BinaryCacheStore: Add index-debug-info option

This integrates the functionality of the index-debuginfo program in
nixos-channel-scripts to maintain an index of DWARF debuginfo files in
a format usable by dwarffs. Thus the debug info index is updated by
Hydra rather than by the channel mirroring script.

Example usage:

  $ nix copy --to 'file:///tmp/binary-cache?index-debug-info=true' /nix/store/vr9mhcch3fljzzkjld3kvkggvpq38cva-nix-2.2.2-debug

  $ cat /tmp/binary-cache/debuginfo/036b210b03bad75ab2d8fc80b7a146f98e7f1ecf.debug
  {"archive":"../nar/0313h2kdhk4v73xna9ysiksp2v8xrsk5xsw79mmwr3rg7byb4ka8.nar.xz","member":"lib/debug/.build-id/03/6b210b03bad75ab2d8fc80b7a146f98e7f1ecf.debug"}

Fixes #3083.

Author: edolstra

src/libstore/binary-cache-store.cc

@@ -10,10 +10,13 @@
 #include "nar-info-disk-cache.hh"
 #include "nar-accessor.hh"
 #include "json.hh"
+#include "thread-pool.hh"
 
 #include <chrono>
-
 #include <future>
+#include <regex>
+
+#include <nlohmann/json.hpp>
 
 namespace nix {
 
@@ -139,6 +142,11 @@ void BinaryCacheStore::addToStore(const ValidPathInfo & info, const ref<std::str
 
     auto accessor_ = std::dynamic_pointer_cast<RemoteFSAccessor>(accessor);
 
+    auto narAccessor = makeNarAccessor(nar);
+
+    if (accessor_)
+        accessor_->addToCache(info.path, *nar, narAccessor);
+
     /* Optionally write a JSON file containing a listing of the
        contents of the NAR. */
     if (writeNARListing) {
@@ -148,11 +156,6 @@ void BinaryCacheStore::addToStore(const ValidPathInfo & info, const ref<std::str
             JSONObject jsonRoot(jsonOut);
             jsonRoot.attr("version", 1);
 
-            auto narAccessor = makeNarAccessor(nar);
-
-            if (accessor_)
-                accessor_->addToCache(info.path, *nar, narAccessor);
-
             {
                 auto res = jsonRoot.placeholder("root");
                 listNar(res, narAccessor, "", true);
@@ -162,11 +165,6 @@ void BinaryCacheStore::addToStore(const ValidPathInfo & info, const ref<std::str
         upsertFile(storePathToHash(info.path) + ".ls", jsonOut.str(), "application/json");
     }
 
-    else {
-        if (accessor_)
-            accessor_->addToCache(info.path, *nar, makeNarAccessor(nar));
-    }
-
     /* Compress the NAR. */
     narInfo->compression = compression;
     auto now1 = std::chrono::steady_clock::now();
@@ -181,12 +179,70 @@ void BinaryCacheStore::addToStore(const ValidPathInfo & info, const ref<std::str
         % ((1.0 - (double) narCompressed->size() / nar->size()) * 100.0)
         % duration);
 
-    /* Atomically write the NAR file. */
     narInfo->url = "nar/" + narInfo->fileHash.to_string(Base32, false) + ".nar"
         + (compression == "xz" ? ".xz" :
            compression == "bzip2" ? ".bz2" :
            compression == "br" ? ".br" :
            "");
+
+    /* Optionally maintain an index of DWARF debug info files
+       consisting of JSON files named 'debuginfo/<build-id>' that
+       specify the NAR file and member containing the debug info. */
+    if (writeDebugInfo) {
+
+        std::string buildIdDir = "/lib/debug/.build-id";
+
+        if (narAccessor->stat(buildIdDir).type == FSAccessor::tDirectory) {
+
+            ThreadPool threadPool(25);
+
+            auto doFile = [&](std::string member, std::string key, std::string target) {
+                checkInterrupt();
+
+                nlohmann::json json;
+                json["archive"] = target;
+                json["member"] = member;
+
+                // FIXME: or should we overwrite? The previous link may point
+                // to a GC'ed file, so overwriting might be useful...
+                if (fileExists(key)) return;
+
+                printMsg(lvlTalkative, "creating debuginfo link from '%s' to '%s'", key, target);
+
+                upsertFile(key, json.dump(), "application/json");
+            };
+
+            std::regex regex1("^[0-9a-f]{2}$");
+            std::regex regex2("^[0-9a-f]{38}\\.debug$");
+
+            for (auto & s1 : narAccessor->readDirectory(buildIdDir)) {
+                auto dir = buildIdDir + "/" + s1;
+
+                if (narAccessor->stat(dir).type != FSAccessor::tDirectory
+                    || !std::regex_match(s1, regex1))
+                    continue;
+
+                for (auto & s2 : narAccessor->readDirectory(dir)) {
+                    auto debugPath = dir + "/" + s2;
+
+                    if (narAccessor->stat(debugPath).type != FSAccessor::tRegular
+                        || !std::regex_match(s2, regex2))
+                        continue;
+
+                    auto buildId = s1 + s2;
+
+                    std::string key = "debuginfo/" + buildId;
+                    std::string target = "../" + narInfo->url;
+
+                    threadPool.enqueue(std::bind(doFile, std::string(debugPath, 1), key, target));
+                }
+            }
+
+            threadPool.process();
+        }
+    }
+
+    /* Atomically write the NAR file. */
     if (repair || !fileExists(narInfo->url)) {
         stats.narWrite++;
         upsertFile(narInfo->url, *narCompressed, "application/x-nix-nar");

src/libstore/binary-cache-store.hh

@@ -17,6 +17,7 @@ public:
 
     const Setting<std::string> compression{this, "xz", "compression", "NAR compression method ('xz', 'bzip2', or 'none')"};
     const Setting<bool> writeNARListing{this, false, "write-nar-listing", "whether to write a JSON file listing the files in each NAR"};
+    const Setting<bool> writeDebugInfo{this, false, "index-debug-info", "whether to index DWARF debug info files by build ID"};
     const Setting<Path> secretKeyFile{this, "", "secret-key", "path to secret key used to sign the binary cache"};
     const Setting<Path> localNarCache{this, "", "local-nar-cache", "path to a local cache of NARs"};
     const Setting<bool> parallelCompression{this, false, "parallel-compression",

src/libstore/local-binary-cache-store.cc

@@ -63,6 +63,8 @@ class LocalBinaryCacheStore : public BinaryCacheStore
 void LocalBinaryCacheStore::init()
 {
     createDirs(binaryCacheDir + "/nar");
+    if (writeDebugInfo)
+        createDirs(binaryCacheDir + "/debuginfo");
     BinaryCacheStore::init();
 }