journal: provide compress_threshold_bytes parameter

Previously the compression threshold was hardcoded to 512, which meant that
smaller values wouldn't be compressed. This left some storage savings on the
table, so instead, we make that number tunable.

9 files changed, 138 insertions, 29 deletions

diff --git a/src/journal/journal-file.c b/src/journal/journal-file.c
index 5643c0578d..5fc61b16e0 100644
--- a/src/journal/journal-file.c
+++ b/src/journal/journal-file.c
@@ -51,7 +51,8 @@
 #define DEFAULT_DATA_HASH_TABLE_SIZE (2047ULL*sizeof(HashItem))
 #define DEFAULT_FIELD_HASH_TABLE_SIZE (333ULL*sizeof(HashItem))
 
-#define COMPRESSION_SIZE_THRESHOLD (512ULL)
+#define DEFAULT_COMPRESS_THRESHOLD (512ULL)
+#define MIN_COMPRESS_THRESHOLD (8ULL)
 
 /* This is the minimum journal file size */
 #define JOURNAL_FILE_SIZE_MIN (512ULL*1024ULL)                 /* 512 KiB */
@@ -1552,7 +1553,7 @@ static int journal_file_append_data(
         o->data.hash = htole64(hash);
 
 #if HAVE_XZ || HAVE_LZ4
-        if (JOURNAL_FILE_COMPRESS(f) && size >= COMPRESSION_SIZE_THRESHOLD) {
+        if (JOURNAL_FILE_COMPRESS(f) && size >= f->compress_threshold_bytes) {
                 size_t rsize = 0;
 
                 compression = compress_blob(data, size, o->data.payload, size - 1, &rsize);
@@ -3211,6 +3212,7 @@ int journal_file_open(
                 int flags,
                 mode_t mode,
                 bool compress,
+                uint64_t compress_threshold_bytes,
                 bool seal,
                 JournalMetrics *metrics,
                 MMapCache *mmap_cache,
@@ -3222,6 +3224,7 @@ int journal_file_open(
         JournalFile *f;
         void *h;
         int r;
+        char bytes[FORMAT_BYTES_MAX];
 
         assert(ret);
         assert(fd >= 0 || fname);
@@ -3247,10 +3250,20 @@ int journal_file_open(
 #elif HAVE_XZ
         f->compress_xz = compress;
 #endif
+
+        if (compress_threshold_bytes == (uint64_t) -1)
+                f->compress_threshold_bytes = DEFAULT_COMPRESS_THRESHOLD;
+        else
+                f->compress_threshold_bytes = MAX(MIN_COMPRESS_THRESHOLD, compress_threshold_bytes);
+
 #if HAVE_GCRYPT
         f->seal = seal;
 #endif
 
+        log_debug("Journal effective settings seal=%s compress=%s compress_threshold_bytes=%s",
+                  yes_no(f->seal), yes_no(JOURNAL_FILE_COMPRESS(f)),
+                  format_bytes(bytes, sizeof(bytes), f->compress_threshold_bytes));
+
         if (mmap_cache)
                 f->mmap = mmap_cache_ref(mmap_cache);
         else {
@@ -3435,7 +3448,7 @@ fail:
         return r;
 }
 
-int journal_file_rotate(JournalFile **f, bool compress, bool seal, Set *deferred_closes) {
+int journal_file_rotate(JournalFile **f, bool compress, uint64_t compress_threshold_bytes, bool seal, Set *deferred_closes) {
         _cleanup_free_ char *p = NULL;
         size_t l;
         JournalFile *old_file, *new_file = NULL;
@@ -3489,7 +3502,9 @@ int journal_file_rotate(JournalFile **f, bool compress, bool seal, Set *deferred
          * we archive them */
         old_file->defrag_on_close = true;
 
-        r = journal_file_open(-1, old_file->path, old_file->flags, old_file->mode, compress, seal, NULL, old_file->mmap, deferred_closes, old_file, &new_file);
+        r = journal_file_open(-1, old_file->path, old_file->flags, old_file->mode, compress,
+                              compress_threshold_bytes, seal, NULL, old_file->mmap, deferred_closes,
+                              old_file, &new_file);
 
         if (deferred_closes &&
             set_put(deferred_closes, old_file) >= 0)
@@ -3506,6 +3521,7 @@ int journal_file_open_reliably(
                 int flags,
                 mode_t mode,
                 bool compress,
+                uint64_t compress_threshold_bytes,
                 bool seal,
                 JournalMetrics *metrics,
                 MMapCache *mmap_cache,
@@ -3517,7 +3533,8 @@ int journal_file_open_reliably(
         size_t l;
         _cleanup_free_ char *p = NULL;
 
-        r = journal_file_open(-1, fname, flags, mode, compress, seal, metrics, mmap_cache, deferred_closes, template, ret);
+        r = journal_file_open(-1, fname, flags, mode, compress, compress_threshold_bytes, seal, metrics, mmap_cache,
+                              deferred_closes, template, ret);
         if (!IN_SET(r,
                     -EBADMSG,           /* Corrupted */
                     -ENODATA,           /* Truncated */
@@ -3559,7 +3576,8 @@ int journal_file_open_reliably(
 
         log_warning_errno(r, "File %s corrupted or uncleanly shut down, renaming and replacing.", fname);
 
-        return journal_file_open(-1, fname, flags, mode, compress, seal, metrics, mmap_cache, deferred_closes, template, ret);
+        return journal_file_open(-1, fname, flags, mode, compress, compress_threshold_bytes, seal, metrics, mmap_cache,
+                                 deferred_closes, template, ret);
 }
 
 int journal_file_copy_entry(JournalFile *from, JournalFile *to, Object *o, uint64_t p, uint64_t *seqnum, Object **ret, uint64_t *offset) {
diff --git a/src/journal/journal-file.h b/src/journal/journal-file.h
index 67abf8da49..04d35f5813 100644
--- a/src/journal/journal-file.h
+++ b/src/journal/journal-file.h
@@ -124,6 +124,7 @@ typedef struct JournalFile {
 
         unsigned last_seen_generation;
 
+        uint64_t compress_threshold_bytes;
 #if HAVE_XZ || HAVE_LZ4
         void *compress_buffer;
         size_t compress_buffer_size;
@@ -153,6 +154,7 @@ int journal_file_open(
                 int flags,
                 mode_t mode,
                 bool compress,
+                uint64_t compress_threshold_bytes,
                 bool seal,
                 JournalMetrics *metrics,
                 MMapCache *mmap_cache,
@@ -169,6 +171,7 @@ int journal_file_open_reliably(
                 int flags,
                 mode_t mode,
                 bool compress,
+                uint64_t compress_threshold_bytes,
                 bool seal,
                 JournalMetrics *metrics,
                 MMapCache *mmap_cache,
@@ -246,7 +249,7 @@ int journal_file_copy_entry(JournalFile *from, JournalFile *to, Object *o, uint6
 void journal_file_dump(JournalFile *f);
 void journal_file_print_header(JournalFile *f);
 
-int journal_file_rotate(JournalFile **f, bool compress, bool seal, Set *deferred_closes);
+int journal_file_rotate(JournalFile **f, bool compress, uint64_t compress_threshold_bytes, bool seal, Set *deferred_closes);
 
 void journal_file_post_change(JournalFile *f);
 int journal_file_enable_post_change_timer(JournalFile *f, sd_event *e, usec_t t);
diff --git a/src/journal/journald-server.c b/src/journal/journald-server.c
index 5cd58e8a77..570450d321 100644
--- a/src/journal/journald-server.c
+++ b/src/journal/journald-server.c
@@ -280,9 +280,9 @@ static int open_journal(
         assert(ret);
 
         if (reliably)
-                r = journal_file_open_reliably(fname, flags, 0640, s->compress, seal, metrics, s->mmap, s->deferred_closes, NULL, &f);
+                r = journal_file_open_reliably(fname, flags, 0640, s->compress, (uint64_t) -1, seal, metrics, s->mmap, s->deferred_closes, NULL, &f);
         else
-                r = journal_file_open(-1, fname, flags, 0640, s->compress, seal, metrics, s->mmap, s->deferred_closes, NULL, &f);
+                r = journal_file_open(-1, fname, flags, 0640, s->compress, (uint64_t) -1, seal, metrics, s->mmap, s->deferred_closes, NULL, &f);
         if (r < 0)
                 return r;
 
@@ -463,7 +463,7 @@ static int do_rotate(
         if (!*f)
                 return -EINVAL;
 
-        r = journal_file_rotate(f, s->compress, seal, s->deferred_closes);
+        r = journal_file_rotate(f, s->compress, (uint64_t) -1, seal, s->deferred_closes);
         if (r < 0) {
                 if (*f)
                         return log_error_errno(r, "Failed to rotate %s: %m", (*f)->path);
diff --git a/src/journal/sd-journal.c b/src/journal/sd-journal.c
index 11dbd83f2d..79ffd9cb3f 100644
--- a/src/journal/sd-journal.c
+++ b/src/journal/sd-journal.c
@@ -1337,7 +1337,7 @@ static int add_any_file(
                 goto finish;
         }
 
-        r = journal_file_open(fd, path, O_RDONLY, 0, false, false, NULL, j->mmap, NULL, NULL, &f);
+        r = journal_file_open(fd, path, O_RDONLY, 0, false, 0, false, NULL, j->mmap, NULL, NULL, &f);
         if (r < 0) {
                 log_debug_errno(r, "Failed to open journal file %s: %m", path);
                 goto finish;
diff --git a/src/journal/test-journal-flush.c b/src/journal/test-journal-flush.c
index f9b7b75ef1..e302fd61f1 100644
--- a/src/journal/test-journal-flush.c
+++ b/src/journal/test-journal-flush.c
@@ -39,7 +39,7 @@ int main(int argc, char *argv[]) {
         assert_se(mkdtemp(dn));
         fn = strappend(dn, "/test.journal");
 
-        r = journal_file_open(-1, fn, O_CREAT|O_RDWR, 0644, false, false, NULL, NULL, NULL, NULL, &new_journal);
+        r = journal_file_open(-1, fn, O_CREAT|O_RDWR, 0644, false, 0, false, NULL, NULL, NULL, NULL, &new_journal);
         assert_se(r >= 0);
 
         r = sd_journal_open(&j, 0);
diff --git a/src/journal/test-journal-interleaving.c b/src/journal/test-journal-interleaving.c
index d87bdbdd32..f8b8d1e6d8 100644
--- a/src/journal/test-journal-interleaving.c
+++ b/src/journal/test-journal-interleaving.c
@@ -52,7 +52,7 @@ _noreturn_ static void log_assert_errno(const char *text, int error, const char
 
 static JournalFile *test_open(const char *name) {
         JournalFile *f;
-        assert_ret(journal_file_open(-1, name, O_RDWR|O_CREAT, 0644, true, false, NULL, NULL, NULL, NULL, &f));
+        assert_ret(journal_file_open(-1, name, O_RDWR|O_CREAT, 0644, true, (uint64_t) -1, false, NULL, NULL, NULL, NULL, &f));
         return f;
 }
 
@@ -217,7 +217,7 @@ static void test_sequence_numbers(void) {
         assert_se(chdir(t) >= 0);
 
         assert_se(journal_file_open(-1, "one.journal", O_RDWR|O_CREAT, 0644,
-                                    true, false, NULL, NULL, NULL, NULL, &one) == 0);
+                                    true, (uint64_t) -1, false, NULL, NULL, NULL, NULL, &one) == 0);
 
         append_number(one, 1, &seqnum);
         printf("seqnum=%"PRIu64"\n", seqnum);
@@ -234,7 +234,7 @@ static void test_sequence_numbers(void) {
         memcpy(&seqnum_id, &one->header->seqnum_id, sizeof(sd_id128_t));
 
         assert_se(journal_file_open(-1, "two.journal", O_RDWR|O_CREAT, 0644,
-                                    true, false, NULL, NULL, NULL, one, &two) == 0);
+                                    true, (uint64_t) -1, false, NULL, NULL, NULL, one, &two) == 0);
 
         assert_se(two->header->state == STATE_ONLINE);
         assert_se(!sd_id128_equal(two->header->file_id, one->header->file_id));
@@ -265,7 +265,7 @@ static void test_sequence_numbers(void) {
         seqnum = 0;
 
         assert_se(journal_file_open(-1, "two.journal", O_RDWR, 0,
-                                    true, false, NULL, NULL, NULL, NULL, &two) == 0);
+                                    true, (uint64_t) -1, false, NULL, NULL, NULL, NULL, &two) == 0);
 
         assert_se(sd_id128_equal(two->header->seqnum_id, seqnum_id));
 
diff --git a/src/journal/test-journal-stream.c b/src/journal/test-journal-stream.c
index 73ed6e5dcb..279301c67d 100644
--- a/src/journal/test-journal-stream.c
+++ b/src/journal/test-journal-stream.c
@@ -93,9 +93,9 @@ int main(int argc, char *argv[]) {
         assert_se(mkdtemp(t));
         assert_se(chdir(t) >= 0);
 
-        assert_se(journal_file_open(-1, "one.journal", O_RDWR|O_CREAT, 0666, true, false, NULL, NULL, NULL, NULL, &one) == 0);
-        assert_se(journal_file_open(-1, "two.journal", O_RDWR|O_CREAT, 0666, true, false, NULL, NULL, NULL, NULL, &two) == 0);
-        assert_se(journal_file_open(-1, "three.journal", O_RDWR|O_CREAT, 0666, true, false, NULL, NULL, NULL, NULL, &three) == 0);
+        assert_se(journal_file_open(-1, "one.journal", O_RDWR|O_CREAT, 0666, true, (uint64_t) -1, false, NULL, NULL, NULL, NULL, &one) == 0);
+        assert_se(journal_file_open(-1, "two.journal", O_RDWR|O_CREAT, 0666, true, (uint64_t) -1, false, NULL, NULL, NULL, NULL, &two) == 0);
+        assert_se(journal_file_open(-1, "three.journal", O_RDWR|O_CREAT, 0666, true, (uint64_t) -1, false, NULL, NULL, NULL, NULL, &three) == 0);
 
         for (i = 0; i < N_ENTRIES; i++) {
                 char *p, *q;
diff --git a/src/journal/test-journal-verify.c b/src/journal/test-journal-verify.c
index fbb75e43e3..89389579f3 100644
--- a/src/journal/test-journal-verify.c
+++ b/src/journal/test-journal-verify.c
@@ -56,7 +56,7 @@ static int raw_verify(const char *fn, const char *verification_key) {
         JournalFile *f;
         int r;
 
-        r = journal_file_open(-1, fn, O_RDONLY, 0666, true, !!verification_key, NULL, NULL, NULL, NULL, &f);
+        r = journal_file_open(-1, fn, O_RDONLY, 0666, true, (uint64_t) -1, !!verification_key, NULL, NULL, NULL, NULL, &f);
         if (r < 0)
                 return r;
 
@@ -89,7 +89,7 @@ int main(int argc, char *argv[]) {
 
         log_info("Generating...");
 
-        assert_se(journal_file_open(-1, "test.journal", O_RDWR|O_CREAT, 0666, true, !!verification_key, NULL, NULL, NULL, NULL, &f) == 0);
+        assert_se(journal_file_open(-1, "test.journal", O_RDWR|O_CREAT, 0666, true, (uint64_t) -1, !!verification_key, NULL, NULL, NULL, NULL, &f) == 0);
 
         for (n = 0; n < N_ENTRIES; n++) {
                 struct iovec iovec;
@@ -112,7 +112,7 @@ int main(int argc, char *argv[]) {
 
         log_info("Verifying...");
 
-        assert_se(journal_file_open(-1, "test.journal", O_RDONLY, 0666, true, !!verification_key, NULL, NULL, NULL, NULL, &f) == 0);
+        assert_se(journal_file_open(-1, "test.journal", O_RDONLY, 0666, true, (uint64_t) -1, !!verification_key, NULL, NULL, NULL, NULL, &f) == 0);
         /* journal_file_print_header(f); */
         journal_file_dump(f);
 
diff --git a/src/journal/test-journal.c b/src/journal/test-journal.c
index 517c9102a6..3b9917b4a8 100644
--- a/src/journal/test-journal.c
+++ b/src/journal/test-journal.c
@@ -43,7 +43,7 @@ static void test_non_empty(void) {
         assert_se(mkdtemp(t));
         assert_se(chdir(t) >= 0);
 
-        assert_se(journal_file_open(-1, "test.journal", O_RDWR|O_CREAT, 0666, true, true, NULL, NULL, NULL, NULL, &f) == 0);
+        assert_se(journal_file_open(-1, "test.journal", O_RDWR|O_CREAT, 0666, true, (uint64_t) -1, true, NULL, NULL, NULL, NULL, &f) == 0);
 
         dual_timestamp_get(&ts);
 
@@ -105,8 +105,8 @@ static void test_non_empty(void) {
 
         assert_se(journal_file_move_to_entry_by_seqnum(f, 10, DIRECTION_DOWN, &o, NULL) == 0);
 
-        journal_file_rotate(&f, true, true, NULL);
-        journal_file_rotate(&f, true, true, NULL);
+        journal_file_rotate(&f, true, (uint64_t) -1, true, NULL);
+        journal_file_rotate(&f, true, (uint64_t) -1, true, NULL);
 
         (void) journal_file_close(f);
 
@@ -132,13 +132,13 @@ static void test_empty(void) {
         assert_se(mkdtemp(t));
         assert_se(chdir(t) >= 0);
 
-        assert_se(journal_file_open(-1, "test.journal", O_RDWR|O_CREAT, 0666, false, false, NULL, NULL, NULL, NULL, &f1) == 0);
+        assert_se(journal_file_open(-1, "test.journal", O_RDWR|O_CREAT, 0666, false, (uint64_t) -1, false, NULL, NULL, NULL, NULL, &f1) == 0);
 
-        assert_se(journal_file_open(-1, "test-compress.journal", O_RDWR|O_CREAT, 0666, true, false, NULL, NULL, NULL, NULL, &f2) == 0);
+        assert_se(journal_file_open(-1, "test-compress.journal", O_RDWR|O_CREAT, 0666, true, (uint64_t) -1, false, NULL, NULL, NULL, NULL, &f2) == 0);
 
-        assert_se(journal_file_open(-1, "test-seal.journal", O_RDWR|O_CREAT, 0666, false, true, NULL, NULL, NULL, NULL, &f3) == 0);
+        assert_se(journal_file_open(-1, "test-seal.journal", O_RDWR|O_CREAT, 0666, false, (uint64_t) -1, true, NULL, NULL, NULL, NULL, &f3) == 0);
 
-        assert_se(journal_file_open(-1, "test-seal-compress.journal", O_RDWR|O_CREAT, 0666, true, true, NULL, NULL, NULL, NULL, &f4) == 0);
+        assert_se(journal_file_open(-1, "test-seal-compress.journal", O_RDWR|O_CREAT, 0666, true, (uint64_t) -1, true, NULL, NULL, NULL, NULL, &f4) == 0);
 
         journal_file_print_header(f1);
         puts("");
@@ -165,6 +165,91 @@ static void test_empty(void) {
         (void) journal_file_close(f4);
 }
 
+#if HAVE_XZ || HAVE_LZ4
+static bool check_compressed(uint64_t compress_threshold, uint64_t data_size) {
+        dual_timestamp ts;
+        JournalFile *f;
+        struct iovec iovec;
+        Object *o;
+        uint64_t p;
+        char t[] = "/tmp/journal-XXXXXX";
+        char data[2048] = {0};
+        bool is_compressed;
+        int r;
+
+        assert_se(data_size <= sizeof(data));
+
+        log_set_max_level(LOG_DEBUG);
+
+        assert_se(mkdtemp(t));
+        assert_se(chdir(t) >= 0);
+
+        assert_se(journal_file_open(-1, "test.journal", O_RDWR|O_CREAT, 0666, true, compress_threshold, true, NULL, NULL, NULL, NULL, &f) == 0);
+
+        dual_timestamp_get(&ts);
+
+        iovec.iov_base = (void*) data;
+        iovec.iov_len = data_size;
+        assert_se(journal_file_append_entry(f, &ts, &iovec, 1, NULL, NULL, NULL) == 0);
+
+#if HAVE_GCRYPT
+        journal_file_append_tag(f);
+#endif
+        journal_file_dump(f);
+
+        /* We have to partially reimplement some of the dump logic, because the normal next_entry does the
+         * decompression for us. */
+        p = le64toh(f->header->header_size);
+        while (true) {
+                r = journal_file_move_to_object(f, OBJECT_UNUSED, p, &o);
+                assert_se(r == 0);
+                if (o->object.type == OBJECT_DATA)
+                        break;
+
+                assert_se(p < le64toh(f->header->tail_object_offset));
+                p = p + ALIGN64(le64toh(o->object.size));
+        }
+
+        is_compressed = (o->object.flags & OBJECT_COMPRESSION_MASK) != 0;
+
+        (void) journal_file_close(f);
+
+        log_info("Done...");
+
+        if (arg_keep)
+                log_info("Not removing %s", t);
+        else {
+                journal_directory_vacuum(".", 3000000, 0, 0, NULL, true);
+
+                assert_se(rm_rf(t, REMOVE_ROOT|REMOVE_PHYSICAL) >= 0);
+        }
+
+        puts("------------------------------------------------------------");
+
+        return is_compressed;
+}
+
+static void test_min_compress_size(void) {
+        /* Note that XZ will actually fail to compress anything under 80 bytes, so you have to choose the limits
+         * carefully */
+
+        /* DEFAULT_MIN_COMPRESS_SIZE is 512 */
+        assert_se(!check_compressed((uint64_t) -1, 255));
+        assert_se(check_compressed((uint64_t) -1, 513));
+
+        /* compress everything */
+        assert_se(check_compressed(0, 96));
+        assert_se(check_compressed(8, 96));
+
+        /* Ensure we don't try to compress less than 8 bytes */
+        assert_se(!check_compressed(0, 7));
+
+        /* check boundary conditions */
+        assert_se(check_compressed(256, 256));
+        assert_se(!check_compressed(256, 255));
+}
+#endif
+
 int main(int argc, char *argv[]) {
         arg_keep = argc > 1;
 
@@ -174,6 +259,9 @@ int main(int argc, char *argv[]) {
 
         test_non_empty();
         test_empty();
+#if HAVE_XZ || HAVE_LZ4
+        test_min_compress_size();
+#endif
 
         return 0;
 }