The nfscl_mtofh() function didn't check for failed operations and, as such,
would have returned EBADRPC for these cases, due to parsing failure.
This patch adds checks, so that it returns with ND_NOMOREDATA set.
This is needed for future use in the pNFS server and acts as a safety
belt in the meantime.
MFC after: 2 weeks
The default uid/gid for NFSv4 are set by the nfsuserd(8) daemon.
However, they were 0 until the nfsuserd(8) was run. Since it is
possible to use NFSv4 without running the nfsuserd(8) daemon, set them
to nobody/nogroup initially.
Without this patch, the values would be set by the nfsuserd(8) daemon
and left changed even if the nfsuserd(8) daemon was killed. The default
values of 0 meant that setting a group to "wheel" would fail even when
done by root.
It also adds a definition of GID_NOGROUP to sys/conf.h.
Discussed on: freebsd-current@
MFC after: 2 weeks
7386 zfs get does not work properly with bookmarks
illumos/illumos-gate@edb901aab9edb901aab9https://www.illumos.org/issues/7386
The zfs get command does not work with the bookmark parameter while it works
properly with both filesystem and snapshot:
# zfs get -t all -r creation rpool/test
NAME PROPERTY VALUE SOURCE
rpool/test creation Fri Sep 16 15:00 2016 -
rpool/test@snap creation Fri Sep 16 15:00 2016 -
rpool/test#bkmark creation Fri Sep 16 15:00 2016 -
# zfs get -t all -r creation rpool/test@snap
NAME PROPERTY VALUE SOURCE
rpool/test@snap creation Fri Sep 16 15:00 2016 -
# zfs get -t all -r creation rpool/test#bkmark
cannot open 'rpool/test#bkmark': invalid dataset name
#
The zfs get command should be modified to work properly with bookmarks too.
Reviewed by: Simon Klinkert <simon.klinkert@gmail.com>
Reviewed by: Paul Dagnelie <pcd@delphix.com>
Approved by: Matthew Ahrens <mahrens@delphix.com>
Author: Marcel Telka <marcel@telka.sk>
remove the former.
All other EGA/VGA methods were already shared, with VGA-only features
mostly not used and no decisions in inner loops to optimize fof VGA,
but this method was split up because it is the only important one and
using VGA methods if possible is about twice as fast. The speed is
mostly not from splitting to reduce branches but from doing half as
many bus accesses, so make this easier to maintain by not splitting.
There is now 1 extra branch in an inner loop where it costs less than
1% of the bus access overhead on Haswell even if the compiler schedules
it poorly.
Before this change it was impossible to set number of PKCS#5v2 iterations,
required to set passphrase, if it has two keys and never had any passphrase.
Due to present metadata format limitations there are still cases when number
of iterations can not be changed, but now it works in cases when it can.
PR: 218512
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
Differential Revision: https://reviews.freebsd.org/D10338
vga planar method (for testing that was supposed to be local that the
former still works). The ega method works on vga but is about twice
as slow. The vga method doesn't work on ega.
Optimize the main vga planar method a little. For changing the
background color (which was otherwise optimized better than most
things), don't switch the write mode from 3 to 0 just to select
the pixel mask of 0xff obscurely by writing 0. Just write 0xff
directly.
-(SYNCOOKIE_LIFETIME + 1) instead of INT64_MIN, since it is
good enough and works when time_t is int32 or int64.
This fixes the issue reported by cy@ on i386.
Reported by: cy
MFC after: 1 week
Sponsored by: Netflix, Inc.
The default uid/gid for NFSv4 are set by the nfsuserd(8) daemon.
However, they were 0 until the nfsuserd(8) was run. Since it is
possible to use NFSv4 without running the nfsuserd(8) daemon, set them
to nobody/nogroup initially.
Without this patch, the values would be set by the nfsuserd(8) daemon
and left changed even if the nfsuserd(8) daemon was killed. Also, the default
values of 0 meant that setting a group to "wheel" would fail even when
done by root and this patch fixes this issue.
MFC after: 2 weeks
7490 real checksum errors are silenced when zinject is on
illumos/illumos-gate@6cedfc397d6cedfc397dhttps://www.illumos.org/issues/7490
When zinject is on, error codes from zfs_checksum_error() can be overwritten
due to an incorrect and overly-complex if condition.
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Paul Dagnelie <pcd@delphix.com>
Reviewed by: Dan Kimmel <dan.kimmel@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Approved by: Robert Mustacchi <rm@joyent.com>
Author: Pavel Zakharov <pavel.zakharov@delphix.com>
7448 ZFS doesn't notice when disk vdevs have no write cache
illumos/illumos-gate@295438ba32295438ba32https://www.illumos.org/issues/7448
I built a SmartOS image with all the NVMe commits including 7372
(support NVMe volatile write cache) and repeated my dd testing:
> #!/bin/bash
> for i in `seq 1 1000`; do
> dd if=/dev/zero of=file00 bs=1M count=102400 oflag=sync &
> dd if=/dev/zero of=file01 bs=1M count=102400 oflag=sync &
> wait
> rm file00 file01
> done
>
Previously each dd command took ~145 seconds to finish, now it takes
~400 seconds.
Eventually I figured out it is 7372 that causes unnecessary
nvme_bd_sync() executions which wasted CPU cycles.
If a NVMe device doesn't support a write cache, the nvme_bd_sync function will
return ENOTSUP to indicate this to upper layers.
It seems this returned value is ignored by ZFS, and as such this bug is not
really specific to NVMe. In vdev_disk_io_start() ZFS sends the flush to the
disk driver (blkdev) with a callback to vdev_disk_ioctl_done(). As nvme filled
in the bd_sync_cache function pointer, blkdev will not return ENOTSUP, as the
nvme driver in general does support cache flush. Instead it will issue an
asynchronous flush to nvme and immediately return 0, and hence ZFS will not set
vdev_nowritecache here. The nvme driver will at some point process the cache
flush command, and if there is no write cache on the device it will return
ENOTSUP, which will be delivered to the vdev_disk_ioctl_done() callback. This
function will not check the error code and not set nowritecache.
The right place to check the error code from the cache flush is in
zio_vdev_io_assess(). This would catch both cases, synchronous and asynchronous
cache flushes. This would also be independent of the implementation detail that
some drivers can return ENOTSUP immediately.
Reviewed by: Dan Fields <dan.fields@nexenta.com>
Reviewed by: Alek Pinchuk <alek.pinchuk@nexenta.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Approved by: Dan McDonald <danmcd@omniti.com>
Author: Hans Rosenfeld <hans.rosenfeld@nexenta.com>
Obtained from: Illumos
The FreeBSD NFSv4 server did not set the attribute bit for TimeAccess in
the reply to an Open with exclusive_create, as required by the RFCs.
(This is required since the FreeBSD NFS server stores the create_verifier
in the va_atime attribute.)
As such, the Linux NFSv4 client did not set the TimeAccess (atime) in
the Setattr done in an RPC after the one with the Open/exclusive_create.
This patch fixes the server to set the TimeAccess bit in the reply.
I believe that storing the create_verifier in an extended attribute for
file systems that support extended attributes might be a good idea,
but I will wait for a discussion of this on the freebsd-fs@ email list
before considering committing a patch to do this.
Reported by: jim@ks.uiuc.edu
Suggested by: dfr
MFC after: 2 weeks
7430 Backfill metadnode more intelligently
illumos/illumos-gate@af346df588af346df588https://www.illumos.org/issues/7430
Description and patch from brought over from the following ZoL commit: https://
github.com/zfsonlinux/zfs/commit/68cbd56e182ab949f58d004778d463aeb3f595c6
Only attempt to backfill lower metadnode object numbers if at least
4096 objects have been freed since the last rescan, and at most once
per transaction group. This avoids a pathology in dmu_object_alloc()
that caused O(N^2) behavior for create-heavy workloads and
substantially improves object creation rates. As summarized by
@mahrens in #4636:
"Normally, the object allocator simply checks to see if the next
object is available. The slow calls happened when dmu_object_alloc()
checks to see if it can backfill lower object numbers. This happens
every time we move on to a new L1 indirect block (i.e. every 32 *
128 = 4096 objects). When re-checking lower object numbers, we use
the on-disk fill count (blkptr_t:blk_fill) to quickly skip over
indirect blocks that don?t have enough free dnodes (defined as an L2
with at least 393,216 of 524,288 dnodes free). Therefore, we may
find that a block of dnodes has a low (or zero) fill count, and yet
we can?t allocate any of its dnodes, because they've been allocated
in memory but not yet written to disk. In this case we have to hold
each of the dnodes and then notice that it has been allocated in
memory.
The end result is that allocating N objects in the same TXG can
require CPU usage proportional to N^2."
Add a tunable dmu_rescan_dnode_threshold to define the number of
objects that must be freed before a rescan is performed. Don't bother
to export this as a module option because testing doesn't show a
compelling reason to change it. The vast majority of the performance
gain comes from limit the rescan to at most once per TXG.
Reviewed by: Alek Pinchuk <alek@nexenta.com>
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Approved by: Gordon Ross <gordon.w.ross@gmail.com>
Author: Ned Bass <bass6@llnl.gov>
Obtained from: Illumos
overflows, syncookies are used.
This patch restricts the usage of syncookies in this case: accept
syncookies only if there was an overflow of the syncache recently.
This mitigates a problem reported in PR217637, where is syncookie was
accepted without any recent drops.
Thanks to glebius@ for suggesting an improvement.
PR: 217637
Reviewed by: gnn, glebius
MFC after: 1 week
Sponsored by: Netflix, Inc.
Differential Revision: https://reviews.freebsd.org/D10272
a pointer to the main ega drawing method which is misoptimized be in
a different function than the main vga planar mode drawing method.
Vga initialization handles everything with no extra code except for
selecting the different function.
corresponding to the gaps between characters. This fixes distortion
of the cursor due to expanding it across the gaps.
Again for character width 9, when the cursor characters are not in the
graphics range (0xb0-0xdf), the gaps were always there (filled in the
background color for the previous char). They still look strange, but
don't cause distortion. When the cursor characters are in the graphics
range, the gaps are filled by repeating the previous line. This gives
distortion with cilia. Removing vertical lines reduces the distortion
to vertical cilia.
Move the default for the cursor characters out of the graphics range.
With character width 9, this gives gaps instead of distortion and
other problems. With character width 8, it just fixes a smaller set
of other problems. Some distortion and other problems can be recovered
using vidcontrol -M. Presumably the default was to fill the gaps
intentionally, but it is much better to leave gaps. The gaps can even
be considered as a feature for text processing -- they give sub-pointers
to character boundaries. The other problems are: (1) with character
width 9, characters near the cursor are moved into the graphics range
and thus distorted if any of their 8th bits is set; (2) conflicts with
national characters in the graphics range.
The default range for the graphics cursor characters is now 8-11. This
doesn't conflict with anything, since the glyphs for the characters in
this range are unreachable.
Use the 10x16 mouse cursor in text mode too (if the font size is >= 14).
When the character width is 9, removal of 1 or 2 vertical lines makes
10x16 cursor no wider than the 9x13 one usually was. We could even
handle cursors 1 pixel wider in 2 character cells and gaps without
more clipping than given by the gaps (the worst case is 1 pixel in the
left cell, 1 removed in the middle gap, 8 in the right cell and 1
removed in the right gap. The pixel in the right gap is removed so
it doesn't matter if it is in the font).
When the character width is 8, we now clip the 10-wide cursor by 1
pixel in the worst case. This clipping is usually invisible since it
is of the border and and the border usually merges with the background
so is invisible. There should be an option to use reverse video to
highlight the border and its tip instead of the interior (graphics
mode can do better using separate colors). This needs the 9x13 cursor
again.
Ideas from: ache (especially about the bad default character range)
Note that KVA mapping of the framebuffer already uses write-combining
mode, so the change, besides improving speed of user mode writes, also
satisfies requirement of the IA32 architecture of using consistent
caching modes for multiple mappings of the same page.
Reported and tested by: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
not consider a "disabled" cpu as a CPU we have to ignore, and we should use
them if they provide a "enable-method".
While I'm there, support "ok" as well as "okay", while ePAPR only accepts
"okay", linux accepts "ok" too so we can expect it to be used.
Reviewed by: andrew (partially)
9 wide.
I only need this to improve the mouse cursor, but it has always been
needed to select and/or adjust fonts.
This is complicated because there are no standard parameter tables
giving this bit of information directly, and the device register bit
giving the information can't be trusted even if it is read from the
hardware. Use a heuristic to guess if the device register can be
trusted. (The device register is normally read from the BIOS mode
table, but on my system where the device register is wrong, the mode
table doesn't match the hardware and is not used; the device registers
are used in this case.)
When forwarding pf tracks the size of the largest fragment in a fragmented
packet, and refragments based on this size.
It failed to ensure that this size was a multiple of 8 (as is required for all
but the last fragment), so it could end up generating incorrect fragments.
For example, if we received an 8 byte and 12 byte fragment pf would emit a first
fragment with 12 bytes of payload and the final fragment would claim to be at
offset 8 (not 12).
We now assert that the fragment size is a multiple of 8 in ip6_fragment(), so
other users won't make the same mistake.
Reported by: Antonios Atlasis <aatlasis at secfu net>
MFC after: 3 days
mode.
Use the general DRAWPIXEL() macro with its bigger case statement
(twice) instead of our big case statement (once). DRAWPIXEL() is more
complicated since it is not missing support for depth 24 or
complications for colors in depth 16 (we currently hard-code black and
white so the complications for colors are not needed). DRAWPIXEL()
also does the bpp calculation in the inner loop. Compilers optimize
DRAWPIXEL() well enough, and the main text drawing method always
depended on this. In direct mode, mouse cursor drawing is now similar
to normal text drawing except it draws in 2 hard-coded colors instead
of 1 variable color.
This also fixes a nested hard-coding of colors. DRAWPIXEL() uses the
palette in all cases, but the direct code didn't use the palette for
its hard-coded black. This only had an effect in depth 8, since
changing the palette is not supported in other depths.
direct mode renderer. I thought that reads were not much slower than
writes, so that the method only tripled the time for the whole function,
but I recently measured that video memory reads can be up to 53 times
slower than writes in tighter loops than here. Loop overheap here
reduces the multiplier to only 16-20 on Haswell.
Start cleaning up and fixing larger bugs in this function. Only replace
the 22-line removal loop by a 3-line one for now, since adjusting the
old loop would have required many palette calculations which are better
done in the DRAW_PIXEL() macro. This also fixes missing support for
depth 24, but only for removal.
Removal is currently sloppy at the right bottom corner. It sometimes
leaks border color into the text window. This is soon cleaned up by the
caller. The planar renderer has complications to clip at the corner.
using a driver-supplied sbuf for printing device discovery
announcements. This helps ensure that messages to the console
will be properly serialized (through sbuf_putbuf) and not be
truncated and interleaved with other messages. The
infrastructure mirrors the existing xpt_announce_periph()
entry point and is opt-in for now. No content or formatting
changes are visible to the operator other than the new coherency.
While here, eliminate the stack usage of the temporary
announcement buffer in some of the drivers. It's moved to the
softc for now, but future work will eliminate it entirely by
making the code flow more linear. Future work will also address
locking so that the sbufs can be dynamically sized.
The scsi_da, scs_cd, scsi_ses, and ata_da drivers are converted
at this point, other drivers can be converted at a later date.
A tunable+sysctl, kern.cam.announce_nosbuf, exists for testing
purposes but will be removed later.
TODO:
Eliminate all of the code duplication and temporary buffers. The
old printf-based methods will be retired, and xpt_announce_periph()
will just be a wrapper that uses a dynamically sized sbuf. This
requires that the register and deregister paths be made malloc-safe,
which they aren't currently.
Sponsored by: Netflix
of our tweaked modes based on it. In practice, this means limiting the
tweaked modes to at most 80x50 based on 80x25, so there are no 90-column,
80x30 or 80x60 modes.
This happens when the the initial mode is is not in the parameter
table. We always detected this case, but assumed that the (necessarily
nonstandard) parameters of the initial mode could be tweaked just as
blindly as the probably-standard parameters of initial modes in the
table.
On 1 laptop system with near-VGA where the initial mode is nonstandard,
this is because the hardware apparently doesn't support 9-bit mode,
but otherwise has standard timing. The initial mode has 8-bit mode
CRTC horizontal parameters similar to those in syscons' 90-column modes
and in EGA modes. Tweaking these values for the 90-column modes has
little effect except to print the extra 10 columns off the screen.
Tweaking from 80x25 to 80x30 requires changing from 400 scan lines to
480. This can probably be made to work, but syscons blindly applies
values based on standard timing. This gives blank output. Tweaking
from 80x25 to 80x50 doesn't change the CRTC timing and works.
kit, CK, in the LinuxKPI.
When threads are pinned to a CPU core or when there is only one CPU,
it can happen that a higher priority thread can call the CK
synchronize function while a lower priority thread holds the read
lock. Because the CK's synchronize is a simple wait loop this can lead
to a deadlock situation. To solve this problem use the recently
introduced CK's wait callback function.
When detecting a CK blocking condition figure out the lowest priority
among the blockers and update the calling thread's priority and
yield. If another CPU core is holding the read lock, pin the thread to
the blocked CPU core and update the priority. The calling threads
priority and CPU bindings are restored before return.
If a thread holding a CK read lock is detected to be sleeping, pause()
will be used instead of yield().
MFC after: 1 week
Sponsored by: Mellanox Technologies
CPUs when allocating a LinuxKPI workqueue. This also ensures that the
created taskqueue always have a non-zero number of worker threads.
MFC after: 1 week
Sponsored by: Mellanox Technologies
According to Warner, multiple TRIM BIOs are collapsed into a single CCB with
NULL bp. It is invalid to biotrack() NULL, and results in a fault. So,
don't do that.
Reported by: asomers@
Sponsored by: Dell EMC Isilon
As the uboot disk interface is using common/disk.c API, we also
should use disk_ioctl() call, this will give us chance to read partition
sizes and have feature parity with UEFI and BIOS implementations.
This does also fix arm boot issue on some systems, reported/tested by Ian,
thanks.
Reported by: ian
Reviewed by: ian
Differential Revision: https://reviews.freebsd.org/D10421
The work to make it possible to avoid bcache via using F_NORA modifier did
miss the fact that not all loader platforms are using the bcache, and so
it is possible the modifier is not cleared, as bcache strategy function is
not used.
For fix, we make sure the checks are dont with masked flag.
This patch does fix boot for platforms which do not use bcache.
Reported by: emaste
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D10422
The smallest device we can have in the pool is 64MB, since we are trying to
walk all four labels to find the most up to date uberblock, this limit will
also give us good method to check if we even should attempt to probe.
Enforcing the check also will make sure we are not getting wrapped while
calculating the label offset.
Also, after label check, we should verify if we actually got any UB or not.
PR: 218473
Reported by: Masachika ISHIZUKA
Reviewed by: allanjude
Differential Revision: https://reviews.freebsd.org/D10381
Add early init handler, which comprises various internal
bus optimisations for Armada 38x SoC's. Magic values used
due to undocumented registers.
Submitted by: Marcin Wojtas <mw@semihalf.com>,
Arnaud Ysmal <arnaud.ysmal@stormshield.eu>
Obtained from: Semihalf, Stormshield
Sponsored by: Stormshield
Differential revision: https://reviews.freebsd.org/D10219
Part of PL310 erratum 727915 in pl310_wbinv_range() was
executed uncoditionally for all possible controllers'
revisions. This patch adds appropriate condition, since
extra operations are required only for revisions between
r2p0 and r3p0.
Submitted by: Marcin Wojtas <mw@semihalf.com>
Obtained from: Semihalf
Sponsored by: Stormshield
Reviewed by: meloun-miracle-cz
Differential revision: https://reviews.freebsd.org/D10221
Introduce machine-dependent part of the arm/pl310 driver for
Armada 38x SoCs. Add prefetch and power savings configuration.
Submitted by: <arnaud.ysmal@stormshield.eu>
Obtained from: Stormshield
Sponsored by: Stormshield
Differential revision: https://reviews.freebsd.org/D10220
Memory space reserved for pmap_kernel_l2dtable_kva and
pmap_kernel_l2ptp_kva has not been taken into account in
original code. All the memory reserved from kernel space by
pmap_alloc_specials() function called in pmap_bootstrap()
should be mapped initially by initarm(). To create initial
mapping initarm() function reserves proper number of l2 page
tables. However the number of the l2 page tables does not take
into account memory for: pmap_kernel_l2ptp_kva,
pmap_kernel_l2dtable_kva, crashdumpmap, etc.
Submitted by: Grzegorz Bernacki <gjb@semihalf.com>
Obtained from: Semihalf
Sponsored by: Stormshield
Reviewed by: meloun-miracle-cz
Differential revision: https://reviews.freebsd.org/D10217
VM_KMEM_SIZE_MAX allows to limit kmem arena size. In our case this was
necessary, as decreasing size of kmem_arena leaves more space for
kernel_arena.
kernel_arena is pool used for contigmalloc (in effect, DMA) allocations,
which failed on Armada38x. This resulted in 'no memory errors'
(e.g. USB_ERR_NOMEM errors) and failure of whole system. The need for
greater size of kernel_arena probably comes from more peripherals making
use of busdma.
Value used as upper limit is half of the default value
(0x1399a000).
Submitted by: Wojciech Macek <wma@semihalf.com>
Obtained from: Semihalf
Sponsored by: Stormshield
Differential revision: https://reviews.freebsd.org/D10216
in place. To do per-cpu stats, convert all fields that previously were
maintained in the vmmeters that sit in pcpus to counter(9).
- Since some vmmeter stats may be touched at very early stages of boot,
before we have set up UMA and we can do counter_u64_alloc(), provide an
early counter mechanism:
o Leave one spare uint64_t in struct pcpu, named pc_early_dummy_counter.
o Point counter(9) fields of vmmeter to pcpu[0].pc_early_dummy_counter,
so that at early stages of boot, before counters are allocated we already
point to a counter that can be safely written to.
o For sparc64 that required a whole dummy pcpu[MAXCPU] array.
Further related changes:
- Don't include vmmeter.h into pcpu.h.
- vm.stats.vm.v_swappgsout and vm.stats.vm.v_swappgsin changed to 64-bit,
to match kernel representation.
- struct vmmeter hidden under _KERNEL, and only vmstat(1) is an exclusion.
This is based on benno@'s 4-year old patch:
https://lists.freebsd.org/pipermail/freebsd-arch/2013-July/014471.html
Reviewed by: kib, gallatin, marius, lidl
Differential Revision: https://reviews.freebsd.org/D10156
messages before accessing message fields that may not be present,
removing dead/duplicate/misleading code along the way.
Document the message format for each routing socket message in
route.h.
Fix a bug in usr.bin/netstat introduced in r287351 that resulted in
pointer computation with essentially random 16-bit offsets and
dereferencing of the results.
Reviewed by: ae
MFC after: 1 month
Differential Revision: https://reviews.freebsd.org/D10330
numbers with Chacha20. Keep the API, though, as that is what the
other *BSD's have done.
Use the boot-time entropy stash (if present) to bootstrap the
in-kernel entropy source.
Reviewed by: delphij,rwatson
Approved by: so(delphij)
MFC after: 2 months
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D10048
is no phy-handle property, fall back to using MII_PHY_ANY.
This still doesn't support an mdio bus with multiple PHYs on it, or the
possibility that the PHY being used by this instance of ffec is on the
mdio bus of some other instance (which is now a possibility with imx6ul).
Adding that support will require changes in fdt_get_phyaddr(), which is
currently making some assumptions that don't work with modern fdt data.
The only thing this phy needs that the ukphy driver doesn't provide is
that the value in the proprietary Phy Control 2 Register must be saved
before doing a soft reset and restored afterwards. Most modern phys have
"sticky bits" for low-level config that survive a reset, but on this one
the values in all registers go back to defaults, wiping out any board-
specific config set up by the bootloader/bios/whatever.
modes if the font size is >= 14.
This is the X cursor XC_left_ptr (#68) (glyph #45 in an X cursor font).
Also found in vt. The old 9x13 cursor is the 10x16 one trimmed not very
well.
8x8 fonts need a smaller cursor instead of a larger one, except when
the pixel size is small. Text mode is still limited to width and height
1 more than the font (so the 9x13 is already 4 pixels too high for it).
access it via pointers (still to only 1 instance, now with a less
generic name).
Restructure the "and" and "or" masks as border and interior masks
(where the "and" mask was for the union of the border and the interior).
"and" and "or" were only a detail in a not very good implementation,
and after fixing that the union was only used to calculate the border
at runtime.
Use the metric data in more places to clip to active pixels earlier.
available starting with T6. The values in the timer holdoff registers
are multiplied by the scaling factor before use.
dev.<nexus>.<n>.holdoff_timers shows the final values of the
timers in microseconds.
MFC after: 1 week
Sponsored by: Chelsio Communications
This fixes some panics after disconnecting mounted disks.
Submitted by: imp (slightly different version, which I've then lost)
Reviewed by: kib, imp, mckusick
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D9674
by the shutdown(2) system call. This ability has been lost as part of the svn
revision 285910.
Reviewed by: ed, rwatson, glebius, hiren
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D10351
At this point we have not rendezvous'ed with the mirror worker thread, and
I/O may still be in flight. Various I/O completion paths expect to be able
to obtain a reference to the mirror softc from the GEOM, so setting it to
NULL may result in various NULL pointer dereferences if the mirror is
stopped with -f or the kernel is shut down while a mirror is
synchronizing. The worker thread will clear the softc pointer before
exiting.
Tested by: pho
MFC after: 2 weeks
Sponsored by: Dell EMC Isilon
We are otherwise susceptible to a race with a concurrent teardown of the
mirror provider, causing the I/O to be left uncompleted after the mirror
started withering.
Tested by: pho
MFC after: 2 weeks
Sponsored by: Dell EMC Isilon
mode.
Direct mode always supported widths up to 32, except for its hard-coded
16s matching the pixmap size. Text mode is still limited to 9 its 2x2
character cell method and missing adjustments for the gap between
characters, if any.
Cursor heights can be almost anything in graphics modes.
Regular I/O requests may be blocked by concurrent synchronization requests
targeted to the same LBAs, in which case they are moved to a holding queue
until the conflicting I/O completes. We therefore want to stop
synchronization before completing pending I/O in g_mirror_destroy_provider()
since this ensures that blocked I/O requests are completed as well.
Tested by: pho
MFC after: 2 weeks
Sponsored by: Dell EMC Isilon
While the former name is easier to read, the "_flags" suffix has a special
meaning for loader(8) and, thus, it was impossible to set the knob via
loader.conf(5). The loader interpreted the setting as flags that should
be passed to a kernel module named "vfs.zfs.debug".
Discussed with: smh
MFC after: 2 weeks
much as possible, by avoiding null ANDs and ORs to the frame buffer.
Mouse cursors are fairly sparse, especially for their frame. Pixels
are written in groups of 8 in planar mode and the per-group sparseness
is not as large, but it still averages about 40% with the current
9x13 mouse cursor. The average drawing time is reduced by about this
amount (from 22 usec constant to 12.5 usec average on Haswell).
This optimization is relatively larger with larger cursors. Width 10
requires 6 frame buffer accesses per line instead of 4 if not done
sparsely, but rarely more than 4 if done sparsely.
The "cred" argument of ncl_flush() is unused and it was confusing to have
the code passing in NULL for this argument in some cases. This patch deletes
this argument.
There is no semantic change because of this patch.
MFC after: 2 weeks
mode.
Don't manually unroll the 2 inner loops. On Haswell, doing so gave a
speedup of about 0.5% (about 4 cycles per iteration out of 1400), but
hard-coded a limit of width 9 and made better better optimizations
harder to see. gcc-4.2.1 -O does the unrolling anyway, unless tricked
with a volatile hack. gcc's unrolling is not very good and gives a
a speedup of about half as much (about 2 cycles per iteration). (All
timing on i386.)
Manual unrolling was only feasible because the inner loop only iterates
once or twice. Usually twice, but a dynamic check is needed to decide,
and was not moved from the second-innermost loop manually or by gcc.
This commit basically adds another dynamic check in the inner loop.
Cursor widths of 10-17 require 3 iterations in the inner loop and this
is not so easy to unroll -- even gcc stops at 2.
The 'pktid' variable is modified while being used twice between
sequence points, probably due to htonl() is macro.
Reported by: PVS-Studio
MFC after: 1 week
to pass rule number and rule set to userland. In r272840 the kernel
internal rule representation was changed and the rulenum field of
struct ip_fw_rule got the type uint32_t, but userlevel representation
still have the type uint16_t. To not overflow the size of pointer
on the systems with 32-bit pointer size use separate variable to
copy rulenum and set.
Reported by: PVS-Studio
MFC after: 1 week
do for streaming sockets.
And do more cleanup in the sbappendaddr_locked_internal() to prevent
leak information from existing mbuf to the one, that will be possible
created later by netgraph.
Suggested by: glebius
Tested by: Irina Liakh <spell at itl ua>
MFC after: 1 week
ipfilter man pages. This also currently restricts keep frags to only when
keep state is used, which is redundant because keep state currently
assumes keep frags. This commit fixes this.
To the user this change means that to maintain the current behaviour
one must add keep frags to any ipfilter keep state rule (as documented
in the man pages).
This patch also allows the flexability to specify and use keep frags
separate from keep state, as documented in an example in ipf.conf.5,
instead of the currently broken behaviour.
Relnotes: yes
Memory is malloc'd, then a search for a match in the fragment table
is made and if the fragment matches, the wrong fragment table is
freed, causing a use after free panic. This commit fixes this.
A symptom of the problem is a kernel page fault in bcopy() called by
ipf_frag_lookup() at line 715 in ip_frag.c. Another symptom is a
kernel page fault in ipf_frag_delete() when called by ipf_frag_expire()
via ipf_slowtimer().
MFC after: 1 week
TXP_CMD_WAIT argument allocates a response buffer. If the allocation
fails, txp_ext_command() returns an error and it's handed in caller.
Found by: PVS-Studio
Some NFSv4.1 servers such as AmazonEFS can only support a small fixed number
of open_owner4s. This patch adds a mount option called "oneopenown" that
can be used for NFSv4.1 mounts to make the client do all Opens with the
same open_owner4 string. This option can only be used with NFSv4.1 and
may not work correctly when Delegations are is use.
Reported by: cperciva
Tested by: cperciva
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D8988
A function called svcpool_close() was added to the server side krpc by
r313735, so that a pool could be closed without destroying the data structures.
This little patch adds a call to it for the callback pool (svcpool_nfscbd),
so that the nfscbd daemon can be killed/restarted and continue to work
correctly.
MFC after: 2 weeks
Some dummynet modules used strcpy() to copy from a larger buffer
(dn_aqm->name) to a smaller buffer (dn_extra_parms->name). It happens that
the lengths of the strings in the dn_aqm buffers were always hardcoded to be
smaller than the dn_extra_parms buffer ("CODEL", "PIE").
Use strlcpy() instead, to appease static checkers. No functional change.
Reported by: Coverity
CIDs: 1356163, 1356165
Sponsored by: Dell EMC Isilon
According to the C standard, it is invalid to copy beyond the end of an
object, even if that object is obviously a member of a larger object (a
struct, in this case).
Appease the standard and Coverity by refactoring the copy in a
straightforward way. No functional change.
Reported by: Coverity (CWE-120)
CIDs: 1007819, 1007820, 1007821, 1007822, 1009668, 1009669
Security: no (false positive detection)
Sponsored by: Dell EMC Isilon
When checksums of received IP and UDP header already checked, UDP uses
sbappendaddr_locked() to pass received data to the socket.
sbappendaddr_locked() uses given mbuf as is, and if NIC supports checksum
offloading, mbuf contains csum_data and csum_flags that were calculated
for already stripped headers. Some NICs support only limited checksums
offloading and do not use CSUM_PSEUDO_HDR flag, and csum_data contains
some value that UDP/TCP should use for pseudo header checksum calculation.
When L2TP is used for tunneling with mpd5, ng_ksocket receives mbuf with
filled csum_flags and csum_data, that were calculated for outer headers.
When L2TP header is stripped, a packet that was tunneled goes to the IP
layer and due to presence of csum_flags (without CSUM_PSEUDO_HDR) and
csum_data, the UDP/TCP checksum check fails for this packet.
Reported by: Irina Liakh <spell at itl ua>
Tested by: Irina Liakh <spell at itl ua>
MFC after: 1 week
Demotions preserve PG_NX, so it is enough to set nx bit for initial
lowest-level paging entries.
Suggested and reviewed by: alc
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
The arm64 binutils only accepts 0 as an offset to the Load-Acquire Register
instructions where llvm will acceps both 0 and 0x0. The thread switching
code uses these with SCHED_ULE to block waiting for a lock to be released.
As the offset of the data to be loaded is zero this is safe, however it is
useful to keep the offset in the instruction to document what is being
loaded.
To work around this issue in binutils only generate the 0x prefix for
non-zero values.
Reported by: kan
Sponsored by: DARPA, AFRL
We need to set the Execute-never bits when mapping device memory as the
hardware may perform speculative instruction fetches.
Set the Privileged Execute-ever bit on userspace memory to stop the kernel
if it is tricked into executing it.
Reviewed by: kib
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D10382
When opening a vdev whose path is unknown, vdev_geom must find a geom
provider with a label whose guids match the desired vdev. However, due to
partitioning, it is possible that two non-synonomous providers will share
some labels. For example, if the first partition starts at the beginning of
the drive, then ada0 and ada0p1 will share the first label. More troubling,
if the last partition runs to the end of the drive, then ada0p3 and ada0
will share the last label. If vdev_geom opens ada0 when it should've opened
ada0p3, then the pool won't be readable. If it opens ada0 when it should've
opened ada0p1, then it will corrupt some other partition when it writes the
3rd and 4th labels.
The easiest way to reproduce this problem is to install a mirrored root pool
with the default partition layout, then swap the positions of the two boot
drives and reboot. Whether the bug manifests depends on the order in which
geom lists its providers, which is arbitrary.
Fix this situation by modifying the search algorithm to prefer geom
providers that have all four labels intact. If no such provider exists, then
open whichever provider has the most.
Reviewed by: mav
MFC after: 3 weeks
Sponsored by: Spectra Logic Corp
Differential Revision: https://reviews.freebsd.org/D10365
When the replay window size is large than UINT8_MAX, add to the request
the SADB_X_EXT_SA_REPLAY extension header that was added in r309144.
Also add support of SADB_X_EXT_NAT_T_TYPE, SADB_X_EXT_NAT_T_SPORT,
SADB_X_EXT_NAT_T_DPORT, SADB_X_EXT_NAT_T_OAI, SADB_X_EXT_NAT_T_OAR,
SADB_X_EXT_SA_REPLAY, SADB_X_EXT_NEW_ADDRESS_SRC, SADB_X_EXT_NEW_ADDRESS_DST
extension headers to the key_debug that is used by `setkey -x`.
Modify kdebug_sockaddr() to use inet_ntop() for IP addresses formatting.
And modify kdebug_sadb_x_policy() to show policy scope and priority.
Reviewed by: gnn, Emeric Poupon
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D10375
from the kernel. Make use of this to restrict accessing userspace to just
the functions that explicitly handle crossing the user kernel boundary.
Reported by: kib
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D10371
The current multiboot loader code doesn't clean the metadata added to the
kernel after the bi_load64 dry run, which breaks accounting of the required
memory for the metadata.
This issue didn't show itself before because all the metadata items where small
(8bytes), but after r316343 there's a big blob in the metadata, which triggers
this. Fix it by cleaning the metadata added to the kernel after the bi_load64
dry run. Also add a comment describing the memory layout when booting using
multiboot (Xen Dom0).
This unbreaks booting a FreeBSD/Xen Dom0 after r316343.
MFC after: 3 weeks
Sponsored by: Citrix Systems R&D
When an mmap'd text file is written and then executed immediately
afterwards, it was possible that the modify time would change after the
text file was executing, resulting in the process executing the file
being killed. This was usually only observed when the file system's
times were set to higher resolution, but could have occurred for any
time resolution.
This was reported on a recent email list discussion.
This patch adds a VOP_SET_TEXT() to the NFS client which flushed all
dirty pages to the NFS server and then makes sure that n_mtime is up
to date to avoid this from occurring.
Thanks go to kib@ and pho@ for their help with developing this patch.
Tested by: pho
Reviewed by: kib
MFC after: 2 weeks
for example not in SYN-SENT.
This patch adds code to check the sysctl variable in other states than
LISTEN.
Thanks to ae and gnn for providing comments.
Reviewed by: gnn
MFC after: 1 week
Sponsored by: Netflix, Inc.
Differential Revision: https://reviews.freebsd.org/D9894
the method a lot.
Reduce the AND mask to the complement of the cursor's frame, so that area
inside the frame is not drawn first in black and then in lightwhite. The
AND-OR method is only directly suitable for the text mouse image, since
it doesn't go to the hardware there. Planar mode Mouse cursor drawing
takes 10-20 usec on my Haswell system (approx. 100 graphics accesses
at 130 nsec each), so the transient was not visible.
The method used the fancy read mode 1 and its color compare and color
don't care registers with value 0 in them so that all colors matched.
All that this did was make byte reads of frame buffer memory return 0xff,
so that the x86 case could obfuscate read+write as "and". The read must
be done for its side effect on the graphics controller but is not used,
except it must return 0xff to avoid affecting the write when the write
is obfuscated as a read-modify-write "and". Perhaps that was a good
optimization for 8088 CPUs where each extra instruction byte took as
long as a byte memory access.
Just use read+write after removing the fancy read mode. Remove x86
ifdefs that did the "and". After removing the "and" in the non-x86
part of the ifdefs, fix 4 of 6 cases where the shift was wrong.
in unusual cases. Optimize and significantly clean up removal in this
renderer. Optimize removal in the vga direct renderer.
Removal only needs to be done in the border (the part with pixels) in
both cases. The planar renderer used the condition scp->xoff > 0 to
test whether a right border exists. This actually tests for a left
border, and when the total horizontal border is 8 pixels, rounding gives
only a right border. This was the unusual broken case. An example
is easy to configure using something like "vidcontrol -f 8x16 iso-8x16
-g 79x25 MODE_27".
Optimize the planar case a little by only removing 9x13 active pixels
out of 16x16. Optimize it a lot by not doing anything if there is no
overlap with the border. Don't unroll the main loop or hard-code so
many assumptions about font sizes in it. On my Haswell system, graphics
memory and i/o accesses takes about 520 cycles each so optimizations from
unrolling are in the noise.
Optimize the direct case to not do anything if there is no overlap with
the border. Do a sanity check on the saveunder's coordinates. This
requires a previous change to pass non-rounded coordinates.
The previous commit also fixed the coordinates passed to the mouse
removal renderer. The coordinates were rounded down to a character
boundary, and thus essentially unusable. The renderer had to keep
track of the previous position, or clear a larger area. The latter
is only safe in the border, which is all that needs special handling
anyway.
I think no renderer depends on the bug. They have the following
handling:
- gfb sparc64: this seems to assume non-rounded coordinates
- gfb other: does nothing (seems to be missing border handling)
- vga text: does nothing (doesn't need border handling)
- vga planar: clears extras in the border, with some bugs. The fixes
will use the precise coordinates to optimize.
- vga direct: clears at the previous position with no check that it
is active, and clears everything. Checking finds this bug.
- others: are there any?
reduce hard-coded assumptions on font sizes so that the cursor size
can be more independent of the font size. Moving the mouse in the
buggy mode left trails of garbage.
The mouse cursor currently has size 9x13 in all modes. This can occupy
2x3 character cells with 8x8 fonts, but the algorithm was hard-coded
for only 2x2 character cells. Rearrange to hard-code only a maximum
cursor size (now 10x16) and to not hard-code in the logic. The number
of cells needed is now over-estimated in some cases.
2x3 character cells should also be used in text mode with 8x8 fonts
(except with large pixels, the cursor size should be reduced), but
hard-coding for 2x2 in the implementation makes it not very easy to
expand, and it practice it shifts out bits to reduce to 2x2.
In graphics modes, expansion is easier and there is no shifting out
for 9x13 cursors (but 9 is a limit for hard-coding for 2 8-bit VGA
cells wide). A previous commit removed the same buggy hard-coding for
removal at a lower level in planar mode. Another previous commit fixed
the much larger code for lower-level removal in direct mode; this is
independent of the font size so worked for 8x8 fonts. Text mode always
depended on the higher-level removal here, and always worked since
everything was hard-coded consistently for 2x2 character cells.
pagetables. This sets both bits when entering an address we know shouldn't
be executed.
I expect we could mark all userspace pages as Privileged execute-never to
ensure the kernel doesn't branch to one of these addresses.
While here add the ARMv8.1 upper attributes.
Reviewed by: alc, kib (previous version)
MFC after: 1 week
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D10360
scteken_init(). Move the internals of scteken_sync() into a local
function to help do this.
scteken_init() reset or adjusted the default attribute and screen
position at least 3 and 5 times, respectively. Warm init shouldn't
do any more than reset the "input" state.
(scterm-sc.c (which still works after minor editing), only resets
the escape state and the saved cursor position, and then does a
nearly-null sync of the current color.)
This mainly broke mode changes, and was most noticeable when the
background color is not teken's default (usually black). Then the
screen gets cleared in the wrong color. vidcontrol restores the
default normal attribute and tries to restore the default reverse
attribute. vidcontrol doesn't clear the screen again after restoring
the attribute(s), and it is too late to do it there without flicker.
Now the default normal attribute is restored before the change affects
the rendering.
When the foreground color is not teken's default, clearing with the
wrong attributes gave strange cursor colors for some cursor types.
The default reverse attribute is not restored since it is unsupported.
2/3 of the clobbering was from 2 resetting window resizing calls. The
second one is needed to restore the size, but must not reset. Window
resizing also sanitizes the cursor position, and after the main reset
resets the window size, the cursor row would often be adjusted from
24 to 23 if it were not already reset to 0. scteken_sync() is good
for restoring the window size and the cursor position in the correct
order, but was unusable at init time since scp->ts is not always
initialized then. Adjust to use its internals.
I didn't notice any problems from the cursor reset. The cursor should
be reset, and a previous fix was to reset it consistently a little
later.
Doing nothing for warm init works almost as well, if not better. It
is not very useful to reset the escape state for mode changes, since
the reset is especially likely to be null then. The escape state is
most likely to be non-initial and corrupted by its most normal uses
-- sloppy non-atomic output where a context switch or just mixing
stdout with stderr splits up escape sequences.
places possible in the kernel. This forces these functions to fail if
userspace is unable to access a given memory location, even if it is in
the user memory range.
This will simplify adding Privileged Access Never support later.
MFC after: 1 week
Sponsored by: DARPA, AFRL
Omit unused rates while initializing / dumping Tx power values.
They were not accessed anywhere (except for debugging), so this is
(mostly) no-op.
Tested with
* RTL8188EU, STA mode.
* RTL8812AU, STA mode.
Found by: PVS-Studio
1. Deadcode in ecore_init_cache_line_size(), qlnx_ioctl() and
qlnx_clean_filters()
2. ARRAY_VS_SINGLETON issue in qlnx_remove_all_mcast_mac() and
qlnx_update_rx_prod()
MFC after:5 days
If the ExchangeID/CreateSession operations done by an NFSv4.1 client
after the server crashes/reboots fails, it is possible that some process/thread
is waiting for an open_owner lock. If the client state is free'd, this
can cause a crash.
This would not normally happen, but has been observed on a mount of the
AmazonEFS service.
Reported by: cperciva
Tested by: cperciva
PR: 216086
MFC after: 2 weeks
during recovery.
If the NFSv4.1 client gets a NFSv4.1 NFSERR_BADSESSION reply to an Open/Lock
operation while recovering from the server crash/reboot, allow the opens
to be retained for a subsequent recovery attempt. Since NFSv4.1 servers
should only reply NFSERR_BADSESSION after a crash/reboot that has lost
state, this case should almost never happen.
However, for the AmazonEFS file service, this has been observed when
the client does a fresh TCP connection for RPCs.
Reported by: cperciva
Tested by: cperciva
PR: 216088
MFC after: 2 weeks
if_vlan(4) interfaces inherit IPv4 checksum offloading flags from the
parent when VLAN_HWCSUM and VLAN_HWTAGGING flags are present on the
parent interface. Do the same for IPv6 checksum offloading flags.
Reported by: Harry Schmalzbauer
Reviewed by: np, gnn
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D10356
With the zfs probe cleanup, the mistake did slip in the probe code;
instead of reading the pool GUID for the actual boot device (partition),
we read GUID for first found pool from the boot disk.
This will break the case when there are both zfs pool and ufs on the boot
disk, and the ufs is used for boot, not zfs.
Reviewed by: smh
Differential Revision: https://reviews.freebsd.org/D10359
administrator.
This restores the behavior that was prior to r274246.
No objection from: #network
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D10215
Instead, issue a diagnostic and return appropriate error if
ncl_flush() was unable to clean buffer queue after the specified
number or retries.
Reviewed by: rmacklem
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
This is in preparation for A-MSDU decap in A-MPDU support.
* refactor out the code to purge a single reorder slot into ampdu_rx_purge_slot().
* refactor out the code to add a frame to the given reorder slot
to ampdu_rx_add_slot().
This should be a big no-op as far as current code is concerned.
Tested:
* QCA9880v2, STA mode (11ac)
* iwn(4), STA mode (11n)
Reviewed by: avos
Differential Revision: https://reviews.freebsd.org/D10328
Larry Rosenman reported a crash on freebsd-current@ which was caused by
a premature release of the krpc backchannel socket structure.
I believe this was caused by a race between the SVC_RELEASE() in clnt_vc.c
and the xprt_unregister() in the higher layer (clnt_rc.c), which tried
to lock the mutex in the xprt structure and crashed.
This patch fixes this by removing the xprt_unregister() in the clnt_vc
layer and allowing this to always be done by the clnt_rc (higher reconnect
layer).
Reported by: ler@lerctr.org
Tested by: ler@letctr.org
MFC after: 2 weeks
The AmazonEFS NFSv4.1 server does not support the FILES_FREE and FILES_TOTAL
attributes. As such, an NFSv4.1 mount to the server would return garbage
for these values. This patch initializes the fields of the nfsstatfs structure,
so that "df" and friends will at least return consistent bogus values.
This patch should have effect when mounting other NFSv4.1 servers.
Reported by: cperciva
MFC after: 2 weeks
We are otherwise susceptible to a race with a concurrent vm_map_wire(),
which may drop the map lock to fault pages into the object chain. In
particular, vm_map_protect() will only copy newly writable wired pages
into the top-level object when MAP_ENTRY_USER_WIRED is set, but
vm_map_wire() only sets this flag after its fault loop. We may thus end
up with a writable wired entry whose top-level object does not contain the
entire range of pages.
Reported and tested by: pho
Reviewed by: kib
MFC after: 1 week
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D10349
This update does add the code to pass partition size to vdev_probe() via
simple callback. Tested via tinderbox build, but not yet with actual boot.
The code can be improved still, but to verify the idea to read media
block size and amedia size has to be confirmed first.
Reviewed by: imp
Differential Revision: https://reviews.freebsd.org/D10302
Entries may be removed and freed if an I/O error occurs during mirror
synchronization, so we cannot assume that all entries of ds_bios are
valid.
Also ensure that a synchronization BIO's array index is preserved after
a successful write.
Reported and tested by: pho
MFC after: 2 weeks
Sponsored by: Dell EMC Isilon
Arm64 pmap interprets accessed writable ptes as modified, since
ARMv8.0 does not track Dirty Bit Modifier in hardware. If writable bit
is removed, page must be marked as dirty for MI VM.
This change is most important for COW, where fork caused losing
content of the dirty pages which were not yet scanned by pagedaemon.
Reviewed by: alc, andrew
Reported and tested by: Mark Millard <markmi@dsl-only.net>
PR: 217138, 217239
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Switched from using timestamps to RTT estimates when performing TCP receive
buffer auto resizing, as not all hosts support / enable TCP timestamps.
Disabled reset of receive buffer auto scaling when not in bulk receive mode,
which gives an extra 20% performance increase.
Also extracted auto resizing to a common method shared between standard and
fastpath modules.
With this AWS S3 downloads at ~17ms latency on a 1Gbps connection jump from
~3MB/s to ~100MB/s using the default settings.
Reviewed by: lstewart, gnn
MFC after: 2 weeks
Relnotes: Yes
Sponsored by: Multiplay
Differential Revision: https://reviews.freebsd.org/D9668
like I hoped, since they are needed for removing parts over the border.
Continue fixing bugs in them.
In the vga planar mode renderer, remove removal of the part of the
image over the text window. This was hard-coded for nearly 8x16 fonts
and in practice didn't remove enough for 8x8 fonts. This used the
wrong attribute over cutmarked regions. The caller refreshes with the
correct attribute later, so the attribute bug only caused flicker.
The caller uses the same hard-coding, so the refreshes fix up all the
spots with the wrong attribute, but keep missing the missed spots.
This still gives trails of bits of cursors for cursor motions in the
affected configurations (mainly depth 4 modes with 8x8) fonts. 8x14
fonts barely escape the problem since although the cursor is drawn
as 16x16, its active part is only 9x13 and the active part fits in
the hard-coded 2x2 character cell window for 8x14 fonts. 8x8 fonts
need a 2x3 window.
In the fb non-sparc64 renderer, the buggy image removal was buggier
and was already avoided by returning before it. Remove it completely
and fix nearby style bugs. It was essentially the same as for the vga
planar mode renderer (obfuscated by swapping x and y). This was buggier
since fb should handle more types of hardware so the hard-coding is
wronger.
The remaining fb image removal is also buggier. It never supported
software cursors drawn into the border, and the hardware cursor is
probably broken by other bugs to be fixed soon.
This patch gives a requestor of the exclusive lock on the client state
in the NFSv4 client priority over shared lock requestors. This avoids
the server crash recovery thread being starved out by other threads doing
RPCs.
Tested by: cperciva
PR: 216087
MFC after: 2 weeks
When the NFSv4 client Commit operation encountered a stale write verifier,
it erroneously mapped that to EIO. This could have caused recently written
data to be lost when a server crashes/reboots between an UNSTABLE write
and the subsequent commit. This patch fixes this.
The bug was only for the NFSv4 client and did not affect NFSv3.
Tested by: cperciva
PR: 215887
MFC after: 2 weeks
For the ReclaimComplete operation, the RPC layer should not loop on
NFSERR_BADSESSION. If it does, the recovery thread (nfscl) can get stuck
looping and will not do a recovery.
This patch fixes it so it does not loop. This bug only affects NFSv4.1 and
only when a server reboots.
Tested by: cperciva
PR: 215886
MFC after: 2 weeks
imx6 based on a single cortex-a7 core. Other changes to imx6 drivers
and support code are needed to fully support the imx6ul.
Also fix an indentation glitch committed in the prior change.
driver for imx6. Some newer dts source puts the GIC node at the root
instead of under /soc, so look in both places. Also, sometimes the GIC
node doesn't list itself as its own interrupt-parent, allow that too.
At this point, INTRNG is not going away (the option may go away, but the
code is not), so we no longer need code to support workarounds that handled
the lack of INTRNG functionality.
If an operation that preceeds a Setattr in an NFSv4 compound fails,
there is no bitmap of attributes to parse. Without this patch, the
parsing would fail and return EBADRPC instead of the correct failure
error. This could break recovery from a server crash/reboot.
Tested by: cperciva
PR: 215883
MFC after: 2 weeks
The work to implement zfs reader to inspect all pool label copies did
miss the userboot, this update does correct this issue.
Since userboot is already using common/disk.c API (disk_open() etc),
the fix is quite simple - we only need to make sure the userdisk_iocl()
will call disk_ioctl(). In addition, the diskioctl callback does return
int, not pointer.
Note, the review request is actually addressing the sparc and userboot,
but as testing the fix for sparc will take some more time, I am posting the
userboot fix now.
This patch is part of the implementation presented in review:
https://reviews.freebsd.org/D10302
Once we have the sparc part tested, we will have the complete fix
for the issue.
Reviewed by: imp
The MFC will include a compat definition of smp_no_rendevous_barrier()
that calls smp_no_rendezvous_barrier().
Reviewed by: gnn, kib
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D10313
- renaming l_ifreq::ifru_metric to l_ifreq::ifru_ivalue;
- adding a definition for ifr_ifindex which points to l_ifreq::ifru_ivalue.
A quick search indicates that Linux already got the above changes since 2.1.14.
Reviewed by: kib, marcel, dchagin
MFC after: 1 week
This is applicable only to the older processors that do not have the AMD
Topology extension.
Opteron 6100-series "Magny-Cours" processors had multiple nodes within a
package and didn't have the Topology extension. Without this change
FreeBSD would assume that those processors have a single L3 cache shared
by all cores while, in fact, each node has its own L3 cache.
Many thanks to Freddie Cash <fjwcash@gmail.com> for providing valuable
hardware information.
MFC after: 2 weeks
(that is, in all supported 8, 15, 16 and 24-color modes). Moving the
mouse cursor while holding down a button (giving cut marking) left a
trail of garbage from misremoved mouse cursors (usually colored
rectangles and not cursor shapes). Cases with a button not held down
worked better and may even have worked.
No renderer support for removing (software) mouse cursors is needed
(and many renderers don't have any), since sc_remove_mouse_image()
marks for update the region containing the image and usually much
more. The mouse cursor can be (partially) over as many as 4 character
cells, and removing it in only the 1-4 cells occupied by it would be
best for efficiency and for avoiding flicker. However,
sc_remove_mouse_image() can only mark a single linear region and
usually marks a full row of cells and 1 more to be sure to cover the
4 cells. It always does this, so using the special rendering method
just wastes even more time and gives even more flicker. The special
methods will be removed soon.
The general method always works. vga_pxlmouse_direct() appeared to
defer to it by returning immediately if !on. However,
vga_pxlmouse_direct() actually did foot-shooting using a disguised
saveunder method. Normal order near a mouse move is:
(1) remove the mouse cursor in the renderer (optional)
(2) remove the mouse cursor again and refresh the screen over the
mouse cursor and much more from the vtb. When the mouse has
actually moved and a button is down, many attributes in this
region are changed to be up to date with the new cut marking
(3) draw the keyboard cursor again if it was clobbered by the update
(4) draw the mouse cursor image in its new position.
The bug was to remove the mouse cursor again in step (4), before the
drawing it again in (4), using a saveunder that was valid in step (1)
at best. The quick fix is to use the saveunder in step (1) and not
in step (4). Using it in step (4) also used it before it was
initialized, initially and after mode and screen switches.
in the vga renderer. Removal used stale attributes and didn't try to
merge with the current attribute for cut marking, so special rendering
of cut marking was lost in many cases. The gfb renderer is too broken
to support special rendering of cut marking at all, so this change is
supposed to be just a style fix for it. Remove all traces of the
saveunder method which was used to implement this bug.
Fix drawing of the cursor image in text mode, only in the vga
renderer. This used a stale attribute from the frame buffer instead
of from the saveunder, but did merge with the current attribute for
cut marking so it caused less obvious bugs (subtle misrendering for
the character under the cursor).
The saveunder method may be good in simpler drivers, but in syscons
the 'under' is already saved in a better way in the vtb. Just redraw
it from there, with visible complications for cut marking and
invisible complications for mouse cursors. Almost all drawing
requests are passed a flag 'flip' which currently means to flip to
reverse video for characters in the cut marking region, but should
mean that the the characters are in the cut marking regions so should
be rendered specially, preferably using something better than reverse
video. The gfb renderer always ignores this flag. The vga renderer
ignored it for removal of the text cursor -- the saveunder gave the
stale rendering at the time the cursor was drawn. Mouse cursors need
even more complicated methods. They are handled by drawing them last
and removing them first. Removing them usually redraws many other
characters with the correct cut marking (but transiently loses the
keyboard cursor, which is redrawn soon). This tended to hide the
saveunder bug for forward motions of the keyboard cursor. But slow
backward motions of the keyboard cursor always lost the cut marking,
and fast backwards motions lost in for about 4 in every 5 characters,
depending on races with the scrn_update() timeout handler. This is
because the forward motions are usually into the region redrawn for
the mouse cursor, while backwards motions rarely are.
Text cursor drawing in the vga renderer used also used a
possibly-stale copy of the character and its attribute. The vga
render has the "optimization" of sometimes reading characters from the
screen instead of from the vtb (this was not so good even in 1990 when
main memory was only a few times faster than video RAM). Due to care
in update orders, the character is never stale, but its attribute
might be (just the cut marking part, again due to care in order).
gfb doesn't have the scp->scr pointer used for the "optimization", and
vga only uses this pointer for text mode. So most cases have to
refresh from the vtb, and we can be sure that the ordering of vtb
updates and drawing is as required for this to work.