search for: filter_regist

...lags, err); else - return f->backend.next->cache (f->backend.next, conn, - count, offset, flags, err); + return b->next->cache (b->next, conn, count, offset, flags, err); } static struct backend filter_functions = { @@ -858,7 +846,7 @@ filter_register (struct backend *next, size_t index, const char *filename, if (f->filename == NULL) goto out_of_memory; f->dl = dl; - debug ("registering filter %s", f->filename); + debug ("registering filter %s", filename); /* Call the initialization function which retu...

This is not intended for v1.14. In fact, we may decide that the second patch is too gross, although the first one still seems like a useful improvement in isolation. I will also point out that all our filters are in-tree, and set the user-controlled field .version to the current release string. We could replace the second patch with a simpler one that just checks ._api_version as an int (as

...@@ filter_cache (struct backend *b, struct connection *conn, static struct backend filter_functions = { .free = filter_free, .thread_model = filter_thread_model, - .name = filter_name, .plugin_name = plugin_name, .usage = filter_usage, .version = filter_version, @@ -826,7 +799,6 @@ filter_register (struct backend *next, size_t index, const char *filename, { struct backend_filter *f; const struct nbdkit_filter *filter; - size_t i, len; union overlay { const char *str; unsigned int api; @@ -834,19 +806,12 @@ filter_register (struct backend *next, size_t index, const char...

...te, + .can_flush = filter_can_flush, + .is_rotational = filter_is_rotational, + .can_trim = filter_can_trim, + .pread = filter_pread, + .pwrite = filter_pwrite, + .flush = filter_flush, + .trim = filter_trim, + .zero = filter_zero, +}; + +/* Register and load a filter. */ +struct backend * +filter_register (struct backend *next, size_t index, const char *filename, + void *dl, struct nbdkit_filter *(*filter_init) (void)) +{ + struct backend_filter *f; + const struct nbdkit_filter *filter; + size_t i, len, size; + + f = calloc (1, sizeof *f); + if (f == NULL) { + out_of_memory:...

...nst char *filename, + void *dl, struct nbdkit_plugin *(*plugin_init) (void)) + __attribute__((__nonnull__ (2, 3, 4))); +extern void set_debug_flags (void *dl, const char *name) + __attribute__((__nonnull__ (1, 2))); /* filters.c */ -extern struct backend *filter_register (struct backend *next, size_t index, const char *filename, void *dl, struct nbdkit_filter *(*filter_init) (void)); +extern struct backend *filter_register (struct backend *next, size_t index, + const char *filename, void *dl, +...

...- count, offset, flags); + debug ("%s: zero count=%" PRIu32 " offset=%" PRIu64 " flags=0x%" PRIx32, + f->name, count, offset, flags); if (f->filter.zero) return f->filter.zero (&next_ops, &nxdata, handle, @@ -719,7 +719,7 @@ filter_register (struct backend *next, size_t index, const char *filename, debug ("registered filter %s (name %s)", f->filename, f->name); /* Call the on-load callback if it exists. */ - debug ("%s: load", f->filename); + debug ("%s: load", f->name); if (f-&...

...tern struct backend *plugin_register (size_t index, const char *filename, void *dl, struct nbdkit_plugin *(*plugin_init) (void)) __attribute__((__nonnull__ (2, 3, 4))); +extern void do_nbdkit_set_error (int err); /* filters.c */ extern struct backend *filter_register (struct backend *next, size_t index, @@ -515,6 +517,11 @@ extern void *threadlocal_buffer (size_t size); extern void threadlocal_set_conn (struct connection *conn); extern struct connection *threadlocal_get_conn (void); +/* public.c */ +extern int do_nbdkit_nanosleep (unsigned sec, unsigned n...

v1 was here: https://www.redhat.com/archives/libguestfs/2019-January/msg00008.html In v2 I have provided two patches: The first patch extends attribute((nonnull)) to most internal functions, but not to the external API. The second patch uses a macro so that attribute((format)) is only used in the public API on GCC or Clang. At least in theory these headers could be used by a C compiler which

...*buf, uint32_t count, uint64_t offset, @@ -582,6 +628,8 @@ static struct backend filter_functions = { .flush = filter_flush, .trim = filter_trim, .zero = filter_zero, + .can_zero = filter_can_zero, + .can_fua = filter_can_fua, }; /* Register and load a filter. */ @@ -626,15 +674,23 @@ filter_register (struct backend *next, size_t index, const char *filename, exit (EXIT_FAILURE); } - /* Since the filter might be much older than the current version of - * nbdkit, only copy up to the self-declared _struct_size of the - * filter and zero out the rest. If the filter is much newer th...

It's easy to use the sh script to demonstrate that nbdkit is inefficiently calling into .get_size, .can_fua, and friends more than necessary. We've also commented on the list in the past that it would be nice to ensure that when filters call into next_ops, they are not violating constraints (as we've have to fix several bugs in the past where we did not have such checking to protect

...te, + .can_flush = filter_can_flush, + .is_rotational = filter_is_rotational, + .can_trim = filter_can_trim, + .pread = filter_pread, + .pwrite = filter_pwrite, + .flush = filter_flush, + .trim = filter_trim, + .zero = filter_zero, +}; + +/* Register and load a filter. */ +struct backend * +filter_register (struct backend *next, size_t index, const char *filename, + void *dl, struct nbdkit_filter *(*filter_init) (void)) +{ + struct backend_filter *f; + const struct nbdkit_filter *filter; + size_t i, len, size; + + f = calloc (1, sizeof *f); + if (f == NULL) { + out_of_memory:...

...te, + .can_flush = filter_can_flush, + .is_rotational = filter_is_rotational, + .can_trim = filter_can_trim, + .pread = filter_pread, + .pwrite = filter_pwrite, + .flush = filter_flush, + .trim = filter_trim, + .zero = filter_zero, +}; + +/* Register and load a filter. */ +struct backend * +filter_register (struct backend *next, size_t index, const char *filename, + void *dl, struct nbdkit_filter *(*filter_init) (void)) +{ + struct backend_filter *f; + const struct nbdkit_filter *filter; + size_t i, len, size; + + f = calloc (1, sizeof *f); + if (f == NULL) { + out_of_memory:...

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

...16f 100644 --- a/server/filters.c +++ b/server/filters.c @@ -102,7 +102,7 @@ filter_version (struct backend *b) { struct backend_filter *f = container_of (b, struct backend_filter, backend); - return f->filter.version; + return f->filter._version; } static void @@ -730,14 +730,24 @@ filter_register (struct backend *next, size_t index, const char *filename, } /* We do not provide API or ABI guarantees for filters, other than - * the ABI position of _api_version that will let us diagnose - * mismatch when the API changes. + * the ABI position and API contents of _api_version and...

v1 was here: https://www.redhat.com/archives/libguestfs/2019-November/msg00003.html This version depends on: https://www.redhat.com/archives/libguestfs/2019-November/msg00004.html and this series: https://www.redhat.com/archives/libguestfs/2019-November/msg00009.html The delta has been reduced slightly because of changes made possible by cleaning up and fixing the quit path in nbdkit. It's

libFuzzer is Clang's fuzzer, and alternative to using AFL: https://llvm.org/docs/LibFuzzer.html I implemented an alternative method of fuzzing for libnbd earlier today and it's pretty simple: https://github.com/libguestfs/libnbd/commit/c19a6fbae9a21a7d4693418706c59e81ed256875 However it's considerably more difficult to use libFuzzer with non-library code -- in this case nbdkit.

This patch isn't complete (patch 6/6 isn't finished) so it's just for discussion, although it does compile and run. This introduces to nbdkit a concept of "filters" which can be placed in front of plugins to modify their behaviour. Some examples where you might use filters: * Serve a subset of the data, such as (offset, range) or a single partition from a disk image.

When I added .can_FOO caching in 1.16, I missed the case that the sh plugin itself was calling .can_flush twice in some situations (in order to default .can_fua). Then right after, I regressed it to call .can_zero twice (in order to default .can_fast_zero). I also missed that .thread_model could use better caching, because at the time, I did not add testsuite coverage. Fix that now. Eric Blake

This is still tentative and needs a lot of work, but: - partition filter works, supporting MBR & GPT - prepare and finalize methods fixed - open method can now be changed (allowing readonly flag to be modified) - thread_model can be limited I believe I made most of the changes which were previously suggested in email. I think the only one I didn't was preventing inclusion of both

Rebased filters patch. Requires current git master + the locks / thread model fix (https://www.redhat.com/archives/libguestfs/2018-January/msg00128.html) So a few changes here since last time: The "introduce filters" and "implement filters" patches are squashed together. I introduced a concept of .prepare and .finalize. These run before and after the data serving phase