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
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
|
RTEMS BSD Library Guide
=======================
:toc:
:icons:
:numbered:
:website: http://www.rtems.org/
== Getting Started
== Issues and TODO
* PCI support on x86 uses a quick and dirty hack, see pci_reserve_map().
* Priority queues are broken with clustered scheduling.
* Per-CPU data should be enabled once the new stack is ready for SMP.
* Per-CPU NETISR(9) should be enabled onece the new stack is ready for SMP.
* Multiple routing tables are not supported. Every FIB value is set to zero
(= BSD_DEFAULT_FIB).
* Process identifiers are not supported. Every PID value is set to zero
(= BSD_DEFAULT_PID).
* User credentials are not supported. The following functions allow the
operation for everyone
- prison_equal_ip4(),
- chgsbsize(),
- cr_cansee(),
- cr_canseesocket() and
- cr_canseeinpcb().
* A basic USB functionality test that is known to work on Qemu is desirable.
* Adapt generic IRQ PIC interface code to Simple Vectored Interrupt Model
so that those architectures can use new TCP/IP and USB code.
* freebsd-userspace/rtems/include/sys/syslog.h is a copy from the old
RTEMS TCP/IP stack. For some reason, the __printflike markers do not
compile in this environment. We may want to use the FreeBSD syslog.h
and get this addressed.
* in_cksum implementations for architectures not supported by FreeBSD.
This will require figuring out where to put implementations that do
not originate from FreeBSD and are populated via the script.
* MAC support functions are not thread-safe ("freebsd/lib/libc/posix1e/mac.c").
* IFCONFIG(8): IEEE80211 support is disabled. This module depends on a XML
parser and mmap().
* get_cyclecount(): The implementation is a security problem.
* What to do with the priority parameter present in the FreeBSD synchronization
primitives and the thread creation functions?
* TASKQUEUE(9): Support spin mutexes.
* ZONE(9): Review allocator lock usage in rtems-bsd-chunk.c.
* KQUEUE(2): Choose proper lock for global kqueue list.
* TIMEOUT(9): Maybe use special task instead of timer server to call
callout_tick().
* sysctl_handle_opaque(): Implement reliable snapshots.
* PING6(8): What to do with SIGALARM?
* <sys/param.h>: Update Newlib to use a MSIZE of 256.
* BPF(4): Add support for zero-copy buffers.
* UNIX(4): Fix race conditions in the area of socket object and file node
destruction. Add support for file descriptor transmission via control
messages.
* PRINTF(9): Add support for log(), the %D format specifier is missing in the
normal printf() family.
* Why is the interrupt server used? The BSD interrupt handlers can block on
synchronization primitives like mutexes. This is in contrast to RTEMS
interrupt service routines. The BSPs using the generic interrupt support must
implement the `bsp_interrupt_vector_enable()` and
`bsp_interrupt_vector_disable()` routines. They normally enable/disable a
particular interrupt source at the interrupt controller. This can be used to
implement the interrupt server. The interrupt server is a task that wakes-up
in case an associated interrupt happens. The interrupt source is disabled in
a generic interrupt handler that wakes-up the interrupt server task. Once the
postponed interrupt processing is performed in the interrupt server the
interrupt source is enabled again.
* Convert all BSP linkcmds to use a linkcmds.base so the sections are
easier to insert.
* NIC Device Drivers
- Only common PCI NIC drivers have been included in the initial set. These
do not include any system on chip or ISA drivers.
- PCI configuration probe does not appear to happen to determine if a
NIC is in I/O or memory space. We have worked around this by using a
static hint to tell the fxp driver the correct mode. But this needs to
be addressed.
- The ISA drivers require more BSD infrastructure to be addressed. This was
outside the scope of the initial porting effort.
The FreeBSD device model is quite elaborated (with follow-ups):
http://www.freebsd.org/cgi/man.cgi?query=driver
The devices form a tree with the Nexus device at a high-level. This Nexus
device is architecture specific in FreeBSD. In RTEMS we have our own Nexus
device, see `rtemsbsd/bsp/bsp-bsd-nexus-devices.c`.
=== SYSCTL_NODE Example
During development, we had an undefined reference to
_bsd_sysctl__net_children that we had trouble tracking down. Thanks to
Chris Johns, we located it. He explained how to read SYSCTL_NODE
definitions. This line from freebsd/netinet/in_proto.c is attempting
to add the "inet" node to the parent node "_net".
----
SYSCTL_NODE(_net, PF_INET, inet, CTLFLAG_RW, 0,
"Internet Family");
----
Our problem was that we could not find where _bsd_sysctl__net_children
was defined. Chris suggested that when in doubt compile with -save-temps
and look at the preprocessed .i files. But he did not need that. He
explained that this the symbol name _bsd_sysctl__net_children was
automatically generated by a SYSCTL_NODE as follows:
* _bsd_ - added by RTEMS modifications to SYSCTL_NODE macro
* sysctl_ - boilerplace added by SYSCTL_NODE macro
* "" - empty string for parent node
* net - name of SYSCTL_NODE
* children - added by SYSCTL macros
This was all generated by a support macro declaring the node as this:
----
struct sysctl_oid_list SYSCTL_NODE_CHILDREN(parent, name);
----
Given this information, we located this SYSCTL_NODE declaration in
kern/kern_mib.c
----
SYSCTL_NODE(, CTL_KERN, kern, CTLFLAG_RW, 0,
"High kernel, proc, limits &c");
----
=== devfs (Device file system) ===
There is a minimal implementation based on IMFS. The mount point is fixed to
"/dev". Note that the devfs is only used by the cdev subsystem. cdev has been
adapted so that the full path (including the leading "/dev") is given to devfs.
This saves some copy operations.
devfs_create() first creates the full path and then creates an IMFS generic node
for the device.
TBD: remove empty paths on devfs_destroy().
== Notes by File ==
altq_subr.c - Arbitrary choices were made in this file that RTEMS would
not support tsc frequency change. Additionally, the clock frequency
for machclk_freq is always measured for RTEMS.
conf.h - In order to add make_dev and destroy_dev, variables in the cdev
structure that were not being used were conditionally compiled out. The
capability of supporting children did not appear to be needed and was
not implemented in the rtems version of these routines.
== Problems to report to FreeBSD ==
The MMAP_NOT_AVAILABLE define is inverted on its usage. When it is
defined the mmap method is called. Additionally, it is not used
thoroughly. It is not used in the unmap portion of the source.
The file rec_open.c uses the define MMAP_NOT_AVAILABLE to wrap
the call to mmap and file rec_close.c uses the munmap method.
|
