@@ -220,6 +220,7 @@ static void nvdimm_write_label_data(NVDIMMDevice *nvdimm, const void *buf,
static Property nvdimm_properties[] = {
DEFINE_PROP_BOOL(NVDIMM_UNARMED_PROP, NVDIMMDevice, unarmed, false),
+ DEFINE_PROP_BOOL(NVDIMM_SYNC_DAX_PROP, NVDIMMDevice, sync_dax, false),
DEFINE_PROP_END_OF_LIST(),
};
@@ -22,6 +22,7 @@
* THE SOFTWARE.
*/
#include "qemu/osdep.h"
+#include "qemu/cutils.h"
#include "qapi/error.h"
#include "hw/ppc/spapr_drc.h"
#include "hw/ppc/spapr_nvdimm.h"
@@ -155,6 +156,11 @@ static int spapr_dt_nvdimm(SpaprMachineState *spapr, void *fdt,
"operating-system")));
_FDT(fdt_setprop(fdt, child_offset, "ibm,cache-flush-required", NULL, 0));
+ if (!nvdimm->sync_dax) {
+ _FDT(fdt_setprop(fdt, child_offset, "ibm,async-flush-required",
+ NULL, 0));
+ }
+
return child_offset;
}
@@ -370,6 +376,78 @@ static target_ulong h_scm_bind_mem(PowerPCCPU *cpu, SpaprMachineState *spapr,
return H_SUCCESS;
}
+typedef struct SCMAsyncFlushData {
+ int fd;
+ uint64_t token;
+} SCMAsyncFlushData;
+
+static int flush_worker_cb(void *opaque)
+{
+ int ret = H_SUCCESS;
+ SCMAsyncFlushData *req_data = opaque;
+
+ /* flush raw backing image */
+ if (qemu_fdatasync(req_data->fd) < 0) {
+ error_report("papr_scm: Could not sync nvdimm to backend file: %s",
+ strerror(errno));
+ ret = H_HARDWARE;
+ }
+
+ g_free(req_data);
+
+ return ret;
+}
+
+static target_ulong h_scm_async_flush(PowerPCCPU *cpu, SpaprMachineState *spapr,
+ target_ulong opcode, target_ulong *args)
+{
+ int ret;
+ uint32_t drc_index = args[0];
+ uint64_t continue_token = args[1];
+ SpaprDrc *drc = spapr_drc_by_index(drc_index);
+ PCDIMMDevice *dimm;
+ HostMemoryBackend *backend = NULL;
+ SCMAsyncFlushData *req_data = NULL;
+
+ if (!drc || !drc->dev ||
+ spapr_drc_type(drc) != SPAPR_DR_CONNECTOR_TYPE_PMEM) {
+ return H_PARAMETER;
+ }
+
+ if (continue_token != 0) {
+ ret = spapr_drc_get_async_hcall_status(drc, continue_token);
+ if (ret == H_BUSY) {
+ args[0] = continue_token;
+ return H_LONG_BUSY_ORDER_1_SEC;
+ }
+
+ return ret;
+ }
+
+ dimm = PC_DIMM(drc->dev);
+ backend = MEMORY_BACKEND(dimm->hostmem);
+
+ req_data = g_malloc0(sizeof(SCMAsyncFlushData));
+ req_data->fd = memory_region_get_fd(&backend->mr);
+
+ continue_token = spapr_drc_get_new_async_hcall_token(drc);
+ if (!continue_token) {
+ g_free(req_data);
+ return H_P2;
+ }
+ req_data->token = continue_token;
+
+ spapr_drc_run_async_hcall(drc, continue_token, &flush_worker_cb, req_data);
+
+ ret = spapr_drc_get_async_hcall_status(drc, continue_token);
+ if (ret == H_BUSY) {
+ args[0] = req_data->token;
+ return ret;
+ }
+
+ return ret;
+}
+
static target_ulong h_scm_unbind_mem(PowerPCCPU *cpu, SpaprMachineState *spapr,
target_ulong opcode, target_ulong *args)
{
@@ -486,6 +564,7 @@ static void spapr_scm_register_types(void)
spapr_register_hypercall(H_SCM_BIND_MEM, h_scm_bind_mem);
spapr_register_hypercall(H_SCM_UNBIND_MEM, h_scm_unbind_mem);
spapr_register_hypercall(H_SCM_UNBIND_ALL, h_scm_unbind_all);
+ spapr_register_hypercall(H_SCM_ASYNC_FLUSH, h_scm_async_flush);
}
type_init(spapr_scm_register_types)
@@ -51,6 +51,7 @@ OBJECT_DECLARE_TYPE(NVDIMMDevice, NVDIMMClass, NVDIMM)
#define NVDIMM_LABEL_SIZE_PROP "label-size"
#define NVDIMM_UUID_PROP "uuid"
#define NVDIMM_UNARMED_PROP "unarmed"
+#define NVDIMM_SYNC_DAX_PROP "sync-dax"
struct NVDIMMDevice {
/* private */
@@ -85,6 +86,15 @@ struct NVDIMMDevice {
*/
bool unarmed;
+ /*
+ * On PPC64,
+ * The 'off' value results in the async-flush-required property set
+ * in the device tree for pseries machines. When 'off', the guest
+ * initiates explicity flush requests to the backend device ensuring
+ * write persistence.
+ */
+ bool sync_dax;
+
/*
* The PPC64 - spapr requires each nvdimm device have a uuid.
*/
@@ -535,8 +535,9 @@ struct SpaprMachineState {
#define H_SCM_BIND_MEM 0x3EC
#define H_SCM_UNBIND_MEM 0x3F0
#define H_SCM_UNBIND_ALL 0x3FC
+#define H_SCM_ASYNC_FLUSH 0x4A0
-#define MAX_HCALL_OPCODE H_SCM_UNBIND_ALL
+#define MAX_HCALL_OPCODE H_SCM_ASYNC_FLUSH
/* The hcalls above are standardized in PAPR and implemented by pHyp
* as well.
When the persistent memory beacked by a file, a cpu cache flush instruction is not sufficient to ensure the stores are correctly flushed to the media. The patch implements the async hcalls for flush operation on demand from the guest kernel. The device option sync-dax is by default off and enables explicit asynchronous flush requests from guest. It can be disabled by setting syn-dax=on. Signed-off-by: Shivaprasad G Bhat <sbhat@linux.ibm.com> --- hw/mem/nvdimm.c | 1 + hw/ppc/spapr_nvdimm.c | 79 +++++++++++++++++++++++++++++++++++++++++++++++ include/hw/mem/nvdimm.h | 10 ++++++ include/hw/ppc/spapr.h | 3 +- 4 files changed, 92 insertions(+), 1 deletion(-)