commands.hx | 137 +++++++++++++ > > sysemu.h | 1 + > > vl.c | 122 ++++++++++++- > > 21 files changed, 1368 insertions(+), 17 deletions(-) > > create mode 100644 docs/specs/acpi_hotplug.txt > > create mode 100644 docs/specs/fwcfg.txt > > create mode 100644 hw/dimm.c > > create mode 100644 hw/dimm.h > > > > Vasilis Liaskovitis (7): > > Add ACPI_EXTRACT_DEVICE* macros > > Add SSDT memory device support > > acpi-dsdt: Implement functions for memory hotplug. > > acpi: generate hotplug memory devices. > > pciinit: Fix pcimem_start value > > acpi_dsdt: Support _OST dimm method > > acpi_dsdt: Revert internal dimm state on _OST failure > > > > Makefile | 2 +- > > src/acpi-dsdt.dsl | 120 ++++++++++++++++++++++++++++++++++++- > > src/acpi.c | 158 +++++++++++++++++++++++++++++++++++++++++++++++-- > > src/pciinit.c | 2 +- > > src/ssdt-mem.dsl | 69 +++++++++++++++++++++ > > tools/acpi_extract.py | 28 +++++++++ > > 6 files changed, 369 insertions(+), 10 deletions(-) > > create mode 100644 src/ssdt-mem.dsl > >

1 0

Re: [SeaBIOS] [Qemu-devel] [RFC PATCH v2 09/21] pc: Add dimm paravirt SRAT info
by Vasilis Liaskovitis 13 Jul '12

13 Jul '12

On Thu, Jul 12, 2012 at 07:48:04PM +0000, Blue Swirl wrote: > On Wed, Jul 11, 2012 at 10:31 AM, Vasilis Liaskovitis > <vasilis.liaskovitis(a)profitbricks.com> wrote: > > The numa_fw_cfg paravirt interface is extended to include SRAT information for > > all hotplug-able dimms. There are 3 words for each hotplug-able memory slot, > > denoting start address, size and node proximity. The new info is appended after > > existing numa info, so that the fw_cfg layout does not break. This information > > is used by Seabios to build hotplug memory device objects at runtime. > > nb_numa_nodes is set to 1 by default (not 0), so that we always pass srat info > > to SeaBIOS. > > > > v1->v2: > > Dimm SRAT info (#dimms) is appended at end of existing numa fw_cfg in order not > > to break existing layout > > Documentation of the new fwcfg layout is included in docs/specs/fwcfg.txt > > > > Signed-off-by: Vasilis Liaskovitis <vasilis.liaskovitis(a)profitbricks.com> > > --- > > docs/specs/fwcfg.txt | 28 ++++++++++++++++++++++++++ > > hw/pc.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++++- > > vl.c | 2 +- > > 3 files changed, 80 insertions(+), 3 deletions(-) > > create mode 100644 docs/specs/fwcfg.txt > > > > diff --git a/docs/specs/fwcfg.txt b/docs/specs/fwcfg.txt > > new file mode 100644 > > index 0000000..e6fcd8f > > --- /dev/null > > +++ b/docs/specs/fwcfg.txt > > @@ -0,0 +1,28 @@ > > +QEMU<->BIOS Paravirt Documentation > > +-------------------------------------- > > + > > +This document describes paravirt data structures passed from QEMU to BIOS. > > + > > +fw_cfg SRAT paravirt info > > +-------------------- > > +The SRAT info passed from QEMU to BIOS has the following layout: > > + > > +----------------------------------------------------------------------------------------------- > > +#nodes | cpu0_pxm | cpu1_pxm | ... | cpulast_pxm | node0_mem | node1_mem | ... | nodelast_mem > > + > > +----------------------------------------------------------------------------------------------- > > +#dimms | dimm0_start | dimm0_sz | dimm0_pxm | ... | dimmlast_start | dimmlast_sz | dimmlast_pxm > > + > > +Entry 0 contains the number of numa nodes (nb_numa_nodes). > > + > > +Entries 1..max_cpus: The next max_cpus entries describe node proximity for each > > +one of the vCPUs in the system. > > + > > +Entries max_cpus+1..max_cpus+nb_numa_nodes+1: The next nb_numa_nodes entries > > +describe the memory size for each one of the NUMA nodes in the system. > > + > > +Entry max_cpus+nb_numa_nodes+1 contains the number of memory dimms (nb_hp_dimms) > > + > > +The last 3 * nb_hp_dimms entries are organized in triplets: Each triplet contains > > +the physical address offset, size (in bytes), and node proximity for the > > +respective dimm. > > The size and endianness are not specified, you are using LE 64 bit > values for each item. thanks, I 'll update. > > > diff --git a/hw/pc.c b/hw/pc.c > > index ef9901a..cf651d0 100644 > > --- a/hw/pc.c > > +++ b/hw/pc.c > > @@ -598,12 +598,15 @@ int e820_add_entry(uint64_t address, uint64_t length, uint32_t type) > > return index; > > } > > > > +static void setup_hp_dimms(uint64_t *fw_cfg_slots); > > + > > static void *bochs_bios_init(void) > > { > > void *fw_cfg; > > uint8_t *smbios_table; > > size_t smbios_len; > > uint64_t *numa_fw_cfg; > > + uint64_t *hp_dimms_fw_cfg; > > int i, j; > > > > register_ioport_write(0x400, 1, 2, bochs_bios_write, NULL); > > @@ -638,8 +641,10 @@ static void *bochs_bios_init(void) > > /* allocate memory for the NUMA channel: one (64bit) word for the number > > * of nodes, one word for each VCPU->node and one word for each node to > > * hold the amount of memory. > > + * Finally one word for the number of hotplug memory slots and three words > > + * for each hotplug memory slot (start address, size and node proximity). > > */ > > - numa_fw_cfg = g_malloc0((1 + max_cpus + nb_numa_nodes) * 8); > > + numa_fw_cfg = g_malloc0((2 + max_cpus + nb_numa_nodes + 3 * nb_hp_dimms) * 8); > > numa_fw_cfg[0] = cpu_to_le64(nb_numa_nodes); > > for (i = 0; i < max_cpus; i++) { > > for (j = 0; j < nb_numa_nodes; j++) { > > @@ -652,8 +657,15 @@ static void *bochs_bios_init(void) > > for (i = 0; i < nb_numa_nodes; i++) { > > numa_fw_cfg[max_cpus + 1 + i] = cpu_to_le64(node_mem[i]); > > } > > + > > + numa_fw_cfg[1 + max_cpus + nb_numa_nodes] = cpu_to_le64(nb_hp_dimms); > > + > > + hp_dimms_fw_cfg = numa_fw_cfg + 2 + max_cpus + nb_numa_nodes; > > + if (nb_hp_dimms) > > + setup_hp_dimms(hp_dimms_fw_cfg); > > Braces. > > > + > > fw_cfg_add_bytes(fw_cfg, FW_CFG_NUMA, (uint8_t *)numa_fw_cfg, > > - (1 + max_cpus + nb_numa_nodes) * 8); > > + (2 + max_cpus + nb_numa_nodes + 3 * nb_hp_dimms) * 8); > > > > return fw_cfg; > > } > > @@ -1223,3 +1235,40 @@ target_phys_addr_t pc_set_hp_memory_offset(uint64_t size) > > > > return ret; > > } > > + > > +static void setup_hp_dimms(uint64_t *fw_cfg_slots) > > +{ > > + int i = 0; > > + Error *err = NULL; > > + DeviceState *dev; > > + DimmState *slot; > > + const char *type; > > + BusChild *kid; > > + BusState *bus = sysbus_get_default(); > > + > > + QTAILQ_FOREACH(kid, &bus->children, sibling) { > > + dev = kid->child; > > + type = object_property_get_str(OBJECT(dev), "type", &err); > > + if (err) { > > + error_free(err); > > + fprintf(stderr, "error getting device type\n"); > > + exit(1); > > + } > > + > > + if (!strcmp(type, "dimm")) { > > + if (!dev->id) { > > + fprintf(stderr, "error getting dimm device id\n"); > > + exit(1); > > + } > > + slot = DIMM(dev); > > + /* determine starting physical address for this memory slot */ > > + assert(slot->start); > > + fw_cfg_slots[3 * slot->idx] = cpu_to_le64(slot->start); > > + fw_cfg_slots[3 * slot->idx + 1] = cpu_to_le64(slot->size); > > + fw_cfg_slots[3 * slot->idx + 2] = cpu_to_le64(slot->node); > > + i++; > > + } > > + } > > + assert(i == nb_hp_dimms); > > +} > > + > > diff --git a/vl.c b/vl.c > > index 0ff8818..37c9798 100644 > > --- a/vl.c > > +++ b/vl.c > > @@ -2335,7 +2335,7 @@ int main(int argc, char **argv, char **envp) > > node_cpumask[i] = 0; > > } > > > > - nb_numa_nodes = 0; > > + nb_numa_nodes = 1; > > nb_nics = 0; > > > > autostart= 1; > > -- > > 1.7.9 > > > >

1 0

Re: [SeaBIOS] [Qemu-devel] [RFC PATCH v2 06/21] dimm: Implement memory device abstraction
by Vasilis Liaskovitis 13 Jul '12

13 Jul '12

Hi, On Thu, Jul 12, 2012 at 07:55:42PM +0000, Blue Swirl wrote: > On Wed, Jul 11, 2012 at 10:31 AM, Vasilis Liaskovitis > <vasilis.liaskovitis(a)profitbricks.com> wrote: > > Each hotplug-able memory slot is a SysBusDevice. A hot-add operation for a > > particular dimm creates a new MemoryRegion of the given physical address > > offset, size and node proximity, and attaches it to main system memory as a > > sub_region. A hot-remove operation detaches and frees the MemoryRegion from > > system memory. > > > > This prototype still lacks proper qdev integration: a separate > > hotplug side-channel is used and main system bus hotplug capability is > > ignored. > > > > Signed-off-by: Vasilis Liaskovitis <vasilis.liaskovitis(a)profitbricks.com> > > --- > > hw/Makefile.objs | 2 +- > > hw/dimm.c | 234 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > hw/dimm.h | 58 +++++++++++++ > > 3 files changed, 293 insertions(+), 1 deletions(-) > > create mode 100644 hw/dimm.c > > create mode 100644 hw/dimm.h > > > > diff --git a/hw/Makefile.objs b/hw/Makefile.objs > > index 3d77259..e2184bf 100644 > > --- a/hw/Makefile.objs > > +++ b/hw/Makefile.objs > > @@ -26,7 +26,7 @@ hw-obj-$(CONFIG_I8254) += i8254_common.o i8254.o > > hw-obj-$(CONFIG_PCSPK) += pcspk.o > > hw-obj-$(CONFIG_PCKBD) += pckbd.o > > hw-obj-$(CONFIG_FDC) += fdc.o > > -hw-obj-$(CONFIG_ACPI) += acpi.o acpi_piix4.o > > +hw-obj-$(CONFIG_ACPI) += acpi.o acpi_piix4.o dimm.o > > hw-obj-$(CONFIG_APM) += pm_smbus.o apm.o > > hw-obj-$(CONFIG_DMA) += dma.o > > hw-obj-$(CONFIG_I82374) += i82374.o > > diff --git a/hw/dimm.c b/hw/dimm.c > > new file mode 100644 > > index 0000000..00c4623 > > --- /dev/null > > +++ b/hw/dimm.c > > @@ -0,0 +1,234 @@ > > +/* > > + * Dimm device for Memory Hotplug > > + * > > + * Copyright ProfitBricks GmbH 2012 > > + * This library is free software; you can redistribute it and/or > > + * modify it under the terms of the GNU Lesser General Public > > + * License as published by the Free Software Foundation; either > > + * version 2 of the License, or (at your option) any later version. > > + * > > + * This library is distributed in the hope that it will be useful, > > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > > + * Lesser General Public License for more details. > > + * > > + * You should have received a copy of the GNU Lesser General Public > > + * License along with this library; if not, see <http://www.gnu.org/licenses/> > > + */ > > + > > +#include "trace.h" > > +#include "qdev.h" > > +#include "dimm.h" > > +#include <time.h> > > +#include "../exec-memory.h" > > +#include "qmp-commands.h" > > + > > +static DeviceState *dimm_hotplug_qdev; > > +static dimm_hotplug_fn dimm_hotplug; > > +static QTAILQ_HEAD(Dimmlist, DimmState) dimmlist; > > Using global state does not look right. It should always be possible > to pass around structures to avoid it. ok, I 'll try to remove the global state. > > > + > > +static Property dimm_properties[] = { > > + DEFINE_PROP_END_OF_LIST() > > +}; > > + > > +void dimm_populate(DimmState *s) > > All functions are global and exported but there does not seem to be > users. Please make all static which you can. will do > > > +{ > > + DeviceState *dev= (DeviceState*)s; > > + MemoryRegion *new = NULL; > > + > > + new = g_malloc(sizeof(MemoryRegion)); > > + memory_region_init_ram(new, dev->id, s->size); > > + vmstate_register_ram_global(new); > > + memory_region_add_subregion(get_system_memory(), s->start, new); > > + s->mr = new; > > + s->populated = true; > > +} > > + > > + > > +void dimm_depopulate(DimmState *s) > > +{ > > + assert(s); > > + if (s->populated) { > > + vmstate_unregister_ram(s->mr, NULL); > > + memory_region_del_subregion(get_system_memory(), s->mr); > > + memory_region_destroy(s->mr); > > + s->populated = false; > > + s->mr = NULL; > > + } > > +} > > + > > +DimmState *dimm_create(char *id, uint64_t size, uint64_t node, uint32_t > > + dimm_idx, bool populated) > > +{ > > + DeviceState *dev; > > + DimmState *mdev; > > + > > + dev = sysbus_create_simple("dimm", -1, NULL); > > + dev->id = id; > > + > > + mdev = DIMM(dev); > > + mdev->idx = dimm_idx; > > + mdev->start = 0; > > + mdev->size = size; > > + mdev->node = node; > > + mdev->populated = populated; > > + QTAILQ_INSERT_TAIL(&dimmlist, mdev, nextdimm); > > + return mdev; > > +} > > + > > +void dimm_register_hotplug(dimm_hotplug_fn hotplug, DeviceState *qdev) > > +{ > > + dimm_hotplug_qdev = qdev; > > + dimm_hotplug = hotplug; > > + dimm_scan_populated(); > > +} > > + > > +void dimm_activate(DimmState *slot) > > +{ > > + dimm_populate(slot); > > + if (dimm_hotplug) > > + dimm_hotplug(dimm_hotplug_qdev, (SysBusDevice*)slot, 1); > > Why the cast? dimm_hotplug accepts SysBusDevice, not DimmState, though that can be changed. > > Also braces, please check your patches with checkpatch.pl. > ok, I 'll do checks with checkpatch.pl. > > +} > > + > > +void dimm_deactivate(DimmState *slot) > > +{ > > + if (dimm_hotplug) > > + dimm_hotplug(dimm_hotplug_qdev, (SysBusDevice*)slot, 0); > > +} > > + > > +DimmState *dimm_find_from_name(char *id) > > const char *id? ok > > > +{ > > + Error *err = NULL; > > + DeviceState *qdev; > > + const char *type; > > + qdev = qdev_find_recursive(sysbus_get_default(), id); > > + if (qdev) { > > + type = object_property_get_str(OBJECT(qdev), "type", &err); > > + if (!type) { > > + return NULL; > > + } > > + if (!strcmp(type, "dimm")) { > > + return DIMM(qdev); > > + } > > + } > > + return NULL; > > +} > > + > > +int dimm_do(Monitor *mon, const QDict *qdict, bool add) > > +{ > > + DimmState *slot = NULL; > > + > > + char *id = (char*) qdict_get_try_str(qdict, "id"); > > Why this cast? unneeded, because id should be declared as const char*. will fix. > > > + if (!id) { > > + fprintf(stderr, "ERROR %s invalid id\n",__FUNCTION__); > > + return 1; > > + } > > + > > + slot = dimm_find_from_name(id); > > + > > + if (!slot) { > > + fprintf(stderr, "%s no slot %s found\n", __FUNCTION__, id); > > + return 1; > > + } > > + > > + if (add) { > > + if (slot->populated) { > > + fprintf(stderr, "ERROR %s slot %s already populated\n", > > + __FUNCTION__, id); > > + return 1; > > + } > > + dimm_activate(slot); > > + } > > + else { > > + if (!slot->populated) { > > + fprintf(stderr, "ERROR %s slot %s is not populated\n", > > + __FUNCTION__, id); > > + return 1; > > + } > > + dimm_deactivate(slot); > > + } > > + > > + return 0; > > +} > > + > > +DimmState *dimm_find_from_idx(uint32_t idx) > > +{ > > + DimmState *slot; > > + > > + QTAILQ_FOREACH(slot, &dimmlist, nextdimm) { > > + if (slot->idx == idx) { > > + return slot; > > + } > > + } > > + return NULL; > > +} > > + > > +/* used to calculate physical address offsets for all dimms */ > > +void dimm_calc_offsets(dimm_calcoffset_fn calcfn) > > +{ > > + DimmState *slot; > > + QTAILQ_FOREACH(slot, &dimmlist, nextdimm) { > > + if (!slot->start) > > + slot->start = calcfn(slot->size); > > + } > > +} > > + > > +/* used to populate and activate dimms at boot time */ > > +void dimm_scan_populated(void) > > +{ > > + DimmState *slot; > > + QTAILQ_FOREACH(slot, &dimmlist, nextdimm) { > > + if (slot->populated && !slot->mr) { > > + dimm_activate(slot); > > + } > > + } > > +} > > + > > +void dimm_notify(uint32_t idx, uint32_t event) > > +{ > > + DimmState *s; > > + s = dimm_find_from_idx(idx); > > + assert(s != NULL); > > + > > + switch(event) { > > + case DIMM_REMOVE_SUCCESS: > > + dimm_depopulate(s); > > + break; > > + default: > > + break; > > + } > > +} > > + > > +static int dimm_init(SysBusDevice *s) > > +{ > > + DimmState *slot; > > + slot = DIMM(s); > > + slot->mr = NULL; > > + slot->populated = false; > > + return 0; > > +} > > + > > +static void dimm_class_init(ObjectClass *klass, void *data) > > +{ > > + SysBusDeviceClass *sc = SYS_BUS_DEVICE_CLASS(klass); > > + DeviceClass *dc = DEVICE_CLASS(klass); > > + > > + dc->props = dimm_properties; > > + sc->init = dimm_init; > > + dimm_hotplug = NULL; > > + QTAILQ_INIT(&dimmlist); > > +} > > + > > +static TypeInfo dimm_info = { > > + .name = "dimm", > > + .parent = TYPE_SYS_BUS_DEVICE, > > + .instance_size = sizeof(DimmState), > > + .class_init = dimm_class_init, > > +}; > > + > > +static void dimm_register_types(void) > > +{ > > + type_register_static(&dimm_info); > > +} > > + > > +type_init(dimm_register_types) > > diff --git a/hw/dimm.h b/hw/dimm.h > > new file mode 100644 > > index 0000000..643f319 > > --- /dev/null > > +++ b/hw/dimm.h > > @@ -0,0 +1,58 @@ > > +#ifndef QEMU_DIMM_H > > +#define QEMU_DIMM_H > > Should be HW_DIMM_H. ok. > > > + > > +#include "qemu-common.h" > > +#include "memory.h" > > +#include "sysbus.h" > > +#include "qapi-types.h" > > +#include "qemu-queue.h" > > +#include "cpus.h" > > +#define MAX_DIMMS 255 > > +#define DIMM_BITMAP_BYTES (MAX_DIMMS + 7) / 8 > > +#define DEFAULT_DIMMSIZE 1024*1024*1024 > > + > > +typedef enum { > > + DIMM_REMOVE_SUCCESS = 0, > > + DIMM_REMOVE_FAIL = 1, > > + DIMM_ADD_SUCCESS = 2, > > + DIMM_ADD_FAIL = 3 > > +} dimm_hp_result_code; > > + > > +#define TYPE_DIMM "dimm" > > +#define DIMM(obj) \ > > + OBJECT_CHECK(DimmState, (obj), TYPE_DIMM) > > +#define DIMM_CLASS(klass) \ > > + OBJECT_CLASS_CHECK(DimmClass, (obj), TYPE_DIMM) > > +#define DIMM_GET_CLASS(obj) \ > > + OBJECT_GET_CLASS(DimmClass, (obj), TYPE_DIMM) > > + > > +typedef struct DimmState { > > + SysBusDevice busdev; > > + uint32_t idx; /* index in memory hotplug register/bitmap */ > > + ram_addr_t start; /* starting physical address */ > > + ram_addr_t size; > > + uint32_t node; /* numa node proximity */ > > + MemoryRegion *mr; /* MemoryRegion for this slot. !NULL only if populated */ > > + bool populated; /* 1 means device has been hotplugged. Default is 0. */ > > + QTAILQ_ENTRY (DimmState) nextdimm; > > +} DimmState; > > + > > +typedef int (*dimm_hotplug_fn)(DeviceState *qdev, SysBusDevice *dev, int add); > > +typedef target_phys_addr_t (*dimm_calcoffset_fn)(uint64_t size); > > + > > +DimmState *dimm_create(char *id, uint64_t size, uint64_t node, uint32_t > > + dimm_idx, bool populated); > > +void dimm_populate(DimmState *s); > > +void dimm_depopulate(DimmState *s); > > +int dimm_do(Monitor *mon, const QDict *qdict, bool add); > > +DimmState *dimm_find_from_idx(uint32_t idx); > > +DimmState *dimm_find_from_name(char *id); > > +void dimm_register_hotplug(dimm_hotplug_fn hotplug, DeviceState *qdev); > > +void dimm_calc_offsets(dimm_calcoffset_fn calcfn); > > +void dimm_activate(DimmState *slot); > > +void dimm_deactivate(DimmState *slot); > > +void dimm_scan_populated(void); > > +void dimm_notify(uint32_t idx, uint32_t event); > > + > > + > > +#endif > > -- > > 1.7.9 > > > >

1 0

Bootorder Failover (Patch)
by Steve Goodrich 13 Jul '12

13 Jul '12

Legacy MBRs on HDDs and USB MSCs allow booting only from device id 0x80. This is OK for platforms that have only one HDD or USB device to boot from, but causes problems if the end user wants to have more than one HDD or USB MSC device in the boot order (and if those devices are present, but not bootable). This patch allows proper failover from one HDD/USB MSC device to the next. Assume the platform has the following devices in its bootorder file: HDD1 (present, no MBR/not bootable) HDD2 (present but unreadable) Floppy (not present) USB1 (present, no MBR/not bootable) USB2 (present, bootable) The current/previous bootorder handling would set HDD1 as drive 0x80, and it would be the only HDD or USB upon which booting would be attempted. Since it has no MBR, the system would not boot (the Floppy is not present, and none of the other drives would be tried). With the code in this patch, the system would behave like this: * Assign HDD1 as device 0x80 and attempt to boot * boot fails because the device is not bootable * Rotate the boot order to be HDD2, Floppy, USB1, USB2, HDD1, making HDD2 device 0x80 * Attempt to boot * boot fails because the device is unreadable * Reset the boot order to its original while we try (and fail) booting from the Floppy * Rotate the boot order to be USB1, USB2, HDD1, HDD2, Floppy, making USB1 device 0x80 * Attempt to boot * boot fails because the device has no MBR * Rotate the boot order to be USB2, HDD1, HDD2, Floppy, USB1, making USB2 device 0x80 * Attempt to boot * boot succeeds with the current device order (USB2=0x80, HDD1=0x81, HDD2= 0x82, Floppy=0x00, USB1=0x83) NOTE: This behavior is controlled via the BOOTORDER_FAILOVER selection in Kconfig. Thoughts? -- Steve G.

1 0

Re: [SeaBIOS] [Qemu-devel] [RFC PATCH v2 00/21] ACPI memory hotplug
by Blue Swirl 12 Jul '12

12 Jul '12

On Wed, Jul 11, 2012 at 10:31 AM, Vasilis Liaskovitis <vasilis.liaskovitis(a)profitbricks.com> wrote: > This is v2 of the ACPI memory hotplug prototype for x86_64 target. I think the concept of DIMMs (what about SIMMs? SODIMMs? I liked memslot) would be useful for most targets, but hotplugging may be limited to x86 only. It would be nice to keep these two separate or as loosely coupled as possible. > > Changes v1->v2 > > - memory map is automatically calculated for hotplug dimms. Dimms are added from > top-of-memory skipping the pci hole at [PCI_HOLE_START, 4G). > - Renamed from "-memslot" to "-dimm". Commands changed to "dimm_add", "dimm_del". > - Seabios ejection array reduced to a byte. Use extraction macros for dimm ssdt. > - additional SRAT paravirt info does not break previous SRAT fw_cfg layout. > - Documentation of new acpi_piix4 registers and paravirt data. > - add ACPI _OST support for _OST enabled guests. This allows qemu to receive > notification for success / failure of memory hot-add and hot-remove operations. > Guest needs to support _OST (https://lkml.org/lkml/2012/6/25/321) > - add monitor info command to report total guest memory (initial + hot-added) > - add command line options and monitor commands for batch dimm creation/population > > Overview: > > Dimm devices are modeled with a new qemu command line > > "-dimm id=name,size=sz,node=pxm,populated=on|off" > > As already mentioned, the starting physical address for all dimms is calculated > automatically from top of memory, skipping the pci hole at [PCI_HOLE_START, 4G). > Node is defining numa proximity for this dimm. When not defined it defaults > to zero. > "-dimm id=dimm0,size=512M,node=0,populated=off" > will define a 512M memory slot belonging to numa node 0. > > Dimms are added or removed with a new hmp command "dimm_add/dimm_del": > Hot-add syntax: "dimm_add id" > Hot-remove syntax: "dimm_del id" > > Issues: > > - Live migration works as long as populated field is changed to "on" for > hotplugged dimms at the destination qemu command line (patch 12/21 lifts > this requirement). The DimmState structure does not yet define a > VMStateDescription, but i assume this is the preferred way to pass state > for migration. > > - Dimms are abstracted as qdevices attached to the main system bus. However, > memory hotplugging has its own side channel ignoring main_system_bus's hotplug > incapability. A cleaner integration is still needed, probably attaching memory > devices as children-links of an acpi-capable device (in the pc case acpi_piix4) > instead of the system bus (TBD). Then device_add/device_del instead of new > commands can hopefully be used. > > Comments/review welcome. > > series is based on uq/master for qemu-kvm, and master for seabios. Can be found > also at: > http://github.com/vliaskov/qemu-kvm/commits/memhp-v2 > http://github.com/vliaskov/seabios/commits/memhp-v2 > > Vasilis Liaskovitis (14): > dimm: Implement memory device abstraction > acpi_piix4: Implement memory device hotplug registers > pc: calculate dimm physical addresses and adjust memory map > pc: Add dimm paravirt SRAT info > Implement "-dimm" command line option > Implement dimm_add and dimm_del commands for hmp and qmp > fix live-migration when "populated=on" is missing > Implement memory hotplug notification lists > acpi_piix4: _OST dimm support > acpi_piix4: Update dimm state on VM reboot > acpi_piix4: Update dimm bitmap state on hot-remove fail > Implement "info memtotal" and "query-memtotal" > Implement -dimms, -dimmspop command line options > Implement mem_increase, mem_decrease hmp/qmp commands > > arch_init.c | 23 ++- > docs/specs/acpi_hotplug.txt | 46 +++++ > docs/specs/fwcfg.txt | 28 +++ > hmp-commands.hx | 67 +++++++ > hmp.c | 24 +++ > hmp.h | 2 + > hw/Makefile.objs | 2 +- > hw/acpi_piix4.c | 131 ++++++++++++- > hw/dimm.c | 449 +++++++++++++++++++++++++++++++++++++++++++ > hw/dimm.h | 72 +++++++ > hw/pc.c | 94 +++++++++- > hw/pc.h | 6 + > hw/pc_piix.c | 18 ++- > monitor.c | 35 ++++ > monitor.h | 5 + > qapi-schema.json | 38 ++++ > qemu-config.c | 70 +++++++ > qemu-options.hx | 15 ++ > qmp-commands.hx | 137 +++++++++++++ > sysemu.h | 1 + > vl.c | 122 ++++++++++++- > 21 files changed, 1368 insertions(+), 17 deletions(-) > create mode 100644 docs/specs/acpi_hotplug.txt > create mode 100644 docs/specs/fwcfg.txt > create mode 100644 hw/dimm.c > create mode 100644 hw/dimm.h > > Vasilis Liaskovitis (7): > Add ACPI_EXTRACT_DEVICE* macros > Add SSDT memory device support > acpi-dsdt: Implement functions for memory hotplug. > acpi: generate hotplug memory devices. > pciinit: Fix pcimem_start value > acpi_dsdt: Support _OST dimm method > acpi_dsdt: Revert internal dimm state on _OST failure > > Makefile | 2 +- > src/acpi-dsdt.dsl | 120 ++++++++++++++++++++++++++++++++++++- > src/acpi.c | 158 +++++++++++++++++++++++++++++++++++++++++++++++-- > src/pciinit.c | 2 +- > src/ssdt-mem.dsl | 69 +++++++++++++++++++++ > tools/acpi_extract.py | 28 +++++++++ > 6 files changed, 369 insertions(+), 10 deletions(-) > create mode 100644 src/ssdt-mem.dsl >

1 0

Re: [SeaBIOS] [Qemu-devel] [RFC PATCH v2 06/21] dimm: Implement memory device abstraction
by Blue Swirl 12 Jul '12

12 Jul '12

On Wed, Jul 11, 2012 at 10:31 AM, Vasilis Liaskovitis <vasilis.liaskovitis(a)profitbricks.com> wrote: > Each hotplug-able memory slot is a SysBusDevice. A hot-add operation for a > particular dimm creates a new MemoryRegion of the given physical address > offset, size and node proximity, and attaches it to main system memory as a > sub_region. A hot-remove operation detaches and frees the MemoryRegion from > system memory. > > This prototype still lacks proper qdev integration: a separate > hotplug side-channel is used and main system bus hotplug capability is > ignored. > > Signed-off-by: Vasilis Liaskovitis <vasilis.liaskovitis(a)profitbricks.com> > --- > hw/Makefile.objs | 2 +- > hw/dimm.c | 234 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ > hw/dimm.h | 58 +++++++++++++ > 3 files changed, 293 insertions(+), 1 deletions(-) > create mode 100644 hw/dimm.c > create mode 100644 hw/dimm.h > > diff --git a/hw/Makefile.objs b/hw/Makefile.objs > index 3d77259..e2184bf 100644 > --- a/hw/Makefile.objs > +++ b/hw/Makefile.objs > @@ -26,7 +26,7 @@ hw-obj-$(CONFIG_I8254) += i8254_common.o i8254.o > hw-obj-$(CONFIG_PCSPK) += pcspk.o > hw-obj-$(CONFIG_PCKBD) += pckbd.o > hw-obj-$(CONFIG_FDC) += fdc.o > -hw-obj-$(CONFIG_ACPI) += acpi.o acpi_piix4.o > +hw-obj-$(CONFIG_ACPI) += acpi.o acpi_piix4.o dimm.o > hw-obj-$(CONFIG_APM) += pm_smbus.o apm.o > hw-obj-$(CONFIG_DMA) += dma.o > hw-obj-$(CONFIG_I82374) += i82374.o > diff --git a/hw/dimm.c b/hw/dimm.c > new file mode 100644 > index 0000000..00c4623 > --- /dev/null > +++ b/hw/dimm.c > @@ -0,0 +1,234 @@ > +/* > + * Dimm device for Memory Hotplug > + * > + * Copyright ProfitBricks GmbH 2012 > + * This library is free software; you can redistribute it and/or > + * modify it under the terms of the GNU Lesser General Public > + * License as published by the Free Software Foundation; either > + * version 2 of the License, or (at your option) any later version. > + * > + * This library is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > + * Lesser General Public License for more details. > + * > + * You should have received a copy of the GNU Lesser General Public > + * License along with this library; if not, see <http://www.gnu.org/licenses/> > + */ > + > +#include "trace.h" > +#include "qdev.h" > +#include "dimm.h" > +#include <time.h> > +#include "../exec-memory.h" > +#include "qmp-commands.h" > + > +static DeviceState *dimm_hotplug_qdev; > +static dimm_hotplug_fn dimm_hotplug; > +static QTAILQ_HEAD(Dimmlist, DimmState) dimmlist; Using global state does not look right. It should always be possible to pass around structures to avoid it. > + > +static Property dimm_properties[] = { > + DEFINE_PROP_END_OF_LIST() > +}; > + > +void dimm_populate(DimmState *s) All functions are global and exported but there does not seem to be users. Please make all static which you can. > +{ > + DeviceState *dev= (DeviceState*)s; > + MemoryRegion *new = NULL; > + > + new = g_malloc(sizeof(MemoryRegion)); > + memory_region_init_ram(new, dev->id, s->size); > + vmstate_register_ram_global(new); > + memory_region_add_subregion(get_system_memory(), s->start, new); > + s->mr = new; > + s->populated = true; > +} > + > + > +void dimm_depopulate(DimmState *s) > +{ > + assert(s); > + if (s->populated) { > + vmstate_unregister_ram(s->mr, NULL); > + memory_region_del_subregion(get_system_memory(), s->mr); > + memory_region_destroy(s->mr); > + s->populated = false; > + s->mr = NULL; > + } > +} > + > +DimmState *dimm_create(char *id, uint64_t size, uint64_t node, uint32_t > + dimm_idx, bool populated) > +{ > + DeviceState *dev; > + DimmState *mdev; > + > + dev = sysbus_create_simple("dimm", -1, NULL); > + dev->id = id; > + > + mdev = DIMM(dev); > + mdev->idx = dimm_idx; > + mdev->start = 0; > + mdev->size = size; > + mdev->node = node; > + mdev->populated = populated; > + QTAILQ_INSERT_TAIL(&dimmlist, mdev, nextdimm); > + return mdev; > +} > + > +void dimm_register_hotplug(dimm_hotplug_fn hotplug, DeviceState *qdev) > +{ > + dimm_hotplug_qdev = qdev; > + dimm_hotplug = hotplug; > + dimm_scan_populated(); > +} > + > +void dimm_activate(DimmState *slot) > +{ > + dimm_populate(slot); > + if (dimm_hotplug) > + dimm_hotplug(dimm_hotplug_qdev, (SysBusDevice*)slot, 1); Why the cast? Also braces, please check your patches with checkpatch.pl. > +} > + > +void dimm_deactivate(DimmState *slot) > +{ > + if (dimm_hotplug) > + dimm_hotplug(dimm_hotplug_qdev, (SysBusDevice*)slot, 0); > +} > + > +DimmState *dimm_find_from_name(char *id) const char *id? > +{ > + Error *err = NULL; > + DeviceState *qdev; > + const char *type; > + qdev = qdev_find_recursive(sysbus_get_default(), id); > + if (qdev) { > + type = object_property_get_str(OBJECT(qdev), "type", &err); > + if (!type) { > + return NULL; > + } > + if (!strcmp(type, "dimm")) { > + return DIMM(qdev); > + } > + } > + return NULL; > +} > + > +int dimm_do(Monitor *mon, const QDict *qdict, bool add) > +{ > + DimmState *slot = NULL; > + > + char *id = (char*) qdict_get_try_str(qdict, "id"); Why this cast? > + if (!id) { > + fprintf(stderr, "ERROR %s invalid id\n",__FUNCTION__); > + return 1; > + } > + > + slot = dimm_find_from_name(id); > + > + if (!slot) { > + fprintf(stderr, "%s no slot %s found\n", __FUNCTION__, id); > + return 1; > + } > + > + if (add) { > + if (slot->populated) { > + fprintf(stderr, "ERROR %s slot %s already populated\n", > + __FUNCTION__, id); > + return 1; > + } > + dimm_activate(slot); > + } > + else { > + if (!slot->populated) { > + fprintf(stderr, "ERROR %s slot %s is not populated\n", > + __FUNCTION__, id); > + return 1; > + } > + dimm_deactivate(slot); > + } > + > + return 0; > +} > + > +DimmState *dimm_find_from_idx(uint32_t idx) > +{ > + DimmState *slot; > + > + QTAILQ_FOREACH(slot, &dimmlist, nextdimm) { > + if (slot->idx == idx) { > + return slot; > + } > + } > + return NULL; > +} > + > +/* used to calculate physical address offsets for all dimms */ > +void dimm_calc_offsets(dimm_calcoffset_fn calcfn) > +{ > + DimmState *slot; > + QTAILQ_FOREACH(slot, &dimmlist, nextdimm) { > + if (!slot->start) > + slot->start = calcfn(slot->size); > + } > +} > + > +/* used to populate and activate dimms at boot time */ > +void dimm_scan_populated(void) > +{ > + DimmState *slot; > + QTAILQ_FOREACH(slot, &dimmlist, nextdimm) { > + if (slot->populated && !slot->mr) { > + dimm_activate(slot); > + } > + } > +} > + > +void dimm_notify(uint32_t idx, uint32_t event) > +{ > + DimmState *s; > + s = dimm_find_from_idx(idx); > + assert(s != NULL); > + > + switch(event) { > + case DIMM_REMOVE_SUCCESS: > + dimm_depopulate(s); > + break; > + default: > + break; > + } > +} > + > +static int dimm_init(SysBusDevice *s) > +{ > + DimmState *slot; > + slot = DIMM(s); > + slot->mr = NULL; > + slot->populated = false; > + return 0; > +} > + > +static void dimm_class_init(ObjectClass *klass, void *data) > +{ > + SysBusDeviceClass *sc = SYS_BUS_DEVICE_CLASS(klass); > + DeviceClass *dc = DEVICE_CLASS(klass); > + > + dc->props = dimm_properties; > + sc->init = dimm_init; > + dimm_hotplug = NULL; > + QTAILQ_INIT(&dimmlist); > +} > + > +static TypeInfo dimm_info = { > + .name = "dimm", > + .parent = TYPE_SYS_BUS_DEVICE, > + .instance_size = sizeof(DimmState), > + .class_init = dimm_class_init, > +}; > + > +static void dimm_register_types(void) > +{ > + type_register_static(&dimm_info); > +} > + > +type_init(dimm_register_types) > diff --git a/hw/dimm.h b/hw/dimm.h > new file mode 100644 > index 0000000..643f319 > --- /dev/null > +++ b/hw/dimm.h > @@ -0,0 +1,58 @@ > +#ifndef QEMU_DIMM_H > +#define QEMU_DIMM_H Should be HW_DIMM_H. > + > +#include "qemu-common.h" > +#include "memory.h" > +#include "sysbus.h" > +#include "qapi-types.h" > +#include "qemu-queue.h" > +#include "cpus.h" > +#define MAX_DIMMS 255 > +#define DIMM_BITMAP_BYTES (MAX_DIMMS + 7) / 8 > +#define DEFAULT_DIMMSIZE 1024*1024*1024 > + > +typedef enum { > + DIMM_REMOVE_SUCCESS = 0, > + DIMM_REMOVE_FAIL = 1, > + DIMM_ADD_SUCCESS = 2, > + DIMM_ADD_FAIL = 3 > +} dimm_hp_result_code; > + > +#define TYPE_DIMM "dimm" > +#define DIMM(obj) \ > + OBJECT_CHECK(DimmState, (obj), TYPE_DIMM) > +#define DIMM_CLASS(klass) \ > + OBJECT_CLASS_CHECK(DimmClass, (obj), TYPE_DIMM) > +#define DIMM_GET_CLASS(obj) \ > + OBJECT_GET_CLASS(DimmClass, (obj), TYPE_DIMM) > + > +typedef struct DimmState { > + SysBusDevice busdev; > + uint32_t idx; /* index in memory hotplug register/bitmap */ > + ram_addr_t start; /* starting physical address */ > + ram_addr_t size; > + uint32_t node; /* numa node proximity */ > + MemoryRegion *mr; /* MemoryRegion for this slot. !NULL only if populated */ > + bool populated; /* 1 means device has been hotplugged. Default is 0. */ > + QTAILQ_ENTRY (DimmState) nextdimm; > +} DimmState; > + > +typedef int (*dimm_hotplug_fn)(DeviceState *qdev, SysBusDevice *dev, int add); > +typedef target_phys_addr_t (*dimm_calcoffset_fn)(uint64_t size); > + > +DimmState *dimm_create(char *id, uint64_t size, uint64_t node, uint32_t > + dimm_idx, bool populated); > +void dimm_populate(DimmState *s); > +void dimm_depopulate(DimmState *s); > +int dimm_do(Monitor *mon, const QDict *qdict, bool add); > +DimmState *dimm_find_from_idx(uint32_t idx); > +DimmState *dimm_find_from_name(char *id); > +void dimm_register_hotplug(dimm_hotplug_fn hotplug, DeviceState *qdev); > +void dimm_calc_offsets(dimm_calcoffset_fn calcfn); > +void dimm_activate(DimmState *slot); > +void dimm_deactivate(DimmState *slot); > +void dimm_scan_populated(void); > +void dimm_notify(uint32_t idx, uint32_t event); > + > + > +#endif > -- > 1.7.9 > >

1 0

Re: [SeaBIOS] [Qemu-devel] [RFC PATCH v2 09/21] pc: Add dimm paravirt SRAT info
by Blue Swirl 12 Jul '12

12 Jul '12

On Wed, Jul 11, 2012 at 10:31 AM, Vasilis Liaskovitis <vasilis.liaskovitis(a)profitbricks.com> wrote: > The numa_fw_cfg paravirt interface is extended to include SRAT information for > all hotplug-able dimms. There are 3 words for each hotplug-able memory slot, > denoting start address, size and node proximity. The new info is appended after > existing numa info, so that the fw_cfg layout does not break. This information > is used by Seabios to build hotplug memory device objects at runtime. > nb_numa_nodes is set to 1 by default (not 0), so that we always pass srat info > to SeaBIOS. > > v1->v2: > Dimm SRAT info (#dimms) is appended at end of existing numa fw_cfg in order not > to break existing layout > Documentation of the new fwcfg layout is included in docs/specs/fwcfg.txt > > Signed-off-by: Vasilis Liaskovitis <vasilis.liaskovitis(a)profitbricks.com> > --- > docs/specs/fwcfg.txt | 28 ++++++++++++++++++++++++++ > hw/pc.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++++- > vl.c | 2 +- > 3 files changed, 80 insertions(+), 3 deletions(-) > create mode 100644 docs/specs/fwcfg.txt > > diff --git a/docs/specs/fwcfg.txt b/docs/specs/fwcfg.txt > new file mode 100644 > index 0000000..e6fcd8f > --- /dev/null > +++ b/docs/specs/fwcfg.txt > @@ -0,0 +1,28 @@ > +QEMU<->BIOS Paravirt Documentation > +-------------------------------------- > + > +This document describes paravirt data structures passed from QEMU to BIOS. > + > +fw_cfg SRAT paravirt info > +-------------------- > +The SRAT info passed from QEMU to BIOS has the following layout: > + > +----------------------------------------------------------------------------------------------- > +#nodes | cpu0_pxm | cpu1_pxm | ... | cpulast_pxm | node0_mem | node1_mem | ... | nodelast_mem > + > +----------------------------------------------------------------------------------------------- > +#dimms | dimm0_start | dimm0_sz | dimm0_pxm | ... | dimmlast_start | dimmlast_sz | dimmlast_pxm > + > +Entry 0 contains the number of numa nodes (nb_numa_nodes). > + > +Entries 1..max_cpus: The next max_cpus entries describe node proximity for each > +one of the vCPUs in the system. > + > +Entries max_cpus+1..max_cpus+nb_numa_nodes+1: The next nb_numa_nodes entries > +describe the memory size for each one of the NUMA nodes in the system. > + > +Entry max_cpus+nb_numa_nodes+1 contains the number of memory dimms (nb_hp_dimms) > + > +The last 3 * nb_hp_dimms entries are organized in triplets: Each triplet contains > +the physical address offset, size (in bytes), and node proximity for the > +respective dimm. The size and endianness are not specified, you are using LE 64 bit values for each item. > diff --git a/hw/pc.c b/hw/pc.c > index ef9901a..cf651d0 100644 > --- a/hw/pc.c > +++ b/hw/pc.c > @@ -598,12 +598,15 @@ int e820_add_entry(uint64_t address, uint64_t length, uint32_t type) > return index; > } > > +static void setup_hp_dimms(uint64_t *fw_cfg_slots); > + > static void *bochs_bios_init(void) > { > void *fw_cfg; > uint8_t *smbios_table; > size_t smbios_len; > uint64_t *numa_fw_cfg; > + uint64_t *hp_dimms_fw_cfg; > int i, j; > > register_ioport_write(0x400, 1, 2, bochs_bios_write, NULL); > @@ -638,8 +641,10 @@ static void *bochs_bios_init(void) > /* allocate memory for the NUMA channel: one (64bit) word for the number > * of nodes, one word for each VCPU->node and one word for each node to > * hold the amount of memory. > + * Finally one word for the number of hotplug memory slots and three words > + * for each hotplug memory slot (start address, size and node proximity). > */ > - numa_fw_cfg = g_malloc0((1 + max_cpus + nb_numa_nodes) * 8); > + numa_fw_cfg = g_malloc0((2 + max_cpus + nb_numa_nodes + 3 * nb_hp_dimms) * 8); > numa_fw_cfg[0] = cpu_to_le64(nb_numa_nodes); > for (i = 0; i < max_cpus; i++) { > for (j = 0; j < nb_numa_nodes; j++) { > @@ -652,8 +657,15 @@ static void *bochs_bios_init(void) > for (i = 0; i < nb_numa_nodes; i++) { > numa_fw_cfg[max_cpus + 1 + i] = cpu_to_le64(node_mem[i]); > } > + > + numa_fw_cfg[1 + max_cpus + nb_numa_nodes] = cpu_to_le64(nb_hp_dimms); > + > + hp_dimms_fw_cfg = numa_fw_cfg + 2 + max_cpus + nb_numa_nodes; > + if (nb_hp_dimms) > + setup_hp_dimms(hp_dimms_fw_cfg); Braces. > + > fw_cfg_add_bytes(fw_cfg, FW_CFG_NUMA, (uint8_t *)numa_fw_cfg, > - (1 + max_cpus + nb_numa_nodes) * 8); > + (2 + max_cpus + nb_numa_nodes + 3 * nb_hp_dimms) * 8); > > return fw_cfg; > } > @@ -1223,3 +1235,40 @@ target_phys_addr_t pc_set_hp_memory_offset(uint64_t size) > > return ret; > } > + > +static void setup_hp_dimms(uint64_t *fw_cfg_slots) > +{ > + int i = 0; > + Error *err = NULL; > + DeviceState *dev; > + DimmState *slot; > + const char *type; > + BusChild *kid; > + BusState *bus = sysbus_get_default(); > + > + QTAILQ_FOREACH(kid, &bus->children, sibling) { > + dev = kid->child; > + type = object_property_get_str(OBJECT(dev), "type", &err); > + if (err) { > + error_free(err); > + fprintf(stderr, "error getting device type\n"); > + exit(1); > + } > + > + if (!strcmp(type, "dimm")) { > + if (!dev->id) { > + fprintf(stderr, "error getting dimm device id\n"); > + exit(1); > + } > + slot = DIMM(dev); > + /* determine starting physical address for this memory slot */ > + assert(slot->start); > + fw_cfg_slots[3 * slot->idx] = cpu_to_le64(slot->start); > + fw_cfg_slots[3 * slot->idx + 1] = cpu_to_le64(slot->size); > + fw_cfg_slots[3 * slot->idx + 2] = cpu_to_le64(slot->node); > + i++; > + } > + } > + assert(i == nb_hp_dimms); > +} > + > diff --git a/vl.c b/vl.c > index 0ff8818..37c9798 100644 > --- a/vl.c > +++ b/vl.c > @@ -2335,7 +2335,7 @@ int main(int argc, char **argv, char **envp) > node_cpumask[i] = 0; > } > > - nb_numa_nodes = 0; > + nb_numa_nodes = 1; > nb_nics = 0; > > autostart= 1; > -- > 1.7.9 > >

1 0

[RFC 0/7+1] QEMU APIC ID + topology bug + CPU hotplug
by Eduardo Habkost 12 Jul '12

12 Jul '12

Hi, This is more a call for discussion than a request for comments in the actual code. Our problem today is: - Lots of QEMU and Seabios code have the assumption that Initial APIC ID == CPU "index" (in other words, that APIC IDs are always contiguous and start at 0); - However, the Initial APIC IDs may be non-contiguous depending on the requested CPU topology (core/thread counts), so we have to break that assumption (see [1]). This series is how a fix could look like if we had the following assumptions: - Seabios can't run any code in hotplugged CPUs; - We don't change the current CPU hotplug interface between the Seabios SSDT code and QEMU (the interface used by method PRSC at acpi-dsdt.dsl in Seabios) Note that I am more than willing to break any of the assumptions above, and get rid of FW_CFG_LAPIC_INFO. But I don't know what's the best way to do it. For the non-hotplug CPUs, it would be quite easy: just make Seabios get the initial APIC ID from the AP processors on boot, at smp.c:smp_ap_boot_code. The hotplug case is a bit more complex: we need to either: - have a mechanism to let the ACPI SSDT code know what's the APIC ID of hotplugged CPUs; or - make Seabios run some code in the hotplugged CPU (I am assuming that this is simply not possible). I am hoping people have suggestions to solve this. I don't know where the interface used by acpi-dsdt.dsl:PRSC comes from, and if it can be easily extended or changed. All that said, I have one question: is it acceptable to apply a fix for the APIC- ID/topology bug that knowingly breaks CPU hotplug, by now? (not exactly the fix in this series, but a fix that simply makes Seabios query the APIC IDs directly from the CPUs). Note that this series is incomplete and not completely tested. Know issues: - It breaks CPU hotplug because the SSDT code in Seabios is still incorrect; - It breaks live-migration because the initial APIC ID will be different even for older machine-types (and the initial APIC ID is not part of the CPU state sent during migration); - The MPTABLE code in Seabios may also need changes to use the right APIC IDs (I am not sure yet). [1] http://software.intel.com/en-us/articles/intel-64-architecture-processor-to… -- Eduardo

4 13

Re: [SeaBIOS] [RFC PATCH v2 05/21] pciinit: Fix pcimem_start value
by Gerd Hoffmann 12 Jul '12

12 Jul '12

On 07/11/12 18:45, Vasilis Liaskovitis wrote: > Hi, > > On Wed, Jul 11, 2012 at 01:56:19PM +0200, Gerd Hoffmann wrote: >> On 07/11/12 12:31, Vasilis Liaskovitis wrote: >>> In order to hotplug memory between RamSize and BUILD_PCIMEM_START, the pci >>> window needs to start at BUILD_PCIMEM_START (0xe0000000). >>> Otherwise, the guest cannot online new dimms at those ranges due to pci_root >>> window conflicts. (workaround for linux guest is booting with pci=nocrs) >> >>> static void pci_bios_map_devices(struct pci_bus *busses) >>> { >>> - pcimem_start = RamSize; >>> + pcimem_start = BUILD_PCIMEM_START; >> >> It isn't that simple. For the 32bit pci window it will work, but will >> leaves address space unused instead of assigning it to the 32bit pci >> window. For the 64bit pci window it will not work. >> >> You have to walk the dimms and figure what the highest used address is, >> for both below-4g and above-4g. Then fill two variable with it and make >> the pci init code use that instead of RamSize and RamSizeOver4G. > > I see. I already have these values values computed in qemu-kvm, so I can pass > them in a paravirt struct, or infer them from the dimm/srat paravirt info that I > already pass to seabios. I'd suggest to infer from the dimm info, to limit the amout of information which needs to be passed from qemu to seabios. > If i understand correctly, we would like the pcimem windows to use the maximum > possible address space (constrained by the exact dimms/ranges which are defined) > instead of leaving unused space. Yes, for the 32bit pci window. The 64bit pci window is mapped above all memory, and it must likewise consider defined+unfilled dimms so the start address doesn't collide with memory hot-plugged above 4G later on. cheers, Gerd

2 1

← Newer
1
...
5
6
7
8
9
10
11
12
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

SeaBIOS July 2012