search for: dba3e24

While working on improving the backend interface to allow filters to handle errors, I noticed that I've introduced some minor incompatibilities for filters that don't quite obey the documentation which states that a callback should return only 0/-1. Prior to my additions, we treated all plugin returns other than -1 as success (sort of makes sense for a positive return, particularly if a

...st char *value) { if (strcmp (key, "file") == 0) { /* See FILENAMES AND PATHS in nbdkit-plugin(3). */ - filename = nbdkit_absolute_path (value); + filename = nbdkit_realpath (value); if (!filename) return -1; } diff --git a/src/plugins.c b/src/plugins.c index dba3e24..595b632 100644 --- a/src/plugins.c +++ b/src/plugins.c @@ -134,7 +134,7 @@ plugin_dump_fields (struct backend *b) struct backend_plugin *p = container_of (b, struct backend_plugin, backend); char *path; - path = nbdkit_absolute_path (p->filename); + path = nbdkit_realpath (p->filen...

...st char *value) { if (strcmp (key, "file") == 0) { /* See FILENAMES AND PATHS in nbdkit-plugin(3). */ - filename = nbdkit_absolute_path (value); + filename = nbdkit_realpath (value); if (!filename) return -1; } diff --git a/src/plugins.c b/src/plugins.c index dba3e24..595b632 100644 --- a/src/plugins.c +++ b/src/plugins.c @@ -134,7 +134,7 @@ plugin_dump_fields (struct backend *b) struct backend_plugin *p = container_of (b, struct backend_plugin, backend); char *path; - path = nbdkit_absolute_path (p->filename); + path = nbdkit_realpath (p->filen...

...(*zero) (struct backend *, struct connection *conn, uint32_t count, uint64_t offset, uint32_t flags); + int (*can_zero) (struct backend *, struct connection *conn); + int (*can_fua) (struct backend *, struct connection *conn); }; /* plugins.c */ diff --git a/src/plugins.c b/src/plugins.c index dba3e24..3468b6d 100644 --- a/src/plugins.c +++ b/src/plugins.c @@ -358,6 +358,26 @@ plugin_can_trim (struct backend *b, struct connection *conn) return p->plugin.trim != NULL; } +static int +plugin_can_zero (struct backend *b, struct connection *conn) +{ + debug ("can_zero"); + + //...

I still need to add testsuite coverage. Perhaps it might be easier if I create a new '--filter=log logfile=foo' filter that produces a log of which commands a client sent, then compare the log using a known client that uses write_zeroes (qemu-io works well) both with and without --filter=nozero to prove that the change in advertisement changes the commands sent over the wire (that would

....dump_fields = filter_dump_fields, .config = filter_config, .config_complete = filter_config_complete, - .errno_is_preserved = plugin_errno_is_preserved, .open = filter_open, .prepare = filter_prepare, .finalize = filter_finalize, diff --git a/src/plugins.c b/src/plugins.c index dba3e24..c49c0f0 100644 --- a/src/plugins.c +++ b/src/plugins.c @@ -227,14 +227,6 @@ plugin_config_complete (struct backend *b) exit (EXIT_FAILURE); } -static int -plugin_errno_is_preserved (struct backend *b) -{ - struct backend_plugin *p = container_of (b, struct backend_plugin, backend); - - re...

Here's what I'm currently playing with; I'm not ready to commit anything until I rebase my FUA work on top of this, as I only want to break filter ABI once between releases. Eric Blake (2): backend: Rework internal/filter error return semantics filters: Add log filter TODO | 2 - docs/nbdkit-filter.pod | 84 +++++++-- docs/nbdkit.pod

Since v1: add the blocksize filter, add testsuite coverage of the log filter, several fixes to the log filter based on what adding tests revealed I'm still working on FUA flag support patches on top of this; the patches should all be committed in the same release, as we want to minimize the number of releases that cause a filter ABI/API bump Eric Blake (3): backend: Rework internal/filter