feat: display content of process memory pages

This commit extends `memparse` sub-command with the ability to display
the content of process memory pages content in a hexdump-like format when
the `--pid` flag is provided. Additionally, the output can be written to a
file using the `--output` flag.

Signed-off-by: Kouame Behouba Manasse <behouba@gmail.com>

checkpointctl.go

@@ -12,17 +12,18 @@ import (
 )
 
 var (
-	name      string
-	version   string
-	format    string
-	stats     bool
-	mounts    bool
-	pID       uint32
-	psTree    bool
-	psTreeCmd bool
-	psTreeEnv bool
-	files     bool
-	showAll   bool
+	name           string
+	version        string
+	format         string
+	stats          bool
+	mounts         bool
+	outputFilePath string
+	pID            uint32
+	psTree         bool
+	psTreeCmd      bool
+	psTreeEnv      bool
+	files          bool
+	showAll        bool
 )
 
 func main() {
@@ -275,6 +276,23 @@ func setupMemParse() *cobra.Command {
 		Args:  cobra.MinimumNArgs(1),
 	}
 
+	flags := cmd.Flags()
+
+	flags.Uint32VarP(
+		&pID,
+		"pid",
+		"p",
+		0,
+		"Specify the PID of a process to analyze",
+	)
+	flags.StringVarP(
+		&outputFilePath,
+		"output",
+		"o",
+		"",
+		"Specify the output file to be written to",
+	)
+
 	return cmd
 }
 
@@ -283,9 +301,19 @@ func memparse(cmd *cobra.Command, args []string) error {
 		metadata.SpecDumpFile, metadata.ConfigDumpFile,
 		filepath.Join(metadata.CheckpointDirectory, "pstree.img"),
 		filepath.Join(metadata.CheckpointDirectory, "core-"),
-		filepath.Join(metadata.CheckpointDirectory, "pagemap-"),
-		filepath.Join(metadata.CheckpointDirectory, "pages-"),
-		filepath.Join(metadata.CheckpointDirectory, "mm-"),
+	}
+
+	if pID == 0 {
+		requiredFiles = append(
+			requiredFiles,
+			filepath.Join(metadata.CheckpointDirectory, "pagemap-"),
+		)
+	} else {
+		requiredFiles = append(
+			requiredFiles,
+			filepath.Join(metadata.CheckpointDirectory, fmt.Sprintf("pagemap-%d.img", pID)),
+			filepath.Join(metadata.CheckpointDirectory, fmt.Sprintf("mm-%d.img", pID)),
+		)
 	}
 
 	tasks, err := createTasks(args, requiredFiles)
@@ -294,5 +322,9 @@ func memparse(cmd *cobra.Command, args []string) error {
 	}
 	defer cleanupTasks(tasks)
 
+	if pID != 0 {
+		return printProcessMemoryPages(tasks[0])
+	}
+
 	return showProcessMemorySizeTables(tasks)
 }

container.go

@@ -348,3 +348,17 @@ func getPsEnvVars(checkpointOutputDir string, pid uint32) (envVars []string, err
 
 	return
 }
+
+// FindPs performs a short-circuiting depth-first search to find
+// a process with a given PID in a process tree.
+func findPs(ps *crit.PsTree, pid uint32) *crit.PsTree {
+	if ps.PID == pid {
+		return ps
+	}
+	for _, child := range ps.Children {
+		if process := findPs(child, pid); process != nil {
+			return process
+		}
+	}
+	return nil
+}

go.mod

@@ -3,7 +3,7 @@ module github.com/checkpoint-restore/checkpointctl
 go 1.18
 
 require (
-	github.com/checkpoint-restore/go-criu/v6 v6.3.1-0.20230705155226-f8745d1ee56a
+	github.com/checkpoint-restore/go-criu/v6 v6.3.1-0.20230803210323-a74ef6248628
 	github.com/containers/storage v1.48.0
 	github.com/olekukonko/tablewriter v0.0.5
 	github.com/opencontainers/runtime-spec v1.1.0
@@ -25,5 +25,5 @@ require (
 	github.com/ulikunitz/xz v0.5.11 // indirect
 	golang.org/x/sys v0.9.0 // indirect
 	golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1 // indirect
-	google.golang.org/protobuf v1.30.0 // indirect
+	google.golang.org/protobuf v1.31.0 // indirect
 )

go.sum

@@ -1,5 +1,5 @@
-github.com/checkpoint-restore/go-criu/v6 v6.3.1-0.20230705155226-f8745d1ee56a h1:I9EbXzk/kRIcXDGk9VAdYjGTicrok3Bu3WZ+SOJ3K2E=
-github.com/checkpoint-restore/go-criu/v6 v6.3.1-0.20230705155226-f8745d1ee56a/go.mod h1:uuIaCJCnWcRzK88xoScsiKHpwBlF8uChRbyhi/BZiZg=
+github.com/checkpoint-restore/go-criu/v6 v6.3.1-0.20230803210323-a74ef6248628 h1:N0LDmW0R1U8ewZFru8ocr6vdwr0XSPpKrttShxJcGs4=
+github.com/checkpoint-restore/go-criu/v6 v6.3.1-0.20230803210323-a74ef6248628/go.mod h1:3eG5SDBmX8R3iY7kAnrj11UrR7DgmHFiJy+ygS9mlNo=
 github.com/containers/storage v1.48.0 h1:wiPs8J2xiFoOEAhxHDRtP6A90Jzj57VqzLRXOqeizns=
 github.com/containers/storage v1.48.0/go.mod h1:pRp3lkRo2qodb/ltpnudoXggrviRmaCmU5a5GhTBae0=
 github.com/cpuguy83/go-md2man/v2 v2.0.2/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o=
@@ -53,8 +53,8 @@ golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8T
 golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1 h1:go1bK/D/BFZV2I8cIQd1NKEZ+0owSTG1fDTci4IqFcE=
 golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
 google.golang.org/protobuf v1.26.0-rc.1/go.mod h1:jlhhOSvTdKEhbULTjvd4ARK9grFBp09yW+WbY/TyQbw=
-google.golang.org/protobuf v1.30.0 h1:kPPoIgf3TsEvrm0PFe15JQ+570QVxYzEvvHqChK+cng=
-google.golang.org/protobuf v1.30.0/go.mod h1:HV8QOd/L58Z+nl8r43ehVNZIU/HEI6OcFqwMG9pJV4I=
+google.golang.org/protobuf v1.31.0 h1:g0LDEJHgrBl9N9r17Ru3sqWhkIx2NB67okBHPwC7hs8=
+google.golang.org/protobuf v1.31.0/go.mod h1:HV8QOd/L58Z+nl8r43ehVNZIU/HEI6OcFqwMG9pJV4I=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
 gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
 gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=

memparse.go

@@ -5,7 +5,9 @@
 package main
 
 import (
+	"bytes"
 	"fmt"
+	"io"
 	"os"
 	"path/filepath"
 
@@ -14,6 +16,10 @@ import (
 	"github.com/olekukonko/tablewriter"
 )
 
+// chunkSize represents the default size of memory chunk (in bytes)
+// to read for each output line when printing memory pages content in hexdump-like format.
+const chunkSize = 16
+
 // Display processes memory sizes within the given container checkpoints.
 func showProcessMemorySizeTables(tasks []task) error {
 	// Initialize the table
@@ -81,3 +87,116 @@ func showProcessMemorySizeTables(tasks []task) error {
 
 	return nil
 }
+
+func printProcessMemoryPages(task task) error {
+	c := crit.New(nil, nil, filepath.Join(task.outputDir, metadata.CheckpointDirectory), false, false)
+	psTree, err := c.ExplorePs()
+	if err != nil {
+		return fmt.Errorf("failed to get process tree: %w", err)
+	}
+
+	// Check if PID exist within the checkpoint
+	if pID != 0 {
+		ps := findPs(psTree, pID)
+		if ps == nil {
+			return fmt.Errorf("no process with PID %d (use `inspect --ps-tree` to view all PIDs)", pID)
+		}
+	}
+
+	memReader, err := crit.NewMemoryReader(
+		filepath.Join(task.outputDir, metadata.CheckpointDirectory),
+		pID, pageSize,
+	)
+	if err != nil {
+		return err
+	}
+
+	// Unpack pages-[pagesID].img file for the given PID
+	if err := untarFiles(
+		task.checkpointFilePath, task.outputDir,
+		[]string{filepath.Join(metadata.CheckpointDirectory, fmt.Sprintf("pages-%d.img", memReader.GetPagesID()))},
+	); err != nil {
+		return err
+	}
+
+	// Write the output to stdout by default
+	var output io.Writer = os.Stdout
+	var compact bool
+
+	if outputFilePath != "" {
+		// Write to ouput to file if --output is specified
+		f, err := os.Create(outputFilePath)
+		if err != nil {
+			return err
+		}
+		defer f.Close()
+		output = f
+		fmt.Printf("\nWriting memory pages content for process ID %d from checkpoint: %s to file: %s...\n",
+			pID, task.checkpointFilePath, outputFilePath,
+		)
+	} else {
+		compact = true // Use a compact format when writing the output to stdout
+		fmt.Printf("\nDisplaying memory pages content for process ID %d from checkpoint: %s\n\n", pID, task.checkpointFilePath)
+	}
+
+	fmt.Fprintln(output, "Address           Hexadecimal                                       ASCII            ")
+	fmt.Fprintln(output, "-------------------------------------------------------------------------------------")
+
+	pagemapEntries := memReader.GetPagemapEntries()
+	for _, entry := range pagemapEntries {
+		start := entry.GetVaddr()
+		end := start + (uint64(pageSize) * uint64(entry.GetNrPages()))
+		buf, err := memReader.GetMemPages(start, end)
+		if err != nil {
+			return err
+		}
+
+		hexdump(output, buf, start, compact)
+	}
+	return nil
+}
+
+// hexdump generates a hexdump of the buffer 'buf' starting at the virtual address 'start'
+// and writes the output to 'out'. If compact is true, consecutive duplicate rows will be represented
+// with an asterisk (*).
+func hexdump(out io.Writer, buf *bytes.Buffer, vaddr uint64, compact bool) {
+	var prevAscii string
+	var isDuplicate bool
+	for buf.Len() > 0 {
+		row := buf.Next(chunkSize)
+		hex, ascii := generateHexAndAscii(row)
+
+		if compact {
+			if prevAscii == ascii {
+				if !isDuplicate {
+					fmt.Fprint(out, "*\n")
+				}
+				isDuplicate = true
+			} else {
+				fmt.Fprintf(out, "%016x  %s |%s|\n", vaddr, hex, ascii)
+				isDuplicate = false
+			}
+		} else {
+			fmt.Fprintf(out, "%016x  %s |%s|\n", vaddr, hex, ascii)
+		}
+
+		vaddr += chunkSize
+		prevAscii = ascii
+	}
+}
+
+// generateHexAndAscii takes a byte slice and generates its hexadecimal and ASCII representations.
+func generateHexAndAscii(data []byte) (string, string) {
+	var hex, ascii string
+	for i := 0; i < len(data); i++ {
+		if data[i] < 32 || data[i] >= 127 {
+			ascii += "."
+			hex += fmt.Sprintf("%02x ", data[i])
+		} else {
+			ascii += string(data[i])
+			hex += fmt.Sprintf("%02x ", data[i])
+		}
+	}
+
+	return hex, ascii
+}

tree.go

@@ -121,21 +121,7 @@ func addDumpStatsToTree(tree treeprint.Tree, dumpStats *stats_pb.DumpStatsEntry)
 func addPsTreeToTree(tree treeprint.Tree, psTree *crit.PsTree, checkpointOutputDir string) error {
 	psRoot := psTree
 	if pID != 0 {
-		// dfs performs a short-circuiting depth-first search.
-		var dfs func(*crit.PsTree) *crit.PsTree
-		dfs = func(root *crit.PsTree) *crit.PsTree {
-			if root.PID == pID {
-				return root
-			}
-			for _, child := range root.Children {
-				if ps := dfs(child); ps != nil {
-					return ps
-				}
-			}
-			return nil
-		}
-
-		ps := dfs(psTree)
+		ps := findPs(psTree, pID)
 		if ps == nil {
 			return fmt.Errorf("no process with PID %d (use `inspect --ps-tree` to view all PIDs)", pID)
 		}