[pacman-dev,v3] Introduce alpm_dbs_update() function for parallel db updates

Message ID 20200326201959.498251-1-anatol.pomozov@gmail.com
State Accepted
Headers show
Series
  • [pacman-dev,v3] Introduce alpm_dbs_update() function for parallel db updates
Related show

Commit Message

Anatol Pomozov March 26, 2020, 8:19 p.m. UTC
This is an equivalent of alpm_db_update but for multiplexed (parallel)
download. The difference is that this function accepts list of
databases to update. And then ALPM internals download it in parallel if
possible.

Add a stub for _alpm_multi_download the function that will do parallel
payloads downloads in the future.

Introduce dload_payload->filepath field that contains url path to the
file we download. It is like fileurl field but does not contain
protocol/server part. The rationale for having this field is that with
the curl multidownload the server retry logic is going to move to a curl
callback. And the callback needs to be able to reconstruct the 'next'
fileurl. One will be able to do it by getting the next server url from
'servers' list and then concat with filepath. Once the 'parallel download'
refactoring is over 'fileurl' field will go away.

Signed-off-by: Anatol Pomozov <anatol.pomozov@gmail.com>
---
 lib/libalpm/alpm.h    |  29 ++++++++++
 lib/libalpm/be_sync.c | 128 ++++++++++++++++++++++++++++++++++++++++++
 lib/libalpm/dload.c   |  12 ++++
 lib/libalpm/dload.h   |   5 ++
 4 files changed, 174 insertions(+)

Comments

Anatol Pomozov April 13, 2020, 6:12 a.m. UTC | #1
Hi Allan and others

Is there a chance you can look at this and other "parallel-download" patches?

On Thu, Mar 26, 2020 at 1:20 PM Anatol Pomozov <anatol.pomozov@gmail.com> wrote:
>
> This is an equivalent of alpm_db_update but for multiplexed (parallel)
> download. The difference is that this function accepts list of
> databases to update. And then ALPM internals download it in parallel if
> possible.
>
> Add a stub for _alpm_multi_download the function that will do parallel
> payloads downloads in the future.
>
> Introduce dload_payload->filepath field that contains url path to the
> file we download. It is like fileurl field but does not contain
> protocol/server part. The rationale for having this field is that with
> the curl multidownload the server retry logic is going to move to a curl
> callback. And the callback needs to be able to reconstruct the 'next'
> fileurl. One will be able to do it by getting the next server url from
> 'servers' list and then concat with filepath. Once the 'parallel download'
> refactoring is over 'fileurl' field will go away.
>
> Signed-off-by: Anatol Pomozov <anatol.pomozov@gmail.com>
> ---
>  lib/libalpm/alpm.h    |  29 ++++++++++
>  lib/libalpm/be_sync.c | 128 ++++++++++++++++++++++++++++++++++++++++++
>  lib/libalpm/dload.c   |  12 ++++
>  lib/libalpm/dload.h   |   5 ++
>  4 files changed, 174 insertions(+)
>
> diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h
> index 5d559db1..2cf20343 100644
> --- a/lib/libalpm/alpm.h
> +++ b/lib/libalpm/alpm.h
> @@ -1049,6 +1049,35 @@ int alpm_db_remove_server(alpm_db_t *db, const char *url);
>   */
>  int alpm_db_update(int force, alpm_db_t *db);
>
> +/** Update package databases
> + *
> + * An update of the package databases in the list \a dbs will be attempted.
> + * Unless \a force is true, the update will only be performed if the remote
> + * databases were modified since the last update.
> + *
> + * This operation requires a database lock, and will return an applicable error
> + * if the lock could not be obtained.
> + *
> + * Example:
> + * @code
> + * alpm_list_t *dbs = alpm_get_syncdbs();
> + * ret = alpm_dbs_update(config->handle, dbs, force);
> + * if(ret < 0) {
> + *     pm_printf(ALPM_LOG_ERROR, _("failed to synchronize all databases (%s)\n"),
> + *         alpm_strerror(alpm_errno(config->handle)));
> + * }
> + * @endcode
> + *
> + * @note After a successful update, the \link alpm_db_get_pkgcache()
> + * package cache \endlink will be invalidated
> + * @param handle the context handle
> + * @param dbs list of package databases to update
> + * @param force if true, then forces the update, otherwise update only in case
> + * the databases aren't up to date
> + * @return 0 on success, -1 on error (pm_errno is set accordingly)
> + */
> +int alpm_dbs_update(alpm_handle_t *handle, alpm_list_t *dbs, int force);
> +
>  /** Get a package entry from a package database.
>   * @param db pointer to the package database to get the package from
>   * @param name of the package
> diff --git a/lib/libalpm/be_sync.c b/lib/libalpm/be_sync.c
> index aafed15d..7ab52301 100644
> --- a/lib/libalpm/be_sync.c
> +++ b/lib/libalpm/be_sync.c
> @@ -301,6 +301,134 @@ int SYMEXPORT alpm_db_update(int force, alpm_db_t *db)
>         return ret;
>  }
>
> +int SYMEXPORT alpm_dbs_update(alpm_handle_t *handle, alpm_list_t *dbs, int force) {
> +       char *syncpath;
> +       const char *dbext = handle->dbext;
> +       alpm_list_t *i;
> +       int ret = -1;
> +       mode_t oldmask;
> +       alpm_list_t *payloads = NULL;
> +
> +       /* Sanity checks */
> +       CHECK_HANDLE(handle, return -1);
> +       ASSERT(dbs != NULL, return -1);
> +       handle->pm_errno = ALPM_ERR_OK;
> +
> +       syncpath = get_sync_dir(handle);
> +       ASSERT(syncpath != NULL, return -1);
> +
> +       /* make sure we have a sane umask */
> +       oldmask = umask(0022);
> +
> +       /* attempt to grab a lock */
> +       if(_alpm_handle_lock(handle)) {
> +               GOTO_ERR(handle, ALPM_ERR_HANDLE_LOCK, cleanup);
> +       }
> +
> +       for(i = dbs; i; i = i->next) {
> +               alpm_db_t *db = i->data;
> +               int dbforce = force;
> +               struct dload_payload *payload = NULL;
> +               size_t len;
> +               int siglevel;
> +
> +               if(!(db->usage & ALPM_DB_USAGE_SYNC)) {
> +                       continue;
> +               }
> +
> +               ASSERT(db != handle->db_local, GOTO_ERR(handle, ALPM_ERR_WRONG_ARGS, cleanup));
> +               ASSERT(db->servers != NULL, GOTO_ERR(handle, ALPM_ERR_SERVER_NONE, cleanup));
> +
> +               /* force update of invalid databases to fix potential mismatched database/signature */
> +               if(db->status & DB_STATUS_INVALID) {
> +                       dbforce = 1;
> +               }
> +
> +               CALLOC(payload, 1, sizeof(*payload), GOTO_ERR(handle, ALPM_ERR_MEMORY, cleanup));
> +
> +               /* set hard upper limit of 128 MiB */
> +               payload->max_size = 128 * 1024 * 1024;
> +               payload->servers = db->servers;
> +
> +               /* print server + filename into a buffer */
> +               len = strlen(db->treename) + strlen(dbext) + 1;
> +               MALLOC(payload->filepath, len, GOTO_ERR(handle, ALPM_ERR_MEMORY, cleanup));
> +               snprintf(payload->filepath, len, "%s%s", db->treename, dbext);
> +               payload->handle = handle;
> +               payload->force = dbforce;
> +               payload->unlink_on_fail = 1;
> +
> +               payloads = alpm_list_add(payloads, payload);
> +
> +               siglevel = alpm_db_get_siglevel(db);
> +               if(siglevel & ALPM_SIG_DATABASE) {
> +                       struct dload_payload *sig_payload;
> +                       CALLOC(sig_payload, 1, sizeof(*sig_payload), GOTO_ERR(handle, ALPM_ERR_MEMORY, cleanup));
> +
> +                       /* print filename into a buffer (leave space for separator and .sig) */
> +                       len = strlen(db->treename) + strlen(dbext) + 5;
> +                       MALLOC(sig_payload->filepath, len, GOTO_ERR(handle, ALPM_ERR_MEMORY, cleanup));
> +                       snprintf(sig_payload->filepath, len, "%s%s.sig", db->treename, dbext);
> +
> +                       sig_payload->handle = handle;
> +                       sig_payload->force = 1;
> +                       sig_payload->errors_ok = (siglevel & ALPM_SIG_DATABASE_OPTIONAL);
> +
> +                       /* set hard upper limit of 16 KiB */
> +                       sig_payload->max_size = 16 * 1024;
> +                       sig_payload->servers = db->servers;
> +
> +                       payloads = alpm_list_add(payloads, sig_payload);
> +               }
> +       }
> +
> +       ret = _alpm_multi_download(handle, payloads, syncpath);
> +       if(ret < 0) {
> +               goto cleanup;
> +       }
> +
> +       for(i = dbs; i; i = i->next) {
> +               alpm_db_t *db = i->data;
> +               if(!(db->usage & ALPM_DB_USAGE_SYNC)) {
> +                       continue;
> +               }
> +
> +               /* Cache needs to be rebuilt */
> +               _alpm_db_free_pkgcache(db);
> +
> +               /* clear all status flags regarding validity/existence */
> +               db->status &= ~DB_STATUS_VALID;
> +               db->status &= ~DB_STATUS_INVALID;
> +               db->status &= ~DB_STATUS_EXISTS;
> +               db->status &= ~DB_STATUS_MISSING;
> +
> +               /* if the download failed skip validation to preserve the download error */
> +               if(sync_db_validate(db) != 0) {
> +                       /* pm_errno should be set */
> +                       ret = -1;
> +               }
> +       }
> +
> +cleanup:
> +       _alpm_handle_unlock(handle);
> +
> +       if(ret == -1) {
> +               /* pm_errno was set by the download code */
> +               _alpm_log(handle, ALPM_LOG_DEBUG, "failed to sync dbs: %s\n",
> +                               alpm_strerror(handle->pm_errno));
> +       } else {
> +               handle->pm_errno = ALPM_ERR_OK;
> +       }
> +
> +       if(payloads) {
> +               alpm_list_free_inner(payloads, (alpm_list_fn_free)_alpm_dload_payload_reset);
> +               FREELIST(payloads);
> +       }
> +       free(syncpath);
> +       umask(oldmask);
> +       return ret;
> +}
> +
>  /* Forward decl so I don't reorganize the whole file right now */
>  static int sync_db_read(alpm_db_t *db, struct archive *archive,
>                 struct archive_entry *entry, alpm_pkg_t **likely_pkg);
> diff --git a/lib/libalpm/dload.c b/lib/libalpm/dload.c
> index 670da03d..7cd3e3a4 100644
> --- a/lib/libalpm/dload.c
> +++ b/lib/libalpm/dload.c
> @@ -636,6 +636,16 @@ int _alpm_download(struct dload_payload *payload, const char *localpath,
>         }
>  }
>
> +int _alpm_multi_download(alpm_handle_t *handle,
> +               alpm_list_t *payloads /* struct dload_payload */,
> +               const char *localpath)
> +{
> +       (void)handle;
> +       (void)payloads;
> +       (void)localpath;
> +       return 0;
> +}
> +
>  static char *filecache_find_url(alpm_handle_t *handle, const char *url)
>  {
>         const char *filebase = strrchr(url, '/');
> @@ -738,6 +748,7 @@ void _alpm_dload_payload_reset(struct dload_payload *payload)
>         FREE(payload->destfile_name);
>         FREE(payload->content_disp_name);
>         FREE(payload->fileurl);
> +       FREE(payload->filepath);
>         *payload = (struct dload_payload){0};
>  }
>
> @@ -746,6 +757,7 @@ void _alpm_dload_payload_reset_for_retry(struct dload_payload *payload)
>         ASSERT(payload, return);
>
>         FREE(payload->fileurl);
> +       FREE(payload->filepath);
>         payload->initial_size += payload->prevprogress;
>         payload->prevprogress = 0;
>         payload->unlink_on_fail = 0;
> diff --git a/lib/libalpm/dload.h b/lib/libalpm/dload.h
> index 1e8f75f3..3eb7fbe1 100644
> --- a/lib/libalpm/dload.h
> +++ b/lib/libalpm/dload.h
> @@ -31,6 +31,7 @@ struct dload_payload {
>         char *destfile_name;
>         char *content_disp_name;
>         char *fileurl;
> +       char *filepath; /* download URL path */
>         alpm_list_t *servers;
>         long respcode;
>         off_t initial_size;
> @@ -53,4 +54,8 @@ void _alpm_dload_payload_reset_for_retry(struct dload_payload *payload);
>  int _alpm_download(struct dload_payload *payload, const char *localpath,
>                 char **final_file, const char **final_url);
>
> +int _alpm_multi_download(alpm_handle_t *handle,
> +               alpm_list_t *payloads /* struct dload_payload */,
> +               const char *localpath);
> +
>  #endif /* ALPM_DLOAD_H */
> --
> 2.26.0
>
Allan McRae April 13, 2020, 6:44 a.m. UTC | #2
On 13/4/20 4:12 pm, Anatol Pomozov wrote:
> Hi Allan and others
> 
> Is there a chance you can look at this and other "parallel-download" patches?
> 

There is a 40% chance I can get to it this week...

A

Patch

diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h
index 5d559db1..2cf20343 100644
--- a/lib/libalpm/alpm.h
+++ b/lib/libalpm/alpm.h
@@ -1049,6 +1049,35 @@  int alpm_db_remove_server(alpm_db_t *db, const char *url);
  */
 int alpm_db_update(int force, alpm_db_t *db);
 
+/** Update package databases
+ *
+ * An update of the package databases in the list \a dbs will be attempted.
+ * Unless \a force is true, the update will only be performed if the remote
+ * databases were modified since the last update.
+ *
+ * This operation requires a database lock, and will return an applicable error
+ * if the lock could not be obtained.
+ *
+ * Example:
+ * @code
+ * alpm_list_t *dbs = alpm_get_syncdbs();
+ * ret = alpm_dbs_update(config->handle, dbs, force);
+ * if(ret < 0) {
+ *     pm_printf(ALPM_LOG_ERROR, _("failed to synchronize all databases (%s)\n"),
+ *         alpm_strerror(alpm_errno(config->handle)));
+ * }
+ * @endcode
+ *
+ * @note After a successful update, the \link alpm_db_get_pkgcache()
+ * package cache \endlink will be invalidated
+ * @param handle the context handle
+ * @param dbs list of package databases to update
+ * @param force if true, then forces the update, otherwise update only in case
+ * the databases aren't up to date
+ * @return 0 on success, -1 on error (pm_errno is set accordingly)
+ */
+int alpm_dbs_update(alpm_handle_t *handle, alpm_list_t *dbs, int force);
+
 /** Get a package entry from a package database.
  * @param db pointer to the package database to get the package from
  * @param name of the package
diff --git a/lib/libalpm/be_sync.c b/lib/libalpm/be_sync.c
index aafed15d..7ab52301 100644
--- a/lib/libalpm/be_sync.c
+++ b/lib/libalpm/be_sync.c
@@ -301,6 +301,134 @@  int SYMEXPORT alpm_db_update(int force, alpm_db_t *db)
 	return ret;
 }
 
+int SYMEXPORT alpm_dbs_update(alpm_handle_t *handle, alpm_list_t *dbs, int force) {
+	char *syncpath;
+	const char *dbext = handle->dbext;
+	alpm_list_t *i;
+	int ret = -1;
+	mode_t oldmask;
+	alpm_list_t *payloads = NULL;
+
+	/* Sanity checks */
+	CHECK_HANDLE(handle, return -1);
+	ASSERT(dbs != NULL, return -1);
+	handle->pm_errno = ALPM_ERR_OK;
+
+	syncpath = get_sync_dir(handle);
+	ASSERT(syncpath != NULL, return -1);
+
+	/* make sure we have a sane umask */
+	oldmask = umask(0022);
+
+	/* attempt to grab a lock */
+	if(_alpm_handle_lock(handle)) {
+		GOTO_ERR(handle, ALPM_ERR_HANDLE_LOCK, cleanup);
+	}
+
+	for(i = dbs; i; i = i->next) {
+		alpm_db_t *db = i->data;
+		int dbforce = force;
+		struct dload_payload *payload = NULL;
+		size_t len;
+		int siglevel;
+
+		if(!(db->usage & ALPM_DB_USAGE_SYNC)) {
+			continue;
+		}
+
+		ASSERT(db != handle->db_local, GOTO_ERR(handle, ALPM_ERR_WRONG_ARGS, cleanup));
+		ASSERT(db->servers != NULL, GOTO_ERR(handle, ALPM_ERR_SERVER_NONE, cleanup));
+
+		/* force update of invalid databases to fix potential mismatched database/signature */
+		if(db->status & DB_STATUS_INVALID) {
+			dbforce = 1;
+		}
+
+		CALLOC(payload, 1, sizeof(*payload), GOTO_ERR(handle, ALPM_ERR_MEMORY, cleanup));
+
+		/* set hard upper limit of 128 MiB */
+		payload->max_size = 128 * 1024 * 1024;
+		payload->servers = db->servers;
+
+		/* print server + filename into a buffer */
+		len = strlen(db->treename) + strlen(dbext) + 1;
+		MALLOC(payload->filepath, len, GOTO_ERR(handle, ALPM_ERR_MEMORY, cleanup));
+		snprintf(payload->filepath, len, "%s%s", db->treename, dbext);
+		payload->handle = handle;
+		payload->force = dbforce;
+		payload->unlink_on_fail = 1;
+
+		payloads = alpm_list_add(payloads, payload);
+
+		siglevel = alpm_db_get_siglevel(db);
+		if(siglevel & ALPM_SIG_DATABASE) {
+			struct dload_payload *sig_payload;
+			CALLOC(sig_payload, 1, sizeof(*sig_payload), GOTO_ERR(handle, ALPM_ERR_MEMORY, cleanup));
+
+			/* print filename into a buffer (leave space for separator and .sig) */
+			len = strlen(db->treename) + strlen(dbext) + 5;
+			MALLOC(sig_payload->filepath, len, GOTO_ERR(handle, ALPM_ERR_MEMORY, cleanup));
+			snprintf(sig_payload->filepath, len, "%s%s.sig", db->treename, dbext);
+
+			sig_payload->handle = handle;
+			sig_payload->force = 1;
+			sig_payload->errors_ok = (siglevel & ALPM_SIG_DATABASE_OPTIONAL);
+
+			/* set hard upper limit of 16 KiB */
+			sig_payload->max_size = 16 * 1024;
+			sig_payload->servers = db->servers;
+
+			payloads = alpm_list_add(payloads, sig_payload);
+		}
+	}
+
+	ret = _alpm_multi_download(handle, payloads, syncpath);
+	if(ret < 0) {
+		goto cleanup;
+	}
+
+	for(i = dbs; i; i = i->next) {
+		alpm_db_t *db = i->data;
+		if(!(db->usage & ALPM_DB_USAGE_SYNC)) {
+			continue;
+		}
+
+		/* Cache needs to be rebuilt */
+		_alpm_db_free_pkgcache(db);
+
+		/* clear all status flags regarding validity/existence */
+		db->status &= ~DB_STATUS_VALID;
+		db->status &= ~DB_STATUS_INVALID;
+		db->status &= ~DB_STATUS_EXISTS;
+		db->status &= ~DB_STATUS_MISSING;
+
+		/* if the download failed skip validation to preserve the download error */
+		if(sync_db_validate(db) != 0) {
+			/* pm_errno should be set */
+			ret = -1;
+		}
+	}
+
+cleanup:
+	_alpm_handle_unlock(handle);
+
+	if(ret == -1) {
+		/* pm_errno was set by the download code */
+		_alpm_log(handle, ALPM_LOG_DEBUG, "failed to sync dbs: %s\n",
+				alpm_strerror(handle->pm_errno));
+	} else {
+		handle->pm_errno = ALPM_ERR_OK;
+	}
+
+	if(payloads) {
+		alpm_list_free_inner(payloads, (alpm_list_fn_free)_alpm_dload_payload_reset);
+		FREELIST(payloads);
+	}
+	free(syncpath);
+	umask(oldmask);
+	return ret;
+}
+
 /* Forward decl so I don't reorganize the whole file right now */
 static int sync_db_read(alpm_db_t *db, struct archive *archive,
 		struct archive_entry *entry, alpm_pkg_t **likely_pkg);
diff --git a/lib/libalpm/dload.c b/lib/libalpm/dload.c
index 670da03d..7cd3e3a4 100644
--- a/lib/libalpm/dload.c
+++ b/lib/libalpm/dload.c
@@ -636,6 +636,16 @@  int _alpm_download(struct dload_payload *payload, const char *localpath,
 	}
 }
 
+int _alpm_multi_download(alpm_handle_t *handle,
+		alpm_list_t *payloads /* struct dload_payload */,
+		const char *localpath)
+{
+	(void)handle;
+	(void)payloads;
+	(void)localpath;
+	return 0;
+}
+
 static char *filecache_find_url(alpm_handle_t *handle, const char *url)
 {
 	const char *filebase = strrchr(url, '/');
@@ -738,6 +748,7 @@  void _alpm_dload_payload_reset(struct dload_payload *payload)
 	FREE(payload->destfile_name);
 	FREE(payload->content_disp_name);
 	FREE(payload->fileurl);
+	FREE(payload->filepath);
 	*payload = (struct dload_payload){0};
 }
 
@@ -746,6 +757,7 @@  void _alpm_dload_payload_reset_for_retry(struct dload_payload *payload)
 	ASSERT(payload, return);
 
 	FREE(payload->fileurl);
+	FREE(payload->filepath);
 	payload->initial_size += payload->prevprogress;
 	payload->prevprogress = 0;
 	payload->unlink_on_fail = 0;
diff --git a/lib/libalpm/dload.h b/lib/libalpm/dload.h
index 1e8f75f3..3eb7fbe1 100644
--- a/lib/libalpm/dload.h
+++ b/lib/libalpm/dload.h
@@ -31,6 +31,7 @@  struct dload_payload {
 	char *destfile_name;
 	char *content_disp_name;
 	char *fileurl;
+	char *filepath; /* download URL path */
 	alpm_list_t *servers;
 	long respcode;
 	off_t initial_size;
@@ -53,4 +54,8 @@  void _alpm_dload_payload_reset_for_retry(struct dload_payload *payload);
 int _alpm_download(struct dload_payload *payload, const char *localpath,
 		char **final_file, const char **final_url);
 
+int _alpm_multi_download(alpm_handle_t *handle,
+		alpm_list_t *payloads /* struct dload_payload */,
+		const char *localpath);
+
 #endif /* ALPM_DLOAD_H */