Re: [Qemu-devel] [PATCH v3.2 01/31] NUMA: move numa related code to new file numa.c

["Michael S. Tsirkin" <mst@redhat.com>]

On Wed, May 14, 2014 at 05:43:05PM +0800, Hu Tao wrote:
> From: Wanlong Gao <gaowanlong@cn.fujitsu.com>
> 
> Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
> Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>

This is also based on bits by Blue Swirl and
Andre Przywara, I think you should add their S.O.B. lines
from the original commits, just in case we need to locate the authors
down the line.

> ---
>  Makefile.target           |   2 +-
>  cpus.c                    |  14 ----
>  include/exec/cpu-all.h    |   2 -
>  include/exec/cpu-common.h |   2 +
>  include/sysemu/cpus.h     |   1 -
>  include/sysemu/sysemu.h   |   3 +
>  numa.c                    | 186 ++++++++++++++++++++++++++++++++++++++++++++++
>  vl.c                      | 139 +---------------------------------
>  8 files changed, 193 insertions(+), 156 deletions(-)
>  create mode 100644 numa.c
> 
> diff --git a/Makefile.target b/Makefile.target
> index 9986047..dd815bb 100644
> --- a/Makefile.target
> +++ b/Makefile.target
> @@ -112,7 +112,7 @@ endif #CONFIG_BSD_USER
>  #########################################################
>  # System emulator target
>  ifdef CONFIG_SOFTMMU
> -obj-y += arch_init.o cpus.o monitor.o gdbstub.o balloon.o ioport.o
> +obj-y += arch_init.o cpus.o monitor.o gdbstub.o balloon.o ioport.o numa.o
>  obj-y += qtest.o
>  obj-y += hw/
>  obj-$(CONFIG_FDT) += device_tree.o
> diff --git a/cpus.c b/cpus.c
> index 7bbe153..7f87adb 100644
> --- a/cpus.c
> +++ b/cpus.c
> @@ -1313,20 +1313,6 @@ static void tcg_exec_all(void)
>      exit_request = 0;
>  }
>  
> -void set_numa_modes(void)
> -{
> -    CPUState *cpu;
> -    int i;
> -
> -    CPU_FOREACH(cpu) {
> -        for (i = 0; i < nb_numa_nodes; i++) {
> -            if (test_bit(cpu->cpu_index, node_cpumask[i])) {
> -                cpu->numa_node = i;
> -            }
> -        }
> -    }
> -}
> -
>  void list_cpus(FILE *f, fprintf_function cpu_fprintf, const char *optarg)
>  {
>      /* XXX: implement xxx_cpu_list for targets that still miss it */
> diff --git a/include/exec/cpu-all.h b/include/exec/cpu-all.h
> index fb649a4..758a928 100644
> --- a/include/exec/cpu-all.h
> +++ b/include/exec/cpu-all.h
> @@ -414,8 +414,6 @@ CPUArchState *cpu_copy(CPUArchState *env);
>  
>  /* memory API */
>  
> -extern ram_addr_t ram_size;
> -
>  /* RAM is pre-allocated and passed into qemu_ram_alloc_from_ptr */
>  #define RAM_PREALLOC_MASK   (1 << 0)
>  
> diff --git a/include/exec/cpu-common.h b/include/exec/cpu-common.h
> index a21b65a..e8c7970 100644
> --- a/include/exec/cpu-common.h
> +++ b/include/exec/cpu-common.h
> @@ -45,6 +45,8 @@ typedef uintptr_t ram_addr_t;
>  #  define RAM_ADDR_FMT "%" PRIxPTR
>  #endif
>  
> +extern ram_addr_t ram_size;
> +
>  /* memory API */
>  
>  typedef void CPUWriteMemoryFunc(void *opaque, hwaddr addr, uint32_t value);
> diff --git a/include/sysemu/cpus.h b/include/sysemu/cpus.h
> index 6502488..4f79081 100644
> --- a/include/sysemu/cpus.h
> +++ b/include/sysemu/cpus.h
> @@ -23,7 +23,6 @@ extern int smp_threads;
>  #define smp_threads 1
>  #endif
>  
> -void set_numa_modes(void);
>  void list_cpus(FILE *f, fprintf_function cpu_fprintf, const char *optarg);
>  
>  #endif
> diff --git a/include/sysemu/sysemu.h b/include/sysemu/sysemu.h
> index ba5c7f8..565c8f6 100644
> --- a/include/sysemu/sysemu.h
> +++ b/include/sysemu/sysemu.h
> @@ -144,6 +144,9 @@ extern QEMUClockType rtc_clock;
>  extern int nb_numa_nodes;
>  extern uint64_t node_mem[MAX_NODES];
>  extern unsigned long *node_cpumask[MAX_NODES];
> +void numa_add(const char *optarg);
> +void set_numa_nodes(void);
> +void set_numa_modes(void);
>  
>  #define MAX_OPTION_ROMS 16
>  typedef struct QEMUOptionRom {

Down the line I think numa.h would be nicer,
not a must to fix.



> diff --git a/numa.c b/numa.c
> new file mode 100644
> index 0000000..395c14f
> --- /dev/null
> +++ b/numa.c
> @@ -0,0 +1,186 @@
> +/*
> + * QEMU System Emulator

Is this the best we can do here? This was reasonable for vl.c

Better  "NUMA parameter parsing"

> + *
> + * Copyright (c) 2013 Fujitsu Ltd.

It's 2014 isn't it?


> + * Author: Wanlong Gao <gaowanlong@cn.fujitsu.com>

These tags are probably best avoided for non original code:
we don't have the resources to keep them up to date as code changes.
Better dropped it for now.

> + *
> + * Permission is hereby granted, free of charge, to any person obtaining a copy
> + * of this software and associated documentation files (the "Software"), to deal
> + * in the Software without restriction, including without limitation the rights
> + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
> + * copies of the Software, and to permit persons to whom the Software is
> + * furnished to do so, subject to the following conditions:
> + *
> + * The above copyright notice and this permission notice shall be included in
> + * all copies or substantial portions of the Software.
> + *
> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
> + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
> + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
> + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
> + * THE SOFTWARE.
> + */
> +
> +#include "sysemu/sysemu.h"
> +#include "exec/cpu-common.h"
> +#include "qemu/bitmap.h"
> +#include "qom/cpu.h"
> +
> +static void numa_node_parse_cpus(int nodenr, const char *cpus)
> +{
> +    char *endptr;
> +    unsigned long long value, endvalue;
> +
> +    /* Empty CPU range strings will be considered valid, they will simply
> +     * not set any bit in the CPU bitmap.
> +     */
> +    if (!*cpus) {
> +        return;
> +    }
> +
> +    if (parse_uint(cpus, &value, &endptr, 10) < 0) {
> +        goto error;
> +    }
> +    if (*endptr == '-') {
> +        if (parse_uint_full(endptr + 1, &endvalue, 10) < 0) {
> +            goto error;
> +        }
> +    } else if (*endptr == '\0') {
> +        endvalue = value;
> +    } else {
> +        goto error;
> +    }
> +
> +    if (endvalue >= MAX_CPUMASK_BITS) {
> +        endvalue = MAX_CPUMASK_BITS - 1;
> +        fprintf(stderr,
> +            "qemu: NUMA: A max of %d VCPUs are supported\n",
> +             MAX_CPUMASK_BITS);
> +    }
> +
> +    if (endvalue < value) {
> +        goto error;
> +    }
> +
> +    bitmap_set(node_cpumask[nodenr], value, endvalue-value+1);
> +    return;
> +
> +error:
> +    fprintf(stderr, "qemu: Invalid NUMA CPU range: %s\n", cpus);
> +    exit(1);
> +}
> +
> +void numa_add(const char *optarg)
> +{
> +    char option[128];
> +    char *endptr;
> +    unsigned long long nodenr;
> +
> +    optarg = get_opt_name(option, 128, optarg, ',');
> +    if (*optarg == ',') {
> +        optarg++;
> +    }
> +    if (!strcmp(option, "node")) {
> +
> +        if (nb_numa_nodes >= MAX_NODES) {
> +            fprintf(stderr, "qemu: too many NUMA nodes\n");
> +            exit(1);
> +        }
> +
> +        if (get_param_value(option, 128, "nodeid", optarg) == 0) {
> +            nodenr = nb_numa_nodes;
> +        } else {
> +            if (parse_uint_full(option, &nodenr, 10) < 0) {
> +                fprintf(stderr, "qemu: Invalid NUMA nodeid: %s\n", option);
> +                exit(1);
> +            }
> +        }
> +
> +        if (nodenr >= MAX_NODES) {
> +            fprintf(stderr, "qemu: invalid NUMA nodeid: %llu\n", nodenr);
> +            exit(1);
> +        }
> +
> +        if (get_param_value(option, 128, "mem", optarg) == 0) {
> +            node_mem[nodenr] = 0;
> +        } else {
> +            int64_t sval;
> +            sval = strtosz(option, &endptr);
> +            if (sval < 0 || *endptr) {
> +                fprintf(stderr, "qemu: invalid numa mem size: %s\n", optarg);
> +                exit(1);
> +            }
> +            node_mem[nodenr] = sval;
> +        }
> +        if (get_param_value(option, 128, "cpus", optarg) != 0) {
> +            numa_node_parse_cpus(nodenr, option);
> +        }
> +        nb_numa_nodes++;
> +    } else {
> +        fprintf(stderr, "Invalid -numa option: %s\n", option);
> +        exit(1);
> +    }
> +}
> +
> +void set_numa_nodes(void)
> +{
> +    if (nb_numa_nodes > 0) {
> +        int i;
> +
> +        if (nb_numa_nodes > MAX_NODES) {
> +            nb_numa_nodes = MAX_NODES;
> +        }
> +
> +        /* If no memory size if given for any node, assume the default case
> +         * and distribute the available memory equally across all nodes
> +         */
> +        for (i = 0; i < nb_numa_nodes; i++) {
> +            if (node_mem[i] != 0) {
> +                break;
> +            }
> +        }
> +        if (i == nb_numa_nodes) {
> +            uint64_t usedmem = 0;
> +
> +            /* On Linux, the each node's border has to be 8MB aligned,
> +             * the final node gets the rest.
> +             */
> +            for (i = 0; i < nb_numa_nodes - 1; i++) {
> +                node_mem[i] = (ram_size / nb_numa_nodes) & ~((1 << 23UL) - 1);
> +                usedmem += node_mem[i];
> +            }
> +            node_mem[i] = ram_size - usedmem;
> +        }
> +
> +        for (i = 0; i < nb_numa_nodes; i++) {
> +            if (!bitmap_empty(node_cpumask[i], MAX_CPUMASK_BITS)) {
> +                break;
> +            }
> +        }
> +        /* assigning the VCPUs round-robin is easier to implement, guest OSes
> +         * must cope with this anyway, because there are BIOSes out there in
> +         * real machines which also use this scheme.
> +         */
> +        if (i == nb_numa_nodes) {
> +            for (i = 0; i < max_cpus; i++) {
> +                set_bit(i, node_cpumask[i % nb_numa_nodes]);
> +            }
> +        }
> +    }
> +}
> +
> +void set_numa_modes(void)
> +{
> +    CPUState *cpu;
> +    int i;
> +
> +    CPU_FOREACH(cpu) {
> +        for (i = 0; i < nb_numa_nodes; i++) {
> +            if (test_bit(cpu->cpu_index, node_cpumask[i])) {
> +                cpu->numa_node = i;
> +            }
> +        }
> +    }
> +}
> diff --git a/vl.c b/vl.c
> index 709d8cd..1dba4a5 100644
> --- a/vl.c
> +++ b/vl.c
> @@ -1265,102 +1265,6 @@ char *get_boot_devices_list(size_t *size, bool ignore_suffixes)
>      return list;
>  }
>  
> -static void numa_node_parse_cpus(int nodenr, const char *cpus)
> -{
> -    char *endptr;
> -    unsigned long long value, endvalue;
> -
> -    /* Empty CPU range strings will be considered valid, they will simply
> -     * not set any bit in the CPU bitmap.
> -     */
> -    if (!*cpus) {
> -        return;
> -    }
> -
> -    if (parse_uint(cpus, &value, &endptr, 10) < 0) {
> -        goto error;
> -    }
> -    if (*endptr == '-') {
> -        if (parse_uint_full(endptr + 1, &endvalue, 10) < 0) {
> -            goto error;
> -        }
> -    } else if (*endptr == '\0') {
> -        endvalue = value;
> -    } else {
> -        goto error;
> -    }
> -
> -    if (endvalue >= MAX_CPUMASK_BITS) {
> -        endvalue = MAX_CPUMASK_BITS - 1;
> -        fprintf(stderr,
> -            "qemu: NUMA: A max of %d VCPUs are supported\n",
> -             MAX_CPUMASK_BITS);
> -    }
> -
> -    if (endvalue < value) {
> -        goto error;
> -    }
> -
> -    bitmap_set(node_cpumask[nodenr], value, endvalue-value+1);
> -    return;
> -
> -error:
> -    fprintf(stderr, "qemu: Invalid NUMA CPU range: %s\n", cpus);
> -    exit(1);
> -}
> -
> -static void numa_add(const char *optarg)
> -{
> -    char option[128];
> -    char *endptr;
> -    unsigned long long nodenr;
> -
> -    optarg = get_opt_name(option, 128, optarg, ',');
> -    if (*optarg == ',') {
> -        optarg++;
> -    }
> -    if (!strcmp(option, "node")) {
> -
> -        if (nb_numa_nodes >= MAX_NODES) {
> -            fprintf(stderr, "qemu: too many NUMA nodes\n");
> -            exit(1);
> -        }
> -
> -        if (get_param_value(option, 128, "nodeid", optarg) == 0) {
> -            nodenr = nb_numa_nodes;
> -        } else {
> -            if (parse_uint_full(option, &nodenr, 10) < 0) {
> -                fprintf(stderr, "qemu: Invalid NUMA nodeid: %s\n", option);
> -                exit(1);
> -            }
> -        }
> -
> -        if (nodenr >= MAX_NODES) {
> -            fprintf(stderr, "qemu: invalid NUMA nodeid: %llu\n", nodenr);
> -            exit(1);
> -        }
> -
> -        if (get_param_value(option, 128, "mem", optarg) == 0) {
> -            node_mem[nodenr] = 0;
> -        } else {
> -            int64_t sval;
> -            sval = strtosz(option, &endptr);
> -            if (sval < 0 || *endptr) {
> -                fprintf(stderr, "qemu: invalid numa mem size: %s\n", optarg);
> -                exit(1);
> -            }
> -            node_mem[nodenr] = sval;
> -        }
> -        if (get_param_value(option, 128, "cpus", optarg) != 0) {
> -            numa_node_parse_cpus(nodenr, option);
> -        }
> -        nb_numa_nodes++;
> -    } else {
> -        fprintf(stderr, "Invalid -numa option: %s\n", option);
> -        exit(1);
> -    }
> -}
> -
>  static QemuOptsList qemu_smp_opts = {
>      .name = "smp-opts",
>      .implied_opt_name = "cpus",
> @@ -4336,48 +4240,7 @@ int main(int argc, char **argv, char **envp)
>      default_drive(default_floppy, snapshot, IF_FLOPPY, 0, FD_OPTS);
>      default_drive(default_sdcard, snapshot, IF_SD, 0, SD_OPTS);
>  
> -    if (nb_numa_nodes > 0) {
> -        int i;
> -
> -        if (nb_numa_nodes > MAX_NODES) {
> -            nb_numa_nodes = MAX_NODES;
> -        }
> -
> -        /* If no memory size if given for any node, assume the default case
> -         * and distribute the available memory equally across all nodes
> -         */
> -        for (i = 0; i < nb_numa_nodes; i++) {
> -            if (node_mem[i] != 0)
> -                break;
> -        }
> -        if (i == nb_numa_nodes) {
> -            uint64_t usedmem = 0;
> -
> -            /* On Linux, the each node's border has to be 8MB aligned,
> -             * the final node gets the rest.
> -             */
> -            for (i = 0; i < nb_numa_nodes - 1; i++) {
> -                node_mem[i] = (ram_size / nb_numa_nodes) & ~((1 << 23UL) - 1);
> -                usedmem += node_mem[i];
> -            }
> -            node_mem[i] = ram_size - usedmem;
> -        }
> -
> -        for (i = 0; i < nb_numa_nodes; i++) {
> -            if (!bitmap_empty(node_cpumask[i], MAX_CPUMASK_BITS)) {
> -                break;
> -            }
> -        }
> -        /* assigning the VCPUs round-robin is easier to implement, guest OSes
> -         * must cope with this anyway, because there are BIOSes out there in
> -         * real machines which also use this scheme.
> -         */
> -        if (i == nb_numa_nodes) {
> -            for (i = 0; i < max_cpus; i++) {
> -                set_bit(i, node_cpumask[i % nb_numa_nodes]);
> -            }
> -        }
> -    }
> +    set_numa_nodes();
>  
>      if (qemu_opts_foreach(qemu_find_opts("mon"), mon_init_func, NULL, 1) != 0) {
>          exit(1);
> -- 
> 1.8.5.2.229.g4448466
>