Sepherosa Ziehau [Wed, 4 Jul 2012 09:32:09 +0000 (17:32 +0800)]
bge: Correct comment
Sepherosa Ziehau [Wed, 4 Jul 2012 09:23:26 +0000 (17:23 +0800)]
bge: Factor our bge_link_poll()
Sepherosa Ziehau [Wed, 4 Jul 2012 09:15:24 +0000 (17:15 +0800)]
bge: Add comment about coalesce parameters during interrupt
Markus Pfeiffer [Wed, 4 Jul 2012 21:27:17 +0000 (21:27 +0000)]
kernel - Correct whitespace in BUS_DMA_KEEP_PG_OFFSET (x86_64)
Markus Pfeiffer [Wed, 4 Jul 2012 20:54:55 +0000 (20:54 +0000)]
mii: add AGERE ET1011 phy
Markus Pfeiffer [Wed, 4 Jul 2012 13:31:38 +0000 (13:31 +0000)]
kernel - Add BUS_DMA_KEEP_PG_OFFSET (i386)
Add BUS_DMA_KEEP_PG_OFFSET from FreeBSD in preparation of import of
FreeBSD's USB stack
Markus Pfeiffer [Wed, 4 Jul 2012 13:21:25 +0000 (13:21 +0000)]
kernel - Add BUS_DMA_KEEP_PG_OFFSET (x86_64)
Add BUS_DMA_KEEP_PG_OFFSET from FreeBSD in preparation of import of
FreeBSD's USB stack
Markus Pfeiffer [Wed, 4 Jul 2012 13:12:46 +0000 (13:12 +0000)]
kernel: add handling for ENODEV in poll_copyout and select_copyout
Peter Avalos [Wed, 4 Jul 2012 07:55:58 +0000 (00:55 -0700)]
Merge branch 'vendor/FILE'
Peter Avalos [Wed, 4 Jul 2012 07:49:04 +0000 (00:49 -0700)]
Import file-5.11.
* Fix CDF parsing issues.
Sepherosa Ziehau [Tue, 3 Jul 2012 09:39:58 +0000 (17:39 +0800)]
bge: Enable "tagged status" on non-BCM5788 and non-BCM5700 chips
By using "tagged status", we could:
- Detect whether the interrupt is for us or not.
- Avoid MAC_STS register read upon every interrupt.
Some implementation note:
- Interrupt must be disabled during interrupt handing, if "tagged status"
is used, else the interrupt will be kept asserted. This is the case at
least on BCM5750 AX/BX.
- Coalesce parameters during interrupt must be properly configured; when
"tagged status" is used we have "in interrupt" state.
Tunable hw.bgeX.status_tag is added to prevent driver from using "tagged
status". hw.bgeX.{rx,tx}_coal_bds_int and hw.bgeX.{rx,tx}_coal_ticks_int
sysctls are added to tune "in interrupt" coalesce parameters.
Matthew Dillon [Tue, 3 Jul 2012 17:42:03 +0000 (10:42 -0700)]
kernel - reduce kernel stack use (2)
* Correct hanging allocation on kldunload from previous commit.
* Rearrange the flow of the code a little for the exit case.
Matthew Dillon [Tue, 3 Jul 2012 17:22:09 +0000 (10:22 -0700)]
kernel - reduce kernel stack use
* fq_dispatcher() was eating 8K+ of kernel stack on x86-64 with a
(512 x sizeof(pointer) * 2) sized array. This is way too much kernel
stack.
* Add a safety to the hz calculation in the lksleep() for hz values < 15.
vkernels already use low hz values of ~20-ish.
Reported-by: swildner
Sepherosa Ziehau [Tue, 3 Jul 2012 08:18:21 +0000 (16:18 +0800)]
mii: Wrong sizeof
Pointed-by: swildner@
Sepherosa Ziehau [Tue, 3 Jul 2012 06:27:19 +0000 (14:27 +0800)]
bge: BGE_PCI_MISC_CTL resides in PCI configuration space
Sepherosa Ziehau [Tue, 3 Jul 2012 05:34:33 +0000 (13:34 +0800)]
bge: Regroup code, no functional changes
- Regroup chip information and bugs setup code
- Regroup MII information setup code
Mihai Carabas [Sun, 1 Jul 2012 19:18:25 +0000 (22:18 +0300)]
sbuf_vprintf: ap is not consistent at second call of kvsnprintf. Use a copy
Sepherosa Ziehau [Mon, 2 Jul 2012 08:43:41 +0000 (16:43 +0800)]
bge: Don't force PCI Express TLP to be 128 bytes
Instead, status errors are cleared and "no snoop" and "relaxed ordering"
are disabled.
Obtained-from: FreeBSD 235821
Sepherosa Ziehau [Mon, 2 Jul 2012 08:42:59 +0000 (16:42 +0800)]
pci: Add more definition for PCI Express device {control,status} registers
Sepherosa Ziehau [Mon, 2 Jul 2012 08:20:05 +0000 (16:20 +0800)]
bge: Workaround mbox write reordering for PCI-X devices
Add per device tunable hw.bgeX.mbox_reorder to control this; it is
off by default.
Based-on: FreeBSD 232848
Sepherosa Ziehau [Mon, 2 Jul 2012 08:14:35 +0000 (16:14 +0800)]
bge: If chip has CPMU, let it handle GPHY power down
Obtained-from: FreeBSD 226804
Sepherosa Ziehau [Mon, 2 Jul 2012 08:11:44 +0000 (16:11 +0800)]
bge: Always enable buffer manager
Also make sure various blocks are stopped in bge_stop()
Obtained-from: NetBSD via FreeBSD 221712
Sepherosa Ziehau [Mon, 2 Jul 2012 07:56:59 +0000 (15:56 +0800)]
bge: Workaround BCM5906 chip bug
When auto-negotiation results in half-duplex operation, excess collision
on the ethernet link may cause internal chip delays that may result in
subsequent valid frames being dropped due to insufficient receive buffer
resources. The workaround is to choose de-pipeline method as a flow
control decision for SDI. De-pipeline method allows only 1 data in TxMbuf
at a time such that a request to RDMA from SDI is made only when TxMbuf
is empty.
Obtained-from: FreeBSD 214219
Sepherosa Ziehau [Mon, 2 Jul 2012 07:49:22 +0000 (15:49 +0800)]
bge: Enable hardware patch for TX MAC lockup on BCM5755+ and BCM5906
Obtained-from: FreeBSD 214216
Sepherosa Ziehau [Mon, 2 Jul 2012 07:45:30 +0000 (15:45 +0800)]
bge: Workaround "short DMA" bug on certain chips
If controllers receive two back-to-back send BDs with less than or equal
to 8 total bytes then the device may hang. The two back-to-back send
BDs must be in the same frame for this failure to occur.
Obtained-from: FreeBSD 214087
Sepherosa Ziehau [Mon, 2 Jul 2012 07:12:20 +0000 (15:12 +0800)]
bge: Avoid accessing invalid internal memory region on BCM5906
Obtained-from: FreeBSD 213747
Sepherosa Ziehau [Mon, 2 Jul 2012 06:59:48 +0000 (14:59 +0800)]
bge: Properly configure receive and send RCBs
- More comment
- Limit receive and RCBs accessing according to chip type
- Always configure send RCB
- Move producer RX ring index resetting to the proper place
Obtained-from: FreeBSD 213495
Sepherosa Ziehau [Mon, 2 Jul 2012 06:19:20 +0000 (14:19 +0800)]
bge: Force interrupt if link status changes
Obtained-from: FreeBSD 178785
Sepherosa Ziehau [Mon, 2 Jul 2012 05:30:04 +0000 (13:30 +0800)]
bge: Rework MII register accessing and auto-polling support
Obtained-from: FreeBSD 213485 213710 213808 213812
Sascha Wildner [Mon, 2 Jul 2012 08:13:02 +0000 (10:13 +0200)]
ixgbe.4: Remove trailing whitespace.
Sepherosa Ziehau [Mon, 2 Jul 2012 02:56:27 +0000 (10:56 +0800)]
bge: Adjust PHY workaround settings
Obtained-from: OpenBSD if_bge.c 1.309
Sepherosa Ziehau [Mon, 2 Jul 2012 02:33:54 +0000 (10:33 +0800)]
mii: Define MII_PROBEMASK_DEFAULT and MII_CAPMASK_DEFAULT
Avoid using magic number
Sepherosa Ziehau [Mon, 2 Jul 2012 02:10:36 +0000 (10:10 +0800)]
bge/brgphy: Split PHY flags out of main flags
Sascha Wildner [Mon, 2 Jul 2012 02:02:46 +0000 (04:02 +0200)]
<net/if_media.h>: Add 10Gb ethernet baudrate descriptions.
Sascha Wildner [Mon, 2 Jul 2012 02:02:31 +0000 (04:02 +0200)]
kernel/net: On i386, make the interface baudrate 64 bits in a few places.
For 10Gb ethernet.
For ifi_baudrate in struct if_data (which is public), NetBSD uses
uint64_t too, so no pkgsrc breakage is expected. However, a few
i386 binary packages we offer might be broken now on master.
Discussed-with: sephe
Sepherosa Ziehau [Mon, 2 Jul 2012 01:49:15 +0000 (09:49 +0800)]
bge: Enable fix for read DMA FIFO overruns on certain controllers
Obtained-from: FreeBSD 213411
Sepherosa Ziehau [Mon, 2 Jul 2012 01:28:13 +0000 (09:28 +0800)]
bge: NetExtreme/NetLink controllers require memory not across 4G boundary
Obtained-from: FreeBSD 226805
Sepherosa Ziehau [Sun, 1 Jul 2012 23:26:11 +0000 (07:26 +0800)]
bge: Utilize new mii_probe()
Thus avoid local hacking about fixed PHY in miibus_{read,write}reg().
Sepherosa Ziehau [Sun, 1 Jul 2012 23:17:30 +0000 (07:17 +0800)]
mii: Allow NIC drivers to pass more information to MII probe
- mii_capmask, which is used to mask out uninstended BMSR bits.
- At which PHYs that MII should perform probing and attching
(mii_probe_args.mii_probemask).
- mii_flags, which will be used for future mii flow control support.
Th he above information could only be passed using new mii_probe(),
it should be used in the following way:
struct mii_probe_args mii_args;
mii_probe_args_init(&mii_args, ifmedia_upd, ifmedia_sts);
/*
* Adjust mii_args if necessary
*/
error = mii_probe(dev, &mii_dev, &mii_args);
Inspired-by: FreeBSD 213878
Sepherosa Ziehau [Sun, 1 Jul 2012 12:11:53 +0000 (20:11 +0800)]
mii: M_WAITOK should be enough
Sascha Wildner [Sun, 1 Jul 2012 20:45:42 +0000 (22:45 +0200)]
kernel/ddb: Remove an unused file.
Sascha Wildner [Sun, 1 Jul 2012 20:41:44 +0000 (22:41 +0200)]
Add a DB_COMMAND(9) manual page (adapted from FreeBSD).
Sascha Wildner [Sun, 1 Jul 2012 12:03:55 +0000 (14:03 +0200)]
makeobjops.awk: Fix comment, it's actually doxygen comments we skip here.
Sascha Wildner [Sun, 1 Jul 2012 11:54:19 +0000 (13:54 +0200)]
makeobjops.awk: Skip C comments (it helps us deal with newer .m files).
Reported-by: ftigeot
Sepherosa Ziehau [Sun, 1 Jul 2012 11:27:10 +0000 (19:27 +0800)]
mii: Cleanup "return", no functional changes
Sepherosa Ziehau [Sun, 1 Jul 2012 11:23:42 +0000 (19:23 +0800)]
mii: Hide miibus_{probe,attach,detach}
While im here, cleanup white space
Sascha Wildner [Sat, 30 Jun 2012 20:00:19 +0000 (22:00 +0200)]
ifmedia.4: Adjust for the last 802.11 upgrade.
Sascha Wildner [Sat, 30 Jun 2012 19:42:02 +0000 (21:42 +0200)]
ixgbe.4: Add a reference to ifmedia(4).
Sascha Wildner [Sat, 30 Jun 2012 19:41:04 +0000 (21:41 +0200)]
ifmedia.4: Add some documentation about 10G settings (adapted from OpenBSD).
Sascha Wildner [Sat, 30 Jun 2012 21:38:10 +0000 (23:38 +0200)]
<net/if_media.h>: Add 10G descriptions (taken from FreeBSD).
Sascha Wildner [Sat, 30 Jun 2012 18:25:49 +0000 (20:25 +0200)]
ifmedia.4: Sync with NetBSD.
François Tigeot [Sat, 30 Jun 2012 14:50:07 +0000 (16:50 +0200)]
ixgbe: Import Intel PRO/10GbE driver from FreeBSD
Local changes:
* Disable LRO and TSO hardware optimizations, commenting out the code
with #if 0 directives
* Disable VLAN hardware acceleration code as well
* Disable MSI-X code, only use one queue per port for now
* Use code from Sascha Wildner to create a per-port sysctl tree
Tested-with: 82599EB
Sascha Wildner [Sat, 30 Jun 2012 08:44:33 +0000 (10:44 +0200)]
Fix more wrong sizeof() usages, part 4/x
We want struct sizes here, not pointer sizes.
Found-with: Coccinelle (http://coccinelle.lip6.fr/)
Sepherosa Ziehau [Sat, 30 Jun 2012 03:25:54 +0000 (11:25 +0800)]
igb: Reduce MSI/legacy interrupt rate limit to 6000
Should work well enough as far as I have tested.
Sepherosa Ziehau [Sat, 30 Jun 2012 03:24:10 +0000 (11:24 +0800)]
igb: Make sure EITR interval is within range
Suggested-by: dillon@
Sepherosa Ziehau [Sun, 17 Jun 2012 08:03:11 +0000 (16:03 +0800)]
igb: Add MSI-X support
- As many RX rings will be enabled, if they are allowed by the number of
available MSI-X vectors and number of CPUs.
- RX MSI-X vectors could be grouped to specific CPU set. However, the
first CPU of the CPU set must be aligned on RX MSI-X vector count; by
default RX MSI-X vectors are grouped according to the device's software
unit number. Tunable hw.igbX.msix.rxoff could be used to regroup RX
MSI-X vector to a specific set of CPUs.
- TX ring uses seperate MSI-X if possible. TX MSI-X vector could be bound
to a specific CPU; bydefault TX MSI-X vectors is bound to the CPU based
on the device's software unit number. Tunable hw.igbX.msix.txoff could
be used to bind TX MSI-X to a specific CPU.
- Per MSI-X vector interrupt moderation sysctl hw.igbX.msiY_rate is added.
The default interrupt rate for TX is 4000, which is enough to generate
1.48Mpps small packets. The default interrupt rate for RX is 6000.
- MSI-X must not be enabled on 82575 according to the specification update
from Intel.
Sepherosa Ziehau [Fri, 29 Jun 2012 10:56:20 +0000 (18:56 +0800)]
bge: Don't prevent bge_init if IFF_RUNNING is set
This probably was old code which intended to guard against if_init()
triggered by SIOCSIFADDR in ether_ioctl(). It no longer makes sense.
Also don't call bge_init() if MTU is changed but device is not running
yet (FreeBSD 213587).
Sepherosa Ziehau [Fri, 29 Jun 2012 10:36:30 +0000 (18:36 +0800)]
bge: Improve 5785 performance
By allowing write DMA to request larger DMA burst size
Obtained-from: OpenBSD via FreeBSD 213333
Sepherosa Ziehau [Fri, 29 Jun 2012 10:31:10 +0000 (18:31 +0800)]
bge: Reduce the number of dropped frames when flow-control is enabled
Set the number of RX frames to receive after RX MBUF low watermark
has reached. Previously it dropped incoming frames once
RX MBUF low watermark has reached. The value used in MAC RX MBUF
low watermark is greater than or equal to 4 so receiving two more
RX frames should not be a problem.
Obtained-from: OpenBSD via FreeBSD 213255
Sepherosa Ziehau [Fri, 29 Jun 2012 10:26:36 +0000 (18:26 +0800)]
bge: Disable UDP hardware checksum
Broadcom NetExtreme controllers always generate 0 UDP checksum.
Obtained-from: FreeBSD 211596
Sepherosa Ziehau [Fri, 29 Jun 2012 10:22:44 +0000 (18:22 +0800)]
bge: Avoid using magic number
Obtained-from: FreeBSD 210152
Sepherosa Ziehau [Fri, 29 Jun 2012 10:19:50 +0000 (18:19 +0800)]
bge: Properly enable Data Fifo Protection for certain chips
Obtained-from: FreeBSD 210083
Sepherosa Ziehau [Fri, 29 Jun 2012 08:26:35 +0000 (16:26 +0800)]
bge: Zero out the status block during initialization
Obtained-from: FreeBSD 209741
Sepherosa Ziehau [Fri, 29 Jun 2012 08:16:54 +0000 (16:16 +0800)]
bge: For dual mode PHY controllers, make sure to enable GMII
Obtained-from: OpenBSD via FreeBSD 202293
Sepherosa Ziehau [Fri, 29 Jun 2012 08:00:09 +0000 (16:00 +0800)]
bge: Don't touch jumbo frame registers on chips which do not support them
Obtained-from: FreeBSD 200228
Unlike above FreeBSD commit, we don't change mini ring handling
Sepherosa Ziehau [Fri, 29 Jun 2012 07:39:10 +0000 (15:39 +0800)]
bge: Improve TX performance on certain low end chips
These chips, e.g. BCM5761, only allow one outstanding DMA read operation,
so on TX path, mbuf chain should be packed to reduce number of DMA reads
from chip to improve performance.
hw.bgeX.force_defrag is added for PCI-E chips to enable this mbuf chain
packing. It is not enabled by default.
Obtained-from: FreeBSD 200088
Sepherosa Ziehau [Fri, 29 Jun 2012 07:06:54 +0000 (15:06 +0800)]
bge: Reduce status block device update size
This reduced the amount of data that device needs to DMAed into
host mmeory update status updating.
Obtained-from: FreeBSD 199679
Changes made to the above FreeBSD commit:
- We don't reduce the amount of host memory allocated for status block
- Properly comment why BCM5700 AX/BX need to have full status block
update setting according to the Broadcom datasheet
Sepherosa Ziehau [Fri, 29 Jun 2012 01:48:30 +0000 (09:48 +0800)]
bge: x86_64 does not have alignment constraint either
Alex Hornung [Thu, 28 Jun 2012 11:11:09 +0000 (12:11 +0100)]
top - Display threads when asked to
Alex Hornung [Thu, 28 Jun 2012 10:53:08 +0000 (11:53 +0100)]
Merge branch 'master' of git://git.dragonflybsd.org/dragonfly
Sepherosa Ziehau [Thu, 28 Jun 2012 06:55:18 +0000 (14:55 +0800)]
bge: Proper configure PCI-X for BCM5703/5704 devices
Set maximum read byte count to 2048. Also disable relaxed ordering
as recommended by data sheet for PCI-X devices. For PCI-X BCM5704,
set maximum outstanding split transactions to 0 as indicated by
datasheet.
Obtained-from: FreeBSD 204978
Sepherosa Ziehau [Thu, 28 Jun 2012 06:45:58 +0000 (14:45 +0800)]
bge: For BCM5703, limit DMA read watermark
For BCM5703 in PCI-X mode, DMA read watermark should be less than
or equal to maximum read byte count configuration. Enforce this
limitation in DMA read watermark configuration.
Obtained-from: FreeBSD 204978
Sepherosa Ziehau [Thu, 28 Jun 2012 06:18:21 +0000 (14:18 +0800)]
bge: Enable hardware fixes for BCM5704 B0 as recommended by datasheet.
Obtained-from: FreeBSD 204975
Sepherosa Ziehau [Thu, 28 Jun 2012 06:10:47 +0000 (14:10 +0800)]
bge: Revert small part of f47afe1, which setup PHY bugs
Sepherosa Ziehau [Thu, 28 Jun 2012 05:54:54 +0000 (13:54 +0800)]
bge: In bge_stop, don't isolate PHY; it is unnecessary
Sepherosa Ziehau [Thu, 28 Jun 2012 05:41:15 +0000 (13:41 +0800)]
bge: Add busdma constraints for old chips
- Most pre BCM5755 controllers have a DMA bug when buffer address
crosses a multiple of the 4GB boundary
- For BCM5714/BCM5715 controllers, limit the DMA address to be within
40bit address space.
Obtained-from: FreeBSD 199670
Sepherosa Ziehau [Thu, 28 Jun 2012 04:45:32 +0000 (12:45 +0800)]
bge: Add missing bit of commit f47afe1
Sepherosa Ziehau [Thu, 28 Jun 2012 03:40:17 +0000 (11:40 +0800)]
bge: Limit BCM5701 B5 to 32-bit mode
This is a workaround for a bug which causes data corruption in
combination with certain bridges.
Obtained-from: FreeBSD 185812
Sepherosa Ziehau [Fri, 15 Jun 2012 09:54:59 +0000 (17:54 +0800)]
tcp: Add XMITNOW which bypasses the Nagle algorithm temporarily
This flag acts differently from ACKNOW that no pure ACK will be sent.
It is currently used by the (extended) limited transmit and the SACK
based fast recovery.
This flag is intended to fix the following bug in the SACK based fast
recovery:
The NextSeg() requires that if the unACKed segments could not pass
IsLost(), previously unsent segment should be selected. In the
application limited period, the size of the previously unsent segment
could be less than the MSS, thus it could not be sent immediately
according to the Nagle algorithm. In our SACK based fast recovery
implementation, if the tcp_output() sends no segments, the current
recovery transmit process will stop immediately. This could stop ACK
clock and cause timeout retransmit, which could be avoided, if the
Nagle algorithm is bypassed temporarily for the small unsent segment
selected by NextSeg().
When this flag is used with (extended) limited transmit, certain amount
of spurious early retranmits could be avoided.
Sascha Wildner [Sun, 24 Jun 2012 17:21:00 +0000 (19:21 +0200)]
librefuse: No puffs_fuse_unlink() function exists.
Sascha Wildner [Sun, 24 Jun 2012 17:15:39 +0000 (19:15 +0200)]
libpuffs: Comment out some info about a non-existant function.
Sascha Wildner [Sun, 24 Jun 2012 17:14:59 +0000 (19:14 +0200)]
libpuffs: Fix some function names in the manpages.
Sascha Wildner [Sat, 23 Jun 2012 12:14:16 +0000 (14:14 +0200)]
find(1): Revert back to using statfs.f_type again for fstype comparison.
It was breaking the weekly locate database update. More investigation
is needed here.
While here, also bring back
00a0ead88e3b856b49bc34aa4c3d8b0d80940600
which was accidentally removed.
Alex Hornung [Thu, 21 Jun 2012 07:34:54 +0000 (08:34 +0100)]
Merge branch 'master' of git://git.dragonflybsd.org/dragonfly
Sascha Wildner [Thu, 21 Jun 2012 07:16:32 +0000 (09:16 +0200)]
kernel: Clean up some Makefiles (custom flags, useless .PATH).
Sascha Wildner [Thu, 21 Jun 2012 02:25:19 +0000 (04:25 +0200)]
Fix more wrong sizeof() usages, part 3/x
Some comments:
* The gzip fix speeds up uncompression of compress(1)'d files with gzip.
A test here took 9s with the fix when it previously took 15s.
* Our OpenSSH is without J-PAKE support, hence the OpenSSH fix is only
cosmetical.
Found-with: Coccinelle (http://coccinelle.lip6.fr/)
Sascha Wildner [Thu, 21 Jun 2012 01:37:28 +0000 (03:37 +0200)]
libedit: Remove a leftover \
Sascha Wildner [Thu, 21 Jun 2012 01:18:09 +0000 (03:18 +0200)]
libedit: Remove documentation of wchar functions (which we don't have).
Sascha Wildner [Wed, 20 Jun 2012 16:50:47 +0000 (18:50 +0200)]
kernel: Remove a bogus .PATH from ntfs_iconv's Makefile.
Markus Pfeiffer [Wed, 20 Jun 2012 21:59:48 +0000 (21:59 +0000)]
kernel: fix a checkloop panic caused by EOPNOTSUPP not being passed down correctly
* When poll(2)ing /dev/tty dev_dkqfilter gets called twice,
on the inner call, since EOPNOTSUPP is returned the outer
call returns ENODEV leading to a checkloop panic.
* A new testcase in test/testcases/io/poll_1 will panic
the system if started without this patch applied.
Alex Hornung [Tue, 19 Jun 2012 13:07:26 +0000 (14:07 +0100)]
puffs - don't lock dvp on nresolve
* locking dvp on nresolve causes a deadlock with the puffsop thread
when the dvp is the root vnode.
* If a lookup/nresolve occurs at the same time as a flush via the
puffsop, then puffsop will not be able to acquire a lock on the root
vnode as the lookup routine already holds it. Similarly the lookup
routine never gets woken up waiting for the message to arrive, since
puffsop is stuck and can't reply.
Antonio Huete Jimenez [Tue, 19 Jun 2012 23:28:23 +0000 (01:28 +0200)]
rc.d - Add a message to initrandom
Previously as no message was shown, it seemed the time was spent in
loading devfs rules. This makes things clearer.
Antonio Huete Jimenez [Tue, 19 Jun 2012 22:05:41 +0000 (00:05 +0200)]
ntfs - Set CNP_PDIRUNLOCK flag if we unlock the parent dir.
If we unlocked the parent dir, we must specify it via CNP_PDIRUNLOCK
flag to avoid further vn_unlock attempts.
DragonFly-bug: http://bugs.dragonflybsd.org/issues/2388
Reported by: swildner, Raimundo Santos
Antonio Huete Jimenez [Tue, 19 Jun 2012 20:56:37 +0000 (22:56 +0200)]
ntfs - Add NTFS_DEBUG kernel option.
Remove also CVS IDs from the files that still have it.
Sascha Wildner [Tue, 19 Jun 2012 14:11:30 +0000 (16:11 +0200)]
commit test
Alex Hornung [Tue, 19 Jun 2012 13:07:26 +0000 (14:07 +0100)]
puffs - don't lock dvp on nresolve
* locking dvp on nresolve causes a deadlock with the puffsop thread
when the dvp is the root vnode.
* If a lookup/nresolve occurs at the same time as a flush via the
puffsop, then puffsop will not be able to acquire a lock on the root
vnode as the lookup routine already holds it. Similarly the lookup
routine never gets woken up waiting for the message to arrive, since
puffsop is stuck and can't reply.
Antonio Huete Jimenez [Mon, 18 Jun 2012 22:24:35 +0000 (00:24 +0200)]
ntfs - Fix build with NTFS_DEBUG
Johannes Hofmann [Tue, 22 May 2012 21:37:55 +0000 (23:37 +0200)]
use rb-tree for directory lookups
* tmpfs directories are structured as lists of directory entries; this
leads to linear lookup costs. Directories with many files become fairly
expensive to operate on.
* This patch uses a rb-tree keyed on the name of the searched file to
speed up lookups
* Besides this rb-tree implementation, a hash version was also tested.
Both gave solid performance enhancements compared to the previous tmpfs
implementation.
The rb-tree version was the faster of the two and thus becomes the
choosen one.
* See issue 2375 for details and performance numbers
https://bugs.dragonflybsd.org/issues/2375
Sascha Wildner [Mon, 18 Jun 2012 06:35:20 +0000 (08:35 +0200)]
cryptdisks(8): Fix a wrong sizeof.
Use the size passed to the alloc_safe_mem() here.
Found-with: Coccinelle (http://coccinelle.lip6.fr/)
Markus Pfeiffer [Sun, 17 Jun 2012 19:38:49 +0000 (19:38 +0000)]
vkernel: revert two commits that break i386 vkernel
revert i386 parts of
commit
c55fa5eeab3eaf11e01222d6412b91dafdcd9e94
commit
1a482e3fea833c34264b54d566ddd0f4276d8bcf
because it breaks build. This is still a kludge.