[PATCH v4 14/15] xl: enable for specifying node-affinity in the config file

[Dario Faggioli <dario.faggioli@citrix.com>]

in a similar way to how it is possible to specify vcpu-affinity.

Signed-off-by: Dario Faggioli <dario.faggioli@citrix.com>
---
Changes from v3:
 * fix typos and language issues in docs and comments, as
   suggested during review;
 * common code to soft and hard affinity parsing factored
   together, as requested uring review.

Changes from v2:
 * use the new libxl API. Although the implementation changed
   only a little bit, I removed IanJ's Acked-by, although I am
   here saying that he did provided it, as requested.
---
 docs/man/xl.cfg.pod.5    |   25 +++++-
 tools/libxl/xl_cmdimpl.c |  192 ++++++++++++++++++++++++++++++----------------
 2 files changed, 146 insertions(+), 71 deletions(-)

diff --git a/docs/man/xl.cfg.pod.5 b/docs/man/xl.cfg.pod.5
index 5e17b5d..9566393 100644
--- a/docs/man/xl.cfg.pod.5
+++ b/docs/man/xl.cfg.pod.5
@@ -144,19 +144,38 @@ run on cpu #3 of the host.
 =back
 
 If this option is not specified, no vcpu to cpu pinning is established,
-and the vcpus of the guest can run on all the cpus of the host.
+and the vcpus of the guest can run on all the cpus of the host. If this
+option is specified, the intersection of the vcpu pinning mask, provided
+here, and the soft affinity mask, provided via B<cpus\_soft=> (if any),
+is utilized to compute the domain node-affinity, for driving memory
+allocations.
 
 If we are on a NUMA machine (i.e., if the host has more than one NUMA
 node) and this option is not specified, libxl automatically tries to
 place the guest on the least possible number of nodes. That, however,
 will not affect vcpu pinning, so the guest will still be able to run on
-all the cpus, it will just prefer the ones from the node it has been
-placed on. A heuristic approach is used for choosing the best node (or
+all the cpus. A heuristic approach is used for choosing the best node (or
 set of nodes), with the goals of maximizing performance for the guest
 and, at the same time, achieving efficient utilization of host cpus
 and memory. See F<docs/misc/xl-numa-placement.markdown> for more
 details.
 
+=item B<cpus_soft="CPU-LIST">
+
+Exactly as B<cpus=>, but specifies soft affinity, rather than pinning
+(also called hard affinity). When using the credit scheduler, this
+means the vcpus of the domain prefers to run on these pcpus.
+
+A C<CPU-LIST> is specified exactly as above, for B<cpus=>.
+
+If this option is not specified, the vcpus of the guest will not have
+any preference regarding on what cpu to run, and the scheduler will
+treat all the cpus where a vcpu can execute (if B<cpus=> is specified),
+or all the host cpus (if not), the same. If this option is specified,
+the intersection of the soft affinity mask, provided here, and the vcpu
+pinning, provided via B<cpus=> (if any), is utilized to compute the
+domain node-affinity, for driving memory allocations.
+
 =back
 
 =head3 CPU Scheduling
diff --git a/tools/libxl/xl_cmdimpl.c b/tools/libxl/xl_cmdimpl.c
index d6fda26..abd3e42 100644
--- a/tools/libxl/xl_cmdimpl.c
+++ b/tools/libxl/xl_cmdimpl.c
@@ -76,8 +76,9 @@ xlchild children[child_max];
 static const char *common_domname;
 static int fd_lock = -1;
 
-/* Stash for specific vcpu to pcpu mappping */
+/* Stash for specific vcpu to pcpu hard and soft mappping */
 static int *vcpu_to_pcpu;
+static int *vcpu_to_pcpu_soft;
 
 static const char savefileheader_magic[32]=
     "Xen saved domain, xl format\n \0 \r";
@@ -687,6 +688,64 @@ static int vcpupin_parse(char *cpu, libxl_bitmap *cpumap)
     return rc;
 }
 
+static int *parse_config_cpumap_list(XLU_ConfigList *cpus,
+                                     libxl_bitmap *cpumap,
+                                     int max_vcpus)
+{
+    int i, n_cpus = 0;
+    int *to_pcpu;
+    const char *buf;
+
+    if (libxl_cpu_bitmap_alloc(ctx, cpumap, 0)) {
+        fprintf(stderr, "Unable to allocate cpumap\n");
+        exit(1);
+    }
+
+    /* Prepare the array for single vcpu to pcpu mappings */
+    to_pcpu = xmalloc(sizeof(int) * max_vcpus);
+    memset(to_pcpu, -1, sizeof(int) * max_vcpus);
+
+    /*
+     * Idea here is to let libxl think all the domain's vcpus
+     * have cpu affinity with all the pcpus on the list. Doing
+     * that ensures memory is allocated on the proper NUMA nodes.
+     * It is then us, here in xl, that matches each single vcpu
+     * to its pcpu (and that's why we need to stash such info in
+     * the to_pcpu array now) after the domain has been created.
+     * This way, we avoid having to pass to libxl some big array
+     * hosting the single mappings.
+     */
+    libxl_bitmap_set_none(cpumap);
+    while ((buf = xlu_cfg_get_listitem(cpus, n_cpus)) != NULL) {
+        i = atoi(buf);
+        if (!libxl_bitmap_cpu_valid(cpumap, i)) {
+            fprintf(stderr, "cpu %d illegal\n", i);
+            exit(1);
+        }
+        libxl_bitmap_set(cpumap, i);
+        if (n_cpus < max_vcpus)
+            to_pcpu[n_cpus] = i;
+        n_cpus++;
+    }
+
+    return to_pcpu;
+}
+
+static void parse_config_cpumap_string(const char *buf, libxl_bitmap *cpumap)
+{
+        char *buf2 = strdup(buf);
+
+        if (libxl_cpu_bitmap_alloc(ctx, cpumap, 0)) {
+            fprintf(stderr, "Unable to allocate cpumap\n");
+            exit(1);
+        }
+
+        libxl_bitmap_set_none(cpumap);
+        if (vcpupin_parse(buf2, cpumap))
+            exit(1);
+        free(buf2);
+}
+
 static void parse_config_data(const char *config_source,
                               const char *config_data,
                               int config_len,
@@ -697,7 +756,8 @@ static void parse_config_data(const char *config_source,
     const char *buf;
     long l;
     XLU_Config *config;
-    XLU_ConfigList *cpus, *vbds, *nics, *pcis, *cvfbs, *cpuids, *vtpms;
+    XLU_ConfigList *cpus, *cpus_soft, *vbds, *nics, *pcis;
+    XLU_ConfigList *cvfbs, *cpuids, *vtpms;
     XLU_ConfigList *ioports, *irqs, *iomem;
     int num_ioports, num_irqs, num_iomem;
     int pci_power_mgmt = 0;
@@ -820,57 +880,24 @@ static void parse_config_data(const char *config_source,
         b_info->max_vcpus = l;
 
     if (!xlu_cfg_get_list (config, "cpus", &cpus, 0, 1)) {
-        int n_cpus = 0;
-
-        if (libxl_cpu_bitmap_alloc(ctx, &b_info->cpumap, 0)) {
-            fprintf(stderr, "Unable to allocate cpumap\n");
-            exit(1);
-        }
-
-        /* Prepare the array for single vcpu to pcpu mappings */
-        vcpu_to_pcpu = xmalloc(sizeof(int) * b_info->max_vcpus);
-        memset(vcpu_to_pcpu, -1, sizeof(int) * b_info->max_vcpus);
-
-        /*
-         * Idea here is to let libxl think all the domain's vcpus
-         * have cpu affinity with all the pcpus on the list.
-         * It is then us, here in xl, that matches each single vcpu
-         * to its pcpu (and that's why we need to stash such info in
-         * the vcpu_to_pcpu array now) after the domain has been created.
-         * Doing it like this saves the burden of passing to libxl
-         * some big array hosting the single mappings. Also, using
-         * the cpumap derived from the list ensures memory is being
-         * allocated on the proper nodes anyway.
-         */
-        libxl_bitmap_set_none(&b_info->cpumap);
-        while ((buf = xlu_cfg_get_listitem(cpus, n_cpus)) != NULL) {
-            i = atoi(buf);
-            if (!libxl_bitmap_cpu_valid(&b_info->cpumap, i)) {
-                fprintf(stderr, "cpu %d illegal\n", i);
-                exit(1);
-            }
-            libxl_bitmap_set(&b_info->cpumap, i);
-            if (n_cpus < b_info->max_vcpus)
-                vcpu_to_pcpu[n_cpus] = i;
-            n_cpus++;
-        }
-
+        vcpu_to_pcpu = parse_config_cpumap_list(cpus, &b_info->cpumap,
+                                                b_info->max_vcpus);
         /* We have a cpumap, disable automatic placement */
         libxl_defbool_set(&b_info->numa_placement, false);
     }
     else if (!xlu_cfg_get_string (config, "cpus", &buf, 0)) {
-        char *buf2 = strdup(buf);
-
-        if (libxl_cpu_bitmap_alloc(ctx, &b_info->cpumap, 0)) {
-            fprintf(stderr, "Unable to allocate cpumap\n");
-            exit(1);
-        }
-
-        libxl_bitmap_set_none(&b_info->cpumap);
-        if (vcpupin_parse(buf2, &b_info->cpumap))
-            exit(1);
-        free(buf2);
+        parse_config_cpumap_string(buf, &b_info->cpumap);
+        libxl_defbool_set(&b_info->numa_placement, false);
+    }
 
+    if (!xlu_cfg_get_list (config, "cpus_soft", &cpus_soft, 0, 1)) {
+        vcpu_to_pcpu_soft =
+            parse_config_cpumap_list(cpus_soft, &b_info->cpumap_soft,
+                                     b_info->max_vcpus);
+        libxl_defbool_set(&b_info->numa_placement, false);
+    }
+    else if (!xlu_cfg_get_string (config, "cpus_soft", &buf, 0)) {
+        parse_config_cpumap_string(buf, &b_info->cpumap_soft);
         libxl_defbool_set(&b_info->numa_placement, false);
     }
 
@@ -1991,6 +2018,40 @@ static void evdisable_disk_ejects(libxl_evgen_disk_eject **diskws,
     }
 }
 
+static inline int set_vcpu_to_pcpu_affinity(uint32_t domid, int *to_pcpu,
+                                            int max_vcpus, int soft)
+{
+    libxl_bitmap vcpu_cpumap;
+    int i, ret = 0;
+
+    ret = libxl_cpu_bitmap_alloc(ctx, &vcpu_cpumap, 0);
+    if (ret)
+        return -1;
+
+    for (i = 0; i < max_vcpus; i++) {
+        if (to_pcpu[i] != -1) {
+            libxl_bitmap_set_none(&vcpu_cpumap);
+            libxl_bitmap_set(&vcpu_cpumap, to_pcpu[i]);
+        } else {
+            libxl_bitmap_set_any(&vcpu_cpumap);
+        }
+        if (!soft &&
+            libxl_set_vcpuaffinity(ctx, domid, i, &vcpu_cpumap, NULL)) {
+            fprintf(stderr, "setting hard affinity failed on vcpu `%d'.\n", i);
+            ret = -1;
+            break;
+        } else if (soft &&
+                   libxl_set_vcpuaffinity(ctx, domid, i, NULL, &vcpu_cpumap)) {
+            fprintf(stderr, "setting soft affinity failed on vcpu `%d'.\n", i);
+            ret = -1;
+            break;
+        }
+    }
+    libxl_bitmap_dispose(&vcpu_cpumap);
+
+    return ret;
+}
+
 static uint32_t create_domain(struct domain_create *dom_info)
 {
     uint32_t domid = INVALID_DOMID;
@@ -2207,31 +2268,26 @@ start:
     if ( ret )
         goto error_out;
 
-    /* If single vcpu to pcpu mapping was requested, honour it */
+    /* If single vcpu pinning or soft affinity was requested, honour it */
     if (vcpu_to_pcpu) {
-        libxl_bitmap vcpu_cpumap;
+        ret = set_vcpu_to_pcpu_affinity(domid, vcpu_to_pcpu,
+                                        d_config.b_info.max_vcpus, 0);
+        free(vcpu_to_pcpu);
 
-        ret = libxl_cpu_bitmap_alloc(ctx, &vcpu_cpumap, 0);
         if (ret)
             goto error_out;
-        for (i = 0; i < d_config.b_info.max_vcpus; i++) {
 
-            if (vcpu_to_pcpu[i] != -1) {
-                libxl_bitmap_set_none(&vcpu_cpumap);
-                libxl_bitmap_set(&vcpu_cpumap, vcpu_to_pcpu[i]);
-            } else {
-                libxl_bitmap_set_any(&vcpu_cpumap);
-            }
-            if (libxl_set_vcpuaffinity(ctx, domid, i, &vcpu_cpumap, NULL)) {
-                fprintf(stderr, "setting affinity failed on vcpu `%d'.\n", i);
-                libxl_bitmap_dispose(&vcpu_cpumap);
-                free(vcpu_to_pcpu);
-                ret = ERROR_FAIL;
-                goto error_out;
-            }
-        }
-        libxl_bitmap_dispose(&vcpu_cpumap);
-        free(vcpu_to_pcpu); vcpu_to_pcpu = NULL;
+        vcpu_to_pcpu = NULL;
+    }
+    if (vcpu_to_pcpu_soft) {
+        ret = set_vcpu_to_pcpu_affinity(domid, vcpu_to_pcpu_soft,
+                                        d_config.b_info.max_vcpus, 1);
+        free(vcpu_to_pcpu_soft);
+
+        if (ret)
+            goto error_out;
+
+        vcpu_to_pcpu_soft = NULL;
     }
 
     ret = libxl_userdata_store(ctx, domid, "xl",