dragonfly.git
7 months agokernel - Fix issue revealed by new yield code master
Matthew Dillon [Sat, 14 Sep 2013 09:08:18 +0000 (02:08 -0700)]
kernel - Fix issue revealed by new yield code

* Make sure the current lwp is released from the scheduler in all cases.
  There were a few race cases where it would not be released.

* Reorder the yield code a little.

7 months agokernel - Misc adjustments used by the vkernel and VMM, misc optimizations
Matthew Dillon [Sat, 14 Sep 2013 02:02:55 +0000 (19:02 -0700)]
kernel - Misc adjustments used by the vkernel and VMM, misc optimizations

* This section committed separately because it is basically independent
  of VMM.

* Improve pfind().  Don't get proc_token if the process being looked up
  is the current process.

* Improve kern_kill().  Do not obtain proc_token any more.  p->p_token
  is sufficient and the process group has its own lock now.

* Call pthread_yield() when spinning on various things.

    x Spinlocks
    x Tokens (spinning in lwkt_switch)
    x cpusync (ipiq)

* Rewrite sched_yield() -> dfly_yield().  dfly_yield() will
  unconditionally round-robin the LWP, ignoring estcpu.  It isn't
  perfect but it works fairly well.

  The dfly scheduler will also no longer attempt to migrate threads
  across cpus when handling yields.  They migrate normally in all
  other circumstances.

  This fixes situations where the vkernel is spinning waiting for multiple
  events from other cpus and in particular when it is doing a global IPI
  for pmap synchronization of the kernel_pmap.

7 months agoDisable making i915kms compilable into the kernel.
Sascha Wildner [Fri, 13 Sep 2013 21:58:26 +0000 (23:58 +0200)]
Disable making i915kms compilable into the kernel.

Right now, on systems with supported cards, it would cause the screen
go black before Xorg is started.

In-discussion-with: ftigeot

7 months agodrm.4: Update a bit to reflect the current state of affairs.
Sascha Wildner [Fri, 13 Sep 2013 21:54:32 +0000 (23:54 +0200)]
drm.4: Update a bit to reflect the current state of affairs.

While here, remove an obsolete MLINK, and add some more MLINKS to cover
module names too.

Based on a patch by ftigeot, with additional input by him.

7 months agotcp: Use mbuf jcluster (4K) in sosendtcp(); improve 10Ge TSO performance
Sepherosa Ziehau [Fri, 13 Sep 2013 09:26:22 +0000 (17:26 +0800)]
tcp: Use mbuf jcluster (4K) in sosendtcp(); improve 10Ge TSO performance

For the Myricom PCIE-8AL-C I have tested, this gives ~900Mbps performance
boost using 1500 MTU when TSO is enabled (from 7.7Gbps to 8.6Gbps).

Using mbuf jcluster could:
- Reduce the number of TX descriptors needed for one TSO packet.
- Let the NIC chip perform longer large data burst.

I believe this is the main reasons for the 10Ge performance boost.

Reduce sosend_agglim from 3 to 2, which means 8K aggregation (was 6K
aggregation) before user thread dispatches the sending buf to netisr.

net.inet.tcp.sosend_jcluster is added to enable this feature; it is
enabled by default.

7 months agombuf: Fix jcluster support
Sepherosa Ziehau [Fri, 13 Sep 2013 09:14:39 +0000 (17:14 +0800)]
mbuf: Fix jcluster support

- Free the jcluster to the correct objcache, which solves "jcluster mbuf"
  objcache exhaustion.
- By default, set the amount of the jclusters to half the amount of the
  normal clusters.  jcluster will be used on TCP sending path in the later
  commit to improve TSO performance for 10Ge.
- Add mbuf stat for jcluster; adjust netstat(1) to show it.
- Add m_getlj(), which will be used on TCP sending path in the later
  commit to improve TSO performnace for 10Ge.

7 months ago<sys/vnode.h>: Remove two unused externs.
Sascha Wildner [Thu, 12 Sep 2013 20:35:53 +0000 (22:35 +0200)]
<sys/vnode.h>: Remove two unused externs.

7 months agodirfs.5: Add missing .Ed
Sascha Wildner [Thu, 12 Sep 2013 16:57:12 +0000 (18:57 +0200)]
dirfs.5: Add missing .Ed

7 months agodrm: Replace the i915 driver by i915kms
François Tigeot [Thu, 12 Sep 2013 07:05:08 +0000 (09:05 +0200)]
drm: Replace the i915 driver by i915kms

i915kms was already an updated version of i915, there's no need to
keep maintaining two separate instances of the same driver.

7 months agofortune - Add yet another prompt tip.
Antonio Huete Jimenez [Thu, 12 Sep 2013 12:13:53 +0000 (14:13 +0200)]
fortune - Add yet another prompt tip.

7 months agomxge: Adjust interrupt moderation and TSO burst parameters
Sepherosa Ziehau [Thu, 12 Sep 2013 07:54:05 +0000 (15:54 +0800)]
mxge: Adjust interrupt moderation and TSO burst parameters

- Increase TSO burst to 48000 (32 x 1500B)
- Reduce interrupt rate from 30000Hz to 6000Hz

7 months agoin6: Fix DoS on INET6 w/ SIOCSIFADDR or SIOCSIFDSTADDR
Sepherosa Ziehau [Thu, 12 Sep 2013 02:36:40 +0000 (10:36 +0800)]
in6: Fix DoS on INET6 w/ SIOCSIFADDR or SIOCSIFDSTADDR

For reference:
http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/netinet6/in6.c?rev=1.104&content-type=text/x-cvsweb-markup&only_with_tag=MAIN

DragonFly-bug: http://bugs.dragonflybsd.org/issues/2581

7 months agotools: Add TCP_SENDFILE test to tcp_stream
Sepherosa Ziehau [Thu, 12 Sep 2013 01:54:32 +0000 (09:54 +0800)]
tools: Add TCP_SENDFILE test to tcp_stream

7 months agoem.4: I217 and I218 are supported by emx
Sepherosa Ziehau [Thu, 12 Sep 2013 01:49:10 +0000 (09:49 +0800)]
em.4: I217 and I218 are supported by emx

7 months agobuild: mxge_lro.c is gone
Sepherosa Ziehau [Thu, 12 Sep 2013 01:45:16 +0000 (09:45 +0800)]
build: mxge_lro.c is gone

7 months agomxge: Sync w/ FreeBSD
Sepherosa Ziehau [Fri, 6 Sep 2013 06:50:24 +0000 (14:50 +0800)]
mxge: Sync w/ FreeBSD

- Update firmware
- Sync w/ FreeBSD if_mxge.c rev 254263

Local changes:
- Adjust output path a little bit:
  o  Utilize bus_dmamap_load_mbuf_defrag()
  o  Utilize the header length fields in mbuf pkthdr instead of
     peeking at the packet data
- Change TSO mode to NDIS mode
- Minor white space changes
- Nuke mxge_lro.c; we won't need it

7 months agovkernel64 - Add newline to the end of options files.
Antonio Huete Jimenez [Wed, 11 Sep 2013 20:55:03 +0000 (22:55 +0200)]
vkernel64 - Add newline to the end of options files.

Spotted by: swildner

7 months agoacpi.4: Remove some unneeded commas, too.
Sascha Wildner [Wed, 11 Sep 2013 19:27:45 +0000 (21:27 +0200)]
acpi.4: Remove some unneeded commas, too.

7 months agoacpi.4: Fix typo, Diabling -> Disabling
Sascha Wildner [Wed, 11 Sep 2013 19:24:20 +0000 (21:24 +0200)]
acpi.4: Fix typo, Diabling -> Disabling

7 months agoifnet - Compute iface statistics for NET_RT_IFLIST
Antonio Huete Jimenez [Tue, 10 Sep 2013 11:11:32 +0000 (04:11 -0700)]
ifnet - Compute iface statistics for NET_RT_IFLIST

  Before the ifnet statistics went per-cpu, they were being stored in
  ifnet->if_data. Some functions, like getifaddrs(3), expect them to
  be there so make sure they are available.

Reported by: Tim Bishop, Jens Rehsack

7 months agosystat - Fix memory leak in -netbw
Matthew Dillon [Tue, 10 Sep 2013 05:26:02 +0000 (22:26 -0700)]
systat - Fix memory leak in -netbw

* Fix memory leak in -netbw.  We weren't checking for deletable
  entries once the display row hit its limit.

Reported-by: Peter Avalos
7 months agokernel - Fix process blocking forever if ^Z'd at the wrong time
Matthew Dillon [Tue, 10 Sep 2013 04:46:55 +0000 (21:46 -0700)]
kernel - Fix process blocking forever if ^Z'd at the wrong time

* tsleep()s with timeouts were testing the wrong state field
  when determining whether a stopped thread should be woken up or not.

* Generally occured under certain circumstances when a ^Z'd process
  was then killed without first continuing it.

7 months agohammer(8) - Migration to libhammer (step 2/many)
Antonio Huete Jimenez [Mon, 9 Sep 2013 14:18:42 +0000 (16:18 +0200)]
hammer(8) - Migration to libhammer (step 2/many)

* Migrate 'bstats' and 'iostats' directives.
* Adjust cmd_stats.c file width to 80 columns

7 months agohammer(8) - Adapt help output to latest changes.
Antonio Huete Jimenez [Mon, 9 Sep 2013 13:33:27 +0000 (15:33 +0200)]
hammer(8) - Adapt help output to latest changes.

* Also adapt hammer.8 manpage.

7 months agolibhammer_get_volinfo.3 - Use commas to separate functions names.
Antonio Huete Jimenez [Mon, 9 Sep 2013 11:44:08 +0000 (13:44 +0200)]
libhammer_get_volinfo.3 - Use commas to separate functions names.

7 months agolibhammer - Add functions to help gather HAMMER statistics.
Antonio Huete Jimenez [Mon, 9 Sep 2013 11:32:41 +0000 (13:32 +0200)]
libhammer - Add functions to help gather HAMMER statistics.

7 months agoem.4: Remove trailing whitespace.
Sascha Wildner [Sun, 8 Sep 2013 17:08:05 +0000 (19:08 +0200)]
em.4: Remove trailing whitespace.

7 months agonrelease: Remove xorg.conf from non-gui images
François Tigeot [Sat, 7 Sep 2013 21:16:00 +0000 (23:16 +0200)]
nrelease: Remove xorg.conf from non-gui images

* Left by itself, Xorg autodetects hardware drivers and other essential
  parameters necessary for its operation

* The shipped xorg.conf file was forcing wrong defaults on many machines,
  leading to broken Xorg installations

* This was a particularly acute problem with i915 hardware

7 months agolibhammer - Fix another memory leak.
Antonio Huete Jimenez [Sat, 7 Sep 2013 19:41:26 +0000 (21:41 +0200)]
libhammer - Fix another memory leak.

* Free a pointer when allocated by asprintf()

Spotted-by: swildner via cppcheck
7 months agolibhammer - Use source directory header instead of the system one.
Antonio Huete Jimenez [Sat, 7 Sep 2013 18:14:18 +0000 (20:14 +0200)]
libhammer - Use source directory header instead of the system one.

* Instead of using libhammer.h from /usr/include, use the one that's
  in the source directory to build the library itself.

7 months agohammer(8) - Allow specifying paths to 'info' directive.
Antonio Huete Jimenez [Sat, 7 Sep 2013 18:06:48 +0000 (20:06 +0200)]
hammer(8) - Allow specifying paths to 'info' directive.

7 months agolibhammer - Few fixes to libhammer_get_volinfo()
Antonio Huete Jimenez [Sat, 7 Sep 2013 17:49:47 +0000 (19:49 +0200)]
libhammer - Few fixes to libhammer_get_volinfo()

* Fix a couple memory leaks.
* Reword some comments.
* While here fix info.c file width to 80 columns.

7 months agohammer(8) - Remove 'status' directive.
Antonio Huete Jimenez [Sat, 7 Sep 2013 16:17:59 +0000 (18:17 +0200)]
hammer(8) - Remove 'status' directive.

  It wasn't working, it was undocumented and directive 'pfs-status'
  supersedes it.

7 months agorconfig - CCD mirror script
Antonio Huete Jimenez [Sat, 7 Sep 2013 11:30:27 +0000 (04:30 -0700)]
rconfig - CCD mirror script

* Add a script to automate installations on a disk mirrored
  setup using ccd(4)

7 months agomxge: Rearrange building
Sepherosa Ziehau [Fri, 6 Sep 2013 06:19:10 +0000 (14:19 +0800)]
mxge: Rearrange building

7 months agoecc: Use __SHIFTOUT macro; add missing line break in error logging
Sepherosa Ziehau [Fri, 6 Sep 2013 05:13:42 +0000 (13:13 +0800)]
ecc: Use __SHIFTOUT macro; add missing line break in error logging

7 months agokernel - Fix minor reporting bug in ecc update
Charlie Root [Fri, 6 Sep 2013 05:04:47 +0000 (22:04 -0700)]
kernel - Fix minor reporting bug in ecc update

* Fix a minor constant comparison test when reporting invalid
  ECC modes.  The constants were normalized to bit 0 but the
  register extraction had not been shifted over.

7 months agoecc: Add E3-1200 v2 and E3-1200 V3 support
Sepherosa Ziehau [Fri, 6 Sep 2013 04:29:27 +0000 (12:29 +0800)]
ecc: Add E3-1200 v2 and E3-1200 V3 support

7 months agokernel/ddb: Fix formatting when printing command tables.
Sascha Wildner [Thu, 5 Sep 2013 20:38:10 +0000 (22:38 +0200)]
kernel/ddb: Fix formatting when printing command tables.

FreeBSD's r163134:

Fixed formatting of printing of command tables.  With the default max
output width of 79, only 6 columns of width 12 each fit, but 7 columns
were printed.

The fix is to pass the width of the next output to db_end_line() and
not assume there that this width is always 1.

Related unfixed bugs:
- 1 character is wasted for a space after the last column
- suppression of trailing spaces used to limit the misformatting, but
  seems to have been lost
- in db_examine(), the width of the next output is not known and is
  still assumed to be 1.

Taken-from: FreeBSD

7 months agokernel - Change time_second to time_uptime for all expiration calculations (2)
Matthew Dillon [Thu, 5 Sep 2013 19:19:25 +0000 (12:19 -0700)]
kernel - Change time_second to time_uptime for all expiration calculations (2)

* Fix userspace issues calculating rmx_expire for display due to
  time_second -> time_uptime change.

  The route, arp, ndp, and rarpd utilities are affected by this change.

7 months agokernel - Change time_second to time_uptime for all expiration calculations
Matthew Dillon [Thu, 5 Sep 2013 19:06:13 +0000 (12:06 -0700)]
kernel - Change time_second to time_uptime for all expiration calculations

* Vet the entire kernel and change use cases for expiration calculations
  using time_second to use time_uptime instead.

* Protects these expiration calculations from step changes in the wall time,
  particularly needed for route table entries.

* Probably requires further variable type adjustments but the use of
  time_uptime instead if time_second is highly unlikely to ever overrun
  any demotions to int still present.

7 months agokernel - ipv6 - Fix broken 'invalid mtu expiration time' message
Matthew Dillon [Thu, 5 Sep 2013 17:00:12 +0000 (10:00 -0700)]
kernel - ipv6 - Fix broken 'invalid mtu expiration time' message

* Fix broken 'invalid mtu expiration time' message on the console.

Reported-by: tuxillo
7 months agotest/vkernel - Fix build.
Antonio Huete Jimenez [Thu, 5 Sep 2013 13:58:27 +0000 (06:58 -0700)]
test/vkernel - Fix build.

7 months agoBring in DIRFS: A filesystem for VKERNELS
Antonio Huete Jimenez [Wed, 22 Aug 2012 12:53:18 +0000 (14:53 +0200)]
Bring in DIRFS: A filesystem for VKERNELS

* What is DIRFS?
  dirfs is a pseudo-filesystem specific for vkernel(7) which allows mounting
  host's directories into the vkernel. It runs directly in the vkernel's VFS
  code, as any other regular filesystem, but it does syscalls (vkernels are
  userland programs) to retrieve or post the information needed on every
  operation requested.

  Needless to say that the operations that you can perform in the host
  directories/files depend on the permissions the user that runs the vkernel.
  For example, you will not be able to 'chflags schg' if you run the vkernel
  with a regular user and not with root.

* How does it work?
  It basically works like any other filesystem. It has its own mount_dirfs
  command that will be called by the system's mount(8) command when needed.

vkernel64 # mount -t dirfs /usr/src2 /mnt
vkernel64 # df -h /mnt
Filesystem        Size   Used  Avail Capacity  Mounted on
dirfs@/usr/src2    47G    36G    12G    75%    /mnt

  Umounting is a normal operation too:

  vkernel64 # mount | fgrep dirfs
dirfs@/usr/src2 on /mnt (dirfs)
vkernel64 # umount /mnt

* What's the current status
  Currently it is in a *experimental* status, with (probably) many bugs and some
  parts missing.

TODO
  - Make dirfs mpsafe.
  - Fix problems with multiple mount points.
  - Implement VOP_NLINK so that hardlinks are possible.
  - Add missing kqueue(2) support.
  - dirfs root so that a vkernel can be booted from it.
  - Locking mechanisms for opened fds between host <-> vkernel.
  - Make sure dirfs is properly restored after vkernel checkpointing
    (upcoming GSoC project).
  - Bug hunting & bug fixing.
  - Any ideas?

7 months agokernel - ipv6 - correct route table callout timer calculation
Matthew Dillon [Wed, 4 Sep 2013 23:27:56 +0000 (16:27 -0700)]
kernel - ipv6 - correct route table callout timer calculation

* Correct the route table callout timer calculation which geometrically
  increases to the point where it isn't called any more.

7 months agode.4: Mention that Microsoft Virtual PC and Hyper-V have de(4) adapters.
Sascha Wildner [Wed, 4 Sep 2013 22:14:37 +0000 (00:14 +0200)]
de.4: Mention that Microsoft Virtual PC and Hyper-V have de(4) adapters.

Therefore, we'd like to keep this driver.

Hyper-V also comes with its specific virtual adapter, but we don't have
support for that yet.

7 months agokernel - IPV6 subnet routing / proxy ND6 (bridge support)
Matthew Dillon [Wed, 4 Sep 2013 20:59:14 +0000 (13:59 -0700)]
kernel - IPV6 subnet routing / proxy ND6 (bridge support)

* Support IPV6 subnet routing and proxy ND6 through bridged interfaces.
  Essentially all the members of the bridge have to be treated as the
  bridge itself for comparison and MAC address handling.

* Prevents the nd6 proxy code from treating two interfaces which are part
  of the same bridge as being different, which would trigger an improper
  proxy ND6.

7 months agolibsmb: Allow libsmb to detect if smbfs.ko is loaded.
Sascha Wildner [Wed, 4 Sep 2013 20:13:04 +0000 (22:13 +0200)]
libsmb: Allow libsmb to detect if smbfs.ko is loaded.

7 months agokernel - Implement IPV6 subnet routing / proxy ND6 (equiv to proxy ARP)
Matthew Dillon [Wed, 4 Sep 2013 17:39:55 +0000 (10:39 -0700)]
kernel - Implement IPV6 subnet routing / proxy ND6 (equiv to proxy ARP)

* Do not require per-host RTF_ANNOUNCE/AF_LINK entries.  They still
  work but they aren't needed any more (and they are such a huge bitch
  to set up anyway... best to avoid them).

* Machine must have net.inet6.ip6.forwarding mode enabled.

* Internet-facing interface must be promiscuous mode.

* Will automatically proxy ND6 any subnets if the interface
  is different from the one receiving the multicast.  So e.g.
  you can route IPV6 which would otherwise have to be switched.

  The subnet interface must currently be different because if it
  were the same the solicitation would be directly received by the
  target host anyway (being a multicast) and we would compete with
  it.  This is also a good safety.

  Example:

  ifconfig igb0 inet6 2999:499:1:555:1::72/80 For DNS
  ifconfig igb0 inet6 2999:499:1:555:1::1/80 For subnet default route
  ifconfig igb1 inet6 2999:499:1:555::2/80 For internet router
  ifconfig igb1 promisc
  route add -inet6 default 2999:499:1:555::1

  The internet router is doing a terminal /64 block, e.g. it's address
  is 2999:499:1:555::1/64, but we want to break the net up further and
  route portions of it instead of switch.

7 months agokernel/drm: Fix AGP detection for Matrox cards.
Sascha Wildner [Wed, 4 Sep 2013 16:18:59 +0000 (18:18 +0200)]
kernel/drm: Fix AGP detection for Matrox cards.

Since DRM drivers attach to vgapci, we need to look at the grandparent.

7 months agoem.4: I217 and I218 are taken by emx
Sepherosa Ziehau [Wed, 4 Sep 2013 13:33:18 +0000 (21:33 +0800)]
em.4: I217 and I218 are taken by emx

7 months agoemx: Add I217/I218
Sepherosa Ziehau [Wed, 4 Sep 2013 12:36:11 +0000 (20:36 +0800)]
emx: Add I217/I218

According to the datasheet, these two chips support 2 RX queues at least.

7 months agotmpfs - Remove duplicated checks in tmpfs_chflags()
Antonio Huete Jimenez [Wed, 4 Sep 2013 10:22:27 +0000 (12:22 +0200)]
tmpfs - Remove duplicated checks in tmpfs_chflags()

* Do not duplicate the checks that vop_helper_setattr_flags() already performed.
* On tmpfs_mount(), root node already has set SF_NOCACHE on its vnode.

7 months agovkernel32 - Change VM_MAX_USER_ADDRESS to match the real kernel.
Antonio Huete Jimenez [Wed, 4 Sep 2013 10:13:45 +0000 (12:13 +0200)]
vkernel32 - Change VM_MAX_USER_ADDRESS to match the real kernel.

Pointed-out-by: dillon
7 months agoem.4: TSO is supported on all PCI-E chips; properly mention I217/I218
Sepherosa Ziehau [Wed, 4 Sep 2013 09:39:10 +0000 (17:39 +0800)]
em.4: TSO is supported on all PCI-E chips; properly mention I217/I218

7 months agore: Add 8401/8402/8106 support
Sepherosa Ziehau [Wed, 4 Sep 2013 09:25:34 +0000 (17:25 +0800)]
re: Add 8401/8402/8106 support

- Fix setup for certain generation of 8105E
- 8101, 8102 and 8105 all need to extract ethernet address from EEPROM

7 months agoem: Move max_frame_size from softc to HAL data struct
Sepherosa Ziehau [Wed, 4 Sep 2013 08:30:29 +0000 (16:30 +0800)]
em: Move max_frame_size from softc to HAL data struct

It is needed to make I217 link status detection work

7 months agore: Add 8411/8168G/8168EP/8168GU/8411B support
Sepherosa Ziehau [Wed, 4 Sep 2013 08:27:09 +0000 (16:27 +0800)]
re: Add 8411/8168G/8168EP/8168GU/8411B support

7 months agore: Remove the macver, which only serves proper EEPROM address handling
Sepherosa Ziehau [Wed, 4 Sep 2013 06:31:14 +0000 (14:31 +0800)]
re: Remove the macver, which only serves proper EEPROM address handling

7 months agore: Add more hardware revision for 10/100 chips
Sepherosa Ziehau [Wed, 4 Sep 2013 06:02:41 +0000 (14:02 +0800)]
re: Add more hardware revision for 10/100 chips

7 months agore: Add more hardware revision for 8411 and newer 8168
Sepherosa Ziehau [Wed, 4 Sep 2013 05:47:35 +0000 (13:47 +0800)]
re: Add more hardware revision for 8411 and newer 8168

7 months agore: Change the hardware revision mask
Sepherosa Ziehau [Wed, 4 Sep 2013 05:17:35 +0000 (13:17 +0800)]
re: Change the hardware revision mask

This keeps us align w/ Realtek's 8168/8101 Linux driver

7 months agoig_hal: Move e1000_phy_i217 close to e1000_phy_82579
Sepherosa Ziehau [Wed, 4 Sep 2013 01:40:43 +0000 (09:40 +0800)]
ig_hal: Move e1000_phy_i217 close to e1000_phy_82579

I217 belongs to the previous generation of Intel PRO/1000

7 months agore: Always log unknown hardware revision
Sepherosa Ziehau [Wed, 4 Sep 2013 01:30:07 +0000 (09:30 +0800)]
re: Always log unknown hardware revision

7 months agoem: Disable EEE by default
Sepherosa Ziehau [Wed, 4 Sep 2013 01:29:45 +0000 (09:29 +0800)]
em: Disable EEE by default

7 months agoRemove old drmn.ko module via 'make upgrade'.
Sascha Wildner [Tue, 3 Sep 2013 17:38:30 +0000 (19:38 +0200)]
Remove old drmn.ko module via 'make upgrade'.

7 months agokernel: Remove drmn leftover.
Sascha Wildner [Tue, 3 Sep 2013 17:33:49 +0000 (19:33 +0200)]
kernel: Remove drmn leftover.

7 months agodisklabel64 - Increased verification of the partition format.
Antonio Huete Jimenez [Tue, 3 Sep 2013 12:02:00 +0000 (14:02 +0200)]
disklabel64 - Increased verification of the partition format.

* Check whether the filesystem type is specified when
  formatting partitions.

  Specifying:
a: * 0

  would cause a SEGFAULT due the filesystem field not being present.

Fix has been taken from disklabel32.

7 months agotest/vkernel - Allow using HAMMER filesystem as vkernel root.
Antonio Huete Jimenez [Tue, 3 Sep 2013 11:54:10 +0000 (13:54 +0200)]
test/vkernel - Allow using HAMMER filesystem as vkernel root.

7 months agokernel/i386: Implement pmap_invalidate_cache_pages()
François Tigeot [Tue, 3 Sep 2013 13:11:02 +0000 (15:11 +0200)]
kernel/i386: Implement pmap_invalidate_cache_pages()

7 months agoigb.4: Mention I354
Sepherosa Ziehau [Tue, 3 Sep 2013 12:28:23 +0000 (20:28 +0800)]
igb.4: Mention I354

7 months agoem.4: Update date
Sepherosa Ziehau [Tue, 3 Sep 2013 12:20:40 +0000 (20:20 +0800)]
em.4: Update date

7 months agoigb: Merge Intel igb-2.3.10
Sepherosa Ziehau [Tue, 3 Sep 2013 12:05:07 +0000 (20:05 +0800)]
igb: Merge Intel igb-2.3.10

- More I210 chips are supported
- Add I354 support
- Fix multicast filtering settings if promisc mode exits

7 months agoem.4: Update
Sepherosa Ziehau [Tue, 3 Sep 2013 08:50:35 +0000 (16:50 +0800)]
em.4: Update

7 months agoem: Enable TSO on all PCI-E chips
Sepherosa Ziehau [Tue, 3 Sep 2013 08:30:58 +0000 (16:30 +0800)]
em: Enable TSO on all PCI-E chips

7 months agoem: Merge Intel em-7.3.8; mainly add i217 and i218 support
Sepherosa Ziehau [Tue, 3 Sep 2013 08:23:58 +0000 (16:23 +0800)]
em: Merge Intel em-7.3.8; mainly add i217 and i218 support

7 months agokernel/x86_64: Unbreak kernel building with drm in the config.
Sascha Wildner [Tue, 3 Sep 2013 06:16:45 +0000 (08:16 +0200)]
kernel/x86_64: Unbreak kernel building with drm in the config.

7 months agoig_hal: Merge Intel em-7.3.8 and igb-2.3.10 hardware abstraction layer
Sepherosa Ziehau [Fri, 30 Aug 2013 08:12:22 +0000 (16:12 +0800)]
ig_hal: Merge Intel em-7.3.8 and igb-2.3.10 hardware abstraction layer

7 months agokernel/ahci: Add support for nvidia MCP79 AHCI.
Sascha Wildner [Mon, 2 Sep 2013 18:25:24 +0000 (20:25 +0200)]
kernel/ahci: Add support for nvidia MCP79 AHCI.

Submitted-by: Ed Berger
Dragonfly-bug: <http://bugs.dragonflybsd.org/issues/2583>

7 months agokernel: Fix compilation errors on i386
François Tigeot [Mon, 2 Sep 2013 07:35:59 +0000 (09:35 +0200)]
kernel: Fix compilation errors on i386

7 months agopppd.8: Fix two references.
Sascha Wildner [Sun, 1 Sep 2013 21:34:31 +0000 (23:34 +0200)]
pppd.8: Fix two references.

7 months agodrm: Merge the drm and drm2 implementations
François Tigeot [Sun, 1 Sep 2013 19:38:11 +0000 (21:38 +0200)]
drm: Merge the drm and drm2 implementations

* Move the drm2 i915 driver to sys/dev/drm/i915kms

* The legacy i915 driver remains untouched

* Only sys/dev/drm/ shall remain

7 months agodrm/mach_64: Fix white spaces.
jkim [Tue, 23 Apr 2013 18:30:33 +0000 (18:30 +0000)]
drm/mach_64: Fix white spaces.

This brings us on part with FreeBSD commit r249816 (2013-04-23 20:30:33 +0200)

7 months agokernel: Rewrite kmem_alloc_attr()
François Tigeot [Sun, 1 Sep 2013 09:20:14 +0000 (11:20 +0200)]
kernel: Rewrite kmem_alloc_attr()

This prevents random kernel panics after Xorg exit with the new
drm scather-gather memory allocation code

7 months agodrm: Partial sync with FreeBSD
François Tigeot [Fri, 30 Aug 2013 06:46:37 +0000 (08:46 +0200)]
drm: Partial sync with FreeBSD

7 months agopppd.8: conversion to mdoc(7)
Franco Fichtner [Sat, 31 Aug 2013 10:34:57 +0000 (12:34 +0200)]
pppd.8: conversion to mdoc(7)

Major overhaul of a well-written manual page.  The content has
been shuffled around and subtly cleaned where appropriate.

Suggested-by: swildner
7 months agokernel: Fix some void functions returning the results of void functions.
Sascha Wildner [Sat, 31 Aug 2013 07:56:00 +0000 (09:56 +0200)]
kernel: Fix some void functions returning the results of void functions.

It was found with -pedantic, since -Wreturn-type will not warn about this
as long as the calling function has a return type of void, too.

7 months agosystat - cleanup netbw/pftop (2)
Matthew Dillon [Fri, 30 Aug 2013 23:34:40 +0000 (16:34 -0700)]
systat - cleanup netbw/pftop (2)

* Cleanup pass, fix a few bugs.

7 months agosystat - cleanup netbw/pftop
Matthew Dillon [Fri, 30 Aug 2013 23:22:57 +0000 (16:22 -0700)]
systat - cleanup netbw/pftop

* Cleanup pass, fix a few bugs.

7 months agosystat - Add -pftop feature to display active PF state table traffic
Matthew Dillon [Fri, 30 Aug 2013 22:30:19 +0000 (15:30 -0700)]
systat - Add -pftop feature to display active PF state table traffic

* Add the -pftop (:pftop) feature.  This feature displays all active
  PF states.  PF (packet filter) must be active for this feature to be
  useful.

  Only those states which are actively passing traffic are displayed,
  which makes this a lot more useful than a raw 'pfctl -s state' dump.

* Display is sorted by highest aggregate receive+transmit bandwidth.

7 months agoclog.8: break line after a sentence
Franco Fichtner [Fri, 30 Aug 2013 08:22:32 +0000 (10:22 +0200)]
clog.8: break line after a sentence

7 months agosyslog - add rexmit to -netbw
Matthew Dillon [Fri, 30 Aug 2013 06:41:08 +0000 (23:41 -0700)]
syslog - add rexmit to -netbw

* Add aggregate byte retransmission rate

7 months agosystat - further enhance -netbw
Matthew Dillon [Fri, 30 Aug 2013 06:36:21 +0000 (23:36 -0700)]
systat - further enhance -netbw

* Print IPs in a more readable fashion
* A better IP:port sort order
* Better alignment
* Add interesting state and TCP flags

7 months agosystat - Add -netbw option
Matthew Dillon [Fri, 30 Aug 2013 04:26:09 +0000 (21:26 -0700)]
systat - Add -netbw option

* Add a new option -netbw (or :netbw when running).  This option displays
  the aggregate and per-connection tcp receive and send data rate.

  Only active connections are shown, making it easy to pick out which
  connections are hogging the line.

7 months agokernel/ed: Fix typo in a variable name.
Sascha Wildner [Fri, 30 Aug 2013 02:48:45 +0000 (04:48 +0200)]
kernel/ed: Fix typo in a variable name.

7 months agotest/vkernel - Additions to leaf target
Antonio Huete Jimenez [Fri, 30 Aug 2013 00:06:38 +0000 (17:06 -0700)]
test/vkernel - Additions to leaf target

* Add nameservers to resolv.conf

7 months agofix mandoc(1) warnings in usr.sbin/
Franco Fichtner [Tue, 27 Aug 2013 20:38:19 +0000 (22:38 +0200)]
fix mandoc(1) warnings in usr.sbin/

7 months agotools/accept_connect: Remove unused .gitignore
Sepherosa Ziehau [Thu, 29 Aug 2013 06:15:36 +0000 (14:15 +0800)]
tools/accept_connect: Remove unused .gitignore

7 months agotcp: Implement random initial msgport
Sepherosa Ziehau [Wed, 28 Aug 2013 07:11:43 +0000 (15:11 +0800)]
tcp: Implement random initial msgport

After e368a6e95e2cd9556a3e0fc43167d2dcf3a8253f, it is found that the
left bottle neck of nonblocking TCP connect(2) performance is that all
socket(2) and initial TCP connect operation (bind laddr and lport) are
all carried out in netisr0; CPU0 is 100% busy during test.

The idea of random initial msgport for TCP is that instead of using
netisr0's msgport as initial msgport, we could use any of the available
netisr msgport to carry out socket(2) and initial TCP connect operation.

Most of parts of TCP are already ready for random initial msgport, only
TCP pru_listen requires trivial modification to fix the socket msgport
to netisr0's msgport (which is required to perform global wild hashtable
updating).

As of this commit, the current CPU's netisr msgport will be selected as
TCP socket's initial msgport, if random initial msgport is enabled.

Sysctl node kern.ipc.rand_initport is added to disable this optimization.
It is enabled by default.

This commit improves both nonblocking TCP connect(2) and blocking TCP
connect(2) performance.

Nonblocking connect(2) performance measurement (i7-2600 w/ bnx(4)), using
tools/tools/netrate/accept_connect/kq_connect_client:

    kq_connect_client -4 SERVADDR -p SERVPORT -i 8 -c 32 -l 30
    (8 processes, each creates 32 connections simultaniously)

16 run average:
    random initial msgport          netisr0 msgport
      263915.17 conns/s            220979.89 conns/s

This commit gives ~19% performance improvement for nonblocking connect(2)

Blocking connect(2) performance measurement (i7-2600 w/ bnx(4)), using
tools/tools/netrate/accept_connect/connect_client:

    connect_client -4 SERVADDR -p SERVPORT -i 256 -l 30
    (256 processes)

16 run average:
    random initial msgport          netisr0 msgport
      240235.23 conns/s            198312.87 conns/s

This commit gives ~21% performance improvement for blocking connect(2)

7 months agoinpcb: Make sure that inp_cpcbinfo always points to the correct pcbinfo
Sepherosa Ziehau [Wed, 28 Aug 2013 06:45:05 +0000 (14:45 +0800)]
inpcb: Make sure that inp_cpcbinfo always points to the correct pcbinfo