I had tested this patch for P9/P8/ guest on powerKVM and powerVM. All
of them works fine.
On Tue, May 22, 2018 at 4:23 PM, Pingfan Liu <kernelfans@xxxxxxxxx> wrote:
> For kexec -p, the boot cpu can be not the cpu0, this causes the problem
> to alloc paca[]. In theory, there is no requirement to assign cpu's logical
> id as its present seq by device tree. But we have something like
> cpu_first_thread_sibling(), which makes assumption on the mapping inside
> a core. Hence partially changing the mapping, i.e. unbind the mapping of
> core while keep the mapping inside a core. After this patch, the core with
> boot-cpu will always be mapped into core 0.
>
> And at present, the code to discovery cpu spreads over two functions:
> early_init_dt_scan_cpus() and smp_setup_cpu_maps().
> This patch tries to fold smp_setup_cpu_maps() into the "previous" one
>
> Signed-off-by: Pingfan Liu <kernelfans@xxxxxxxxx>
> ---
> v5 -> v6:
> simplify the loop logic (Hope it can answer Benjamin's concern)
> concentrate the cpu recovery code to early stage (Hope it can answer Michael's concern)
> Todo: (if this method is accepted)
> fold the whole smp_setup_cpu_maps()
>
> arch/powerpc/include/asm/smp.h | 1 +
> arch/powerpc/kernel/prom.c | 123 ++++++++++++++++++++++++++++---------
> arch/powerpc/kernel/setup-common.c | 58 ++---------------
> drivers/of/fdt.c | 2 +-
> include/linux/of_fdt.h | 2 +
> 5 files changed, 103 insertions(+), 83 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/smp.h b/arch/powerpc/include/asm/smp.h
> index fac963e..80c7693 100644
> --- a/arch/powerpc/include/asm/smp.h
> +++ b/arch/powerpc/include/asm/smp.h
> @@ -30,6 +30,7 @@
> #include <asm/percpu.h>
>
> extern int boot_cpuid;
> +extern int threads_in_core;
> extern int spinning_secondaries;
>
> extern void cpu_die(void);
> diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
> index 4922162..2ae0b4a 100644
> --- a/arch/powerpc/kernel/prom.c
> +++ b/arch/powerpc/kernel/prom.c
> @@ -77,7 +77,6 @@ unsigned long tce_alloc_start, tce_alloc_end;
> u64 ppc64_rma_size;
> #endif
> static phys_addr_t first_memblock_size;
> -static int __initdata boot_cpu_count;
>
> static int __init early_parse_mem(char *p)
> {
> @@ -305,6 +304,14 @@ static void __init check_cpu_feature_properties(unsigned long node)
> }
> }
>
> +struct bootinfo {
> + int boot_thread_id;
> + unsigned int cpu_cnt;
> + int cpu_hwids[NR_CPUS];
> + bool avail[NR_CPUS];
> +};
> +static struct bootinfo *bt_info;
> +
> static int __init early_init_dt_scan_cpus(unsigned long node,
> const char *uname, int depth,
> void *data)
> @@ -312,10 +319,12 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
> const char *type = of_get_flat_dt_prop(node, "device_type", NULL);
> const __be32 *prop;
> const __be32 *intserv;
> - int i, nthreads;
> + int i, nthreads, maxidx;
> int len;
> - int found = -1;
> - int found_thread = 0;
> + int found_thread = -1;
> + struct bootinfo *info = data;
> + bool avail;
> + int rotate_cnt, id;
>
> /* We are scanning "cpu" nodes only */
> if (type == NULL || strcmp(type, "cpu") != 0)
> @@ -325,8 +334,15 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
> intserv = of_get_flat_dt_prop(node, "ibm,ppc-interrupt-server#s", &len);
> if (!intserv)
> intserv = of_get_flat_dt_prop(node, "reg", &len);
> + avail = of_fdt_device_is_available(initial_boot_params, node);
> +#if 0
> + //todo
> + if (!avail)
> + avail = !of_fdt_property_match_string(node,
> + "enable-method", "spin-table");
> +#endif
>
> - nthreads = len / sizeof(int);
> + threads_in_core = nthreads = len / sizeof(int);
>
> /*
> * Now see if any of these threads match our boot cpu.
> @@ -338,9 +354,10 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
> * booted proc.
> */
> if (fdt_version(initial_boot_params) >= 2) {
> + info->cpu_hwids[info->cpu_cnt] =
> + be32_to_cpu(intserv[i]);
> if (be32_to_cpu(intserv[i]) ==
> fdt_boot_cpuid_phys(initial_boot_params)) {
> - found = boot_cpu_count;
> found_thread = i;
> }
> } else {
> @@ -351,22 +368,37 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
> */
> if (of_get_flat_dt_prop(node,
> "linux,boot-cpu", NULL) != NULL)
> - found = boot_cpu_count;
> + found_thread = info->cpu_cnt;
> }
> + info->avail[info->cpu_cnt] = avail;
> +
> #ifdef CONFIG_SMP
> /* logical cpu id is always 0 on UP kernels */
> - boot_cpu_count++;
> + info->cpu_cnt++;
> #endif
> }
>
> /* Not the boot CPU */
> - if (found < 0)
> + if (found_thread < 0)
> return 0;
>
> - DBG("boot cpu: logical %d physical %d\n", found,
> + /* always mapping boot-core to core 0 to cope with kexec -p */
> + maxidx = info->cpu_cnt - 1;
> + rotate_cnt = nthreads;
> + while (rotate_cnt-- > 0) {
> + avail = info->avail[maxidx];
> + id = info->cpu_hwids[maxidx];
> + for (i = maxidx; i > 0; i--) {
> + info->avail[i] = info->avail[i - 1];
> + info->cpu_hwids[i] = info->cpu_hwids[i - 1];
> + }
> + info->avail[i] = avail;
> + info->cpu_hwids[i] = id;
> + }
> +
> + info->boot_thread_id = found_thread;
> + DBG("boot cpu: logical %d physical %d\n", found_thread,
> be32_to_cpu(intserv[found_thread]));
> - boot_cpuid = found;
> - set_hard_smp_processor_id(found, be32_to_cpu(intserv[found_thread]));
>
> /*
> * PAPR defines "logical" PVR values for cpus that
> @@ -675,6 +707,55 @@ static void __init tm_init(void)
> static void tm_init(void) { }
> #endif /* CONFIG_PPC_TRANSACTIONAL_MEM */
>
> +static void early_setup_cpu_mapping(void)
> +{
> + unsigned int cpu, cnt;
> + int nr_cpus_aligned;
> +
> + bt_info = __va(memblock_alloc(sizeof(struct bootinfo),
> + sizeof(unsigned long)));
> + memset(bt_info, 0, sizeof(struct bootinfo));
> + bt_info->boot_thread_id = -1;
> + /* Retrieve CPU related informations from the flat tree
> + * (altivec support, boot CPU ID, ...)
> + */
> + of_scan_flat_dt(early_init_dt_scan_cpus, bt_info);
> +
> + if (bt_info->boot_thread_id < 0) {
> + pr_err("Failed to identify boot CPU !\n");
> + BUG();
> + }
> +
> + boot_cpuid = bt_info->boot_thread_id;
> + /* work around subcore mode */
> + nr_cpus_aligned = _ALIGN_UP(nr_cpu_ids, threads_in_core);
> + if (nr_cpus_aligned != nr_cpu_ids) {
> + pr_info("nr_cpus is forced to be aligned up from: %d to: %d\n",
> + nr_cpu_ids, nr_cpus_aligned);
> + nr_cpu_ids = nr_cpus_aligned;
> + }
> + cnt = (nr_cpu_ids < bt_info->cpu_cnt) ? nr_cpu_ids
> + : bt_info->cpu_cnt;
> +
> + allocate_pacas();
> + for (cpu = 0; cpu < cnt; cpu++) {
> + set_cpu_present(cpu, bt_info->avail[cpu]);
> + DBG("set cpu present: %d -> hwid:%d\n",
> + cpu, bt_info->cpu_hwids[cpu]);
> + set_hard_smp_processor_id(cpu, bt_info->cpu_hwids[cpu]);
> + set_cpu_possible(cpu, true);
> + }
> +
> +#if defined(CONFIG_SMP) && defined(CONFIG_PPC64)
> + /* We'll later wait for secondaries to check in; there are
> + * NCPUS-1 non-boot CPUs :-)
> + */
> + spinning_secondaries = bt_info->cpu_cnt - 1;
> +#endif
> + memblock_free(__pa(bt_info), sizeof(struct bootinfo));
> + bt_info = NULL;
> +}
> +
> void __init early_init_devtree(void *params)
> {
> phys_addr_t limit;
> @@ -745,27 +826,11 @@ void __init early_init_devtree(void *params)
> * FIXME .. and the initrd too? */
> move_device_tree();
>
> - allocate_pacas();
> -
> DBG("Scanning CPUs ...\n");
>
> dt_cpu_ftrs_scan();
>
> - /* Retrieve CPU related informations from the flat tree
> - * (altivec support, boot CPU ID, ...)
> - */
> - of_scan_flat_dt(early_init_dt_scan_cpus, NULL);
> - if (boot_cpuid < 0) {
> - printk("Failed to identify boot CPU !\n");
> - BUG();
> - }
> -
> -#if defined(CONFIG_SMP) && defined(CONFIG_PPC64)
> - /* We'll later wait for secondaries to check in; there are
> - * NCPUS-1 non-boot CPUs :-)
> - */
> - spinning_secondaries = boot_cpu_count - 1;
> -#endif
> + early_setup_cpu_mapping();
>
> mmu_early_init_devtree();
>
> diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c
> index 66f7cc6..46d034a 100644
> --- a/arch/powerpc/kernel/setup-common.c
> +++ b/arch/powerpc/kernel/setup-common.c
> @@ -86,7 +86,9 @@ struct machdep_calls *machine_id;
> EXPORT_SYMBOL(machine_id);
>
> int boot_cpuid = -1;
> +int threads_in_core = 1;
> EXPORT_SYMBOL_GPL(boot_cpuid);
> +EXPORT_SYMBOL_GPL(threads_in_core);
>
> /*
> * These are used in binfmt_elf.c to put aux entries on the stack
> @@ -460,61 +462,11 @@ void __init smp_setup_cpu_maps(void)
> {
> struct device_node *dn;
> int cpu = 0;
> - int nthreads = 1;
> -
> - DBG("smp_setup_cpu_maps()\n");
> -
> - for_each_node_by_type(dn, "cpu") {
> - const __be32 *intserv;
> - __be32 cpu_be;
> - int j, len;
> -
> - DBG(" * %pOF...\n", dn);
> -
> - intserv = of_get_property(dn, "ibm,ppc-interrupt-server#s",
> - &len);
> - if (intserv) {
> - DBG(" ibm,ppc-interrupt-server#s -> %d threads\n",
> - nthreads);
> - } else {
> - DBG(" no ibm,ppc-interrupt-server#s -> 1 thread\n");
> - intserv = of_get_property(dn, "reg", &len);
> - if (!intserv) {
> - cpu_be = cpu_to_be32(cpu);
> - intserv = &cpu_be; /* assume logical == phys */
> - len = 4;
> - }
> - }
> -
> - nthreads = len / sizeof(int);
> -
> - for (j = 0; j < nthreads && cpu < nr_cpu_ids; j++) {
> - bool avail;
> -
> - DBG(" thread %d -> cpu %d (hard id %d)\n",
> - j, cpu, be32_to_cpu(intserv[j]));
> -
> - avail = of_device_is_available(dn);
> - if (!avail)
> - avail = !of_property_match_string(dn,
> - "enable-method", "spin-table");
> -
> - set_cpu_present(cpu, avail);
> - set_hard_smp_processor_id(cpu, be32_to_cpu(intserv[j]));
> - set_cpu_possible(cpu, true);
> - cpu++;
> - }
> -
> - if (cpu >= nr_cpu_ids) {
> - of_node_put(dn);
> - break;
> - }
> - }
>
> /* If no SMT supported, nthreads is forced to 1 */
> if (!cpu_has_feature(CPU_FTR_SMT)) {
> DBG(" SMT disabled ! nthreads forced to 1\n");
> - nthreads = 1;
> + threads_in_core = 1;
> }
>
> #ifdef CONFIG_PPC64
> @@ -539,7 +491,7 @@ void __init smp_setup_cpu_maps(void)
>
> /* Double maxcpus for processors which have SMT capability */
> if (cpu_has_feature(CPU_FTR_SMT))
> - maxcpus *= nthreads;
> + maxcpus *= threads_in_core;
>
> if (maxcpus > nr_cpu_ids) {
> printk(KERN_WARNING
> @@ -565,7 +517,7 @@ void __init smp_setup_cpu_maps(void)
> * every CPU in the system. If that is not the case, then some code
> * here will have to be reworked
> */
> - cpu_init_thread_core_maps(nthreads);
> + cpu_init_thread_core_maps(threads_in_core);
>
> /* Now that possible cpus are set, set nr_cpu_ids for later use */
> setup_nr_cpu_ids();
> diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c
> index 84aa9d6..16d6b02 100644
> --- a/drivers/of/fdt.c
> +++ b/drivers/of/fdt.c
> @@ -130,7 +130,7 @@ bool of_fdt_is_big_endian(const void *blob, unsigned long node)
> return false;
> }
>
> -static bool of_fdt_device_is_available(const void *blob, unsigned long node)
> +bool of_fdt_device_is_available(const void *blob, unsigned long node)
> {
> const char *status = fdt_getprop(blob, node, "status", NULL);
>
> diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h
> index b9cd9eb..28756c5 100644
> --- a/include/linux/of_fdt.h
> +++ b/include/linux/of_fdt.h
> @@ -30,6 +30,8 @@ extern void *of_fdt_get_property(const void *blob,
> int *size);
> extern bool of_fdt_is_big_endian(const void *blob,
> unsigned long node);
> +extern bool of_fdt_device_is_available(const void *blob,
> + unsigned long node);
> extern int of_fdt_match(const void *blob, unsigned long node,
> const char *const *compat);
> extern void *of_fdt_unflatten_tree(const unsigned long *blob,
> --
> 2.7.4
>
_______________________________________________
kexec mailing list
kexec@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/kexec
