Bring i386 up to date with amd64 and others.
The macros for PCPU can be slightly simplified, which makes the
resulting tangle qa lot easier to understand when trying to read them.
New counters now exist for:
requests sent
replies sent
requests received
replies received
packets received
total packets dropped due to no ARP entry
entrys timed out
Duplicate IPs seen
The new statistics are seen in the netstat command
when it is given the -s command line switch.
Warner Losh [Thu, 3 Sep 2009 17:04:42 +0000 (17:04 +0000)]
Time for house-cleaning:
o remove all entries before RELENG_7 was branched, as is tradition[*].
o Update examples... nobody cares about 5.x upgrades.
o minor format tweaking in a few places.
o update copyright (although at best I hold an editors copyright these days).
o Remove giving people permission to buy me beer. I don't do enough for
this document for that anymore...
Do the first step in removing lukemftpd from the base system. Disconnect
it from the build.
If you are using the FTP daemon, please consider using the port ftp/tnftpd
which is the same FTP server, but newer and might have more/better
functionality.
This results in us providing only one ftp daemon by default.
Ed Schouten [Thu, 3 Sep 2009 16:31:11 +0000 (16:31 +0000)]
Expose the TF_REVERSE flag to the console driver.
Right now libteken processes TF_REVERSE internally and returns the
toggled colors to the console driver. This isn't entirely correct. This
means that the bold flag is always processed by the foreground color,
while reversing should be done after the foreground color has been set
to a brighter version by the bold flag.
This is no problem with the syscons driver, because with VGA it only
supports 16 foreground and 8 background colors. My WIP console driver
reconfigures the graphics hardware to disable the blink functionality
and uses 16 foreground and 16 background colors. This means that this
driver will handle the TF_REVERSE flag a little different from what
syscons does right now.
Add intermediate states for attaching and detaching that will be
reused by the enhached newbus locking once it is checked in.
This change can be easilly MFCed to STABLE_8 at the appropriate moment.
Reviewed by: jhb, scottl
Tested by: Giovanni Trematerra <giovanni dot trematerra at gmail dot com>
Ed Schouten [Thu, 3 Sep 2009 09:33:57 +0000 (09:33 +0000)]
Move libteken out of the syscons directory.
I initially committed libteken to sys/dev/syscons/teken, but now that
I'm working on a console driver myself, I noticed this was not a good
decision. Move it to sys/teken to make it easier for other drivers to
use a terminal emulator.
Also list teken.c in sys/conf/files, instead of listing it in all the
files.arch files separately.
Fix regression introduced in r196712 - the 'name' string needs
to be rewritten for each file we want to check ACL on. Without
this change, ls(1) would check only the ACL on the first file
to list.
In the lockmgr support:
- GIANT_RESTORE() is just called when the sleep finishes, so the current
code can ends up into a giant unlock problem. Fix it by appropriately
call GIANT_RESTORE() when needed. Note that this is not exactly ideal
because for any interation of the adaptive spinning we drop and restore
Giant, but the overhead should be not a factor.
- In the lock held in exclusive mode case, after the adaptive spinning is
brought to completition, we should just retry to acquire the lock
instead to fallthrough. Fix that.
- Fix a style nit
In the sx support:
- Call GIANT_SAVE() before than looping. This saves some overhead because
in the current code GIANT_SAVE() is called several times.
Tested by: Giovanni Trematerra <giovanni dot trematerra at gmail dot com>
Jung-uk Kim [Wed, 2 Sep 2009 16:02:48 +0000 (16:02 +0000)]
- Work around ACPI mode transition problem for recent NVIDIA 9400M chipset
based Intel Macs. Since r189055, these platforms started freezing when
ACPI is being initialized for unknown reason. For these platforms, we just
use the old PAT layout. Note this change is not enough to boot fully on
these platforms because of other problems but it makes debugging possible.
Note MacBook5,2 may be affected as well but it was not added here because
of lack of hardware to test.
- Initialize PAT MSR fully instead of reading and modifying it for safety.
- Remove BSD.{x11,x11-4}.dist as we merged prefixes a while ago and those
files aren't used anymore.
- Remove BSD.local.dist as the file moved to ports/Templates when we merged
prefixes.
- Bump PKG_INSTALL_VERSION to 20090902 after dougb's changes.
- Change the comment to say that version must be changed when a
non-cosmetic change is made.
Add NFSv4 ACL support to cp(1) and fix a few memory leaks.
Note that this changes error reporting behaviour somewhat - before,
no error was reported if ACL couldn't be copied because the target
filesystem doesn't support ACLs. Now, it will be reported - of course,
only if there actually is an ACL to copy.
1) Use isprint() instead of hardcoded values to detect non-printable.
2) Use (unsigned char) cast in waddch() calls.
It fix highlighting bug: sign extension of 8bit to the attributes area.
3) Use setlocale() in any case.
John Baldwin [Wed, 2 Sep 2009 00:39:59 +0000 (00:39 +0000)]
Don't attempt to bind the current thread to the CPU an IRQ is bound to
when removing an interrupt handler from an IRQ during shutdown. During
shutdown we are already bound to CPU 0 and this was triggering a panic.
Fix regression introduced with NFSv4 ACL support - make acl_to_text(3)
and acl_calc_mask(3) return error instead of crashing when acl passed
to them is NULL.
Submitted by: markus
Reviewed by: rwatson
MFC after: 3 days
Add counters for the i7 architecture which were accidentally left
out of the original commit of i7 support. These are all the counters
on pages A-32 and A-33 of the _Intel(R) 64 and IA32 Architectures
Software Developer's Manual Vol 3B_, June 2009. Almost all
of these counters relate to operations on the L2 cache.
In case an upper layer protocol tries to send a packet but the
L2 code does not have the ethernet address for the destination
within the broadcast domain in the table, we remember the
original mbuf in `la_hold' in arpresolve() and send out a
different packet with an arp request.
In case there will be more upper layer packets to send we will
free an earlier one held in `la_hold' and queue the new one.
Once we get a packet in, with which we can perfect our arp table
entry we send out the original 'on hold' packet, should there
be any.
Rather than continuing to process the packet that we received,
we returned without freeing the packet that came in, which
basically means that we leaked an mbuf for every arp request
we sent.
Rather than freeing the received packet and returning, continue
to process the incoming arp packet as well.
This should (a) improve some setups, also proxy-arp, in case it was an
incoming arp request and (b) resembles the behaviour FreeBSD had
from day 1, which alignes with RFC826 "Packet reception" (merge case).
Rename 'm0' to 'hold' to make the code more understandable as
well as diffable to earlier versions more easily.
Handle the link-layer entry 'la' lock comepletely in the block
where needed and release it as early as possible, rather than
holding it longer, down to the end of the function.
Found by: pointyhat, ns1
Bug hunting session with: erwin, simon, rwatson
Tested by: simon on cluster machines
Reviewed by: ratson, kmacy, julian
MFC after: 3 days
Reintroduce the r196640, after fixing the problem with my testing.
Remove the altkstacks, instead instantiate threads with kernel stack
allocated with the right size from the start. For the thread that has
kernel stack cached, verify that requested stack size is equial to the
actual, and reallocate the stack if sizes differ [1].
This fixes the bug introduced by r173361 that was committed several days
after r173004 and consisted of kthread_add(9) ignoring the non-default
kernel stack size.
Also, r173361 removed the caching of the kernel stacks for a non-first
thread in the process. Introduce separate kernel stack cache that keeps
some limited amount of preallocated kernel stacks to lower the latency
of thread allocation. Add vm_lowmem handler to prune the cache on
low memory condition. This way, system with reasonable amount of the
threads get lower latency of thread creation, while still not exhausting
significant portion of KVA for unused kstacks.
Submitted by: peter [1]
Discussed with: jhb, julian, peter
Reviewed by: jhb
Tested by: pho (and retested according to new test scenarious)
MFC after: 1 week
Pyun YongHyeon [Mon, 31 Aug 2009 22:09:48 +0000 (22:09 +0000)]
Make sure rx descriptor ring align on 16 bytes. I guess the
alignment requirement could be multiple of 4 bytes but I think
using descriptor size would make intention clearer.
Previously the size of rx descriptor was not power of 2 so it
caused panic in bus_dmamem_alloc(9).
Reported by: Jeff Blank (jb000003 <> mr-happy dot com)
MFC after: 3 days
Sam Leffler [Mon, 31 Aug 2009 21:25:49 +0000 (21:25 +0000)]
On resume in sta mode program the beacon timers so when roaming (and
the previous ap is no longer in range) the device will deliver bmiss
interrupts and trigger the state machine. Also arrange to sync the
beacon timers on the next received beacon frame so that when we don't
roam we re-synchronize with the ap.
Qing Li [Mon, 31 Aug 2009 21:02:48 +0000 (21:02 +0000)]
This patch fixes the following issues:
- Routing messages are not generated when adding and removing
interface address aliases.
- Loopback route installed for an interface address alias is
not deleted from the routing table when that address alias
is removed from the associated interface.
- Function in_ifscrub() is called extraneously.
John Baldwin [Mon, 31 Aug 2009 18:41:13 +0000 (18:41 +0000)]
Simplify pmap_change_attr() a bit:
- Always calculate the cache bits instead of doing it on-demand.
- Always set changed to TRUE rather than only doing it if it is false.
John Baldwin [Mon, 31 Aug 2009 17:42:52 +0000 (17:42 +0000)]
Improve pmap_change_attr() so that it is able to demote a large (2/4MB)
page into 4KB pages as needed. This should be fairly rare in practice
on i386. This includes merging the following changes from the amd64 pmap:
180430, 180485, 180845, 181043, 181077, and 196318.
- Add basic support for changing attributes on PDEs to pmap_change_attr()
similar to the support in the initial version of pmap_change_attr() on
amd64 including inlines for pmap_pde_attr() and pmap_pte_attr().
- Extend pmap_demote_pde() to include the ability to instantiate a new page
table page where none existed before.
- Enhance pmap_change_attr(). Use pmap_demote_pde() to demote a 2/4MB page
mapping to 4KB page mappings when the specified attribute change only
applies to a portion of the 2/4MB page. Previously, in such cases,
pmap_change_attr() gave up and returned an error.
- Correct a critical accounting error in pmap_demote_pde().
Make the mnt_writeopcount and mnt_secondary_writes counters,
used by the suspension code, not greater then mnt_ref reference
counter value. Increment mnt_ref together with write counter
in vn_start_write()/ vn_start_secondary_write(), releasing in
vn_finished_write/vn_finished_secondary_write().
Since r186197, unmount code requires that no writers occured after all
references are expired. We still could get write counter incremented
for freed or reused struct mount, but it seems to be innocent, since
corresponding vnode should be referenced and reclaimed then.
Qing Li [Mon, 31 Aug 2009 00:14:37 +0000 (00:14 +0000)]
As part of r196609, a call to "rtalloc" did not take the fib into
account. So call the appropriate "rtalloc_ign_fib()" instead of
calling "rtalloc_ign()".
Reviewed by:i pointed out by bz
MFC after: immediately
Bjoern A. Zeeb [Sun, 30 Aug 2009 14:38:17 +0000 (14:38 +0000)]
Make sure FreeBSD binaries without .note.ABI-tag section work
correctly and do not match a colliding Debian GNU/kFreeBSD
brandinfo statements.
For this mark the Debian GNU/kFreeBSD brandinfo that it must have
an .note.ABI-tag section and ignore the old EI_OSABI brandinfo
when comparing a possibly colliding set of options.
Due to SYSINIT we add the brandinfo in a non-deterministic order,
so native FreeBSD is not always first. We may want to consider
to force native FreeBSD to come first as well.
The only way a problem could currently be noticed is when running an
i386 binary without the .note.ABI-tag on amd64 and the Debian GNU/kFreeBSD
brandinfo was matched first, as the fallback to ld-elf32.so.1 does
not exist in that case.
Reported and tested by: ticso
In collaboration with: kib
MFC after: 3 days
Qing Li [Sun, 30 Aug 2009 02:07:23 +0000 (02:07 +0000)]
Prefix on-link verification is being performed on statically
configured prefixes. Since these statically configured prefixes
do not have any associated advertising routers, these prefixes
are treated as unreachable and those prefix routes are deleted
from the routing table. Therefore bypass prefixes that are not
learned from router advertisements during prefix on-link check.
Remove the altkstacks, instead instantiate threads with kernel stack
allocated with the right size from the start. For the thread that has
kernel stack cached, verify that requested stack size is equial to the
actual, and reallocate the stack if sizes differ [1].
This fixes the bug introduced by r173361 that was committed several days
after r173004 and consisted of kthread_add(9) ignoring the non-default
kernel stack size.
Also, r173361 removed the caching of the kernel stacks for a non-first
thread in the process. Introduce separate kernel stack cache that keeps
some limited amount of preallocated kernel stacks to lower the latency
of thread allocation. Add vm_lowmem handler to prune the cache on
low memory condition. This way, system with reasonable amount of the
threads get lower latency of thread creation, while still not exhausting
significant portion of KVA for unused kstacks.
Submitted by: peter [1]
Discussed with: jhb, julian, peter
Reviewed by: jhb
Tested by: pho
MFC after: 1 week
John Baldwin [Sat, 29 Aug 2009 02:17:40 +0000 (02:17 +0000)]
Mark the fake pages constructed by the OBJT_SG pager valid. This was
accidentally lost at one point during the PAT development. Without this
fix vm_pager_get_pages() was zeroing each of the pages.
Warner Losh [Sat, 29 Aug 2009 01:34:42 +0000 (01:34 +0000)]
Connect bwi up to the build. While there are some problems with this
driver still, it generally works well for most people most of the
time. It is still too green for GENERIC, however.
Submitted by: many (latest being kwm@)
MFC after: 2 days (before RC1 if possible)
Jilles Tjoelker [Fri, 28 Aug 2009 22:41:25 +0000 (22:41 +0000)]
sh: Fix crash with empty functions (f() { }) introduced in r196483
Empty pairs of braces are represented by a NULL node pointer, just like
empty lines at the top level.
Support for empty pairs of braces may be removed later. They make the code
more complex, have inconsistent behaviour (may or may not change $?), are
not specified by POSIX and are not allowed by some other shells like bash,
dash and ksh93.
Marko Zec [Fri, 28 Aug 2009 22:30:55 +0000 (22:30 +0000)]
Introduce a separate sx lock for protecting lists of vnet sysinit
and sysuninit handlers.
Previously, sx_vnet, which is a lock designated for protecting
the vnet list, was (ab)used for protecting vnet sysinit / sysuninit
handler lists as well. Holding exclusively the sx_vnet lock while
invoking sysinit and / or sysuninit handlers turned out to be
problematic, since some of the handlers may attempt to wake up
another thread and wait for it to walk over the vnet list, hence
acquire a shared lock on sx_vnet, which in turn leads to a deadlock.
Protecting vnet sysinit / sysuninit lists with a separate lock
mitigates this issue, which was first observed with
flowtable_flush() / flowtable_cleaner() in sys/net/flowtable.c.
John Baldwin [Fri, 28 Aug 2009 14:06:55 +0000 (14:06 +0000)]
Extend the device pager to support different memory attributes on different
pages in an object.
- Add a new variant of d_mmap() currently called d_mmap2() which accepts
an additional in/out parameter that is the memory attribute to use for
the requested page.
- A driver either uses d_mmap() or d_mmap2() for all requests but not both.
The current implementation uses a flag in the cdevsw (D_MMAP2) to indicate
that the driver provides a d_mmap2() handler instead of d_mmap(). This
is done to make the change ABI compatible with existing drivers and
MFC'able to 7 and 8.
Qing Li [Fri, 28 Aug 2009 07:01:09 +0000 (07:01 +0000)]
In ip_output(), the flow-table module must not try to cache L2/L3
information for interface of IFF_POINTOPOINT or IFF_LOOPBACK type.
Since the L2 information (rt_lle) is invalid for these interface
types, accidental caching attempt will trigger panic when the invalid
rt_lle reference is accessed.
When installing a new route, or when updating an existing route, the
user supplied gateway address may be an interface address (this is
particularly true for point-to-point interface related modules such
as ppp, if_tun, if_gif). Currently the routing command handler always
set the RTF_GATEWAY flag if the gateway address is given as part of the
command paramters. Therefore the gateway address must be verified against
interface addresses or else the route would be treated as an indirect
route, thus making that route unusable.
Reviewed by: kmacy, julia, rwatson
Verified by: marcus
MFC after: 3 days
Qing Li [Fri, 28 Aug 2009 05:37:31 +0000 (05:37 +0000)]
Do not try to free the rt_lle entry of the cached route in
ip_output() if the cached route was not initialized from the
flow-table. The rt_lle entry is invalid unless it has been
initialized through the flow-table.