without bpf

.github/workflows/cmake.yml

@@ -2,9 +2,9 @@ name: CMake
 
 on:
   push:
-    branches: [ "main" ]
+    branches: [ "main", "asplos-dev" ]
   pull_request:
-    branches: [ "main" ]
+    branches: [ "main", "asplos-dev"  ]
 
 env:
   # Customize the CMake build type here (Release, Debug, RelWithDebInfo, etc.)

artifact/gen_workloads.sh

@@ -0,0 +1 @@
+#!/bin/bash

artifact/mlc-sapphirerapids.txt

@@ -0,0 +1,55 @@
+Intel(R) Memory Latency Checker - v3.10
+*** Unable to modify prefetchers (try executing 'modprobe msr')
+*** So, enabling random access for latency measurements
+Measuring idle latencies for random access (in ns)...
+                Numa node   Numa node
+Numa node             0           1
+       0          106.3       437.5
+
+Measuring Peak Injection Memory Bandwidths for the system
+Bandwidths are in MB/sec (1 MB/sec = 1,000,000 Bytes/sec)
+Using all the threads from each core if Hyper-threading is enabled
+Using traffic with the following read-write ratios
+ALL Reads        :      28611.9
+3:1 Reads-Writes :      25057.1
+2:1 Reads-Writes :      25078.0
+1:1 Reads-Writes :      23965.9
+Stream-triad like:      24943.3
+
+Measuring Memory Bandwidths between nodes within system 
+Bandwidths are in MB/sec (1 MB/sec = 1,000,000 Bytes/sec)
+Using all the threads from each core if Hyper-threading is enabled
+Using Read-only traffic type
+                Numa node   Numa node
+Numa node             0            1
+       0        28612.0      19216.8
+
+Measuring Loaded Latencies for the system
+Using all the threads from each core if Hyper-threading is enabled
+Using Read-only traffic type
+Inject  Latency Bandwidth
+Delay   (ns)    MB/sec
+==========================
+00000   370.12   28393.2
+00002   369.29   28435.4
+00008   378.41   28490.5
+00015   354.32   28414.2
+00050   313.07   28323.0
+00100   238.51   28010.5
+00200   125.13   14566.0
+00300   119.69   10232.0
+00400   116.76   7905.3
+00500   115.33   6500.4
+00700   113.89   4858.3
+01000   113.03   3594.6
+01300   112.57   2906.9
+01700   112.09   2363.9
+02500   111.51   1798.9
+03500   111.21   1520.8
+05000   110.77   1193.2
+09000   110.38   922.3
+20000   110.14   735.6
+
+Measuring cache-to-cache transfer latency (in ns)...
+Local Socket L2->L2 HIT  latency        67.3
+Local Socket L2->L2 HITM latency        67.5

src/incore.cpp

@@ -103,13 +103,12 @@ int Incore::read_cpu_elems(struct CPUElem *elem) {
 }
 Incore::Incore(const pid_t pid, const int cpu, struct PerfConfig *perf_config) : perf_config(perf_config) {
     /* reset all pmc values */
-    // this->init_all_dram_rds(pid, cpu);
+    this->init_all_dram_rds(pid, cpu);
     this->init_cpu_mem_read(pid, cpu);
     this->init_cpu_l2stall(pid, cpu);
     this->init_cpu_llcl_hits(pid, cpu);
     this->init_cpu_llcl_miss(pid, cpu);
-    this->init_cpu_ebpf(pid, cpu);
-    // this->init_cpu_mem_write(pid, cpu);
+    this->init_cpu_mem_write(pid, cpu);
 }
 bool get_cpu_info(struct CPUInfo *cpu_info) {
     char buffer[1024];
@@ -127,7 +126,7 @@ bool get_cpu_info(struct CPUInfo *cpu_info) {
     buf.ibuf[1] = cpuinfo.array[3];
     buf.ibuf[2] = cpuinfo.array[2];
 
-    if (strncmp(buf.cbuf, "GenuineIntel", 4 * 3) != 0) {
+    if (strncmp(buf.cbuf, "GenuineIntel", 12) != 0) {
         LOG(ERROR) << fmt::format("We only Support Intel CPU\n");
         return false;
     }