davidcs [Thu, 15 Jun 2017 02:45:43 +0000 (02:45 +0000)]
Upgrade STORMFW to 8.30.0.0 and ecore version to 8.30.0.0
Add support for pci deviceID 0x8070 for QLE41xxx product line which
supports 10GbE/25GbE/40GbE
emaste [Wed, 14 Jun 2017 18:56:33 +0000 (18:56 +0000)]
lld: Fix weak symbols on arm and aarch64
Given
.weak target
.global _start
_start:
b target
The intention is that the branch goes to the instruction after the
branch, effectively turning it on a nop. The branch adds the runtime
PC, but we were adding it statically too.
I noticed the oddity by inspection, but llvm-objdump seems to agree,
since it now prints things like:
https://www.illumos.org/issues/8311
Description:
There was a misunderstanding about the enforcement details of the "Read-only"
flag introduced for SMB/CIFS compatibility, way back in 2007 in the Sun PSARC
2007/315 case.
The original authors thought enforcement of the READONLY flag should work
similarly as the IMMUTABLE flag. Unfortunately, that enforcement is
incompatible with the expectations of Windows applications using this feature
through the SMB service. Applications assume (and the MS File System Algorithms
MS-FSA confirms they should) that an SMB client can:
(a) Open an SMB handle on a file with read/write access,
(b) Set the DOS attributes to include the READONLY flag,
(c) continue to have write access via that handle.
This access model is essentially the same as a Unix/POSIX application that
creates a file (with read/write access), uses fchmod() to change the file mode
to something not granting write access (i.e. 0444), and then continues to write
that file using the open handle it got before the mode change.
Currently, the SMB server works-around this problem in a way that will become
difficult to maintain as we implement support for SMB3 persistent handles, so
SMB depends on this fix.
I've written a test program that can be used to demonstrate this problem, and
added it to zfs-tests (tests/functional/acl/cifs/cifs_attr_004_pos).
It currently fails, but will pass when this problem fixed.
Steps to Reproduce:
Run the test program on a ZFS file system.
Expected Results:
Pass
Actual Results:
Fail.
Reviewed by: Sanjay Nadkarni <sanjay.nadkarni@nexenta.com>
Reviewed by: Yuri Pankov <yuri.pankov@nexenta.com>
Reviewed by: Andrew Stormont <andyjstormont@gmail.com>
Reviewed by: Matt Ahrens <mahrens@delphix.com>
Reviewed by: John Kennedy <john.kennedy@delphix.com>
Approved by: Prakash Surya <prakash.surya@delphix.com>
Author: Gordon Ross <gwr@nexenta.com>
MFC after: 2 weeks
Most of the upstream change is not applicable to FreeBSD.
Only the renaming of strtonum to zfs_strtonum is relevant to us.
And we already had it partially done.
Reviewed by: Robert Mustacchi <rm@joyent.com>
Approved by: Joshua M. Clulow <josh@sysmgr.org>
Author: Yuri Pankov <yuri.pankov@nexenta.com>
MFC after: 1 week
jhb [Wed, 14 Jun 2017 13:34:09 +0000 (13:34 +0000)]
Don't try to assign interrupts to a CPU on single-CPU systems.
All interrupts are routed to the sole CPU in that case implicitly.
This is a regression in EARLY_AP_STARTUP. Previously the 'assign_cpu'
variable was only set when a multi-CPU system finished booting, so
it's value both meant that interrupts could be assigned and that
there was more than one CPU.
PR: 219882
Reported by: ota@j.email.ne.jp
MFC after: 3 days
jhibbits [Wed, 14 Jun 2017 04:26:37 +0000 (04:26 +0000)]
Actually add the mpc85xx_get_platform_clock() function.
Follow up r319935 by actually committing the mpc85xx_get_platform_clock()
function. This function was created to facilitate other development, and I
thought I had committed it earlier.
Some blocks depend on the platform clock rather than the system clock.
The System clock is derived from the platform clock as one-half the
platform clock. Rewrite mpc85xx_get_system_clock() to use the new
function.
markj [Wed, 14 Jun 2017 03:55:11 +0000 (03:55 +0000)]
Don't call vm_pager_page_unswapped() when writing or deleting a dirty page.
The swap space backing a clean page is released when it is first dirtied,
so there's no need to attempt to release swap space when the page is
already dirty.
markj [Wed, 14 Jun 2017 03:50:02 +0000 (03:50 +0000)]
Free the request page if an I/O error occurs while reading from swap.
After such a failure, the page is invalid, so there's point in keeping it
around. Moreover, such pages were not being inserted into the active queue,
making them unreclaimable until a subsequent write or delete made them
valid.
markj [Wed, 14 Jun 2017 03:45:26 +0000 (03:45 +0000)]
Fix handling of subpage BIO_WRITE and BIO_DELETE requests on swap MDs.
Such requests would previously mark the entire page as valid, which was
incorrect since nothing guaranteed that the page's contents had been
initialized. This change also modifies subpage BIO_DELETEs so that the
entire page is marked dirty, rather than only a subrange. There is no
benefit to creating partially dirty swap pages.
Reviewed by: alc, kib (previous version)
MFC after: 3 days
zbb [Tue, 13 Jun 2017 18:55:21 +0000 (18:55 +0000)]
Enable HWPMC overflow IRQ on both CPUs in MPIC
This commit enables usage of HWPMC interrupts for the
Marvell SoCs, which use MPIC (Armada38x and ArmadaXP).
Those interrupts require extra unmasking, comparing to
others. Also, in order to process counters per-CPU,
they are masked/unmasked using separate registers' sets
for each core.
Submitted by: Michal Mazur <mkm@semihalf.com>
Marcin Wojtas <mw@semihalf.com>
Obtained from: Semihalf
Sponsored by: Stormshield, Netgate
Differential revision: https://reviews.freebsd.org/D10913
zbb [Tue, 13 Jun 2017 18:53:56 +0000 (18:53 +0000)]
Fix INVARIANTS debug code in HWPMC
When HWPMC stops sampling, ps_pmc may be freed before samples
are processed. In such situation treat PMC as stopped.
Add "ifdef" to fix build without INVARIANTS code.
zbb [Tue, 13 Jun 2017 18:52:39 +0000 (18:52 +0000)]
Fix event table for Cortex A9.
Removed events 0x8 (INSTR_EXECUTED), 0xE (PC_PROC_RETURN) and
0x13-0x1d not supported on Cortex A9.
Add events 0x68 and 0x6E which replaced 0x8 and 0xE.
zbb [Tue, 13 Jun 2017 18:46:29 +0000 (18:46 +0000)]
Introduce Armada 38x/XP network controller support
This patch contains a new driver for the network unit of Marvell
Armada 38x/XP SoCs, called NETA. This support was thoroughly tested
and optimised in terms of stability and performance. Additional
hardware features, like Buffer Management (BM) or Parser and Classifier
(PnC) will be progressively supported as needed.
Submitted by: Fabien Thomas <fabien.thomas@stormshield.eu>
Arnaud Ysmal <arnaud.ysmal@stormshield.eu>
Zbigniew Bodek <zbb@semihalf.com>
Michal Mazur <mkm@semihalf.com>
Bartosz Szczepanek <bsz@semihalf.com>
Marcin Wojtas <mw@semihalf.com>
zbb [Tue, 13 Jun 2017 18:35:14 +0000 (18:35 +0000)]
Prevent multiple lock initialization in e6000sw probe
r319886 ("Add the initial support for the Marvell 88E6141
and 88E6341 switches.") unveiled a problem with possible
multiple lock creation. Move its initialization
to the driver attach and for obtaining the switch ID
create a temprorary one, which is immediately destroyed
after the check.
Submitted by: Zbigniew Bodek <zbb@semihalf.com>
Marcin Wojtas <mw@semihalf.com>
Obtained from: Semihalf
alc [Tue, 13 Jun 2017 17:49:49 +0000 (17:49 +0000)]
Reduce the frequency of hint updates on allocation without incurring
additional allocation overhead. Previously, blst_meta_alloc() updated the
hint after every successful allocation. However, these "eager" hint
updates are of no actual benefit if, instead, the "lazy" hint update at
the start of blst_meta_alloc() is generalized to handle all cases where
the number of available blocks is less than the requested allocation.
Previously, the lazy hint update at the start of blst_meta_alloc() only
handled the ALL-FULL case. (I would also note that this change provides
consistency between blist_alloc() and blist_fill() in that their hint
maintenance is now entirely lazy.)
Eliminate unnecessary checks for terminators in blst_meta_alloc() and
blst_meta_fill() when handling ALL-FREE meta nodes.
Eliminate the field "bl_free" from struct blist. It is redundant. Unless
the entire radix tree is a single leaf, the count of free blocks is stored
in the root node. Instead, provide a function blist_avail() for obtaining
the number of free blocks.
In blst_meta_alloc(), perform a sanity check on the allocation once rather
than repeating it in a loop over the meta node's children.
In blst_leaf_fill(), use the optimized bitcount*() function instead of a
loop to count the blocks being allocated.
mmel [Tue, 13 Jun 2017 12:07:18 +0000 (12:07 +0000)]
Implement tunable CPU quirks.
These quirks are intended for optimizing CPU performance, not for
applying errata workarounds. Nobody can expect that CPU with unfixed
errata is stable enough to execute the kernel until quirks are applied.
emaste [Tue, 13 Jun 2017 01:25:19 +0000 (01:25 +0000)]
lld: revert accidentally committed change from r319887
This change is a portion of LLD rev 305212 which accidentally ended
up in my svn tree. We do want to backport the change to LLD 4.0, but
it needs additional work and was not supposed to be included in
r319887.
ian [Tue, 13 Jun 2017 01:12:37 +0000 (01:12 +0000)]
By popular demand: change MAKE_GENERIC_KERNELS to MAKE_LINT_KERNELS.
It appears that the same arches that lack GENERIC kernel configs also lack
LINT. But enough different arches get built to ensure a kernel change
should build everywhere (32 and 64 bit, clang and old gcc, little and big
endian).
emaste [Tue, 13 Jun 2017 00:31:16 +0000 (00:31 +0000)]
lld: ELF: Fix ICF crash on absolute symbol relocations.
If two sections contained relocations to absolute symbols with the same
value we would crash when trying to access their sections. Add a check that
both symbols point to sections before accessing their sections, and treat
absolute symbols as equal if their values are equal.
Obtained from: LLD commit r292578
MFC after: 3 days
rmacklem [Mon, 12 Jun 2017 23:41:20 +0000 (23:41 +0000)]
Define NFS_MAXXDR as the upper bound on XDR overhead in an NFS RPC.
This definition is a part of the maxiotune2 patch that will be
committed soon. It is being committed separately to ease merging
with the pNFS projects subversion trees.
kib [Mon, 12 Jun 2017 21:03:23 +0000 (21:03 +0000)]
Move struct syscall_args syscall arguments parameters container into
struct thread.
For all architectures, the syscall trap handlers have to allocate the
structure on the stack. The structure takes 88 bytes on 64bit arches
which is not negligible. Also, it cannot be easily found by other
code, which e.g. caused duplication of some members of the structure
to struct thread already. The change removes td_dbg_sc_code and
td_dbg_sc_nargs which were directly copied from syscall_args.
The structure is put into the copied on fork part of the struct thread
to make the syscall arguments information correct in the child after
fork.
This move will also allow several more uses shortly.
Reviewed by: jhb (previous version)
Sponsored by: The FreeBSD Foundation
MFC after: 3 weeks
X-Differential revision: https://reviews.freebsd.org/D11080
emaste [Mon, 12 Jun 2017 19:45:05 +0000 (19:45 +0000)]
remove stale dependencies for utimens* wrappers removed in r319663
Use a similar approach to r318957 (which was for ptrace dependencies):
grep the .depend file for the old source file and delete the stale
dependency if found.
Reviewed by: bdrewery
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D11091
ian [Mon, 12 Jun 2017 17:22:23 +0000 (17:22 +0000)]
Add support for "make universe_kernels -DMAKE_GENERIC_KERNELS" to build
just the GENERIC kernels for each arch (including variations such as
GENERIC-NODEBUG, GENERIC64, etc).
This helps with quickly doing a test build for all[*] arches without
building dozens of variant kernels for the arches that have lots of
hardware/board/system variations.
[*] Not all arches have a generic kernel (but they probably should for
test-building purposes, even if it can't boot on any real hardware).
ngie [Mon, 12 Jun 2017 16:38:37 +0000 (16:38 +0000)]
Add a testcase for `ln -sF`
The testcase fails today, so mark it with atf_expect_fail: in
particular, the target (B) isn't being unlinked and the documentation
doesn't suggest special handling for directories. Thus, there's either
a doc or an implementation bug in ln(1) that needs to be resolved.
ngie [Mon, 12 Jun 2017 16:31:32 +0000 (16:31 +0000)]
Use readlink(1)/stat(1) to query symlinks instead of file(1)
file(1) can be compiled out of the system via MK_FILE == no, and the
output isn't guaranteed to be stable. It's better to use stat(1)/readlink(1)
instead to query symlink/file paths.
delphij [Mon, 12 Jun 2017 09:11:31 +0000 (09:11 +0000)]
Fix buffer lengths.
After r319369, the RPC code validates caller supplied buffer length in
taddr2uaddr. When no -h is specified, the sizeof(ai_addr) is used,
which is always smaller than the required size and therefore uaddr
would be NULL, causing the kernel to copyin() from userland NULL
and fail with EFAULT.
dchagin [Mon, 12 Jun 2017 07:35:59 +0000 (07:35 +0000)]
Since r318735 (ino64 project) the size of the native struct dirent is
equal or greater than the size of Linux struct dirent or struct dirent64.
So, remove LINUX_RECLEN_RATIO magic as useless.
ngie [Mon, 12 Jun 2017 02:38:37 +0000 (02:38 +0000)]
Remove stdlib.h #include added in r319844
A previous iteration of the tests I added in r319844 involved free(3), but
that attempt didn't pan out, so I switched to stack allocated buffers instead
of heap allocated ones, making the #include unnecessary.
ngie [Mon, 12 Jun 2017 00:43:14 +0000 (00:43 +0000)]
getbsize(3): clarify that underflow/overflow warnings in regard to $BLOCKSIZE
gets output via warnx(3)
This helps set expectations for how one might deal with those messages, i.e.,
mute output from /dev/stderr today, since that's where vwarn(3) outputs messages
to today.
ngie [Mon, 12 Jun 2017 00:21:55 +0000 (00:21 +0000)]
Add initial tests for stat(1)
Testcases for -H, -L, and -f haven't been implemented yet, in part due
to additional complexity needed to validate the features:
* -H and -f will require an external "helper" program to display/modify
the state/permissions for a given path.
* -L is being covered partially via the -n testcase today.
jilles [Sun, 11 Jun 2017 19:06:07 +0000 (19:06 +0000)]
rc.subr: Optimize repeated sourcing.
When /etc/rc runs all /etc/rc.d scripts, it has already loaded /etc/rc.subr
but each /etc/rc.d script sources it again (since /etc/rc.d scripts must
also work when started stand-alone).
Therefore, if rc.subr is already loaded, return so sh need not parse the
rest of the file.
A second effect is that there is no longer a compound command around most of
rc.subr. This reduces memory usage while sh is loading rc.subr for the first
time (but this memory is free()d once rc.subr is loaded).
For purposes of porting this to other systems, I do not recommend porting
this to systems with shells that do not have the change to the return
special builtin like in r255215 (before FreeBSD 10.0-RELEASE). This change
ensures that return in the top level of a dot script returns from the dot
script, even if the dot script was sourced from a function.
A comparison of CPU time on an amd64 bhyve virtual machine from a times
command added near the end of /etc/rc, all four values summed:
x orig1
+ quickreturn
+--------------------------------------------------------------------------+
| + + + x x x|
||______M__A_________| |______M___A__________| |
+--------------------------------------------------------------------------+
N Min Max Median Avg Stddev
x 3 1.704 1.802 1.726 1.744 0.051419841
+ 3 1.467 1.559 1.487 1.5043333 0.048387326
Difference at 95.0% confidence
-0.239667 +/- 0.113163
-13.7424% +/- 6.48873%
(Student's t, pooled s = 0.0499266)
projects / FreeBSD / FreeBSD.git / log