1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
|
=head1 NAME
nbd_pread_structured - read from the NBD server
=head1 SYNOPSIS
#include <libnbd.h>
typedef struct {
int (*callback) (void *user_data, const void *subbuf,
size_t count, uint64_t offset,
unsigned status, int *error);
void *user_data;
void (*free) (void *user_data);
} nbd_chunk_callback;
int nbd_pread_structured (
struct nbd_handle *h, void *buf, size_t count,
uint64_t offset, nbd_chunk_callback chunk_callback,
uint32_t flags
);
=head1 DESCRIPTION
Issue a read command to the NBD server for the range starting
at C<offset> and ending at C<offset> + C<count> - 1. The server's
response may be subdivided into chunks which may arrive out of order
before reassembly into the original buffer; the C<chunk> callback
is used for notification after each chunk arrives, and may perform
additional sanity checking on the server's reply. The callback cannot
call C<nbd_*> APIs on the same handle since it holds the handle lock
and will cause a deadlock. If the callback returns C<-1>, and no
earlier error has been detected, then the overall read command will
fail with any non-zero value stored into the callback's C<error>
parameter (with a default of C<EPROTO>); but any further chunks will
still invoke the callback.
The C<chunk> function is called once per chunk of data received, with
the C<user_data> passed to this function. The
C<subbuf> and C<count> parameters represent the subset of the original
buffer which has just been populated by results from the server (in C,
C<subbuf> always points within the original C<buf>; but this guarantee
may not extend to other language bindings). The C<offset> parameter
represents the absolute offset at which C<subbuf> begins within the
image (note that this is not the relative offset of C<subbuf> within
the original buffer C<buf>). Changes to C<error> on output are ignored
unless the callback fails. The input meaning of the C<error> parameter
is controlled by the C<status> parameter, which is one of
=over 4
=item C<LIBNBD_READ_DATA> = 1
C<subbuf> was populated with C<count> bytes of data. On input, C<error>
contains the errno value of any earlier detected error, or zero.
=item C<LIBNBD_READ_HOLE> = 2
C<subbuf> represents a hole, and contains C<count> NUL bytes. On input,
C<error> contains the errno value of any earlier detected error, or zero.
=item C<LIBNBD_READ_ERROR> = 3
C<count> is 0, so C<subbuf> is unusable. On input, C<error> contains the
errno value reported by the server as occurring while reading that
C<offset>, regardless if any earlier error has been detected.
=back
Future NBD extensions may permit other values for C<status>, but those
will not be returned to a client that has not opted in to requesting
such extensions. If the server is non-compliant, it is possible for
the C<chunk> function to be called more times than you expect or with
C<count> 0 for C<LIBNBD_READ_DATA> or C<LIBNBD_READ_HOLE>. It is also
possible that the C<chunk> function is not called at all (in
particular, C<LIBNBD_READ_ERROR> is used only when an error is
associated with a particular offset, and not when the server reports a
generic error), but you are guaranteed that the callback was called at
least once if the overall read succeeds. Libnbd does not validate that
the server obeyed the requirement that a read call must not have
overlapping chunks and must not succeed without enough chunks to cover
the entire request.
Note that libnbd currently enforces a maximum read buffer of 64MiB,
even if the server would permit a larger buffer in a single transaction;
attempts to exceed this will result in an C<ERANGE> error. The server
may enforce a smaller limit, which can be learned with
L<nbd_get_block_size(3)>.
The C<flags> parameter may be C<0> for no flags, or may contain
C<LIBNBD_CMD_FLAG_DF> meaning that the server should not reply with
more than one fragment (if that is supported - some servers cannot do
this, see L<nbd_can_df(3)>). Libnbd does not validate that the server
actually obeys the flag.
Note that if this command fails, and L<nbd_get_pread_initialize(3)>
returns true, then libnbd sanitized C<buf>, but it is unspecified
whether the contents of C<buf> will read as zero or as partial results
from the server. If L<nbd_get_pread_initialize(3)> returns false,
then libnbd did not sanitize C<buf>, and the contents are undefined
on failure.
By default, libnbd will reject attempts to use this function with
parameters that are likely to result in server failure, such as
requesting an unknown command flag. The L<nbd_set_strict_mode(3)>
function can be used to alter which scenarios should await a server
reply rather than failing fast.
=head1 RETURN VALUE
If the call is successful the function returns C<0>.
=head1 ERRORS
On error C<-1> is returned.
Refer to L<libnbd(3)/ERROR HANDLING>
for how to get further details of the error.
The following parameters must not be NULL: C<h>, C<buf>.
For more information see L<libnbd(3)/Non-NULL parameters>.
=head1 HANDLE STATE
nbd_pread_structured
can be called when the handle is in the following state:
┌─────────────────────────────────────┬─────────────────────────┐
│ Handle created, before connecting │ ❌ error │
│ Connecting │ ❌ error │
│ Connecting & handshaking (opt_mode) │ ❌ error │
│ Connected to the server │ ✅ allowed │
│ Connection shut down │ ❌ error │
│ Handle dead │ ❌ error │
└─────────────────────────────────────┴─────────────────────────┘
=head1 VERSION
This function first appeared in libnbd 1.0.
If you need to test if this function is available at compile time
check if the following macro is defined:
#define LIBNBD_HAVE_NBD_PREAD_STRUCTURED 1
=head1 SEE ALSO
L<nbd_aio_pread_structured(3)>,
L<nbd_can_df(3)>,
L<nbd_create(3)>,
L<nbd_get_block_size(3)>,
L<nbd_get_pread_initialize(3)>,
L<nbd_pread(3)>,
L<nbd_set_pread_initialize(3)>,
L<nbd_set_request_block_size(3)>,
L<nbd_set_strict_mode(3)>,
L<libnbd(3)>.
=head1 AUTHORS
Eric Blake
Richard W.M. Jones
=head1 COPYRIGHT
Copyright Red Hat
|
