Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
187,922 Commits 606 Branches 0 Tags 3.1 GiB
753d396585
Commit Graph

1122 Commits

Author SHA1 Message Date
pooka
93256256ef add cd@scsi and cd@atapi (untested. if you have a usb cd drive, let me know) 2010-02-09 19:02:19 +00:00
pooka
84e94fc428 Now that there is a working bus.h for all architectures, move rump 
bus dma out of a wip place and into rumpdev where it belongs.
 2010-02-09 18:57:03 +00:00
pooka
bd6771a8d0 * UR_CBI_ADSC ctrl req fails with EIO sometime for unknown reasons. 
Let it instead of panicking.  Doesn't seem to affect functionality.
* process pipe queue only until it is empty
 2010-02-09 18:27:17 +00:00
pooka
cefcf2133f Add a bus.h blanket header which forces definition of bus space / 
bus dma interfaces as functions and therefore makes it possible to
use rump on archs which suffer from macro maladies, i.e. everything
except i386 and amd64.

(build-tested on sparc64 and vax, which are usually the two extremes
of weirdness in these cases)
 2010-02-09 17:33:31 +00:00
pooka
2f3605b79c Store l_name for kernel threads. 2010-02-09 16:53:13 +00:00
pooka
a1e25c6e46 IOCONFIG comes from .CURDIR 2010-02-09 16:16:26 +00:00
joerg
d621e29eca Remove separate mb_map. The nmbclusters is computed at boot time based 
on the amount of physical memory and limited by NMBCLUSTERS if present.
Architectures without direct mapping also limit it based on the kmem_map
size, which is used as backing store. On i386 and ARM, the maximum KVA
used for mbuf clusters is limited to 64MB by default.

The old default limits and limits based on GATEWAY have been removed.
key_registered_sb_max is hard-wired to a value derived from 2048
clusters.
 2010-02-08 19:02:25 +00:00
pooka
9d22319fa9 Move the net80211 support component into the proper place in "net" 
and install as rumpnet_net80211.
 2010-02-05 22:01:38 +00:00
pooka
f5b877c68d Put common stuff in a single file which can be included. 2010-02-04 00:37:08 +00:00
pooka
665d28b452 Convert to match current ioconf fashion. 2010-02-03 21:35:22 +00:00
pooka
c192756108 Convert to ioconf new world order: define attachments using config(1) 
grammar instead of wildly copypasted, hand-edited and glue-sniffed
C files.
 2010-02-03 21:18:38 +00:00
pooka
b68d6b80c0 add a rule for locators.h too 2010-02-03 21:15:39 +00:00
pooka
61cce04cfe If IOCONFIG is defined, autogenerate ioconf.c and locators.h 
from ${IOCONFIG}.
 2010-02-03 21:11:24 +00:00
pooka
3cf97d3909 one more request type to pass through (seriously, i'll make it pass 
all of them through very soon.  just want to debug a little more)
 2010-02-03 18:14:56 +00:00
pooka
6c9f003a76 Include newly-created subr_device.c and remove few special case 
device accessor copypastes.  This makes it possible to link static
binaries which use -lrumpdev.
 2010-01-31 15:12:42 +00:00
pooka
1fe7a9e1bb If RUMP_BLKSECTSHIFT is set in the environment, use that as 
device sector size instead of DEV_BSHIFT.
 2010-01-31 13:15:08 +00:00
pooka
f91185803c Add sysmon rump component. 2010-01-31 03:06:59 +00:00
pooka
8844790318 include kern_hook.c 2010-01-31 02:08:36 +00:00
snj
58e4422c47 Some might argue that it is benefi_c_ial to spell words correctly. 2010-01-31 00:54:22 +00:00
pooka
5412e3f097 Make (no pun intended) it possible to perform component symbol 
protection from a shell prompt with the command line:

    make -f /sys/rump/Makefile RUMP_SYMREN=module.kobj

This is benefitial especially on i386, where rump ABI matches the
kernel ABI (sans the renaming isolation).  In the above example
module.kobj can then be loaded in a rump kernel either statically
at link-time simply by giving module.kobj as an object to the
linker, or dynamically at runtime by calling rump_sys_modctl(MODCTL_LOAD).
 2010-01-31 00:38:36 +00:00
pooka
edbb5bb953 no longer used 2010-01-30 05:09:24 +00:00
pooka
ffdaceb3e7 * add comment explaining why we randomize the last source ip octet 
* fix bug in randomization (well, at least with the current configuration ..)
 2010-01-29 12:34:17 +00:00
pooka
2da39fecaa Typecast memwinsize to off_t before taking the complement to be 
&'ed with an off_t.  Otherwise things go a bit wrong with >4GB
images ...
 2010-01-27 22:03:11 +00:00
pooka
580ff1a88f ignore pmf threads if RUMP_THREADS == 0 2010-01-27 20:16:16 +00:00
martti
40b313cfaa Use tabs instead of 8 spaces. 2010-01-27 13:30:45 +00:00
pooka
c52d5453ad support bpf from sockin too (i.e. DLT_NULL/AF_UNSPEC) 2010-01-26 17:52:21 +00:00
pooka
ac4d3e46b5 let attach via module space 2010-01-26 17:50:02 +00:00
pooka
3694a1482e support bpf 2010-01-26 17:41:20 +00:00
pooka
daa3b2e4c7 need rumpvfs for etfs_register() and sys_mknod() 2010-01-25 22:26:19 +00:00
pooka
d70c91c8de In case opening /dev/bpf fails, try to load bpf kmod and try again. 2010-01-25 22:25:38 +00:00
pooka
51ebf87267 Set attached cpus as running, otherwise (evil) state-probing code 
will not execute properly.
 2010-01-25 18:37:51 +00:00
pooka
89a3fc440c Add components necessary for bpf support (commented out just for 
the sheer heck of it).
 2010-01-19 22:47:46 +00:00
pooka
fcad67762f Add bpf rump component (and put in some wscons stuff i forgot to earlier). 2010-01-19 22:38:21 +00:00
pooka
b014350f7f Redefine bpf linkage through an always present op vector, i.e. 
#if NBPFILTER is no longer required in the client.  This change
doesn't yet add support for loading bpf as a module, since drivers
can register before bpf is attached.  However, callers of bpf can
now be modularized.

Dynamically loadable bpf could probably be done fairly easily with
coordination from the stub driver and the real driver by registering
attachments in the stub before the real driver is loaded and doing
a handoff.  ... and I'm not going to ponder the depths of unload
here.

Tested with i386/MONOLITHIC, modified MONOLITHIC without bpf and rump.
 2010-01-19 22:08:16 +00:00
pooka
23a987d2a2 Make build with PF_INET-is-the-"NIC" support again. 2010-01-19 17:51:03 +00:00
pooka
fe3e79c05a * add option to use bpf to dump response 
* randomize local address to avoid poor confused peers due to us
  "rebooting" quite rapidly when testing.
 2010-01-16 20:47:01 +00:00
pooka
c3ac0e0ea8 move bpfilter.h to a more central location 2010-01-16 18:02:19 +00:00
pooka
29ca934b44 support bpf 2010-01-16 17:55:50 +00:00
pooka
c396ac1512 Accomplish messy stack with slightly less messy code. 2010-01-15 20:39:46 +00:00
pooka
e5bb7fe875 Use subr_percpu.c instead of homegrown implementation. ...except 
when using malloc(3)-relegated allocators (happens in production
e.g. on Linux), since subr_percpu.c uses vmem and i don't want to
reimplement vmem.
 2010-01-15 19:01:04 +00:00
pooka
435aa34303 print mbuf statistics a few times. incidentally, can be used to 
test percpu.
 2010-01-15 18:38:16 +00:00
pooka
065afcb61a Minimize unnecessary differences in rump. 2010-01-13 01:53:38 +00:00
pooka
8bb03ea96b Add a few symbols used by the tty code. 2010-01-13 01:18:51 +00:00
pooka
e0714677a2 cgd is attached via the module interface now instead of pseudodevs. 2010-01-13 00:31:57 +00:00
pooka
a25612b2d0 If RUMP_VERBOSE is set in the env, set boothowto to AB_VERBOSE. 2010-01-13 00:07:40 +00:00
stacktic
1b7d9846f4 Fixed typo. 2010-01-11 19:19:57 +00:00
pooka
2d1ed1666f decend into ums, ukbd and wscons 2010-01-11 02:15:38 +00:00
pooka
6a31202221 Support ums (usb mouse) and ukbd (usb keyboard) drivers in rump. 
(both components include tty support and uhid currently.  they should
be split out to separate ones.  i'll do that soon'ish).
 2010-01-11 02:15:20 +00:00
pooka
20d948900e Add wscons rump component. 2010-01-11 02:12:51 +00:00
pooka
4f48ec16a6 Add a simple hack to make interrupt pipes work (can't close them 
yet ... except with SIGINT.  "muah").
 2010-01-10 22:13:18 +00:00