search for: nbdkit_fua_emulate

On 3/18/20 8:21 PM, Eric Blake wrote: > In commit c306fa93ab and neighbors (v1.15.1), a concerted effort went > into caching the results of .can_FOO callbacks, with commit messages > demonstrating that a plugin with a slow callback should not have that > delay magnified multiple times. But nothing was added to the > testsuite at the time, and with the sh and eval plugins, we still

...connections that are not read-only. For now, this -function has no counterpart in plugins, although it will be added. -Unlike other feature functions with just two success states, this one -returns three success values: C<NBDKIT_FUA_NONE> to avoid advertising -FUA support to the client, C<NBDKIT_FUA_EMULATE> if FUA is emulated by -nbdkit calling flush after a write transaction, and +only relevant for connections that are not read-only. This intercepts +the corresponding plugin method, to control feature bits advertised to +the client. Unlike other feature functions with just two success +states,...

On Wed, Apr 11, 2018 at 12:03:41AM -0500, Eric Blake wrote: > +=item C<can_fua> > + > +(Optional) > + > + def can_fua(h): > + # return a boolean > + > +Unlike the C counterpart, the Python callback does not need a > +tri-state return value, because Python introspection is sufficient to > +learn whether callbacks support FUA. Thus, this function only returns

...tant (m, "FLAG_REQ_ONE", NBDKIT_FLAG_REQ_ONE); > + PyModule_AddIntConstant (m, "FLAG_FAST_ZERO", NBDKIT_FLAG_FAST_ZERO); > + > + PyModule_AddIntConstant (m, "FUA_NONE", NBDKIT_FUA_NONE); > + PyModule_AddIntConstant (m, "FUA_EMULATE", NBDKIT_FUA_EMULATE); > + PyModule_AddIntConstant (m, "FUA_NATIVE", NBDKIT_FUA_NATIVE); > + > + PyModule_AddIntConstant (m, "CACHE_NONE", NBDKIT_CACHE_NONE); > + PyModule_AddIntConstant (m, "CACHE_EMULATE", NBDKIT_CACHE_EMULATE); > + PyModule_AddIntConstant (m,...

...-git a/src/plugins.c b/src/plugins.c index d44e724..e583fac 100644 --- a/src/plugins.c +++ b/src/plugins.c @@ -389,7 +389,7 @@ plugin_can_fua (struct backend *b, struct connection *conn) if (p->plugin.can_fua) { r = p->plugin.can_fua (connection_get_handle (conn, 0)); - if (r > NBDKIT_FUA_EMULATE && p->plugin._api_version == 1) + if (r > NBDKIT_FUA_EMULATE && !p->plugin.pread) r = NBDKIT_FUA_EMULATE; return r; } @@ -655,7 +655,7 @@ plugin_register (size_t index, const char *filename, } /* Check for incompatible future versions. */ - if (plu...

On 11/22/19 3:14 PM, Richard W.M. Jones wrote: >>> @@ -54,20 +61,20 @@ def get_size(h): >>> return len(disk) >>> -def pread(h, count, offset): >>> +def pread(h, count, offset, flags): >>> global disk >>> return disk[offset:offset+count] >> >> Do we really want to be passing 'flags' as an integer that the

...KIT_FLAG_REQ_ONE); > > > + PyModule_AddIntConstant (m, "FLAG_FAST_ZERO", NBDKIT_FLAG_FAST_ZERO); > > > + > > > + PyModule_AddIntConstant (m, "FUA_NONE", NBDKIT_FUA_NONE); > > > + PyModule_AddIntConstant (m, "FUA_EMULATE", NBDKIT_FUA_EMULATE); > > > + PyModule_AddIntConstant (m, "FUA_NATIVE", NBDKIT_FUA_NATIVE); > > > + > > > + PyModule_AddIntConstant (m, "CACHE_NONE", NBDKIT_CACHE_NONE); > > > + PyModule_AddIntConstant (m, "CACHE_EMULATE", NBDKIT_CACHE_EMULA...

After more than a month since v2 [1], I've finally got my FUA support series polished. This is all of my outstanding patches, even though some of them were originally posted in separate threads from the original FUA post [2], [3] [1] https://www.redhat.com/archives/libguestfs/2018-January/msg00113.html [2] https://www.redhat.com/archives/libguestfs/2018-January/msg00219.html [3]

...lter: +when splitting a write request that requested FUA from the client, if +C<next_ops-E<gt>can_fua> returns C<NBDKIT_FUA_NATIVE>, then the filter +should pass the FUA flag on to each sub-request; while if it is known +that FUA is emulated by a flush because of a return of +C<NBDKIT_FUA_EMULATE>, it is more efficient to only flush once after +all sub-requests have completed (often by passing C<NBDKIT_FLAG_FUA> +on to only the final sub-request, or by dropping the flag and ending +with a direct call to C<next_ops-E<gt>flush>). + If there is an error, the callback shou...

...int r; + + switch (call_read (&s, &slen, args)) { + case OK: + if (slen > 0 && s[slen-1] == '\n') + s[slen-1] = '\0'; + if (strcmp (s, "none") == 0) + r = NBDKIT_FUA_NONE; + else if (strcmp (s, "emulate") == 0) + r = NBDKIT_FUA_EMULATE; + else if (strcmp (s, "native") == 0) + r = NBDKIT_FUA_NATIVE; + else { + nbdkit_error ("%s: could not parse output from can_fua method: %s", + script, s); + free (s); + return -1; + } + free (s); + return r; + + case MISSI...

...int r; + + switch (call_read (&s, &slen, args)) { + case OK: + if (slen > 0 && s[slen-1] == '\n') + s[slen-1] = '\0'; + if (strcmp (s, "none") == 0) + r = NBDKIT_FUA_NONE; + else if (strcmp (s, "emulate") == 0) + r = NBDKIT_FUA_EMULATE; + else if (strcmp (s, "native") == 0) + r = NBDKIT_FUA_NATIVE; + else { + nbdkit_error ("%s: could not parse output from can_fua method: %s", + script, s); + free (s); + return -1; + } + free (s); + return r; + + case MISSI...

..., argpos))) +#else +#define ATTRIBUTE_FORMAT_PRINTF(fmtpos, argpos) +#endif + #define NBDKIT_THREAD_MODEL_SERIALIZE_CONNECTIONS 0 #define NBDKIT_THREAD_MODEL_SERIALIZE_ALL_REQUESTS 1 #define NBDKIT_THREAD_MODEL_SERIALIZE_REQUESTS 2 @@ -57,11 +64,9 @@ extern "C" { #define NBDKIT_FUA_EMULATE 1 #define NBDKIT_FUA_NATIVE 2 -extern void nbdkit_error (const char *msg, ...) - __attribute__((__format__ (__printf__, 1, 2))); +extern void nbdkit_error (const char *msg, ...) ATTRIBUTE_FORMAT_PRINTF (1, 2); extern void nbdkit_verror (const char *msg, va_list args); -extern void nbdki...

...{ + int writeflags = 0; + bool need_flush = false; + assert (zeromode != NONE); flags &= ~NBDKIT_FLAG_MAY_TRIM; if (zeromode == NOTRIM) return next_ops->zero (nxdata, count, offs, flags, err); + if (flags & NBDKIT_FLAG_FUA) { + if (next_ops->can_fua (nxdata) == NBDKIT_FUA_EMULATE) + need_flush = true; + else + writeflags = NBDKIT_FLAG_FUA; + } + while (count) { /* Always contains zeroes, but we can't use const or else gcc 9 * will use .rodata instead of .bss and inflate the binary size. @@ -115,7 +125,10 @@ nozero_zero (struct nbdkit_next_ops...

...PyModule_AddIntConstant (m, "FLAG_REQ_ONE", NBDKIT_FLAG_REQ_ONE); + PyModule_AddIntConstant (m, "FLAG_FAST_ZERO", NBDKIT_FLAG_FAST_ZERO); + + PyModule_AddIntConstant (m, "FUA_NONE", NBDKIT_FUA_NONE); + PyModule_AddIntConstant (m, "FUA_EMULATE", NBDKIT_FUA_EMULATE); + PyModule_AddIntConstant (m, "FUA_NATIVE", NBDKIT_FUA_NATIVE); + + PyModule_AddIntConstant (m, "CACHE_NONE", NBDKIT_CACHE_NONE); + PyModule_AddIntConstant (m, "CACHE_EMULATE", NBDKIT_CACHE_EMULATE); + PyModule_AddIntConstant (m, "CACHE_NATIVE"...

...REQ_ONE", NBDKIT_FLAG_REQ_ONE); > > + PyModule_AddIntConstant (m, "FLAG_FAST_ZERO", NBDKIT_FLAG_FAST_ZERO); > > + > > + PyModule_AddIntConstant (m, "FUA_NONE", NBDKIT_FUA_NONE); > > + PyModule_AddIntConstant (m, "FUA_EMULATE", NBDKIT_FUA_EMULATE); > > + PyModule_AddIntConstant (m, "FUA_NATIVE", NBDKIT_FUA_NATIVE); > > + > > + PyModule_AddIntConstant (m, "CACHE_NONE", NBDKIT_CACHE_NONE); > > + PyModule_AddIntConstant (m, "CACHE_EMULATE", NBDKIT_CACHE_EMULATE); > > + Py...

...nctionObjArgs (fn, obj, NULL); + Py_DECREF (fn); + if (check_python_failure ("can_fua") == -1) + return -1; + if (r == Py_False) + fua = NBDKIT_FUA_NONE; + else if (pwrite_has_fua || zero_has_fua || trim_has_fua) + fua = NBDKIT_FUA_NATIVE; + else + fua = NBDKIT_FUA_EMULATE; + Py_DECREF (r); + } + else if (pwrite_has_fua || zero_has_fua || trim_has_fua) + fua = NBDKIT_FUA_NATIVE; + else if (callback_defined ("flush", NULL)) + fua = NBDKIT_FUA_EMULATE; + else + fua = NBDKIT_FUA_NONE; + + return fua; +} + #define py_config_help \ "scr...

..._ops->can_flush (nxdata); + case NONE: + case EMULATE: + case NATIVE: + case PASS: + return next_ops->can_flush (nxdata); + } + abort (); } /* Advertise desired fua mode. */ @@ -136,6 +148,7 @@ fua_can_fua (struct nbdkit_next_ops *next_ops, void *nxdata, void *handle) return NBDKIT_FUA_EMULATE; case NATIVE: case FORCE: + case DISCARD: return NBDKIT_FUA_NATIVE; case PASS: return next_ops->can_fua (nxdata); @@ -167,6 +180,9 @@ fua_pwrite (struct nbdkit_next_ops *next_ops, void *nxdata, case FORCE: flags |= NBDKIT_FLAG_FUA; break; + case DISCARD: + fl...

v1 was here: https://www.redhat.com/archives/libguestfs/2018-December/thread.html#00123 v2: - Document "-" instead of "script=-" and use it in the test; and verify this also works on FreeBSD; and verify that it doesn't depend on the particular behaviour of our wrapper script and should work with installed nbdkit too. - Fix handling of zero flags parameter. -

Another couple things I noticed that are worth improving, but aren't strictly related to implementing fast zero support. Eric Blake (2): server: Assert sane error responses nozero: More efficient FUA handling filters/nozero/nozero.c | 17 +++++++++++-- server/filters.c | 56 +++++++++++++++++++++++++++++++++-------- server/protocol.c | 32 +++++++++++++++++------ 3 files

...plugin.pod b/docs/nbdkit-plugin.pod index cfa14b0..c72ea0d 100644 --- a/docs/nbdkit-plugin.pod +++ b/docs/nbdkit-plugin.pod @@ -586,6 +586,26 @@ handle FUA requests. If omitted, nbdkit checks whether C<.flush> exists, and behaves as if this function returns C<NBDKIT_FUA_NONE> or C<NBDKIT_FUA_EMULATE> as appropriate. +=head2 C<.can_multi_conn> + + int can_multi_conn (void *handle); + +This is called during the option negotiation phase to find out if the +plugin can handle multiple connections from a single client. + +Specifically it means that either the plugin does not cache request...