Add SLOF-based partition firmware for pSeries machine, allowing more boot options

Currently, the emulated pSeries machine requires the use of the
-kernel parameter in order to explicitly load a guest kernel.  This
means booting from the virtual disk, cdrom or network is not possible.

This patch addresses this limitation by inserting a within-partition
firmware image (derived from the "SLOF" free Open Firmware project).
If -kernel is not specified, qemu will now load the SLOF image, which
has access to the qemu boot device list through the device tree, and
can boot from any of the usual virtual devices.

In order to support the new firmware, an extension to the emulated
machine/hypervisor is necessary.  Unlike Linux, which expects
multi-CPU entry to be handled kexec() style, the SLOF firmware expects
only one CPU to be active at entry, and to use a hypervisor RTAS
method to enable the other CPUs one by one.

This patch also implements this 'start-cpu' method, so that SLOF can
start the secondary CPUs and marshal them into the kexec() holding
pattern ready for entry into the guest OS.  Linux should, and in the
future might directly use the start-cpu method to enable initially
disabled CPUs, but for now it does require kexec() entry.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Alexander Graf <agraf@suse.de>

2 files changed, 109 insertions, 4 deletions

diff --git a/hw/spapr.c b/hw/spapr.c
index 876768631b..1152a25b2e 100644
--- a/hw/spapr.c
+++ b/hw/spapr.c
@@ -44,6 +44,10 @@
 #define INITRD_LOAD_ADDR        0x02800000
 #define FDT_MAX_SIZE            0x10000
 #define RTAS_MAX_SIZE           0x10000
+#define FW_MAX_SIZE             0x400000
+#define FW_FILE_NAME            "slof.bin"
+
+#define MIN_RAM_SLOF		512UL
 
 #define TIMEBASE_FREQ           512000000ULL
 
@@ -57,6 +61,7 @@ static void *spapr_create_fdt(int *fdt_size, ram_addr_t ramsize,
                               sPAPREnvironment *spapr,
                               target_phys_addr_t initrd_base,
                               target_phys_addr_t initrd_size,
+                              const char *boot_device,
                               const char *kernel_cmdline,
                               target_phys_addr_t rtas_addr,
                               target_phys_addr_t rtas_size,
@@ -105,6 +110,7 @@ static void *spapr_create_fdt(int *fdt_size, ram_addr_t ramsize,
                        &start_prop, sizeof(start_prop))));
     _FDT((fdt_property(fdt, "linux,initrd-end",
                        &end_prop, sizeof(end_prop))));
+    _FDT((fdt_property_string(fdt, "qemu,boot-device", boot_device)));
 
     _FDT((fdt_end_node(fdt)));
 
@@ -261,7 +267,7 @@ static void ppc_spapr_init(ram_addr_t ram_size,
     ram_addr_t ram_offset;
     target_phys_addr_t fdt_addr, rtas_addr;
     uint32_t kernel_base, initrd_base;
-    long kernel_size, initrd_size, htab_size, rtas_size;
+    long kernel_size, initrd_size, htab_size, rtas_size, fw_size;
     long pteg_shift = 17;
     int fdt_size;
     char *filename;
@@ -392,13 +398,33 @@ static void ppc_spapr_init(ram_addr_t ram_size,
             initrd_size = 0;
         }
     } else {
-        fprintf(stderr, "pSeries machine needs -kernel for now");
-        exit(1);
+        if (ram_size < (MIN_RAM_SLOF << 20)) {
+            fprintf(stderr, "qemu: pSeries SLOF firmware requires >= "
+                    "%ldM guest RAM\n", MIN_RAM_SLOF);
+            exit(1);
+        }
+        filename = qemu_find_file(QEMU_FILE_TYPE_BIOS, "slof.bin");
+        fw_size = load_image_targphys(filename, 0, FW_MAX_SIZE);
+        if (fw_size < 0) {
+            hw_error("qemu: could not load LPAR rtas '%s'\n", filename);
+            exit(1);
+        }
+        qemu_free(filename);
+        kernel_base = 0x100;
+        initrd_base = 0;
+        initrd_size = 0;
+
+        /* SLOF will startup the secondary CPUs using RTAS,
+           rather than expecting a kexec() style entry */
+        for (i = 0; i < smp_cpus; i++) {
+            envs[i]->halted = 1;
+        }
     }
 
     /* Prepare the device tree */
     fdt = spapr_create_fdt(&fdt_size, ram_size, cpu_model, envs, spapr,
-                           initrd_base, initrd_size, kernel_cmdline,
+                           initrd_base, initrd_size,
+                           boot_device, kernel_cmdline,
                            rtas_addr, rtas_size, pteg_shift + 7);
     assert(fdt != NULL);
 
@@ -409,6 +435,7 @@ static void ppc_spapr_init(ram_addr_t ram_size,
     envs[0]->gpr[3] = fdt_addr;
     envs[0]->gpr[5] = 0;
     envs[0]->hreset_vector = kernel_base;
+    envs[0]->halted = 0;
 }
 
 static QEMUMachine spapr_machine = {
diff --git a/hw/spapr_rtas.c b/hw/spapr_rtas.c
index 72268537a8..16b65422b6 100644
--- a/hw/spapr_rtas.c
+++ b/hw/spapr_rtas.c
@@ -90,6 +90,81 @@ static void rtas_power_off(sPAPREnvironment *spapr,
     rtas_st(rets, 0, 0);
 }
 
+static void rtas_query_cpu_stopped_state(sPAPREnvironment *spapr,
+                                         uint32_t token, uint32_t nargs,
+                                         target_ulong args,
+                                         uint32_t nret, target_ulong rets)
+{
+    target_ulong id;
+    CPUState *env;
+
+    if (nargs != 1 || nret != 2) {
+        rtas_st(rets, 0, -3);
+        return;
+    }
+
+    id = rtas_ld(args, 0);
+    for (env = first_cpu; env; env = env->next_cpu) {
+        if (env->cpu_index != id) {
+            continue;
+        }
+
+        if (env->halted) {
+            rtas_st(rets, 1, 0);
+        } else {
+            rtas_st(rets, 1, 2);
+        }
+
+        rtas_st(rets, 0, 0);
+        return;
+    }
+
+    /* Didn't find a matching cpu */
+    rtas_st(rets, 0, -3);
+}
+
+static void rtas_start_cpu(sPAPREnvironment *spapr,
+                           uint32_t token, uint32_t nargs,
+                           target_ulong args,
+                           uint32_t nret, target_ulong rets)
+{
+    target_ulong id, start, r3;
+    CPUState *env;
+
+    if (nargs != 3 || nret != 1) {
+        rtas_st(rets, 0, -3);
+        return;
+    }
+
+    id = rtas_ld(args, 0);
+    start = rtas_ld(args, 1);
+    r3 = rtas_ld(args, 2);
+
+    for (env = first_cpu; env; env = env->next_cpu) {
+        if (env->cpu_index != id) {
+            continue;
+        }
+
+        if (!env->halted) {
+            rtas_st(rets, 0, -1);
+            return;
+        }
+
+        env->msr = (1ULL << MSR_SF) | (1ULL << MSR_ME);
+        env->nip = start;
+        env->gpr[3] = r3;
+        env->halted = 0;
+
+        qemu_cpu_kick(env);
+
+        rtas_st(rets, 0, 0);
+        return;
+    }
+
+    /* Didn't find a matching cpu */
+    rtas_st(rets, 0, -3);
+}
+
 static struct rtas_call {
     const char *name;
     spapr_rtas_fn fn;
@@ -196,5 +271,8 @@ static void register_core_rtas(void)
     spapr_rtas_register("display-character", rtas_display_character);
     spapr_rtas_register("get-time-of-day", rtas_get_time_of_day);
     spapr_rtas_register("power-off", rtas_power_off);
+    spapr_rtas_register("query-cpu-stopped-state",
+                        rtas_query_cpu_stopped_state);
+    spapr_rtas_register("start-cpu", rtas_start_cpu);
 }
 device_init(register_core_rtas);