Translation Lookaside Buffer Demonstration

plot_tlb.R

required_packages <- c("ggplot2", "scales")

for (pkg in required_packages) {
  if (!requireNamespace(pkg, quietly = TRUE)) {
    cat(sprintf("Installing missing package: %s\n", pkg))
    install.packages(pkg)
  }
  library(pkg, character.only = TRUE)
}

n <- 300

run_test <- function(mode) {
  times <- numeric(n)
  for (i in 1:n) {
    output <- system2("./tlb_test", args = mode, stdout = TRUE, stderr = TRUE)
    # Extract the elapsed time using regex
    time_line <- grep("Elapsed time", output, value = TRUE)
    time_sec <- as.numeric(sub(".*Elapsed time: ([0-9.]+) seconds.*", "\\1", time_line))
    times[i] <- time_sec
  }
  return(data.frame(time = times, mode = mode))
}

set.seed(42)  # For reproducibility
cat("Running hit test...\n")
hit_data <- run_test("hit")

cat("Running miss test...\n")
miss_data <- run_test("miss")

cat("Plotting data...\n")
all_data <- rbind(hit_data, miss_data)

ggplot_obj <- ggplot(all_data, aes(x = time, fill = mode)) +
  geom_density(alpha = 0.6) +
  scale_y_continuous(trans = pseudo_log_trans(base = 10)) +
  labs(title = "TLB Timing Distribution (Hit vs Miss)",
       x = "Elapsed Time (seconds)",
       y = "Log-like Density")

ggsave("tlb_plot.png", plot = ggplot_obj, width = 8, height = 5)
cat("Plot saved to tlb_plot.png\n")

ggplot_obj <- ggplot(all_data, aes(x = mode, y = time, fill = mode)) +
  geom_boxplot(outlier.shape = 21, alpha = 0.7, width = 0.5) +
  labs(title = "TLB Access Time Comparison",
       x = "Mode",
       y = "Elapsed Time (seconds)")

ggsave("tlb_boxplot.png", plot = ggplot_obj, width = 8, height = 5)
cat("Plot saved to tlb_boxplot.png\n")

run_tlb_test.sh

#!/bin/bash

set -e  # Exit on error

echo "🔨 Compiling tlb_test.c..."
clang -O2 -o tlb_test tlb.c

echo "📈 Running R benchmark and plot..."
Rscript plot_tlb.R

echo "🖼️ Opening plot..."
open tlb_boxplot.png

tlb.c

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <sys/mman.h>
#include <time.h>

#define PAGE_SIZE 4096
#define NUM_PAGES_HIT 4
#define NUM_PAGES_MISS 8192
#define TOTAL_ACCESSES 8192*2

double get_time_sec() {
    struct timespec ts;
    clock_gettime(CLOCK_MONOTONIC, &ts);
    return ts.tv_sec + ts.tv_nsec / 1e9;
}

int main(int argc, char *argv[]) {
    if (argc != 2) {
        fprintf(stderr, "Usage: %s [hit|miss]\n", argv[0]);
        return 1;
    }

    int num_pages = 0;
    if (strcmp(argv[1], "hit") == 0) {
        num_pages = NUM_PAGES_HIT;
    } else if (strcmp(argv[1], "miss") == 0) {
        num_pages = NUM_PAGES_MISS;
    } else {
        fprintf(stderr, "Invalid mode. Use 'hit' or 'miss'.\n");
        return 1;
    }

    size_t region_size = num_pages * PAGE_SIZE;
    char *region = mmap(NULL, region_size, PROT_READ | PROT_WRITE,
                        MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
    if (region == MAP_FAILED) {
        perror("mmap");
        return 1;
    }

    // Touch each page to ensure it's mapped into memory
    for (int i = 0; i < num_pages; i++) {
        region[i * PAGE_SIZE] = 1;
    }

    volatile char sink;
    double start = get_time_sec();

    if (num_pages == NUM_PAGES_HIT) {
        // Access a few pages repeatedly
        for (int i = 0; i < TOTAL_ACCESSES; i++) {
            sink = region[(i % num_pages) * PAGE_SIZE];
        }
    } else {
        // Access many unique pages
        for (int i = 0; i < TOTAL_ACCESSES; i++) {
            sink = region[(i % num_pages) * PAGE_SIZE];

        }
    }

    double end = get_time_sec();
    printf("Mode: %s | Elapsed time: %.6f seconds\n", argv[1], end - start);

    munmap(region, region_size);
    return 0;
}