28284 firstnsanci/pr28284

Signed-off-by: Noah Sanci <nsanci@redhat.com>

4 files changed, 210 insertions, 37 deletions

diff --git a/debuginfod/debuginfod-client.c b/debuginfod/debuginfod-client.c
index 88e45567..e01a11e1 100644
--- a/debuginfod/debuginfod-client.c
+++ b/debuginfod/debuginfod-client.c
@@ -128,6 +128,7 @@ struct debuginfod_client
      handle data, etc. So those don't have to be reparsed and
      recreated on each request.  */
   char * winning_headers;
+  size_t winning_size;
 };
 
 /* The cache_clean_interval_s file within the debuginfod cache specifies
@@ -187,6 +188,10 @@ struct handle_data
   /* Response http headers for this client handle, sent from the server */
   char *response_data;
   size_t response_data_size;
+
+  /* synonymous with 'describe', but used in the context of the 
+   * http response. */
+  char head_request;
 };
 
 static size_t
@@ -504,8 +509,7 @@ default_progressfn (debuginfod_client *c, long a, long b)
 
 /* This is a callback function that receives http response headers in buffer for use
  * in this program. https://curl.se/libcurl/c/CURLOPT_HEADERFUNCTION.html is the
- * online documentation.
- */
+ * online documentation.  */
 static size_t
 header_callback (char * buffer, size_t size, size_t numitems, void * userdata)
 {
@@ -514,6 +518,22 @@ header_callback (char * buffer, size_t size, size_t numitems, void * userdata)
   /* Temporary buffer for realloc */
   char *temp = NULL;
   struct handle_data *data = (struct handle_data *) userdata;
+  curl_off_t resp;
+  curl_easy_getinfo(data->handle, CURLINFO_RESPONSE_CODE, &resp);
+  if (*data->target_handle == NULL && 
+      (resp == 200 || resp == 406) && 
+      data->head_request)
+    {
+      *data->target_handle = data->handle;
+      /* update the client object */
+      const char *url = NULL;
+      (void) curl_easy_getinfo (data->handle, CURLINFO_EFFECTIVE_URL, &url);
+      if (url)
+        {
+          free (data->client->url);
+          data->client->url = strdup(url); /* ok if fails */
+        }
+    }
   if (data->response_data == NULL)
     {
       temp = malloc(numitems+1);
@@ -544,6 +564,7 @@ debuginfod_query_server (debuginfod_client *c,
 			 const unsigned char *build_id,
                          int build_id_len,
                          const char *type,
+                         const char describe,
                          const char *filename,
                          char **path)
 {
@@ -748,7 +769,10 @@ debuginfod_query_server (debuginfod_client *c,
     }
 
   xalloc_str (target_cache_dir, "%s/%s", cache_path, build_id_bytes);
-  xalloc_str (target_cache_path, "%s/%s%s", target_cache_dir, type, suffix);
+  if (describe)
+    xalloc_str (target_cache_path, "%s/%s_describe%s", target_cache_dir, type, suffix);
+  else
+    xalloc_str (target_cache_path, "%s/%s%s", target_cache_dir, type, suffix);
   xalloc_str (target_cache_tmppath, "%s.XXXXXX", target_cache_path);
 
   /* XXX combine these */
@@ -773,6 +797,26 @@ debuginfod_query_server (debuginfod_client *c,
       /* Success!!!! */
       if (path != NULL)
         *path = strdup(target_cache_path);
+      if (describe)
+        {
+          FILE * cachefp = fopen(target_cache_path, "r");
+          if (cachefp == NULL)
+            fprintf(stderr, "Could not open cache: %s, for reading",target_cache_path);
+          else
+            {
+              fseek(cachefp, 0, SEEK_END);
+              size_t cachelen = ftell(cachefp);
+              fseek(cachefp, 0, SEEK_SET);
+              char *str = malloc(cachelen+1);
+              memset(str, '\0', cachelen+1);
+              size_t read = fread(str, 1, cachelen, cachefp);
+              if (read != cachelen)
+                fprintf(stderr, "Could not properly read file\n");
+              fclose(cachefp);
+              fprintf(stderr, "Describe cache content:\n%s \n", str);
+              free(str);
+            }
+        }
       rc = fd;
       goto out;
     }
@@ -972,6 +1016,11 @@ debuginfod_query_server (debuginfod_client *c,
         snprintf(data[i].url, PATH_MAX, "%s/%s/%s", server_url, build_id_bytes, type);
       if (vfd >= 0)
 	dprintf (vfd, "url %d %s\n", i, data[i].url);
+      data[i].head_request = describe;
+      /* PR28284 when describing simply send HEAD request
+       * (done with CURLOPT_NOBODY) */
+      if (describe)
+        curl_easy_setopt(data[i].handle, CURLOPT_NOBODY, 1);
 
       curl_easy_setopt(data[i].handle, CURLOPT_URL, data[i].url);
       if (vfd >= 0)
@@ -1019,6 +1068,8 @@ debuginfod_query_server (debuginfod_client *c,
   int committed_to = -1;
   bool verbose_reported = false;
   struct timespec start_time, cur_time;
+  /* PR28284 saved later if describing to remember what server has the file */
+  char *committed_url;
   c->winning_headers = NULL;
   if ( maxtime > 0 && clock_gettime(CLOCK_MONOTONIC_RAW, &start_time) == -1)
     {
@@ -1048,6 +1099,7 @@ debuginfod_query_server (debuginfod_client *c,
       curl_multi_wait(curlm, NULL, 0, 1000, NULL);
       CURLMcode curlm_res = curl_multi_perform(curlm, &still_running);
 
+
       /* If the target file has been found, abort the other queries.  */
       if (target_handle != NULL)
 	{
@@ -1058,15 +1110,18 @@ debuginfod_query_server (debuginfod_client *c,
               {
 	        committed_to = i;
                 if (c->winning_headers == NULL)
-                  {
-                    c->winning_headers = data[committed_to].response_data;
-                    if (vfd >= 0 && c->winning_headers != NULL)
-                      dprintf(vfd, "\n%s", c->winning_headers);
-                    data[committed_to].response_data = NULL;
+                    {
+                      c->winning_headers = data[i].response_data;
+                      c->winning_size = data[i].response_data_size;
+                      /* PR28284 if verbose or describing and winning headers exist
+                       * print them. */
+                      if ((vfd >= 0 || describe) && c->winning_headers != NULL)
+                        dprintf(vfd, "%s\n", c->winning_headers);
+                      data[i].response_data = NULL;
                   }
-
               }
-	}
+
+        }
 
       if (vfd >= 0 && !verbose_reported && committed_to >= 0)
 	{
@@ -1077,6 +1132,8 @@ debuginfod_query_server (debuginfod_client *c,
 	    c->default_progressfn_printed_p = 0;
 	  verbose_reported = true;
 	}
+      if (committed_to >=0 && data[committed_to].url)
+        xalloc_str(committed_url, "Retrieved from: %s\n", data[committed_to].url);
 
       if (curlm_res != CURLM_OK)
         {
@@ -1347,7 +1404,23 @@ debuginfod_query_server (debuginfod_client *c,
       goto out2;
       /* Perhaps we need not give up right away; could retry or something ... */
     }
-
+  if (describe)
+    {
+      /* PR28284 write info to cache file */
+      FILE *cache_file = fdopen(fd, "w");
+      if (cache_file == NULL)
+        {
+          rc = -errno;
+          goto out2;
+        }
+      if (c->winning_headers && c->winning_size)
+        fwrite(c->winning_headers, c->winning_size, 1, cache_file);
+      else dprintf(vfd, "Could not write headers to cache");
+      if (committed_url)
+        fwrite(committed_url, strlen(committed_url), 1, cache_file);
+      else dprintf(vfd, "Could not urls to cache");
+      fclose(cache_file);
+    }
   /* remove all handles from multi */
   for (int i = 0; i < num_urls; i++)
     {
@@ -1380,6 +1453,7 @@ debuginfod_query_server (debuginfod_client *c,
       free (data[i].response_data);
     }
 
+  free(committed_url);
   unlink (target_cache_tmppath);
   close (fd); /* before the rmdir, otherwise it'll fail */
   (void) rmdir (target_cache_dir); /* nop if not empty */
@@ -1407,7 +1481,7 @@ debuginfod_query_server (debuginfod_client *c,
   if (c->default_progressfn_printed_p)
     dprintf(STDERR_FILENO, "\n");
 
-  if (vfd >= 0)
+  if (vfd >= 0 && !describe)
     {
       if (rc < 0)
 	dprintf (vfd, "not found %s (err=%d)\n", strerror (-rc), rc);
@@ -1500,7 +1574,7 @@ debuginfod_find_debuginfo (debuginfod_client *client,
                            char **path)
 {
   return debuginfod_query_server(client, build_id, build_id_len,
-                                 "debuginfo", NULL, path);
+                                 "debuginfo", 0, NULL, path);
 }
 
 
@@ -1511,7 +1585,7 @@ debuginfod_find_executable(debuginfod_client *client,
                            char **path)
 {
   return debuginfod_query_server(client, build_id, build_id_len,
-                                 "executable", NULL, path);
+                                 "executable", 0, NULL, path);
 }
 
 /* See debuginfod.h  */
@@ -1520,10 +1594,38 @@ int debuginfod_find_source(debuginfod_client *client,
                            const char *filename, char **path)
 {
   return debuginfod_query_server(client, build_id, build_id_len,
-                                 "source", filename, path);
+                                 "source", 0, filename, path);
+}
+
+/*  See debuginfod.h */
+int
+debuginfod_find_describe_debuginfo (debuginfod_client *client,
+			   const unsigned char *build_id, int build_id_len,
+                           char **path)
+{
+  return debuginfod_query_server(client, build_id, build_id_len,
+                                 "debuginfo", 1, NULL, path);
 }
 
 
+/* See debuginfod.h  */
+int
+debuginfod_find_describe_executable(debuginfod_client *client,
+			   const unsigned char *build_id, int build_id_len,
+                           char **path)
+{
+  return debuginfod_query_server(client, build_id, build_id_len,
+                                 "executable", 1, NULL, path);
+}
+
+/* See debuginfod.h  */
+int debuginfod_find_describe_source(debuginfod_client *client,
+			   const unsigned char *build_id, int build_id_len,
+                           const char *filename, char **path)
+{
+  return debuginfod_query_server(client, build_id, build_id_len,
+                                 "source", 1, filename, path);
+}
 /* Add an outgoing HTTP header.  */
 int debuginfod_add_http_header (debuginfod_client *client, const char* header)
 {
diff --git a/debuginfod/debuginfod-find.c b/debuginfod/debuginfod-find.c
index 3e8ab203..cee8d689 100644
--- a/debuginfod/debuginfod-find.c
+++ b/debuginfod/debuginfod-find.c
@@ -48,7 +48,11 @@ static const char args_doc[] = N_("debuginfo BUILDID\n"
                                   "executable BUILDID\n"
                                   "executable PATH\n"
                                   "source BUILDID /FILENAME\n"
-                                  "source PATH /FILENAME\n");
+                                  "source PATH /FILENAME\n"
+                                  "describe BUILDID ARTIFACT\n"
+                                  "describe PATH ARTIFACT\n"
+                                  "describe BUILDID source /FILENAME\n"
+                                  "describe PATH source /FILENAME\n");
 
 
 /* Definitions of arguments for argp functions.  */
@@ -138,7 +142,6 @@ main(int argc, char** argv)
       argp_help (&argp, stderr, ARGP_HELP_USAGE, argv[0]);
       return 1;
     }
-
   /* If we were passed an ELF file name in the BUILDID slot, look in there. */
   unsigned char* build_id = (unsigned char*) argv[remaining+1];
   int build_id_len = 0; /* assume text */
@@ -182,7 +185,7 @@ main(int argc, char** argv)
         fprintf (stderr, "Cannot extract build-id from %s: %s\n", build_id, elf_errmsg(-1));
     }
 
-  char *cache_name;
+  char *cache_name = NULL;
   int rc = 0;
 
   /* Check whether FILETYPE is valid and call the appropriate
@@ -207,6 +210,34 @@ main(int argc, char** argv)
                                   build_id, build_id_len,
 				  argv[remaining+2], &cache_name);
     }
+  else if (strcmp(argv[remaining], "describe") == 0)
+    {
+      if (remaining+2 == argc)
+        {
+          fprintf(stderr, "Needs ARTIFACT. Options: executable, debuginfo, and source");
+          return 1;
+        }
+      /* Check the artifact slots of argv */
+      else if (strcmp(argv[remaining+2], "debuginfo") == 0)
+        rc = debuginfod_find_describe_debuginfo(client,
+                                             build_id, build_id_len,
+                                             &cache_name);
+      else if (strcmp(argv[remaining+2], "executable") == 0)
+        rc = debuginfod_find_describe_executable(client,
+                                                 build_id, build_id_len,
+                                                 &cache_name);
+      else if (strcmp(argv[remaining+2], "source") == 0)
+        {
+          if (remaining+3 == argc || argv[remaining+3][0] != '/')
+            {
+              fprintf(stderr, "If FILETYPE is \"source\" then absolute /FILENAME must be given\n");
+              return 1;
+            }
+          rc = debuginfod_find_describe_source(client,
+                                  build_id, build_id_len,
+				  argv[remaining+3], &cache_name);
+        }
+    }
   else
     {
       argp_help (&argp, stderr, ARGP_HELP_USAGE, argv[0]);
@@ -232,7 +263,11 @@ main(int argc, char** argv)
       return 1;
     }
 
-  printf("%s\n", cache_name);
+  // PR28284 Previously, this could print junk. Now it only prints
+  // when the cache is present
+  if (cache_name)
+    printf("%s\n", cache_name);
+
   free (cache_name);
 
   return 0;
diff --git a/debuginfod/debuginfod.cxx b/debuginfod/debuginfod.cxx
index 2b9a1c41..88c56927 100644
--- a/debuginfod/debuginfod.cxx
+++ b/debuginfod/debuginfod.cxx
@@ -1047,6 +1047,7 @@ add_mhd_last_modified (struct MHD_Response *resp, time_t mtime)
 
 static struct MHD_Response*
 handle_buildid_f_match (bool internal_req_t,
+                        bool describe,
                         int64_t b_mtime,
                         const string& b_source0,
                         int *result_fd)
@@ -1075,21 +1076,26 @@ handle_buildid_f_match (bool internal_req_t,
       close(fd);
       return 0;
     }
-
+  struct MHD_Response* r;
+  if (describe)
+    r = MHD_create_response_from_buffer (0, (void*)"", MHD_RESPMEM_PERSISTENT);
+  /* create a response with the file's body */
+  else
+    r = MHD_create_response_from_fd ((uint64_t) s.st_size, fd);
   inc_metric ("http_responses_total","result","file");
-  struct MHD_Response* r = MHD_create_response_from_fd ((uint64_t) s.st_size, fd);
+
   if (r == 0)
     {
       if (verbose)
-        obatched(clog) << "cannot create fd-response for " << b_source0 << endl;
+        obatched(clog) << "cannot create response for " << b_source0 << endl;
       close(fd);
     }
   else
     {
       MHD_add_response_header (r, "Content-Type", "application/octet-stream");
       std::string file = b_source0.substr(b_source0.find_last_of("/")+1, b_source0.length());
+      MHD_add_response_header (r, "X-DEBUGINFOD-FILE", file.c_str());
       MHD_add_response_header (r, "X-DEBUGINFOD-SIZE", to_string(s.st_size).c_str() );
-      MHD_add_response_header (r, "X-DEBUGINFOD-FILE", file.c_str() );
       add_mhd_last_modified (r, s.st_mtime);
       if (verbose > 1)
         obatched(clog) << "serving file " << b_source0 << endl;
@@ -1516,6 +1522,7 @@ string canonicalized_archive_entry_pathname(struct archive_entry *e)
 
 static struct MHD_Response*
 handle_buildid_r_match (bool internal_req_p,
+                        bool describe,
                         int64_t b_mtime,
                         const string& b_source0,
                         const string& b_source1,
@@ -1546,12 +1553,15 @@ handle_buildid_r_match (bool internal_req_p,
           fdcache.clear(b_source0, b_source1);
           break; // branch out of if "loop", to try new libarchive fetch attempt
         }
-
-      struct MHD_Response* r = MHD_create_response_from_fd (fs.st_size, fd);
+      struct MHD_Response* r;
+      if (describe)
+        r = MHD_create_response_from_buffer (0, (void*)"", MHD_RESPMEM_PERSISTENT);
+      else
+        r = MHD_create_response_from_fd (fs.st_size, fd);
       if (r == 0)
         {
           if (verbose)
-            obatched(clog) << "cannot create fd-response for " << b_source0 << endl;
+            obatched(clog) << "cannot create response for " << b_source0 << endl;
           close(fd);
           break; // branch out of if "loop", to try new libarchive fetch attempt
         }
@@ -1559,9 +1569,9 @@ handle_buildid_r_match (bool internal_req_p,
       inc_metric ("http_responses_total","result","archive fdcache");
 
       MHD_add_response_header (r, "Content-Type", "application/octet-stream");
+      MHD_add_response_header (r, "X-DEBUGINFOD-FILE", b_source1.c_str());
       MHD_add_response_header (r, "X-DEBUGINFOD-SIZE", to_string(fs.st_size).c_str());
       MHD_add_response_header (r, "X-DEBUGINFOD-ARCHIVE", b_source0.c_str());
-      MHD_add_response_header (r, "X-DEBUGINFOD-FILE", b_source1.c_str());
       add_mhd_last_modified (r, fs.st_mtime);
       if (verbose > 1)
         obatched(clog) << "serving fdcache archive " << b_source0 << " file " << b_source1 << endl;
@@ -1725,6 +1735,7 @@ handle_buildid_r_match (bool internal_req_p,
 
 static struct MHD_Response*
 handle_buildid_match (bool internal_req_p,
+                      bool describe,
                       int64_t b_mtime,
                       const string& b_stype,
                       const string& b_source0,
@@ -1734,9 +1745,9 @@ handle_buildid_match (bool internal_req_p,
   try
     {
       if (b_stype == "F")
-        return handle_buildid_f_match(internal_req_p, b_mtime, b_source0, result_fd);
+        return handle_buildid_f_match(internal_req_p, describe, b_mtime, b_source0, result_fd);
       else if (b_stype == "R")
-        return handle_buildid_r_match(internal_req_p, b_mtime, b_source0, b_source1, result_fd);
+        return handle_buildid_r_match(internal_req_p, describe, b_mtime, b_source0, b_source1, result_fd);
     }
   catch (const reportable_exception &e)
     {
@@ -1803,6 +1814,7 @@ static struct MHD_Response*
 handle_buildid (MHD_Connection* conn,
                 const string& buildid /* unsafe */,
                 const string& artifacttype /* unsafe */,
+                const bool describe,
                 const string& suffix /* unsafe */,
                 int *result_fd)
 {
@@ -1813,7 +1825,7 @@ handle_buildid (MHD_Connection* conn,
   else if (artifacttype == "source") atype_code = "S";
   else throw reportable_exception("invalid artifacttype");
 
-  inc_metric("http_requests_total", "type", artifacttype);
+  inc_metric("http_requests_total", "type", artifacttype /*, "method", describe ? "HEAD" : "GET" */);
   
   if (atype_code == "S" && suffix == "")
      throw reportable_exception("invalid source suffix");
@@ -1883,11 +1895,11 @@ handle_buildid (MHD_Connection* conn,
 
       if (verbose > 1)
         obatched(clog) << "found mtime=" << b_mtime << " stype=" << b_stype
-             << " source0=" << b_source0 << " source1=" << b_source1 << endl;
+             << " source0=" << b_source0 << " source1=" << b_source1 << " describe="<< describe << endl;
 
       // Try accessing the located match.
       // XXX: in case of multiple matches, attempt them in parallel?
-      auto r = handle_buildid_match (conn ? false : true,
+      auto r = handle_buildid_match (conn ? false : true, describe,
                                      b_mtime, b_stype, b_source0, b_source1, result_fd);
       if (r)
         return r;
@@ -1956,6 +1968,7 @@ and will not query the upstream servers");
 				     (const unsigned char*) buildid.c_str(),
 				     0, suffix.c_str(), NULL);
     }
+
   else
     fd = -errno; /* Set by debuginfod_begin.  */
   debuginfod_pool_end (client);
@@ -1967,7 +1980,12 @@ and will not query the upstream servers");
       int rc = fstat (fd, &s);
       if (rc == 0)
         {
-          auto r = MHD_create_response_from_fd ((uint64_t) s.st_size, fd);
+          struct MHD_Response *r = 0;
+          if (describe)
+            r = MHD_create_response_from_buffer(0, (void*)"", MHD_RESPMEM_PERSISTENT);
+          else
+            r = MHD_create_response_from_fd ((uint64_t) s.st_size, fd);
+
           if (r)
             {
               MHD_add_response_header (r, "Content-Type", "application/octet-stream");
@@ -2165,11 +2183,12 @@ handler_cb (void * /*cls*/,
   struct timespec ts_start, ts_end;
   clock_gettime (CLOCK_MONOTONIC, &ts_start);
   double afteryou = 0.0;
+  bool describe = (string(method) == "HEAD");
 
   try
     {
-      if (string(method) != "GET")
-        throw reportable_exception(400, "we support GET only");
+      if (string(method) != "GET" && string(method) != "HEAD")
+        throw reportable_exception(400, "we only support GET and HEAD.");
 
       /* Start decoding the URL. */
       size_t slash1 = url_copy.find('/', 1);
@@ -2217,7 +2236,7 @@ handler_cb (void * /*cls*/,
 
           // get the resulting fd so we can report its size
           int fd;
-          r = handle_buildid(connection, buildid, artifacttype, suffix, &fd);
+          r = handle_buildid(connection, buildid, artifacttype, describe, suffix, &fd);
           if (r)
             {
               struct stat fs;
@@ -2250,6 +2269,7 @@ handler_cb (void * /*cls*/,
         }
 
       rc = MHD_queue_response (connection, MHD_HTTP_OK, r);
+      obatched(clog) << "describe" << describe<< "response "<<r<< " connection "<<connection<<endl;
       http_code = MHD_HTTP_OK;
       MHD_destroy_response (r);
     }
@@ -2334,7 +2354,7 @@ dwarf_extract_source_paths (Elf *elf, set<string>& debug_sourcefiles)
           struct MHD_Response *r = 0;
           try
             {
-              r = handle_buildid (0, buildid, "debuginfo", "", &alt_fd);
+              r = handle_buildid (0, buildid, "debuginfo", false, "", &alt_fd);
             }
           catch (const reportable_exception& e)
             {
diff --git a/debuginfod/debuginfod.h.in b/debuginfod/debuginfod.h.in
index c358df4d..4f57efff 100644
--- a/debuginfod/debuginfod.h.in
+++ b/debuginfod/debuginfod.h.in
@@ -78,6 +78,22 @@ int debuginfod_find_source (debuginfod_client *client,
                             const char *filename,
                             char **path);
 
+int debuginfod_find_describe_debuginfo (debuginfod_client *client,
+			       const unsigned char *build_id,
+                               int build_id_len,
+                               char **path);
+
+int debuginfod_find_describe_executable (debuginfod_client *client,
+				const unsigned char *build_id,
+                                int build_id_len,
+                                char **path);
+
+int debuginfod_find_describe_source (debuginfod_client *client,
+			    const unsigned char *build_id,
+                            int build_id_len,
+                            const char *filename,
+                            char **path);
+
 typedef int (*debuginfod_progressfn_t)(debuginfod_client *c, long a, long b);
 void debuginfod_set_progressfn(debuginfod_client *c,
 			       debuginfod_progressfn_t fn);