generate code to branch based on an address in a register. This allows us
to have binaries larger than the 32MiB limit of a branch instruction.
The main use of this is with clang. Clang 3.8.0 has been shown to be larger
than the above limit.
up to now.
The new sendfile is the code that Netflix uses to send their multiple tens
of gigabits of data per second. The new implementation features asynchronous
I/O, when I/O operations are launched, but not awaited to be complete. An
explanation of why such behavior is beneficial compared to old one is
going to be too long for a commit message, so we will skip it here.
Additional features of new syscall are extra flags, which provide an
application more control over data sent. The SF_NOCACHE flag tells
kernel that data shouldn't be cached after it was sent. The SF_READAHEAD()
macro allows to specify readahead size in pages.
The new syscalls is a drop in replacement. No modifications are required
to applications. One can take nginx binary for stable/10 and run it
successfully on head. Although SF_NODISKIO lost its original sense, as now
sendfile doesn't block, and now means something completely different (tm),
using the new sendfile the old way is absolutely safe.
Celebrates: Netflix global launch!
Sponsored by: Nginx, Inc.
Sponsored by: Netflix
Relnotes: yes
Although we don't use it in tree yet libstand is installed as user-
facing /usr/liblibstand.a, and some work in progress makes use of it.
Instead of conflicting with ongoing libstand Makefile deduplication,
just add it now.
Set _PATH_DEFPATH to
/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin:/usr/local/bin. This is the
path in the default class in the default /etc/login.conf,
excluding ~/bin which would not be expanded properly in a string
constant.
For normal logins, _PATH_DEFPATH is overridden by /etc/login.conf,
~/.login_conf or shell startup files. _PATH_DEFPATH is still used as a
default by execlp(), execvp(), posix_spawnp() and sh if PATH is not set, and
by cron. Especially the latter is a common trap (most recently in PR
204813).
PR: 204813
Reviewed by: secteam (delphij), alfred
Just remove it completely from the test as it's initialized but unused apart
from the free(3) call
Differential Revision: https://reviews.freebsd.org/D4769 (part of larger diff)
MFC after: 5 days
Reported by: cppcheck
Reviewed by: oshogbo
Sponsored by: EMC / Isilon Storage Division
the nvlists
Convert some of the variables from int to unsigned int to squelch -Wsign-compare
warnings when converting hardcoded values to nitems(..)
Differential Revision: https://reviews.freebsd.org/D4769 (part of larger diff)
MFC after: 5 days
Reviewed by: oshogbo
Sponsored by: EMC / Isilon Storage Division
Rename the `nitems` variable to `num_items` to avoid collisions with the
macro in sys/param.h for counting elements in an array
Similarly, rename `string` to `string_arr` to avoid future collisions with
potential keywords, as well as make it clear that `string_arr` isn't a char*
value, but instead a char** value.
Differential Revision: https://reviews.freebsd.org/D4769 (part of larger diff)
MFC after: 5 days
Reviewed by: oshogbo
Sponsored by: EMC / Isilon Storage Division
libstand is built in three places (lib/libstand, sys/boot/libstand32,
and sys/boot/userboot/libstand). Reduce Makefile duplication by
.including libstand/Makefile from sys/boot/libstand32/Makefile.
sys/boot/userboot/libstand/Makefile will be addressed later, as it
contains additional differences yet to be handled.
This change also switches libstand32 to use the new uuid_from_string
and uuid_to_string, which was not included in r292473.
Reviewed by: jhb
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D4747
relocation fixups unsupported by the self_reloc() code, and don't optimize
memcpy/memset using floating point registers, because in a standalone
environment nothing has initialized the fpu hardware.
aligned on a int64_t boundary. However, when we allocate the array of
these structures, we use ALIGNBYTES which defaults to sizeof(int) on
arm, i386 and others. The i386 stuff can handle unaligned accesses
seemlessly. However, arm cannot. Take this into account when creating
the array of FILEs, and add some comments about why.
Differential Revision: https://reviews.freebsd.org/D4708
cperciva's libmd implementation is 5-30% faster
The same was done for SHA256 previously in r263218
cperciva's implementation was lacking SHA-384 which I implemented, validated against OpenSSL and the NIST documentation
Extend sbin/md5 to create sha384(1)
Chase dependancies on sys/crypto/sha2/sha2.{c,h} and replace them with sha512{c.c,.h}
Reviewed by: cperciva, des, delphij
Approved by: secteam, bapt (mentor)
MFC after: 2 weeks
Sponsored by: ScaleEngine Inc.
Differential Revision: https://reviews.freebsd.org/D3929
(CLOCK_REALTIME case) system calls is non negative.
This commit hides a kernel panic in atrtc_settime() as the clock_ts_to_ct()
does not properly convert negative tv_sec.
ps. in my opinion clock_ts_to_ct() should be rewritten to properly handle
negative tv_sec values.
Differential Revision: https://reviews.freebsd.org/D4714
Reviewed by: kib
MFC after: 1 week
Traditionally the hcreate() function creates a hash table that uses
chaining, using a fixed user-provided size. The problem with this
approach is that this often either wastes memory (table too big) or
yields bad performance (table too small). For applications it may not
always be easy to estimate the right hash table size. A fixed number
only increases performance compared to a linked list by a constant
factor.
This problem can be solved easily by dynamically resizing the hash
table. If the size of the hash table is at least doubled, this has no
negative on the running time complexity. If a dynamically sized hash
table is used, we can also switch to using open addressing instead of
chaining, which has the advantage of just using a single allocation for
the entire table, instead of allocating many small objects.
Finally, a problem with the existing implementation is that its
deterministic algorithm for hashing makes it possible to come up with
fixed patterns to trigger an excessive number of collisions. We can
easily solve this by using FNV-1a as a hashing algorithm in combination
with a randomly generated offset basis.
Measurements have shown that this implementation is about 20-25% faster
than the existing implementation (even if the existing implementation is
given an excessive number of buckets). Though it allocates more memory
through malloc() than the old implementation (between 4-8 pointers per
used entry instead of 3), process memory use is similar to the old
implementation as if the estimated size was underestimated by a factor
10. This is due to the fact that malloc() needs to perform less
bookkeeping.
Reviewed by: jilles, pfg
Obtained from: https://github.com/NuxiNL/cloudlibc
Differential Revision: https://reviews.freebsd.org/D4644
bugfix-only release, with no new features.
Please note that from 3.5.0 onwards, clang and llvm require C++11
support to build; see UPDATING for more information.
POSIX requires for the c99 compiler.
(In fact, our c99(1) already ignores -lxnet; but our make(1) doesn't set
${CC} correctly, and our cc(1) treats xnet like any other library.)
Reviewed by: kib
450 seconds
This is required on slower network connections, and on older releases
(stable/10 seems to be slower as far as name resolution goes.. not sure
why yet).
Remove an outdated comment in the Makefile from when I was working on
this code over a year ago on github
MFC after: 1 week
Sponsored by: EMC / Isilon Storage Division
i386 is the only current FreeBSD architecture that ever used a.out
format.
Reviewed by: kib
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D4687
sysdecode_ioctlname() function. This function matches the behavior
of the truss variant in that it returns a pointer to a string description
for known ioctls. The caller is responsible for displaying unknown
ioctl requests. For kdump this meant moving the logic to handle unknown
ioctl requests out of the generated function and into an ioctlname()
function in kdump.c instead.
Differential Revision: https://reviews.freebsd.org/D4610
The existing implementations of POSIX tsearch() and tdelete() don't
attempt to perform any balancing at all. Testing reveals that inserting
100k nodes into a tree sequentially takes approximately one minute on my
system.
Though most other BSDs also don't use any balanced tree internally, C
libraries like glibc and musl do provide better implementations. glibc
uses a red-black tree and musl uses an AVL tree.
Red-black trees have the advantage over AVL trees that they only require
O(1) rotations after insertion and deletion, but have the disadvantage
that the tree has a maximum depth of 2*log2(n) instead of 1.44*log2(n).
My take is that it's better to focus on having a lower maximum depth,
for the reason that in the case of tsearch() the invocation of the
comparator likely dominates the running time.
This change replaces the tsearch() and tdelete() functions by versions
that create an AVL tree. Compared to musl's implementation, this version
is different in two different ways:
- We don't keep track of heights; just balances. This is sufficient.
This has the advantage that it reduces the number of nodes that are
being accessed. Storing heights requires us to also access all of the
siblings along the path.
- Don't use any recursion at all. We know that the tree cannot 2^64
elements in size, so the height of the tree can never be larger than
96. Use a 128-bit bitmask to keep track of the path that is computed.
This allows us to iterate over the same path twice, meaning we can
apply rotations from top to bottom.
Inserting 100k nodes into a tree now only takes 0.015 seconds. Insertion
seems to be twice as fast as glibc, whereas deletion has about the same
performance. Unlike glibc, it uses a fixed amount of memory.
I also experimented with both recursive and iterative bottom-up
implementations of the same algorithm. This iterative top-down version
performs similar to the recursive bottom-up version in terms of speed
and code size.
For some reason, the iterative bottom-up algorithm was actually 30%
faster for deletion, but has a quadratic memory complexity to keep track
of all the parent pointers.
Reviewed by: jilles
Obtained from: https://github.com/NuxiNL/cloudlibc
Differential Revision: https://reviews.freebsd.org/D4412
variables in loader(8) and other libstand applications.
Sometimes a dhcp server provides incorrect information along with the IP
address. It would be useful to have a way to override this with
locally-supplied information, such as command line parameters passed from a
prior-stage bootloader. This change allows pre-existing env vars to take
precedence over values delivered by the dhcp or bootp server.
The bootp/dhcp code in libstand automatically creates environment variables
from the data provided by the server (dhcp.root-path, dhcp.domain-name,
etc). It also transcribes the values to some global variables such as
'rootpath' and 'hostname'.
This change does two things:
When adding dhcp.* vars to the environment, don't replace existing
vars/values.
When setting the global vars rootpath and hostname, use the
dhcp.root-path and dhcp.host-name env var values if they exist.
This allows the platform-specific part of loader(8) to obtain override
values in some platform-specific way and store them in the environment
before opening the network device. The set of values that can be overriden
is currently limited to just string options. The values that are delivered
as binary data are things that probably shouldn't be overridden (IP,
netmask, gateway, etc).
The original patch this evolved from was submitted by martymac@
PR: 202098
Differential Revision: https://reviews.freebsd.org/D4559
Depending on system configuration and parameters, clock_gettime() and
gettimeofday() may not be system calls. If so, passing an invalid pointer
will cause a signal and not an [EFAULT] error.
From a standards perspective, this is OK since passing an invalid pointer is
undefined behaviour.
MFC after: 1 week
