Add automatic garbage collection

Nix can now automatically run the garbage collector during builds or
while adding paths to the store. The option "min-free = <bytes>"
specifies that Nix should run the garbage collector whenever free
space in the Nix store drops below <bytes>. It will then delete
garbage until "max-free" bytes are available.

Garbage collection during builds is asynchronous; running builds are
not paused and new builds are not blocked. However, there also is a
synchronous GC run prior to the first build/substitution.

Currently, no old GC roots are deleted (as in "nix-collect-garbage
-d").

Author: edolstra

doc/manual/release-notes/rl-1.12.xml

@@ -393,6 +393,10 @@ configureFlags = "--prefix=${placeholder "out"} --includedir=${placeholder "dev"
     package repository.</para>
   </listitem>
 
+  <listitem>
+    <para>Automatic garbage collection.</para>
+  </listitem>
+
 </itemizedlist>
 
 <para>This release has contributions from TBD.</para>

src/libstore/build.cc

@@ -3957,6 +3957,8 @@ void Worker::run(const Goals & _topGoals)
 
         checkInterrupt();
 
+        store.autoGC(false);
+
         /* Call every wake goal (in the ordering established by
            CompareGoalPtrs). */
         while (!awake.empty() && !topGoals.empty()) {
@@ -4014,6 +4016,9 @@ void Worker::waitForInput()
        is a build timeout, then wait for input until the first
        deadline for any child. */
     auto nearest = steady_time_point::max(); // nearest deadline
+    if (settings.minFree.get() != 0)
+        // Periodicallty wake up to see if we need to run the garbage collector.
+        nearest = before + std::chrono::seconds(10);
     for (auto & i : children) {
         if (!i.respectTimeouts) continue;
         if (0 != settings.maxSilentTime)

src/libstore/gc.cc

@@ -1,6 +1,7 @@
 #include "derivations.hh"
 #include "globals.hh"
 #include "local-store.hh"
+#include "finally.hh"
 
 #include <functional>
 #include <queue>
@@ -9,6 +10,7 @@
 
 #include <sys/types.h>
 #include <sys/stat.h>
+#include <sys/statvfs.h>
 #include <errno.h>
 #include <fcntl.h>
 #include <unistd.h>
@@ -845,4 +847,72 @@ void LocalStore::collectGarbage(const GCOptions & options, GCResults & results)
 }
 
 
+void LocalStore::autoGC(bool sync)
+{
+    auto getAvail = [this]() {
+        struct statvfs st;
+        if (statvfs(realStoreDir.c_str(), &st))
+            throw SysError("getting filesystem info about '%s'", realStoreDir);
+
+        return (uint64_t) st.f_bavail * st.f_bsize;
+    };
+
+    std::shared_future<void> future;
+
+    {
+        auto state(_state.lock());
+
+        if (state->gcRunning) {
+            future = state->gcFuture;
+            debug("waiting for auto-GC to finish");
+            goto sync;
+        }
+
+        auto now = std::chrono::steady_clock::now();
+
+        if (now < state->lastGCCheck + std::chrono::seconds(5)) return;
+
+        auto avail = getAvail();
+
+        state->lastGCCheck = now;
+
+        if (avail >= settings.minFree || avail >= settings.maxFree) return;
+
+        if (avail > state->availAfterGC * 0.97) return;
+
+        state->gcRunning = true;
+
+        std::promise<void> promise;
+        future = state->gcFuture = promise.get_future().share();
+
+        std::thread([promise{std::move(promise)}, this, avail, getAvail]() mutable {
+
+            /* Wake up any threads waiting for the auto-GC to finish. */
+            Finally wakeup([&]() {
+                auto state(_state.lock());
+                state->gcRunning = false;
+                state->lastGCCheck = std::chrono::steady_clock::now();
+                promise.set_value();
+            });
+
+            printInfo("running auto-GC to free %d bytes", settings.maxFree - avail);
+
+            GCOptions options;
+            options.maxFreed = settings.maxFree - avail;
+
+            GCResults results;
+
+            collectGarbage(options, results);
+
+            _state.lock()->availAfterGC = getAvail();
+
+        }).detach();
+    }
+
+ sync:
+    // Wait for the future outside of the state lock.
+    if (sync) future.get();
+}
+
+
 }

src/libstore/globals.hh

@@ -4,8 +4,9 @@
 #include "config.hh"
 
 #include <map>
-#include <sys/types.h>
+#include <limits>
 
+#include <sys/types.h>
 
 namespace nix {
 
@@ -342,6 +343,13 @@ public:
 
     Setting<Strings> hashedMirrors{this, {"http://tarballs.nixos.org/"}, "hashed-mirrors",
         "A list of servers used by builtins.fetchurl to fetch files by hash."};
+
+    Setting<uint64_t> minFree{this, 0, "min-free",
+        "Automatically run the garbage collector when free disk space drops below the specified amount."};
+
+    Setting<uint64_t> maxFree{this, std::numeric_limits<uint64_t>::max(), "max-free",
+        "Stop deleting garbage when free disk space is above the specified amount."};
+
 };
 
 

src/libstore/local-store.cc

@@ -244,6 +244,18 @@ LocalStore::LocalStore(const Params & params)
 
 LocalStore::~LocalStore()
 {
+    std::shared_future<void> future;
+
+    {
+        auto state(_state.lock());
+        if (state->gcRunning)
+            future = state->gcFuture;
+    }
+
+    if (future.valid()) {
+        printError("waiting for auto-GC to finish on exit...");
+        future.get();
+    }
 
     try {
         auto state(_state.lock());
@@ -991,6 +1003,8 @@ void LocalStore::addToStore(const ValidPathInfo & info, const ref<std::string> &
             StringSource source(*nar);
             restorePath(realPath, source);
 
+            autoGC();
+
             canonicalisePathMetaData(realPath, -1);
 
             optimisePath(realPath); // FIXME: combine with hashPath()
@@ -1025,6 +1039,8 @@ Path LocalStore::addToStoreFromDump(const string & dump, const string & name,
 
             deletePath(realPath);
 
+            autoGC();
+
             if (recursive) {
                 StringSource source(dump);
                 restorePath(realPath, source);
@@ -1097,6 +1113,8 @@ Path LocalStore::addTextToStore(const string & name, const string & s,
 
             deletePath(realPath);
 
+            autoGC();
+
             writeFile(realPath, s);
 
             canonicalisePathMetaData(realPath, -1);

src/libstore/local-store.hh

@@ -7,6 +7,8 @@
 #include "sync.hh"
 #include "util.hh"
 
+#include <chrono>
+#include <future>
 #include <string>
 #include <unordered_set>
 
@@ -60,6 +62,21 @@ private:
 
         /* The file to which we write our temporary roots. */
         AutoCloseFD fdTempRoots;
+
+        /* The last time we checked whether to do an auto-GC, or an
+           auto-GC finished. */
+        std::chrono::time_point<std::chrono::steady_clock> lastGCCheck;
+
+        /* Whether auto-GC is running. If so, get gcFuture to wait for
+           the GC to finish. */
+        bool gcRunning = false;
+        std::shared_future<void> gcFuture;
+
+        /* How much disk space was available after the previous
+           auto-GC. If the current available disk space is below
+           minFree but not much below availAfterGC, then there is no
+           point in starting a new GC. */
+        uint64_t availAfterGC = std::numeric_limits<uint64_t>::max();
     };
 
     Sync<State, std::recursive_mutex> _state;
@@ -196,6 +213,10 @@ public:
 
     void addSignatures(const Path & storePath, const StringSet & sigs) override;
 
+    /* If free disk space in /nix/store if below minFree, delete
+       garbage until it exceeds maxFree. */
+    void autoGC(bool sync = true);
+
 private:
 
     int getSchema();