aboutsummaryrefslogtreecommitdiff
path: root/hw
diff options
context:
space:
mode:
authorPeter Maydell <peter.maydell@linaro.org>2015-01-10 19:50:21 +0000
committerPeter Maydell <peter.maydell@linaro.org>2015-01-10 19:50:21 +0000
commit97052d64e4fb934bcf25f4a6b42dc06f6ecbf9eb (patch)
treec4a72adc118c1cc1ca6e0c51c4072158ce3d1bc1 /hw
parente77d927f1a1876ef341b4df287939b0c935703d5 (diff)
parent75c74ccbe1d4175edb79c6f810c2207dcf5edb22 (diff)
Merge remote-tracking branch 'remotes/agraf/tags/signed-ppc-for-upstream' into staging
Patch queue for ppc - 2015-01-07 New year's release. This time's highlights: - E500: More RAM support - pseries: New SLOF release - Migration fixes - Simplify USB spawning logic, removes support for explicit usb=off - TCG: Simple untansactional TM emulation # gpg: Signature made Wed 07 Jan 2015 15:19:37 GMT using RSA key ID 03FEDC60 # gpg: Good signature from "Alexander Graf <agraf@suse.de>" # gpg: aka "Alexander Graf <alex@csgraf.de>" * remotes/agraf/tags/signed-ppc-for-upstream: (37 commits) hw/ppc/mac_newworld: simplify usb controller creation logic hw/ppc/spapr: simplify usb controller creation logic hw/ppc/mac_newworld: QOMified mac99 machines hw/usb: simplified usb_enabled hw/machine: added machine_usb wrapper hw/ppc: modified the condition for usb controllers to be created for some ppc machines target-ppc: Cast ssize_t to size_t before printing with %zx target-ppc: Mark SR() and gen_sync_exception() as !CONFIG_USER_ONLY PPC: e500: Fix GPIO controller interrupt number target-ppc: Introduce Privileged TM Noops target-ppc: Introduce tcheck target-ppc: Introduce TM Noops target-ppc: Introduce tbegin target-ppc: Introduce TEXASRU Bit Fields target-ppc: Power8 Supports Transactional Memory target-ppc: Introduce tm_enabled Bit to CPU State target-ppc: Introduce Feature Flag for Transactional Memory target-ppc: Introduce Instruction Type for Transactional Memory pseries: Update SLOF firmware image to 20141202 PPC: Fix crash on spapr_tce_table_finalize() ... Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Diffstat (limited to 'hw')
-rw-r--r--hw/pci-host/ppce500.c113
-rw-r--r--hw/ppc/e500.c36
-rw-r--r--hw/ppc/e500.h5
-rw-r--r--hw/ppc/e500plat.c5
-rw-r--r--hw/ppc/mpc8544ds.c11
-rw-r--r--hw/ppc/ppc.c4
-rw-r--r--hw/ppc/spapr.c62
-rw-r--r--hw/ppc/spapr_iommu.c6
8 files changed, 202 insertions, 40 deletions
diff --git a/hw/pci-host/ppce500.c b/hw/pci-host/ppce500.c
index 1b4c0f0023..574f8b2efb 100644
--- a/hw/pci-host/ppce500.c
+++ b/hw/pci-host/ppce500.c
@@ -62,11 +62,19 @@
#define PPCE500_PCI_NR_POBS 5
#define PPCE500_PCI_NR_PIBS 3
+#define PIWAR_EN 0x80000000 /* Enable */
+#define PIWAR_PF 0x20000000 /* prefetch */
+#define PIWAR_TGI_LOCAL 0x00f00000 /* target - local memory */
+#define PIWAR_READ_SNOOP 0x00050000
+#define PIWAR_WRITE_SNOOP 0x00005000
+#define PIWAR_SZ_MASK 0x0000003f
+
struct pci_outbound {
uint32_t potar;
uint32_t potear;
uint32_t powbar;
uint32_t powar;
+ MemoryRegion mem;
};
struct pci_inbound {
@@ -74,6 +82,7 @@ struct pci_inbound {
uint32_t piwbar;
uint32_t piwbear;
uint32_t piwar;
+ MemoryRegion mem;
};
#define TYPE_PPC_E500_PCI_HOST_BRIDGE "e500-pcihost"
@@ -91,10 +100,13 @@ struct PPCE500PCIState {
uint32_t irq_num[PCI_NUM_PINS];
uint32_t first_slot;
uint32_t first_pin_irq;
+ AddressSpace bm_as;
+ MemoryRegion bm;
/* mmio maps */
MemoryRegion container;
MemoryRegion iomem;
MemoryRegion pio;
+ MemoryRegion busmem;
};
#define TYPE_PPC_E500_PCI_BRIDGE "e500-host-bridge"
@@ -181,6 +193,71 @@ static uint64_t pci_reg_read4(void *opaque, hwaddr addr,
return value;
}
+/* DMA mapping */
+static void e500_update_piw(PPCE500PCIState *pci, int idx)
+{
+ uint64_t tar = ((uint64_t)pci->pib[idx].pitar) << 12;
+ uint64_t wbar = ((uint64_t)pci->pib[idx].piwbar) << 12;
+ uint64_t war = pci->pib[idx].piwar;
+ uint64_t size = 2ULL << (war & PIWAR_SZ_MASK);
+ MemoryRegion *address_space_mem = get_system_memory();
+ MemoryRegion *mem = &pci->pib[idx].mem;
+ MemoryRegion *bm = &pci->bm;
+ char *name;
+
+ if (memory_region_is_mapped(mem)) {
+ /* Before we modify anything, unmap and destroy the region */
+ memory_region_del_subregion(bm, mem);
+ object_unparent(OBJECT(mem));
+ }
+
+ if (!(war & PIWAR_EN)) {
+ /* Not enabled, nothing to do */
+ return;
+ }
+
+ name = g_strdup_printf("PCI Inbound Window %d", idx);
+ memory_region_init_alias(mem, OBJECT(pci), name, address_space_mem, tar,
+ size);
+ memory_region_add_subregion_overlap(bm, wbar, mem, -1);
+ g_free(name);
+
+ pci_debug("%s: Added window of size=%#lx from PCI=%#lx to CPU=%#lx\n",
+ __func__, size, wbar, tar);
+}
+
+/* BAR mapping */
+static void e500_update_pow(PPCE500PCIState *pci, int idx)
+{
+ uint64_t tar = ((uint64_t)pci->pob[idx].potar) << 12;
+ uint64_t wbar = ((uint64_t)pci->pob[idx].powbar) << 12;
+ uint64_t war = pci->pob[idx].powar;
+ uint64_t size = 2ULL << (war & PIWAR_SZ_MASK);
+ MemoryRegion *mem = &pci->pob[idx].mem;
+ MemoryRegion *address_space_mem = get_system_memory();
+ char *name;
+
+ if (memory_region_is_mapped(mem)) {
+ /* Before we modify anything, unmap and destroy the region */
+ memory_region_del_subregion(address_space_mem, mem);
+ object_unparent(OBJECT(mem));
+ }
+
+ if (!(war & PIWAR_EN)) {
+ /* Not enabled, nothing to do */
+ return;
+ }
+
+ name = g_strdup_printf("PCI Outbound Window %d", idx);
+ memory_region_init_alias(mem, OBJECT(pci), name, &pci->busmem, tar,
+ size);
+ memory_region_add_subregion(address_space_mem, wbar, mem);
+ g_free(name);
+
+ pci_debug("%s: Added window of size=%#lx from CPU=%#lx to PCI=%#lx\n",
+ __func__, size, wbar, tar);
+}
+
static void pci_reg_write4(void *opaque, hwaddr addr,
uint64_t value, unsigned size)
{
@@ -199,18 +276,22 @@ static void pci_reg_write4(void *opaque, hwaddr addr,
case PPCE500_PCI_OW3:
case PPCE500_PCI_OW4:
idx = (addr >> 5) & 0x7;
- switch (addr & 0xC) {
+ switch (addr & 0x1F) {
case PCI_POTAR:
pci->pob[idx].potar = value;
+ e500_update_pow(pci, idx);
break;
case PCI_POTEAR:
pci->pob[idx].potear = value;
+ e500_update_pow(pci, idx);
break;
case PCI_POWBAR:
pci->pob[idx].powbar = value;
+ e500_update_pow(pci, idx);
break;
case PCI_POWAR:
pci->pob[idx].powar = value;
+ e500_update_pow(pci, idx);
break;
default:
break;
@@ -221,18 +302,22 @@ static void pci_reg_write4(void *opaque, hwaddr addr,
case PPCE500_PCI_IW2:
case PPCE500_PCI_IW1:
idx = ((addr >> 5) & 0x3) - 1;
- switch (addr & 0xC) {
+ switch (addr & 0x1F) {
case PCI_PITAR:
pci->pib[idx].pitar = value;
+ e500_update_piw(pci, idx);
break;
case PCI_PIWBAR:
pci->pib[idx].piwbar = value;
+ e500_update_piw(pci, idx);
break;
case PCI_PIWBEAR:
pci->pib[idx].piwbear = value;
+ e500_update_piw(pci, idx);
break;
case PCI_PIWAR:
pci->pib[idx].piwar = value;
+ e500_update_piw(pci, idx);
break;
default:
break;
@@ -349,13 +434,20 @@ static int e500_pcihost_bridge_initfn(PCIDevice *d)
return 0;
}
+static AddressSpace *e500_pcihost_set_iommu(PCIBus *bus, void *opaque,
+ int devfn)
+{
+ PPCE500PCIState *s = opaque;
+
+ return &s->bm_as;
+}
+
static int e500_pcihost_initfn(SysBusDevice *dev)
{
PCIHostState *h;
PPCE500PCIState *s;
PCIBus *b;
int i;
- MemoryRegion *address_space_mem = get_system_memory();
h = PCI_HOST_BRIDGE(dev);
s = PPC_E500_PCI_HOST_BRIDGE(dev);
@@ -369,12 +461,22 @@ static int e500_pcihost_initfn(SysBusDevice *dev)
}
memory_region_init(&s->pio, OBJECT(s), "pci-pio", PCIE500_PCI_IOLEN);
+ memory_region_init(&s->busmem, OBJECT(s), "pci bus memory", UINT64_MAX);
+
+ /* PIO lives at the bottom of our bus space */
+ memory_region_add_subregion_overlap(&s->busmem, 0, &s->pio, -2);
b = pci_register_bus(DEVICE(dev), NULL, mpc85xx_pci_set_irq,
- mpc85xx_pci_map_irq, s, address_space_mem,
- &s->pio, PCI_DEVFN(s->first_slot, 0), 4, TYPE_PCI_BUS);
+ mpc85xx_pci_map_irq, s, &s->busmem, &s->pio,
+ PCI_DEVFN(s->first_slot, 0), 4, TYPE_PCI_BUS);
h->bus = b;
+ /* Set up PCI view of memory */
+ memory_region_init(&s->bm, OBJECT(s), "bm-e500", UINT64_MAX);
+ memory_region_add_subregion(&s->bm, 0x0, &s->busmem);
+ address_space_init(&s->bm_as, &s->bm, "pci-bm");
+ pci_setup_iommu(b, e500_pcihost_set_iommu, s);
+
pci_create_simple(b, 0, "e500-host-bridge");
memory_region_init(&s->container, OBJECT(h), "pci-container", PCIE500_ALL_SIZE);
@@ -388,7 +490,6 @@ static int e500_pcihost_initfn(SysBusDevice *dev)
memory_region_add_subregion(&s->container, PCIE500_CFGDATA, &h->data_mem);
memory_region_add_subregion(&s->container, PCIE500_REG_BASE, &s->iomem);
sysbus_init_mmio(dev, &s->container);
- sysbus_init_mmio(dev, &s->pio);
pci_bus_set_route_irq_fn(b, e500_route_intx_pin_to_irq);
return 0;
diff --git a/hw/ppc/e500.c b/hw/ppc/e500.c
index 2832fc0da4..7e17d180c6 100644
--- a/hw/ppc/e500.c
+++ b/hw/ppc/e500.c
@@ -51,21 +51,16 @@
#define RAM_SIZES_ALIGN (64UL << 20)
/* TODO: parameterize */
-#define MPC8544_CCSRBAR_BASE 0xE0000000ULL
#define MPC8544_CCSRBAR_SIZE 0x00100000ULL
#define MPC8544_MPIC_REGS_OFFSET 0x40000ULL
#define MPC8544_MSI_REGS_OFFSET 0x41600ULL
#define MPC8544_SERIAL0_REGS_OFFSET 0x4500ULL
#define MPC8544_SERIAL1_REGS_OFFSET 0x4600ULL
#define MPC8544_PCI_REGS_OFFSET 0x8000ULL
-#define MPC8544_PCI_REGS_BASE (MPC8544_CCSRBAR_BASE + \
- MPC8544_PCI_REGS_OFFSET)
#define MPC8544_PCI_REGS_SIZE 0x1000ULL
-#define MPC8544_PCI_IO 0xE1000000ULL
#define MPC8544_UTIL_OFFSET 0xe0000ULL
-#define MPC8544_SPIN_BASE 0xEF000000ULL
#define MPC8XXX_GPIO_OFFSET 0x000FF000ULL
-#define MPC8XXX_GPIO_IRQ 43
+#define MPC8XXX_GPIO_IRQ 47
struct boot_info
{
@@ -293,12 +288,12 @@ static int ppce500_load_device_tree(MachineState *machine,
int len;
uint32_t pci_ranges[14] =
{
- 0x2000000, 0x0, 0xc0000000,
- 0x0, 0xc0000000,
+ 0x2000000, 0x0, params->pci_mmio_bus_base,
+ params->pci_mmio_base >> 32, params->pci_mmio_base,
0x0, 0x20000000,
0x1000000, 0x0, 0x0,
- 0x0, 0xe1000000,
+ params->pci_pio_base >> 32, params->pci_pio_base,
0x0, 0x10000,
};
QemuOpts *machine_opts = qemu_get_machine_opts();
@@ -389,7 +384,7 @@ static int ppce500_load_device_tree(MachineState *machine,
CPUState *cpu;
PowerPCCPU *pcpu;
char cpu_name[128];
- uint64_t cpu_release_addr = MPC8544_SPIN_BASE + (i * 0x20);
+ uint64_t cpu_release_addr = params->spin_base + (i * 0x20);
cpu = qemu_get_cpu(i);
if (cpu == NULL) {
@@ -426,7 +421,7 @@ static int ppce500_load_device_tree(MachineState *machine,
qemu_fdt_add_subnode(fdt, "/aliases");
/* XXX These should go into their respective devices' code */
- snprintf(soc, sizeof(soc), "/soc@%llx", MPC8544_CCSRBAR_BASE);
+ snprintf(soc, sizeof(soc), "/soc@%"PRIx64, params->ccsrbar_base);
qemu_fdt_add_subnode(fdt, soc);
qemu_fdt_setprop_string(fdt, soc, "device_type", "soc");
qemu_fdt_setprop(fdt, soc, "compatible", compatible_sb,
@@ -434,7 +429,7 @@ static int ppce500_load_device_tree(MachineState *machine,
qemu_fdt_setprop_cell(fdt, soc, "#address-cells", 1);
qemu_fdt_setprop_cell(fdt, soc, "#size-cells", 1);
qemu_fdt_setprop_cells(fdt, soc, "ranges", 0x0,
- MPC8544_CCSRBAR_BASE >> 32, MPC8544_CCSRBAR_BASE,
+ params->ccsrbar_base >> 32, params->ccsrbar_base,
MPC8544_CCSRBAR_SIZE);
/* XXX should contain a reasonable value */
qemu_fdt_setprop_cell(fdt, soc, "bus-frequency", 0);
@@ -493,7 +488,8 @@ static int ppce500_load_device_tree(MachineState *machine,
qemu_fdt_setprop_cell(fdt, msi, "phandle", msi_ph);
qemu_fdt_setprop_cell(fdt, msi, "linux,phandle", msi_ph);
- snprintf(pci, sizeof(pci), "/pci@%llx", MPC8544_PCI_REGS_BASE);
+ snprintf(pci, sizeof(pci), "/pci@%llx",
+ params->ccsrbar_base + MPC8544_PCI_REGS_OFFSET);
qemu_fdt_add_subnode(fdt, pci);
qemu_fdt_setprop_cell(fdt, pci, "cell-index", 0);
qemu_fdt_setprop_string(fdt, pci, "compatible", "fsl,mpc8540-pci");
@@ -512,8 +508,10 @@ static int ppce500_load_device_tree(MachineState *machine,
}
qemu_fdt_setprop_cell(fdt, pci, "fsl,msi", msi_ph);
qemu_fdt_setprop(fdt, pci, "ranges", pci_ranges, sizeof(pci_ranges));
- qemu_fdt_setprop_cells(fdt, pci, "reg", MPC8544_PCI_REGS_BASE >> 32,
- MPC8544_PCI_REGS_BASE, 0, 0x1000);
+ qemu_fdt_setprop_cells(fdt, pci, "reg",
+ (params->ccsrbar_base + MPC8544_PCI_REGS_OFFSET) >> 32,
+ (params->ccsrbar_base + MPC8544_PCI_REGS_OFFSET),
+ 0, 0x1000);
qemu_fdt_setprop_cell(fdt, pci, "clock-frequency", 66666666);
qemu_fdt_setprop_cell(fdt, pci, "#interrupt-cells", 1);
qemu_fdt_setprop_cell(fdt, pci, "#size-cells", 2);
@@ -841,7 +839,7 @@ void ppce500_init(MachineState *machine, PPCE500Params *params)
irqs[i][OPENPIC_OUTPUT_INT] = input[PPCE500_INPUT_INT];
irqs[i][OPENPIC_OUTPUT_CINT] = input[PPCE500_INPUT_CINT];
env->spr_cb[SPR_BOOKE_PIR].default_value = cs->cpu_index = i;
- env->mpic_iack = MPC8544_CCSRBAR_BASE +
+ env->mpic_iack = params->ccsrbar_base +
MPC8544_MPIC_REGS_OFFSET + 0xa0;
ppc_booke_timers_init(cpu, 400000000, PPC_TIMER_E500);
@@ -875,7 +873,7 @@ void ppce500_init(MachineState *machine, PPCE500Params *params)
qdev_init_nofail(dev);
ccsr = CCSR(dev);
ccsr_addr_space = &ccsr->ccsr_space;
- memory_region_add_subregion(address_space_mem, MPC8544_CCSRBAR_BASE,
+ memory_region_add_subregion(address_space_mem, params->ccsrbar_base,
ccsr_addr_space);
mpic = ppce500_init_mpic(params, ccsr_addr_space, irqs);
@@ -917,8 +915,6 @@ void ppce500_init(MachineState *machine, PPCE500Params *params)
if (!pci_bus)
printf("couldn't create PCI controller!\n");
- sysbus_mmio_map(SYS_BUS_DEVICE(dev), 1, MPC8544_PCI_IO);
-
if (pci_bus) {
/* Register network interfaces. */
for (i = 0; i < nb_nics; i++) {
@@ -927,7 +923,7 @@ void ppce500_init(MachineState *machine, PPCE500Params *params)
}
/* Register spinning region */
- sysbus_create_simple("e500-spin", MPC8544_SPIN_BASE, NULL);
+ sysbus_create_simple("e500-spin", params->spin_base, NULL);
if (cur_base < (32 * 1024 * 1024)) {
/* u-boot occupies memory up to 32MB, so load blobs above */
diff --git a/hw/ppc/e500.h b/hw/ppc/e500.h
index 9f61ab2b1c..ef224ea5e6 100644
--- a/hw/ppc/e500.h
+++ b/hw/ppc/e500.h
@@ -17,6 +17,11 @@ typedef struct PPCE500Params {
hwaddr platform_bus_size;
int platform_bus_first_irq;
int platform_bus_num_irqs;
+ hwaddr ccsrbar_base;
+ hwaddr pci_pio_base;
+ hwaddr pci_mmio_base;
+ hwaddr pci_mmio_bus_base;
+ hwaddr spin_base;
} PPCE500Params;
void ppce500_init(MachineState *machine, PPCE500Params *params);
diff --git a/hw/ppc/e500plat.c b/hw/ppc/e500plat.c
index d50ae000ee..14b14eaa7d 100644
--- a/hw/ppc/e500plat.c
+++ b/hw/ppc/e500plat.c
@@ -41,6 +41,11 @@ static void e500plat_init(MachineState *machine)
.platform_bus_size = (128ULL * 1024 * 1024),
.platform_bus_first_irq = 5,
.platform_bus_num_irqs = 10,
+ .ccsrbar_base = 0xFE0000000ULL,
+ .pci_pio_base = 0xFE1000000ULL,
+ .pci_mmio_base = 0xC00000000ULL,
+ .pci_mmio_bus_base = 0xE0000000ULL,
+ .spin_base = 0xFEF000000ULL,
};
/* Older KVM versions don't support EPR which breaks guests when we announce
diff --git a/hw/ppc/mpc8544ds.c b/hw/ppc/mpc8544ds.c
index b99f74af75..3a3b141e43 100644
--- a/hw/ppc/mpc8544ds.c
+++ b/hw/ppc/mpc8544ds.c
@@ -15,6 +15,7 @@
#include "hw/boards.h"
#include "sysemu/device_tree.h"
#include "hw/ppc/openpic.h"
+#include "qemu/error-report.h"
static void mpc8544ds_fixup_devtree(PPCE500Params *params, void *fdt)
{
@@ -33,8 +34,18 @@ static void mpc8544ds_init(MachineState *machine)
.pci_nr_slots = 2,
.fixup_devtree = mpc8544ds_fixup_devtree,
.mpic_version = OPENPIC_MODEL_FSL_MPIC_20,
+ .ccsrbar_base = 0xE0000000ULL,
+ .pci_mmio_base = 0xC0000000ULL,
+ .pci_mmio_bus_base = 0xC0000000ULL,
+ .pci_pio_base = 0xE1000000ULL,
+ .spin_base = 0xEF000000ULL,
};
+ if (machine->ram_size > 0xc0000000) {
+ error_report("The MPC8544DS board only supports up to 3GB of RAM");
+ exit(1);
+ }
+
ppce500_init(machine, &params);
}
diff --git a/hw/ppc/ppc.c b/hw/ppc/ppc.c
index bec82cd7a9..5ce565d5ec 100644
--- a/hw/ppc/ppc.c
+++ b/hw/ppc/ppc.c
@@ -844,7 +844,7 @@ static void timebase_pre_save(void *opaque)
return;
}
- tb->time_of_the_day_ns = get_clock_realtime();
+ tb->time_of_the_day_ns = qemu_clock_get_ns(QEMU_CLOCK_HOST);
/*
* tb_offset is only expected to be changed by migration so
* there is no need to update it from KVM here
@@ -873,7 +873,7 @@ static int timebase_post_load(void *opaque, int version_id)
* We try to adjust timebase by downtime if host clocks are not
* too much out of sync (1 second for now).
*/
- host_ns = get_clock_realtime();
+ host_ns = qemu_clock_get_ns(QEMU_CLOCK_HOST);
ns_diff = MAX(0, host_ns - tb_remote->time_of_the_day_ns);
migration_duration_ns = MIN(NSEC_PER_SEC, ns_diff);
migration_duration_tb = muldiv64(migration_duration_ns, freq, NSEC_PER_SEC);
diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
index 53c4116ed3..b560459e83 100644
--- a/hw/ppc/spapr.c
+++ b/hw/ppc/spapr.c
@@ -819,9 +819,16 @@ static void emulate_spapr_hypercall(PowerPCCPU *cpu)
}
}
+#define HPTE(_table, _i) (void *)(((uint64_t *)(_table)) + ((_i) * 2))
+#define HPTE_VALID(_hpte) (tswap64(*((uint64_t *)(_hpte))) & HPTE64_V_VALID)
+#define HPTE_DIRTY(_hpte) (tswap64(*((uint64_t *)(_hpte))) & HPTE64_V_HPTE_DIRTY)
+#define CLEAN_HPTE(_hpte) ((*(uint64_t *)(_hpte)) &= tswap64(~HPTE64_V_HPTE_DIRTY))
+#define DIRTY_HPTE(_hpte) ((*(uint64_t *)(_hpte)) |= tswap64(HPTE64_V_HPTE_DIRTY))
+
static void spapr_reset_htab(sPAPREnvironment *spapr)
{
long shift;
+ int index;
/* allocate hash page table. For now we always make this 16mb,
* later we should probably make it scale to the size of guest
@@ -833,6 +840,11 @@ static void spapr_reset_htab(sPAPREnvironment *spapr)
/* Kernel handles htab, we don't need to allocate one */
spapr->htab_shift = shift;
kvmppc_kern_htab = true;
+
+ /* Tell readers to update their file descriptor */
+ if (spapr->htab_fd >= 0) {
+ spapr->htab_fd_stale = true;
+ }
} else {
if (!spapr->htab) {
/* Allocate an htab if we don't yet have one */
@@ -841,6 +853,10 @@ static void spapr_reset_htab(sPAPREnvironment *spapr)
/* And clear it */
memset(spapr->htab, 0, HTAB_SIZE(spapr));
+
+ for (index = 0; index < HTAB_SIZE(spapr) / HASH_PTE_SIZE_64; index++) {
+ DIRTY_HPTE(HPTE(spapr->htab, index));
+ }
}
/* Update the RMA size if necessary */
@@ -867,6 +883,28 @@ static int find_unknown_sysbus_device(SysBusDevice *sbdev, void *opaque)
return 0;
}
+/*
+ * A guest reset will cause spapr->htab_fd to become stale if being used.
+ * Reopen the file descriptor to make sure the whole HTAB is properly read.
+ */
+static int spapr_check_htab_fd(sPAPREnvironment *spapr)
+{
+ int rc = 0;
+
+ if (spapr->htab_fd_stale) {
+ close(spapr->htab_fd);
+ spapr->htab_fd = kvmppc_get_htab_fd(false);
+ if (spapr->htab_fd < 0) {
+ error_report("Unable to open fd for reading hash table from KVM: "
+ "%s", strerror(errno));
+ rc = -1;
+ }
+ spapr->htab_fd_stale = false;
+ }
+
+ return rc;
+}
+
static void ppc_spapr_reset(void)
{
PowerPCCPU *first_ppc_cpu;
@@ -986,11 +1024,6 @@ static const VMStateDescription vmstate_spapr = {
},
};
-#define HPTE(_table, _i) (void *)(((uint64_t *)(_table)) + ((_i) * 2))
-#define HPTE_VALID(_hpte) (tswap64(*((uint64_t *)(_hpte))) & HPTE64_V_VALID)
-#define HPTE_DIRTY(_hpte) (tswap64(*((uint64_t *)(_hpte))) & HPTE64_V_HPTE_DIRTY)
-#define CLEAN_HPTE(_hpte) ((*(uint64_t *)(_hpte)) &= tswap64(~HPTE64_V_HPTE_DIRTY))
-
static int htab_save_setup(QEMUFile *f, void *opaque)
{
sPAPREnvironment *spapr = opaque;
@@ -1005,6 +1038,7 @@ static int htab_save_setup(QEMUFile *f, void *opaque)
assert(kvm_enabled());
spapr->htab_fd = kvmppc_get_htab_fd(false);
+ spapr->htab_fd_stale = false;
if (spapr->htab_fd < 0) {
fprintf(stderr, "Unable to open fd for reading hash table from KVM: %s\n",
strerror(errno));
@@ -1037,7 +1071,7 @@ static void htab_save_first_pass(QEMUFile *f, sPAPREnvironment *spapr,
/* Consume valid HPTEs */
chunkstart = index;
- while ((index < htabslots)
+ while ((index < htabslots) && (index - chunkstart < USHRT_MAX)
&& HPTE_VALID(HPTE(spapr->htab, index))) {
index++;
CLEAN_HPTE(HPTE(spapr->htab, index));
@@ -1089,7 +1123,7 @@ static int htab_save_later_pass(QEMUFile *f, sPAPREnvironment *spapr,
chunkstart = index;
/* Consume valid dirty HPTEs */
- while ((index < htabslots)
+ while ((index < htabslots) && (index - chunkstart < USHRT_MAX)
&& HPTE_DIRTY(HPTE(spapr->htab, index))
&& HPTE_VALID(HPTE(spapr->htab, index))) {
CLEAN_HPTE(HPTE(spapr->htab, index));
@@ -1099,7 +1133,7 @@ static int htab_save_later_pass(QEMUFile *f, sPAPREnvironment *spapr,
invalidstart = index;
/* Consume invalid dirty HPTEs */
- while ((index < htabslots)
+ while ((index < htabslots) && (index - invalidstart < USHRT_MAX)
&& HPTE_DIRTY(HPTE(spapr->htab, index))
&& !HPTE_VALID(HPTE(spapr->htab, index))) {
CLEAN_HPTE(HPTE(spapr->htab, index));
@@ -1157,6 +1191,11 @@ static int htab_save_iterate(QEMUFile *f, void *opaque)
if (!spapr->htab) {
assert(kvm_enabled());
+ rc = spapr_check_htab_fd(spapr);
+ if (rc < 0) {
+ return rc;
+ }
+
rc = kvmppc_save_htab(f, spapr->htab_fd,
MAX_KVM_BUF_SIZE, MAX_ITERATION_NS);
if (rc < 0) {
@@ -1188,6 +1227,11 @@ static int htab_save_complete(QEMUFile *f, void *opaque)
assert(kvm_enabled());
+ rc = spapr_check_htab_fd(spapr);
+ if (rc < 0) {
+ return rc;
+ }
+
rc = kvmppc_save_htab(f, spapr->htab_fd, MAX_KVM_BUF_SIZE, -1);
if (rc < 0) {
return rc;
@@ -1438,7 +1482,7 @@ static void ppc_spapr_init(MachineState *machine)
}
if (spapr->rtas_size > RTAS_MAX_SIZE) {
hw_error("RTAS too big ! 0x%zx bytes (max is 0x%x)\n",
- spapr->rtas_size, RTAS_MAX_SIZE);
+ (size_t)spapr->rtas_size, RTAS_MAX_SIZE);
exit(1);
}
g_free(filename);
diff --git a/hw/ppc/spapr_iommu.c b/hw/ppc/spapr_iommu.c
index 6c91d8edd8..da474740c0 100644
--- a/hw/ppc/spapr_iommu.c
+++ b/hw/ppc/spapr_iommu.c
@@ -173,9 +173,9 @@ sPAPRTCETable *spapr_tce_new_table(DeviceState *owner, uint32_t liobn,
return tcet;
}
-static void spapr_tce_table_finalize(Object *obj)
+static void spapr_tce_table_unrealize(DeviceState *dev, Error **errp)
{
- sPAPRTCETable *tcet = SPAPR_TCE_TABLE(obj);
+ sPAPRTCETable *tcet = SPAPR_TCE_TABLE(dev);
QLIST_REMOVE(tcet, list);
@@ -420,6 +420,7 @@ static void spapr_tce_table_class_init(ObjectClass *klass, void *data)
DeviceClass *dc = DEVICE_CLASS(klass);
dc->init = spapr_tce_table_realize;
dc->reset = spapr_tce_reset;
+ dc->unrealize = spapr_tce_table_unrealize;
QLIST_INIT(&spapr_tce_tables);
@@ -435,7 +436,6 @@ static TypeInfo spapr_tce_table_info = {
.parent = TYPE_DEVICE,
.instance_size = sizeof(sPAPRTCETable),
.class_init = spapr_tce_table_class_init,
- .instance_finalize = spapr_tce_table_finalize,
};
static void register_types(void)