journal: by default do not decompress dat objects larger than 64K

This introduces a new data threshold setting for sd_journal objects
which controls the maximum size of objects to decompress. This is
relieves the library from having to decompress full data objects even
if a client program is only interested in the initial part of them.

This speeds up "systemd-coredumpctl" drastically when invoked without
parameters.

11 files changed, 78 insertions, 20 deletions

diff --git a/src/journal/compress.c b/src/journal/compress.c
index 75e70c5ffa..a4427be75a 100644
--- a/src/journal/compress.c
+++ b/src/journal/compress.c
@@ -24,6 +24,7 @@
 #include <string.h>
 #include <lzma.h>
 
+#include "macro.h"
 #include "compress.h"
 
 bool compress_blob(const void *src, uint64_t src_size, void *dst, uint64_t *dst_size) {
@@ -66,10 +67,11 @@ fail:
 }
 
 bool uncompress_blob(const void *src, uint64_t src_size,
-                     void **dst, uint64_t *dst_alloc_size, uint64_t* dst_size) {
+                     void **dst, uint64_t *dst_alloc_size, uint64_t* dst_size, uint64_t dst_max) {
 
         lzma_stream s = LZMA_STREAM_INIT;
         lzma_ret ret;
+        uint64_t space;
         bool b = false;
 
         assert(src);
@@ -98,7 +100,8 @@ bool uncompress_blob(const void *src, uint64_t src_size,
         s.avail_in = src_size;
 
         s.next_out = *dst;
-        s.avail_out = *dst_alloc_size;
+        space = dst_max > 0 ? MIN(*dst_alloc_size, dst_max) : *dst_alloc_size;
+        s.avail_out = space;
 
         for (;;) {
                 void *p;
@@ -111,18 +114,23 @@ bool uncompress_blob(const void *src, uint64_t src_size,
                 if (ret != LZMA_OK)
                         goto fail;
 
-                p = realloc(*dst, *dst_alloc_size*2);
+                if (dst_max > 0 && (space - s.avail_out) >= dst_max)
+                        break;
+
+                p = realloc(*dst, space*2);
                 if (!p)
                         goto fail;
 
                 s.next_out = (uint8_t*) p + ((uint8_t*) s.next_out - (uint8_t*) *dst);
-                s.avail_out += *dst_alloc_size;
+                s.avail_out += space;
+
+                space *= 2;
 
                 *dst = p;
-                *dst_alloc_size *= 2;
+                *dst_alloc_size = space;
         }
 
-        *dst_size = *dst_alloc_size - s.avail_out;
+        *dst_size = space - s.avail_out;
         b = true;
 
 fail:
diff --git a/src/journal/compress.h b/src/journal/compress.h
index b6f1aa0ed7..2b87e73a26 100644
--- a/src/journal/compress.h
+++ b/src/journal/compress.h
@@ -27,7 +27,7 @@
 bool compress_blob(const void *src, uint64_t src_size, void *dst, uint64_t *dst_size);
 
 bool uncompress_blob(const void *src, uint64_t src_size,
-                     void **dst, uint64_t *dst_alloc_size, uint64_t* dst_size);
+                     void **dst, uint64_t *dst_alloc_size, uint64_t* dst_size, uint64_t dst_max);
 
 bool uncompress_startswith(const void *src, uint64_t src_size,
                            void **buffer, uint64_t *buffer_size,
diff --git a/src/journal/coredumpctl.c b/src/journal/coredumpctl.c
index 4adc9236f1..f9cebb7a42 100644
--- a/src/journal/coredumpctl.c
+++ b/src/journal/coredumpctl.c
@@ -342,6 +342,11 @@ static int dump_list(sd_journal *j) {
 
         assert(j);
 
+        /* The coredumps are likely to compressed, and for just
+         * listing them we don#t need to decompress them, so let's
+         * pick a fairly low data threshold here */
+        sd_journal_set_data_threshold(j, 4096);
+
         SD_JOURNAL_FOREACH(j) {
                 if (field)
                         print_field(stdout, j);
@@ -381,6 +386,9 @@ static int dump_core(sd_journal* j) {
 
         assert(j);
 
+        /* We want full data, nothing truncated. */
+        sd_journal_set_data_threshold(j, 0);
+
         r = focus(j);
         if (r < 0)
                 return r;
@@ -428,6 +436,8 @@ static int run_gdb(sd_journal *j) {
 
         assert(j);
 
+        sd_journal_set_data_threshold(j, 0);
+
         r = focus(j);
         if (r < 0)
                 return r;
diff --git a/src/journal/journal-file.c b/src/journal/journal-file.c
index 3df099dbd2..13fc8edea9 100644
--- a/src/journal/journal-file.c
+++ b/src/journal/journal-file.c
@@ -780,7 +780,7 @@ int journal_file_find_data_object_with_hash(
 
                         l -= offsetof(Object, data.payload);
 
-                        if (!uncompress_blob(o->data.payload, l, &f->compress_buffer, &f->compress_buffer_size, &rsize))
+                        if (!uncompress_blob(o->data.payload, l, &f->compress_buffer, &f->compress_buffer_size, &rsize, 0))
                                 return -EBADMSG;
 
                         if (rsize == size &&
@@ -2591,7 +2591,6 @@ int journal_file_open_reliably(
                                  metrics, mmap_cache, template, ret);
 }
 
-
 int journal_file_copy_entry(JournalFile *from, JournalFile *to, Object *o, uint64_t p, uint64_t *seqnum, Object **ret, uint64_t *offset) {
         uint64_t i, n;
         uint64_t q, xor_hash = 0;
@@ -2645,7 +2644,7 @@ int journal_file_copy_entry(JournalFile *from, JournalFile *to, Object *o, uint6
 #ifdef HAVE_XZ
                         uint64_t rsize;
 
-                        if (!uncompress_blob(o->data.payload, l, &from->compress_buffer, &from->compress_buffer_size, &rsize))
+                        if (!uncompress_blob(o->data.payload, l, &from->compress_buffer, &from->compress_buffer_size, &rsize, 0))
                                 return -EBADMSG;
 
                         data = from->compress_buffer;
diff --git a/src/journal/journal-internal.h b/src/journal/journal-internal.h
index 75a4129e5b..97de0e75ff 100644
--- a/src/journal/journal-internal.h
+++ b/src/journal/journal-internal.h
@@ -121,6 +121,8 @@ struct sd_journal {
         uint64_t unique_offset;
 
         bool on_network;
+
+        size_t data_threshold;
 };
 
 char *journal_make_match_string(sd_journal *j);
diff --git a/src/journal/journal-verify.c b/src/journal/journal-verify.c
index 1a67d5a04b..ed28b45737 100644
--- a/src/journal/journal-verify.c
+++ b/src/journal/journal-verify.c
@@ -69,7 +69,7 @@ static int journal_file_object_verify(JournalFile *f, Object *o) {
 
                         if (!uncompress_blob(o->data.payload,
                                              le64toh(o->object.size) - offsetof(Object, data.payload),
-                                             &b, &alloc, &b_size))
+                                             &b, &alloc, &b_size, 0))
                                 return -EBADMSG;
 
                         h2 = hash64(b, b_size);
diff --git a/src/journal/journald-server.c b/src/journal/journald-server.c
index 3eaf0d3ee2..43ffe75560 100644
--- a/src/journal/journald-server.c
+++ b/src/journal/journald-server.c
@@ -934,6 +934,8 @@ int server_flush_to_var(Server *s) {
                 return r;
         }
 
+        sd_journal_set_data_threshold(j, 0);
+
         SD_JOURNAL_FOREACH(j) {
                 Object *o = NULL;
                 JournalFile *f;
diff --git a/src/journal/libsystemd-journal.sym b/src/journal/libsystemd-journal.sym
index 17b5bf80d9..7b602f59cb 100644
--- a/src/journal/libsystemd-journal.sym
+++ b/src/journal/libsystemd-journal.sym
@@ -86,4 +86,6 @@ global:
         sd_journal_fd_reliable;
         sd_journal_get_catalog;
         sd_journal_get_catalog_for_message_id;
+        sd_journal_set_data_threshold;
+        sd_journal_get_data_threshold;
 } LIBSYSTEMD_JOURNAL_195;
diff --git a/src/journal/sd-journal.c b/src/journal/sd-journal.c
index fe0478f166..095fbb249c 100644
--- a/src/journal/sd-journal.c
+++ b/src/journal/sd-journal.c
@@ -47,6 +47,8 @@
 
 #define REPLACE_VAR_MAX 256
 
+#define DEFAULT_DATA_THRESHOLD (64*1024)
+
 static void detach_location(sd_journal *j) {
         Iterator i;
         JournalFile *f;
@@ -1560,6 +1562,7 @@ static sd_journal *journal_new(int flags, const char *path) {
 
         j->inotify_fd = -1;
         j->flags = flags;
+        j->data_threshold = DEFAULT_DATA_THRESHOLD;
 
         if (path) {
                 j->path = strdup(path);
@@ -1838,7 +1841,8 @@ _public_ int sd_journal_get_data(sd_journal *j, const char *field, const void **
                                 uint64_t rsize;
 
                                 if (!uncompress_blob(o->data.payload, l,
-                                                     &f->compress_buffer, &f->compress_buffer_size, &rsize))
+                                                     &f->compress_buffer, &f->compress_buffer_size, &rsize,
+                                                     j->data_threshold))
                                         return -EBADMSG;
 
                                 *data = f->compress_buffer;
@@ -1862,7 +1866,7 @@ _public_ int sd_journal_get_data(sd_journal *j, const char *field, const void **
                         *data = o->data.payload;
                         *size = t;
 
-                        return 0;
+                        return 1;
                 }
 
                 r = journal_file_move_to_object(f, OBJECT_ENTRY, f->current_offset, &o);
@@ -1873,7 +1877,7 @@ _public_ int sd_journal_get_data(sd_journal *j, const char *field, const void **
         return -ENOENT;
 }
 
-static int return_data(JournalFile *f, Object *o, const void **data, size_t *size) {
+static int return_data(sd_journal *j, JournalFile *f, Object *o, const void **data, size_t *size) {
         size_t t;
         uint64_t l;
 
@@ -1888,7 +1892,7 @@ static int return_data(JournalFile *f, Object *o, const void **data, size_t *siz
 #ifdef HAVE_XZ
                 uint64_t rsize;
 
-                if (!uncompress_blob(o->data.payload, l, &f->compress_buffer, &f->compress_buffer_size, &rsize))
+                if (!uncompress_blob(o->data.payload, l, &f->compress_buffer, &f->compress_buffer_size, &rsize, j->data_threshold))
                         return -EBADMSG;
 
                 *data = f->compress_buffer;
@@ -1942,7 +1946,7 @@ _public_ int sd_journal_enumerate_data(sd_journal *j, const void **data, size_t
         if (le_hash != o->data.hash)
                 return -EBADMSG;
 
-        r = return_data(f, o, data, size);
+        r = return_data(j, f, o, data, size);
         if (r < 0)
                 return r;
 
@@ -2339,7 +2343,7 @@ _public_ int sd_journal_enumerate_unique(sd_journal *j, const void **data, size_
                 if (o->object.type != OBJECT_DATA)
                         return -EBADMSG;
 
-                r = return_data(j->unique_file, o, &odata, &ol);
+                r = return_data(j, j->unique_file, o, &odata, &ol);
                 if (r < 0)
                         return r;
 
@@ -2371,7 +2375,7 @@ _public_ int sd_journal_enumerate_unique(sd_journal *j, const void **data, size_
                 if (found)
                         continue;
 
-                r = return_data(j->unique_file, o, data, l);
+                r = return_data(j, j->unique_file, o, data, l);
                 if (r < 0)
                         return r;
 
@@ -2456,3 +2460,21 @@ _public_ int sd_journal_get_catalog_for_message_id(sd_id128_t id, char **ret) {
 
         return catalog_get(id, ret);
 }
+
+_public_ int sd_journal_set_data_threshold(sd_journal *j, size_t sz) {
+        if (!j)
+                return -EINVAL;
+
+        j->data_threshold = sz;
+        return 0;
+}
+
+_public_ int sd_journal_get_data_threshold(sd_journal *j, size_t *sz) {
+        if (!j)
+                return -EINVAL;
+        if (!sz)
+                return -EINVAL;
+
+        *sz = j->data_threshold;
+        return 0;
+}
diff --git a/src/shared/logs-show.c b/src/shared/logs-show.c
index cb93761bd1..ca5ad43b63 100644
--- a/src/shared/logs-show.c
+++ b/src/shared/logs-show.c
@@ -91,7 +91,7 @@ static bool shall_print(const char *p, size_t l, OutputFlags flags) {
         if (flags & OUTPUT_SHOW_ALL)
                 return true;
 
-        if (l > PRINT_THRESHOLD)
+        if (l >= PRINT_THRESHOLD)
                 return false;
 
         if (!utf8_is_printable_n(p, l))
@@ -119,6 +119,8 @@ static int output_short(
         assert(f);
         assert(j);
 
+        sd_journal_set_data_threshold(j, flags & OUTPUT_SHOW_ALL ? 0 : PRINT_THRESHOLD);
+
         SD_JOURNAL_FOREACH_DATA(j, data, length) {
 
                 r = parse_field(data, length, "PRIORITY=", &priority, &priority_len);
@@ -308,6 +310,8 @@ static int output_verbose(
         assert(f);
         assert(j);
 
+        sd_journal_set_data_threshold(j, 0);
+
         r = sd_journal_get_realtime_usec(j, &realtime);
         if (r < 0) {
                 log_error("Failed to get realtime timestamp: %s", strerror(-r));
@@ -368,6 +372,8 @@ static int output_export(
 
         assert(j);
 
+        sd_journal_set_data_threshold(j, 0);
+
         r = sd_journal_get_realtime_usec(j, &realtime);
         if (r < 0) {
                 log_error("Failed to get realtime timestamp: %s", strerror(-r));
@@ -441,7 +447,7 @@ void json_escape(
         assert(f);
         assert(p);
 
-        if (!(flags & OUTPUT_SHOW_ALL) && l > JSON_THRESHOLD)
+        if (!(flags & OUTPUT_SHOW_ALL) && l >= JSON_THRESHOLD)
 
                 fputs("null", f);
 
@@ -502,6 +508,8 @@ static int output_json(
 
         assert(j);
 
+        sd_journal_set_data_threshold(j, flags & OUTPUT_SHOW_ALL ? 0 : JSON_THRESHOLD);
+
         r = sd_journal_get_realtime_usec(j, &realtime);
         if (r < 0) {
                 log_error("Failed to get realtime timestamp: %s", strerror(-r));
@@ -714,6 +722,8 @@ static int output_cat(
         assert(j);
         assert(f);
 
+        sd_journal_set_data_threshold(j, 0);
+
         r = sd_journal_get_data(j, "MESSAGE", &data, &l);
         if (r < 0) {
                 /* An entry without MESSAGE=? */
diff --git a/src/systemd/sd-journal.h b/src/systemd/sd-journal.h
index fd9c0f5620..2e8d2d882f 100644
--- a/src/systemd/sd-journal.h
+++ b/src/systemd/sd-journal.h
@@ -97,6 +97,9 @@ int sd_journal_next_skip(sd_journal *j, uint64_t skip);
 int sd_journal_get_realtime_usec(sd_journal *j, uint64_t *ret);
 int sd_journal_get_monotonic_usec(sd_journal *j, uint64_t *ret, sd_id128_t *ret_boot_id);
 
+int sd_journal_set_data_threshold(sd_journal *j, size_t sz);
+int sd_journal_get_data_threshold(sd_journal *j, size_t *sz);
+
 int sd_journal_get_data(sd_journal *j, const char *field, const void **data, size_t *l);
 int sd_journal_enumerate_data(sd_journal *j, const void **data, size_t *l);
 void sd_journal_restart_data(sd_journal *j);