lib/cpuinfo: Increase the file descriptors limit to handle more CPUs

The pqos tool fails with the following errors on systems with 300 or more
CPU cores.
$pqos
NOTE:  Mixed use of MSR and kernel interfaces to manage
       CAT or CMT & MBM may lead to unexpected behavior.
ERROR: Could not open /sys/fs/resctrl directory
ERROR: Failed to stop resctrl events
ERROR: Failed to start all selected OS monitoring events
Monitoring start error on core(s) 339, status 1

By default, the file descriptor limit is set to 1024 for a session. pqos
monitor uses 3 descriptors for each CPU for perf monitoring. So, it runs
out of limit(1024) on systems with 300 or more CPUs.

Fix the issue by detecting the number of CPUs in the system and increasing
the descriptor limit using system call getrlimit and setrlimit respectively.
Increase the limit to 4 times the number of CPUs to take care of open files
limit.

Signed-off-by: Babu Moger <[email protected]>

lib/common.c

@@ -43,6 +43,7 @@
 #include <stdlib.h>
 #include <string.h>
 #include <sys/mman.h>
+#include <sys/resource.h>
 #include <unistd.h>
 
 FILE *
@@ -392,3 +393,24 @@ pqos_read(int fd, void *buf, size_t count)
 
         return count;
 }
+
+int
+pqos_set_no_files_limit(unsigned long max_core_count)
+{
+        struct rlimit files_limit;
+
+        if (getrlimit(RLIMIT_NOFILE, &files_limit))
+                return PQOS_RETVAL_ERROR;
+
+        /* Check if rlim_max can handle the numner of CPUs */
+        if (files_limit.rlim_max < max_core_count * 4)
+                return PQOS_RETVAL_ERROR;
+
+        if (files_limit.rlim_cur < max_core_count * 4) {
+                files_limit.rlim_cur = max_core_count * 4;
+                if (setrlimit(RLIMIT_NOFILE, &files_limit))
+                        return PQOS_RETVAL_ERROR;
+        }
+
+        return PQOS_RETVAL_OK;
+}

lib/common.h

@@ -200,6 +200,17 @@ PQOS_LOCAL void pqos_munmap(void *mem, const uint64_t size);
  */
 PQOS_LOCAL ssize_t pqos_read(int fd, void *buf, size_t count);
 
+/**
+ * @brief Increase the number of open files limit to handle more
+ * than 256 CPUs.
+ *
+ * @param [in] Max CPUs on the system
+ *
+ * @return PQOS_RETVAL_OK for success
+ * @retval PQOS_RETVAL_ERROR for failure
+ */
+PQOS_LOCAL int pqos_set_no_files_limit(unsigned long max_core_count);
+
 #ifdef __cplusplus
 }
 #endif

lib/cpuinfo.c

@@ -39,6 +39,7 @@
 
 #include "allocation.h"
 #include "cap.h"
+#include "common.h"
 #include "cpu_registers.h"
 #include "log.h"
 #include "machine.h"
@@ -57,6 +58,7 @@
 #endif
 #ifdef __linux__
 #include <sched.h> /* sched affinity */
+#include <sys/resource.h>
 #include <sys/syscall.h>
 #endif
 
@@ -452,6 +454,11 @@ cpuinfo_build_topo(struct apic_info *apic)
                 return NULL;
         }
 
+        if (pqos_set_no_files_limit(max_core_count)) {
+                LOG_ERROR("Open files limit not sufficient!\n");
+                return NULL;
+        }
+
         const size_t mem_sz =
             sizeof(*l_cpu) + (max_core_count * sizeof(struct pqos_coreinfo));
 

lib/os_cpuinfo.c

@@ -43,6 +43,7 @@
 #include <limits.h>
 #include <stdlib.h>
 #include <string.h>
+#include <sys/resource.h>
 
 #define SYSTEM_CPU  "/sys/devices/system/cpu"
 #define SYSTEM_NODE "/sys/devices/system/node"
@@ -267,6 +268,11 @@ os_cpuinfo_topology(void)
                 return NULL;
         }
 
+        if (pqos_set_no_files_limit(max_core_count)) {
+                LOG_ERROR("Open files limit not sufficient!\n");
+                return NULL;
+        }
+
         const size_t mem_sz =
             sizeof(*cpu) + (max_core_count * sizeof(struct pqos_coreinfo));
 

pqos/alloc.c

@@ -840,7 +840,7 @@ set_allocation_assoc(const struct pqos_devinfo *dev)
 static void
 fill_core_tab(char *str)
 {
-        unsigned max_cores_count = 128;
+        unsigned max_cores_count = sysconf(_SC_NPROCESSORS_CONF);
         uint64_t *cores = calloc(max_cores_count, sizeof(uint64_t));
         unsigned i = 0, n = 0, cos = 0;
         char *p = NULL;