Part 3 of many ...
The VPC framework relies heavily on cloning pseudo interfaces
(vmnics, vpc switch, vcpswitch port, hostif, vxlan if, etc).
This pulls in that piece. Some ancillary changes get pulled
in as a side effect.
Reviewed by: shurd@
Approved by: sbruno@
Sponsored by: Joyent, Inc.
Differential Revision: https://reviews.freebsd.org/D15347
This patch adds a new socket option, SO_REUSEPORT_LB, which allow multiple
programs or threads to bind to the same port and incoming connections will be
load balanced using a hash function.
Most of the code was copied from a similar patch for DragonflyBSD.
However, in DragonflyBSD, load balancing is a global on/off setting and can not
be set per socket. This patch allows for simultaneous use of both the current
SO_REUSEPORT and the new SO_REUSEPORT_LB options on the same system.
Required changes to structures
Globally change so_options from 16 to 32 bit value to allow for more options.
Add hashtable in pcbinfo to hold all SO_REUSEPORT_LB sockets.
Limitations
As DragonflyBSD, a load balance group is limited to 256 pcbs
(256 programs or threads sharing the same socket).
Submitted by: Johannes Lundberg <johanlun0@gmail.com>
Sponsored by: Limelight Networks
Differential Revision: https://reviews.freebsd.org/D11003
Use an accessor to access ifgr_group and ifgr_groups.
Use an macro CASE_IOC_IFGROUPREQ(cmd) in place of case statements such
as "case SIOCAIFGROUP:". This avoids poluting the switch statements
with large numbers of #ifdefs.
Reviewed by: kib
Obtained from: CheriBSD
MFC after: 1 week
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D14960
This fixes 32-bit compat (no ioctl command defintions are required
as struct ifreq is the same size). This is believed to be sufficent to
fully support ifconfig on 32-bit systems.
Reviewed by: kib
Obtained from: CheriBSD
MFC after: 1 week
Relnotes: yes
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D14900
Add a new "interleave" allocation policy which stripes pages across
domains with a stride or width keeping contiguity within a multi-page
region.
Move the kernel to the dedicated numbered cpuset #2 making it possible
to assign kernel threads and memory policy separately from user. This
also eliminates the need for the complicated interrupt binding code.
Add a sysctl API for viewing and manipulating domainsets. Refactor some
of the cpuset_t manipulation code using the generic bitset type so that
it can be used for both. This probably belongs in a dedicated subr file.
Attempt to improve the include situation.
Reviewed by: kib
Discussed with: jhb (cpuset parts)
Tested by: pho (before review feedback)
Sponsored by: Netflix, Dell/EMC Isilon
Differential Revision: https://reviews.freebsd.org/D14839
Make all kernel accesses to ifru_buffer go via access functions
which take the process ABI into account and use an appropriate union
to access members in the correct place in struct ifreq.
Reviewed by: kib
Obtained from: CheriBSD
MFC after: 1 week
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D14846
According to 802.1Q-2014, VLAN tagged packets with VLAN id 0 should be
considered as untagged, and only PCP and DEI values from the VLAN tag
are meaningful. See for instance
https://www.cisco.com/c/en/us/td/docs/switches/connectedgrid/cg-switch-sw-master/software/configuration/guide/vlan0/b_vlan_0.html.
Make it possible to specify PCP value for outgoing packets on an
ethernet interface. When PCP is supplied, the tag is appended, VLAN
id set to 0, and PCP is filled by the supplied value. The code to do
VLAN tag encapsulation is refactored from the if_vlan.c and moved into
if_ethersubr.c.
Drivers might have issues with filtering VID 0 packets on
receive. This bug should be fixed for each driver.
Reviewed by: ae (previous version), hselasky, melifaro
Sponsored by: Mellanox Technologies
MFC after: 2 weeks
Differential revision: https://reviews.freebsd.org/D14702
Include _uio.h instead of uio.h in several headers to reduce header
polution.
Fix a few places that relied on header polution to get the uio.h header.
I have not moved struct uio as many more things that use it rely on
header polution to get other definitions from uio.h.
Reviewed by: cem, kib, markj
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D14811
which we discussed at the developer summits at BSDCan and BSDCam in 2017.
The TCP Blackbox Recorder allows you to capture events on a TCP connection
in a ring buffer. It stores metadata with the event. It optionally stores
the TCP header associated with an event (if the event is associated with a
packet) and also optionally stores information on the sockets.
It supports setting a log ID on a TCP connection and using this to correlate
multiple connections that share a common log ID.
You can log connections in different modes. If you are doing a coordinated
test with a particular connection, you may tell the system to put it in
mode 4 (continuous dump). Or, if you just want to monitor for errors, you
can put it in mode 1 (ring buffer) and dump all the ring buffers associated
with the connection ID when we receive an error signal for that connection
ID. You can set a default mode that will be applied to a particular ratio
of incoming connections. You can also manually set a mode using a socket
option.
This commit includes only basic probes. rrs@ has added quite an abundance
of probes in his TCP development work. He plans to commit those soon.
There are user-space programs which we plan to commit as ports. These read
the data from the log device and output pcapng files, and then let you
analyze the data (and metadata) in the pcapng files.
Reviewed by: gnn (previous version)
Obtained from: Netflix, Inc.
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D11085
These macros take an existing ioctl(2) command and replace the length
with the specified length or length of the specified type respectively.
These can be used to define commands for 32-bit compatibility with fewer
opportunities for cut-and-paste errors then a whole new definition.
Reviewed by: cem, kib
Obtained from: CheriBSD
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D14706
[RFC7413]. It also includes a pre-shared key mode of operation in which
the server requires the client to be in possession of a shared secret in
order to successfully open TFO connections with that server.
The names of some existing fastopen sysctls have changed (e.g.,
net.inet.tcp.fastopen.enabled -> net.inet.tcp.fastopen.server_enable).
Reviewed by: tuexen
MFC after: 1 month
Sponsored by: Limelight Networks
Differential Revision: https://reviews.freebsd.org/D14047
that had the IPv6 fragmentation header:
o Neighbor Solicitation
o Neighbor Advertisement
o Router Solicitation
o Router Advertisement
o Redirect
Introduce M_FRAGMENTED mbuf flag, and set it after IPv6 fragment reassembly
is completed. Then check the presence of this flag in correspondig ND6
handling routines.
PR: 224247
MFC after: 2 weeks
Mainly focus on files that use BSD 2-Clause license, however the tool I
was using misidentified many licenses so this was mostly a manual - error
prone - task.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
Mainly focus on files that use BSD 3-Clause license.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
Special thanks to Wind River for providing access to "The Duke of
Highlander" tool: an older (2014) run over FreeBSD tree was useful as a
starting point.
Mainly focus on files that use BSD 3-Clause license.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
Special thanks to Wind River for providing access to "The Duke of
Highlander" tool: an older (2014) run over FreeBSD tree was useful as a
starting point.
for SO_TIMESTAMP and other similar socket options.
Provide new control message SCM_TIME_INFO to supply information about
timestamp. Currently it indicates that the timestamp was
hardware-assisted and high-precision, for software timestamps the
message is not returned. Reserved fields are added to ABI to report
additional info about it, it is expected that raw hardware clock value
might be useful for some applications.
Reviewed by: gallatin (previous version), hselasky
Sponsored by: Mellanox Technologies
MFC after: 2 weeks
X-Differential revision: https://reviews.freebsd.org/D12638
in nanoseconds from boot for the received packets.
The rcv_tstmp field overlaps the place of Ln header length indicators,
not used by received packets. The basic pkthdr rearrangement change
in sys/mbuf.h was provided by gallatin.
There are two accompanying M_ flags: M_TSTMP means that there is the
timestamp (and it was generated by hardware).
Another flag M_TSTMP_HPREC indicates that the timestamp is
high-precision. Practically M_TSTMP_HPREC means that hardware
provided additional precision comparing with the stamps when the flag
is not set. E.g., for ConnectX all packets are stamped by hardware
when PCIe transaction to write out the completion descriptor is
performed, but PTP packet are stamped on port. For Intel cards, when
PTP assist is enabled, only PTP packets are stamped in the limited
number of registers, so if Intel cards ever start support this
mechanism, they would always set M_TSTMP | M_TSTMP_HPREC if hardware
timestamp is present for the given packet.
Add IFCAP_HWRXTSTMP interface capability to indicate the support for
hardware rx timestamping, and ifconfig(8) command to toggle it.
Based on the patch by: gallatin
Reviewed by: gallatin (previous version), hselasky
Sponsored by: Mellanox Technologies
MFC after: 2 weeks (? mbuf KBI issue)
X-Differential revision: https://reviews.freebsd.org/D12638
It will be needed by hn(4) to configure its RSS key and hash
type/function in the transparent VF mode in order to match VF's
RSS settings. The description of the transparent VF mode and
the RSS hash value issue are here:
https://svnweb.freebsd.org/base?view=revision&revision=322299https://svnweb.freebsd.org/base?view=revision&revision=322485
These are generic enough to promise two independent IOCs instead
of abusing SIOCGDRVSPEC.
Setting RSS key and hash type/function is a different story,
which probably requires more discussion.
Comment about UDP_{IPV4,IPV6,IPV6_EX} were only in the patch
in the review request; these hash types are standardized now.
Reviewed by: gallatin
MFC after: 1 week
Sponsored by: Microsoft
Differential Revision: https://reviews.freebsd.org/D12174
They are defined by XSI or newer SUS.
This is a follow-up to r318780.
Reported by: jbeich
Obtained from: DragonflyBSD commit e08b3836c962
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Guard, requested by the MAP_GUARD mmap(2) flag, prevents the reuse of
the allocated address space, but does not allow instantiation of the
pages in the range. It is useful for more explicit support for usual
two-stage reserve then commit allocators, since it prevents accidental
instantiation of the mapping, e.g. by mprotect(2).
Use guards to reimplement stack grow code. Explicitely track stack
grow area with the guard, including the stack guard page. On stack
grow, trivial shift of the guard map entry and stack map entry limits
makes the stack expansion. Move the code to detect stack grow and
call vm_map_growstack(), from vm_fault() into vm_map_lookup().
As result, it is impossible to get random mapping to occur in the
stack grow area, or to overlap the stack guard page.
Enable stack guard page by default.
Reviewed by: alc, markj
Man page update reviewed by: alc, bjk, emaste, markj, pho
Tested by: pho, Qualys
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D11306 (man pages)
o Separate fields of struct socket that belong to listening from
fields that belong to normal dataflow, and unionize them. This
shrinks the structure a bit.
- Take out selinfo's from the socket buffers into the socket. The
first reason is to support braindamaged scenario when a socket is
added to kevent(2) and then listen(2) is cast on it. The second
reason is that there is future plan to make socket buffers pluggable,
so that for a dataflow socket a socket buffer can be changed, and
in this case we also want to keep same selinfos through the lifetime
of a socket.
- Remove struct struct so_accf. Since now listening stuff no longer
affects struct socket size, just move its fields into listening part
of the union.
- Provide sol_upcall field and enforce that so_upcall_set() may be called
only on a dataflow socket, which has buffers, and for listening sockets
provide solisten_upcall_set().
o Remove ACCEPT_LOCK() global.
- Add a mutex to socket, to be used instead of socket buffer lock to lock
fields of struct socket that don't belong to a socket buffer.
- Allow to acquire two socket locks, but the first one must belong to a
listening socket.
- Make soref()/sorele() to use atomic(9). This allows in some situations
to do soref() without owning socket lock. There is place for improvement
here, it is possible to make sorele() also to lock optionally.
- Most protocols aren't touched by this change, except UNIX local sockets.
See below for more information.
o Reduce copy-and-paste in kernel modules that accept connections from
listening sockets: provide function solisten_dequeue(), and use it in
the following modules: ctl(4), iscsi(4), ng_btsocket(4), ng_ksocket(4),
infiniband, rpc.
o UNIX local sockets.
- Removal of ACCEPT_LOCK() global uncovered several races in the UNIX
local sockets. Most races exist around spawning a new socket, when we
are connecting to a local listening socket. To cover them, we need to
hold locks on both PCBs when spawning a third one. This means holding
them across sonewconn(). This creates a LOR between pcb locks and
unp_list_lock.
- To fix the new LOR, abandon the global unp_list_lock in favor of global
unp_link_lock. Indeed, separating these two locks didn't provide us any
extra parralelism in the UNIX sockets.
- Now call into uipc_attach() may happen with unp_link_lock hold if, we
are accepting, or without unp_link_lock in case if we are just creating
a socket.
- Another problem in UNIX sockets is that uipc_close() basicly did nothing
for a listening socket. The vnode remained opened for connections. This
is fixed by removing vnode in uipc_close(). Maybe the right way would be
to do it for all sockets (not only listening), simply move the vnode
teardown from uipc_detach() to uipc_close()?
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D9770
INHERIT_ZERO is an OpenBSD feature.
When a page is marked as such, it would be zeroed
upon fork().
This would be used in new arc4random(3) functions.
PR: 182610
Reviewed by: kib (earlier version)
MFC after: 1 month
Differential Revision: https://reviews.freebsd.org/D427
Renumber cluase 4 to 3, per what everybody else did when BSD granted
them permission to remove clause 3. My insistance on keeping the same
numbering for legal reasons is too pedantic, so give up on that point.
Submitted by: Jan Schaumann <jschauma@stevens.edu>
Pull Request: https://github.com/freebsd/freebsd/pull/96
Our mprotect() function seems to take a "const void *" address to the
pages whose permissions need to be adjusted. POSIX uses "void *". Simply
stick to the POSIX one to prevent us from writing unportable code.
PR: 211423 (exp-run)
Tested by: antoine@ (Thanks!)
for libthr.so.3, without breaking the ABI. Special value is stored in
the lock pointer to indicate shared lock, and offline page in the shared
memory is allocated to store the actual lock.
Reviewed by: vangyzen (previous version)
Discussed with: deischen, emaste, jhb, rwatson,
Martin Simmons <martin@lispworks.com>
Tested by: pho
Sponsored by: The FreeBSD Foundation
operations. File type-specific logic is now placed in the mmap hook
implementation rather than requiring it to be placed in
sys/vm/vm_mmap.c. This hook allows new file types to support mmap() as
well as potentially allowing mmap() for existing file types that do not
currently support any mapping.
The vm_mmap() function is now split up into two functions. A new
vm_mmap_object() function handles the "back half" of vm_mmap() and accepts
a referenced VM object to map rather than a (handle, handle_type) tuple.
vm_mmap() is now reduced to converting a (handle, handle_type) tuple to a
a VM object and then calling vm_mmap_object() to handle the actual mapping.
The vm_mmap() function remains for use by other parts of the kernel
(e.g. device drivers and exec) but now only supports mapping vnodes,
character devices, and anonymous memory.
The mmap() system call invokes vm_mmap_object() directly with a NULL object
for anonymous mappings. For mappings using a file descriptor, the
descriptors fo_mmap() hook is invoked instead. The fo_mmap() hook is
responsible for performing type-specific checks and adjustments to
arguments as well as possibly modifying mapping parameters such as flags
or the object offset. The fo_mmap() hook routines then call
vm_mmap_object() to handle the actual mapping.
The fo_mmap() hook is optional. If it is not set, then fo_mmap() will
fail with ENODEV. A fo_mmap() hook is implemented for regular files,
character devices, and shared memory objects (created via shm_open()).
While here, consistently use the VM_PROT_* constants for the vm_prot_t
type for the 'prot' variable passed to vm_mmap() and vm_mmap_object()
as well as the vm_mmap_vnode() and vm_mmap_cdev() helper routines.
Previously some places were using the mmap()-specific PROT_* constants
instead. While this happens to work because PROT_xx == VM_PROT_xx,
using VM_PROT_* is more correct.
Differential Revision: https://reviews.freebsd.org/D2658
Reviewed by: alc (glanced over), kib
MFC after: 1 month
Sponsored by: Chelsio
and MAP_NORESERVE flags to mmap(2). Older binaries are still permitted
to use these flags.
PR: 193961 (exp-run in ports)
Differential Revision: https://reviews.freebsd.org/D848
Reviewed by: kib
to add type-specific information to struct kinfo_file. - Move the
various fill_*_info() methods out of kern_descrip.c and into the various
file type implementations. - Rework the support for kinfo_ofile to
generate a suitable kinfo_file object for each file and then convert
that to a kinfo_ofile structure rather than keeping a second, different
set of code that directly manipulates type-specific file information. -
Remove the shm_path() and ksem_info() layering violations.
Differential Revision: https://reviews.freebsd.org/D775
Reviewed by: kib, glebius (earlier version)
It should be combined with MAP_FIXED, and prevents the request from
deleting existing mappings in the region, failing instead.
Reviewed by: alc
Discussed with: jhb
Tested by: markj, pho (previous version, as part of the bigger patch)
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
to request that a mapping use an address in the first 2GB of the
process's address space. This flag should have the same semantics as the
same flag on Linux.
To facilitate this, add a new parameter to vm_map_find() that specifies an
optional maximum virtual address. While here, fix several callers of
vm_map_find() to use a VMFS_* constant for the findspace argument instead of
TRUE and FALSE.
Reviewed by: alc
Approved by: re (kib)
for posix shmfd. Add MAC framework entries for posix shm read and write.
Do not allow implicit extension of the underlying memory segment past
the limit set by ftruncate(2) by either of the syscalls. Read and
write returns short i/o, lseek(2) fails with EINVAL when resulting
offset does not fit into the limit.
Discussed with: alc
Tested by: pho
Sponsored by: The FreeBSD Foundation
Add __nl_item to <sys/_types.h> for FreeBSD compatibility. Use it in
<langinfo.h> and the Cygwin <nl_types.h>. Make the enum __nl_item in
<langinfo.h> anonymous.
Signed-off-by: Sebastian Huber <sebastian.huber@embedded-brains.de>
Add __tls_get_addr() for all targets to crt0. This is not only used on
ARM. In particular, it is used on RISC-V. This helps to adequately
support the GCC libgomp.
Signed-off-by: Sebastian Huber <sebastian.huber@embedded-brains.de
strtof ("-nan") returned positive NaN instead of negative NaN.
strtod ("-nan") and strtold ("-nan") return negative NaN.
Linux glibc has been fixed
that strto{f|d|ld} ("-nan") returns negative NaN.
https://sourceware.org/bugzilla/show_bug.cgi?id=23007
This commit makes strtof preserves the negative sign bit
when parsing "-nan" like glibc.
By previous commit, strto{d|ld} ("nan")
does not use the definition of NaN.
There is no other function that uses the definitions.
This commit remove the definitions.
The definition of qNaN for x86_64 and i386 was wrong.
strto{d|ld} ("nan") returned wrong negative NaN
instead of correct positive NaN
since it used the wrong definition.
On the other hand, strtof ("nan") returns correct positive NaN
since it uses nanf ("") instead of the wrong definition.
This commit makes strto{d|ld} ("nan") uses {nan|nanl} ("")
like strtof ("nan") using.
So strto{d|ld} ("nan") returns positive NaN.
Improve comments in sincosf implementation to make the code easier
to understand. Rename the constant pi64 to pi63 since it's actually
PI * 2^-63. Add comments for fields of sincos_t structure. Add comments
describing implementation details to reduce_fast.
wordexp uses fprintf in a dangerous way. It uses an unchecked
input string as format string, rather than as parameter to a %s.
Replace fprintf with fputs.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Introduce new host configuration variable "have_init_fini" which is set
to "yes" by default. Override it for RISC-V to "no".
Signed-off-by: Sebastian Huber <sebastian.huber@embedded-brains.de>
Restore FreeBSD compatibility for __alloc_size() and __alloc_align().
This is a follow-up to commit e494b56035.
Signed-off-by: Sebastian Huber <sebastian.huber@embedded-brains.de>
At least on GCC7 calling __alloc_size(x) twice is not equivalent to
calling using the attribute once with two arguments. The later is the
documented use in GCC documentation so add a new alloc_size(n, x)
alternative to cover for the few places where it is used: basically:
calloc(3), reallocarray(3) and mallocarray(9).
Submitted by: Mark Millard
MFC after: 3 days
Reference:
http://docs.freebsd.org/cgi/mid.cgi?F227842D-6BE2-4680-82E7-07906AF61CD7
Mainly focus on files that use BSD 3-Clause license.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
Special thanks to Wind River for providing access to "The Duke of
Highlander" tool: an older (2014) run over FreeBSD tree was useful as a
starting point.
GCC only activates C11 keywords in C mode, not C++ mode. This means
that when targeting an older C++ standard, we cannot fall back to using
_Static_assert(). In this case, do define _Static_assert() as a macro
that uses a typedef'ed array.
Discussed in: r322875 commit thread
Reported by: Mark MIllard
MFC after: 1 month
The previous version genenerated the following GCC note:
towctrans_l.c:44:1: note: offset of packed bit-field 'diff' has changed in GCC 4.4
caseconv_table [] = {
^~~~~~~~~~~~~~
Signed-off-by: Sebastian Huber <sebastian.huber@embedded-brains.de>
The commit 46ba1675c4 accidently changed a
bit-field from signed to unsigned. The caseconv_entry::delta must be a
signed integer, see also "newlib/libc/ctype/caseconv.t".
Unfortunately, a standard GCC/Newlib build is done without
-Wsign-conversion. Using this warning option would have helped to avoid
this bug:
caseconv.t:2:22: warning: unsigned conversion from 'int' to 'unsigned int:17' changes value from '-32' to '131040' [-Wsign-conversion]
{0x0061, 25, TOUP, -32},
Signed-off-by: Sebastian Huber <sebastian.huber@embedded-brains.de>
This prevents errors like this:
newlib/libc/ctype/categories.c:6:3: error: width of 'first' exceeds its type
unsigned int first: 24;
^
Signed-off-by: Sebastian Huber <sebastian.huber@embedded-brains.de>
Exotic RTEMS targets can define this back to int32_t as an exception if
there are good reasons.
Signed-off-by: Sebastian Huber <sebastian.huber@embedded-brains.de>
Replace the simple byte-wise compare in the misaligned case with a
dword compare with page boundary checks in place. For simplicity I've
chosen a 4K page boundary so that we don't have to query the actual
page size on the system.
This results in up to 3x improvement in performance in the unaligned
case on falkor and about 2.5x improvement on mustang as measured using
bench-strcmp in glibc.
This improved memcmp provides a fast path for compares up to 16 bytes
and then compares 16 bytes at a time, thus optimizing loads from both
sources. The glibc memcmp microbenchmark retains performance (with an
error of ~1ns) for smaller compare sizes and reduces up to 31% of
execution time for compares up to 4K on the APM Mustang. On Qualcomm
Falkor this improves to almost 48%, i.e. it is almost 2x improvement
for sizes of 2K and above.
The mutually misaligned inputs on aarch64 are compared with a simple
byte copy, which is not very efficient. Enhance the comparison
similar to strcmp by loading a double-word at a time. The peak
performance improvement (i.e. 4k maxlen comparisons) due to this on
the strncmp microbenchmark in glibc is as follows:
falkor: 3.5x (up to 72% time reduction)
cortex-a73: 3.5x (up to 71% time reduction)
cortex-a53: 3.5x (up to 71% time reduction)
All mutually misaligned inputs from 16 bytes maxlen onwards show
upwards of 15% improvement and there is no measurable effect on the
performance of aligned/mutually aligned inputs.
PREFER_FLOAT_COMPARISON setting was not correct as it could raise
spurious exceptions. Fixing it is easy: just use ISLESS(x, y) instead
of abstop12(x) < abstop12(y) with appropriate non-signaling definition
for ISLESS. However it seems this setting is not very useful (there is
only minor performance difference on various architectures), so remove
this option for now.
The !HAVE_FAST_FMA code path split r = z/c - 1 into r = rhi + rlo such
that when z = 1-tiny and c = 1 then rlo and rhi could have much larger
magnitude than r which later caused large rounding errors.
So do a nearest rounding instead of truncation at the split.
In newlib with default settings this was observable on some arm targets
that enable the new math code but has no fma.
The roundtoint and converttoint internal functions are only called with small
values, so 32 bit result is enough for converttoint and it is a signed int
conversion so the natural return type is int32_t.
The original idea was to help the compiler keeping the result in uint64_t,
then it's clear that no sign extension is needed and there is no accidental
undefined or implementation defined signed int arithmetics.
But it turns out gcc does a good job with inlining so changing the type has
no overhead and the semantics of the conversion is less surprising this way.
Since we want to allow the asuint64 (x + 0x1.8p52) style conversion, the top
bits were never usable and the existing code ensures that only the bottom
32 bits of the conversion result are used.
In newlib with default settings only aarch64 is affected and there is no
significant code generation change with gcc after the patch.
Synchronize code style and comments with Arm Optimized Routines, there
are no code changes in this patch. This ensures different projects using
the same code have consistent code style so bug fix patches can be applied
more easily.
This fix is for some platforms which do not have writev().
*perror.c: Use _write_r() instead of writev().
*psignal.c: Use write() insetad of writev().
Revise commit: d4f4e7ae1b
The new implementation is provided under !__OBSOLETE_MATH, it uses
ISO C99 code. With default settings the worst case error in nearest
rounding mode is 0.54 ULP with inlined fma and fma contraction. It uses
a 4 KB lookup table in addition to the table in exp_data.c, on aarch64
.text+.rodata size of libm.a is increased by 2295 bytes.
Improvements on Cortex-A72:
latency: 3.3x
thruput: 4.9x
The new implementation is provided under !__OBSOLETE_MATH, it uses
ISO C99 code. With default settings the worst case error in nearest
rounding mode is 0.547 ULP with inlined fma and fma contraction. It uses
a 1 KB lookup table, on aarch64 .text+.rodata size of libm.a is increased
by 1584 bytes.
Note that the math.h header defines log2(x) to be log(x)/Ln2, this is
not changed, so the new code is only used if that macro is suppressed.
Improvements on Cortex-A72:
latency: 2.0x
thruput: 2.2x
The new implementations are provided under !__OBSOLETE_MATH, it uses
ISO C99 code. With default settings the worst case error in nearest
rounding mode is 0.519 ULP with inlined fma and fma contraction. It uses
a 2 KB lookup table, on aarch64 .text+.rodata size of libm.a is increased
by 1703 bytes. The w_log.c wrapper is disabled since error handling is
inline in the new code.
New __HAVE_FAST_FMA and __HAVE_FAST_FMA_DEFAULT feature macros were
added to enable selecting between the code path that uses fma and the
one that does not. Targets supposed to set __HAVE_FAST_FMA_DEFAULT
if they have single instruction fma and the compiler can actually
inline it (gcc has __FP_FAST_FMA macro but that does not guarantee
inlining with -fno-builtin-fma).
Improvements on Cortex-A72:
latency: 1.9x
thruput: 2.3x
The new implementations are provided under !__OBSOLETE_MATH, they use
ISO C99 code. There are several settings, with the default one the
worst case error in nearest rounding mode is 0.509 ULP for exp and
0.507 ULP for exp2 when a multiply and add is contracted into an fma.
They use a shared 2 KB lookup table, on aarch64 .text+.rodata size
of libm.a is increased by 1868 bytes. The w_*.c wrappers are disabled
for the new code as it takes care of error handling inline.
The old exp2(x) code used to be just pow(2,x) so the speedup there
is more significant.
The file name has no special prefix to avoid any name collision with
existing files.
Improvements on Cortex-A72:
exp latency: 3.2x
exp thruput: 4.1x
exp2 latency: 7.8x
exp2 thruput: 18.8x
This change is equivalent to the commit
c65db17340
and only affects code that is from the Arm optimized-routines project.
It does not affect the observable behaviour, but the code generation
can be different on 64bit targets. The intention is to make the
portable semantics of the code obvious by using a fixed size type.
* (mkcategories): Fix a bug that outputs incorrect Unicode category
table for code point ranges.
* (categories.t): Rebuild it using the bug-fixed mkcategories.
This fixes the problem reported in the following post.
https://cygwin.com/ml/cygwin/2018-06/msg00248.html
Here is the correct patch with both filenames and int cast fixed:
This patch is a complete rewrite of sinf, cosf and sincosf. The new version
is significantly faster, as well as simple and accurate.
The worst-case ULP is 0.56072, maximum relative error is 0.5303p-23 over all
4 billion inputs. In non-nearest rounding modes the error is 1ULP.
The algorithm uses 3 main cases: small inputs which don't need argument
reduction, small inputs which need a simple range reduction and large inputs
requiring complex range reduction. The code uses approximate integer
comparisons to quickly decide between these cases - on some targets this may
be slow, so this can be configured to use floating point comparisons.
The small range reducer uses a single reduction step to handle values up to
120.0. It is fastest on targets which support inlined round instructions.
The large range reducer uses integer arithmetic for simplicity. It does a
32x96 bit multiply to compute a 64-bit modulo result. This is more than
accurate enough to handle the worst-case cancellation for values close to
an integer multiple of PI/4. It could be further optimized, however it is
already much faster than necessary.
Simple benchmark showing speedup factor on AArch64 for various ranges:
range 0.7853982 sinf 1.7 cosf 2.2 sincosf 2.8
range 1.570796 sinf 1.9 cosf 1.9 sincosf 2.7
range 3.141593 sinf 2.0 cosf 2.0 sincosf 3.5
range 6.283185 sinf 2.3 cosf 2.3 sincosf 4.2
range 125.6637 sinf 2.9 cosf 3.0 sincosf 5.1
range 1.1259e15 sinf 26.8 cosf 26.8 sincosf 45.2
ChangeLog:
2018-05-18 Wilco Dijkstra <wdijkstr@arm.com>
* newlib/libm/common/Makefile.in: Regenerated.
* newlib/libm/common/Makefile.am: Add sinf.c, cosf.c, sincosf.c
sincosf.h, sincosf_data.c. Add -fbuiltin -fno-math-errno to CFLAGS.
* newlib/libm/common/math_config.h: Add HAVE_FAST_ROUND, HAVE_FAST_LROUND,
roundtoint, converttoint, force_eval_float, force_eval_double, eval_as_float,
eval_as_double, likely, unlikely.
* newlib/libm/common/cosf.c: New file.
* newlib/libm/common/sinf.c: Likewise.
* newlib/libm/common/sincosf.h: Likewise.
* newlib/libm/common/sincosf.c: Likewise.
* newlib/libm/common/sincosf_data.c: Likewise.
* newlib/libm/math/sf_cos.c: Add #if to build conditionally.
* newlib/libm/math/sf_sin.c: Likewise.
* newlib/libm/math/wf_sincos.c: Likewise.
--
This patch is a complete rewrite of sinf, cosf and sincosf. The new version
is significantly faster, as well as simple and accurate.
The worst-case ULP is 0.56072, maximum relative error is 0.5303p-23 over all
4 billion inputs. In non-nearest rounding modes the error is 1ULP.
The algorithm uses 3 main cases: small inputs which don't need argument
reduction, small inputs which need a simple range reduction and large inputs
requiring complex range reduction. The code uses approximate integer
comparisons to quickly decide between these cases - on some targets this may
be slow, so this can be configured to use floating point comparisons.
The small range reducer uses a single reduction step to handle values up to
120.0. It is fastest on targets which support inlined round instructions.
The large range reducer uses integer arithmetic for simplicity. It does a
32x96 bit multiply to compute a 64-bit modulo result. This is more than
accurate enough to handle the worst-case cancellation for values close to
an integer multiple of PI/4. It could be further optimized, however it is
already much faster than necessary.
Simple benchmark showing speedup factor on AArch64 for various ranges:
range 0.7853982 sinf 1.7 cosf 2.2 sincosf 2.8
range 1.570796 sinf 1.9 cosf 1.9 sincosf 2.7
range 3.141593 sinf 2.0 cosf 2.0 sincosf 3.5
range 6.283185 sinf 2.3 cosf 2.3 sincosf 4.2
range 125.6637 sinf 2.9 cosf 3.0 sincosf 5.1
range 1.1259e15 sinf 26.8 cosf 26.8 sincosf 45.2
ChangeLog:
2018-06-18 Wilco Dijkstra <wdijkstr@arm.com>
* newlib/libm/common/Makefile.in: Regenerated.
* newlib/libm/common/Makefile.am: Add sinf.c, cosf.c, sincosf.c
sincosf.h, sincosf_data.c. Add -fbuiltin -fno-math-errno to CFLAGS.
* newlib/libm/common/math_config.h: Add HAVE_FAST_ROUND, HAVE_FAST_LROUND,
roundtoint, converttoint, force_eval_float, force_eval_double, eval_as_float,
eval_as_double, likely, unlikely.
* newlib/libm/common/cosf.c: New file.
* newlib/libm/common/sinf.c: Likewise.
* newlib/libm/common/sincosf.h: Likewise.
* newlib/libm/common/sincosf.c: Likewise.
* newlib/libm/common/sincosf_data.c: Likewise.
* newlib/libm/math/sf_cos.c: Add #if to build conditionally.
* newlib/libm/math/sf_sin.c: Likewise.
* newlib/libm/math/wf_sincos.c: Likewise.
--
Previously, "test 1 2 3 -a -b -c" was permuted to "test -a -b -c 1 2 3",
but "test 1 2 3 -abc" was left as "test 1 2 3 -abc".
Signed-off-by: Thomas Kindler <mail+newlib@t-kindler.de>
- when calculating a correction to align next brk to page boundary,
ensure that the correction is less than a page size
- if allocating the correction fails, ensure that the top size is
set to brk + sbrk_size (minus any front alignment made)
Signed-off-by: Jeff Johnston <jjohnstn@redhat.com>
When converting number of days since epoch (32-bits) to seconds,
calculations using 32-bit `long` overflow for years above 2038. Solve
this by casting number of days to `time_t` just before final
multiplication.
Signed-off-by: Freddie Chopin <freddie.chopin@gmail.com>
- From: Cesar Philippidis <cesar@codesourcery.com>
Date: Tue, 10 Apr 2018 14:43:42 -0700
Subject: [PATCH] nvptx port
This port adds support for Nvidia GPU's, which are primarily used as
offload accelerators in OpenACC and OpenMP.
The gdtoa implementation uses the type long, defined as Long, in lots
of code. For historical reason newlib defines Long as int32_t instead.
This works fine, as long as floating point exceptions are not enabled.
The conversion to 32 bit int can lead to a FE_INVALID situation.
Example:
const char *str = "121645100408832000.0";
char *ptr;
feenableexcept (FE_INVALID);
strtod (str, &ptr);
This leads to the following situation in strtod
double aadj;
Long L;
[...]
L = (Long)aadj;
For instance, on x86_64 the code here is
cvttsd2si %xmm0,%eax
At this point, aadj is 2529648000.0 in our example. The conversion to
32 bit %eax results in a negative int value, thus the conversion is
invalid. With feenableexcept (FE_INVALID), a SIGFPE is raised.
Fix this by always using 64 bit ints here if double is not a 32 bit type
to avoid this type of FP exceptions.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Classical function call recursion wastes a lot of stack space.
Each recursion level requires a full stack frame comprising all
local variables and additional space as dictated by the
processor calling convention.
This implementation instead stores the variables that are unique
for each recursion level in a parameter stack array, and uses
iteration to emulate recursion. Function call recursion is not
used until the array is full.
To ensure the stack consumption isn't worsened by this design, the
size of the parameter stack array is chosen to be similar to the
stack frame excluding the array. Each function call recursion level
can handle 8 iterative recursion levels.
Stack consumption will worsen when sorting tiny arrays that do not
need recursion (of 6 elements or less). It will be about equal for
up to 15 elements, and be an improvement for larger arrays. The best
case improvement is a stack size reduction down to about one quarter
of the stack consumption before the change.
A design where the parameter stack array is large enough for the
worst case recursion level was rejected because it would worsen
the stack consumption when sorting arrays smaller than about 1500
elements. The worst case is 31 levels on a 32-bit system.
A design with a dynamic parameter array size was rejected because
of limitations in some compilers.
The qsort algorithm splits the input array in three parts. The
left and right parts may need further sorting. One of them is
sorted by recursion, the other by iteration. This update ensures
that it is the smaller part that is chosen for recursion.
By choosing the smaller part, each recursion level will handle
less than half the array of the previous recursion level. Hence
the recursion depth is bounded to be less than log2(n) i.e. 1
level per significant bit in the array size n.
The update also includes code comments explaining the algorithm.
Newlib has a build configuration where syscalls can be directly
embedded in the newlib library rather than relying on libgloss.
This configuration was broken recently by an update to the libgloss
support for Arm that was not propagated to the syscalls interface in
newlib itself. This patch restores the build. It's essentially a
copy of https://sourceware.org/ml/newlib/2018/msg00128.html but there
are some other minor cleanups and changes that I've made at the same
time. None of those cleanups affect functionality.
The prototypes of the following functions have been updated: _link,
_sbrk, _getpid, _write, _swiwrite, _lseek, _swilseek, _read and
_swiread.
Signed-off-by: Richard Earnshaw <Richard.Earnshaw@arm.com>
E.g. arm ABI requires -fshort-enums for bare-metal toolchains.
Given there are only 29 category enums, the compiler chooses an
8 bit enum type, so a size of 11 bits for the bitfield leads to
a compile time error:
error: width of 'cat' exceeds its type
enum category cat: 11;
^~~
Fix this by aligning the size of the category members to byte
borders.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Scripts do not try to acquire Unicode data by best-effort magic anymore.
Options supported:
-h for help
-i to copy Unicode data from /usr/share/unicode/ucd first
-u to download Unicode data from unicode.org first
If (despite of -i or -u if given) the necessary Unicode files are not
available locally, table generation is skipped, but no error code is
returned, so not to obstruct the build process if called from a Makefile.
E.g. arm ABI requires -fshort-enums for bare-metal toolchains.
Given there are only 29 category enums, the compiler chooses an
8 bit enum type, so a size of 11 bits for the bitfield leads to
a compile time error:
error: width of 'cat' exceeds its type
enum category cat: 11;
^~~
Fix this by aligning the size of the category members to byte
borders.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
touupper and toulower didn't return a value in all cases. Worse,
this only broke Cygwin when building without optimization for debug
purposes.
Why GCC neglects to notice this is a mystery.
While at it, fix formatting.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
The tow* functions use an included case conversion table which can be
generated from Unicode data.
The isw* functions use a character categories table (provided by
categories.c) which can be generated from Unicode data.
Delegation between current-locale and specific-locale-dependent functions
was reverted towards the generic locale-dependent functions (*_l.c);
this is however only relevant on systems with non-Unicode wide character
locales, thus not on Cygwin.
Table categories.t and tag enumeration categories.cat provide
character class data for most of the isw* functions.
These data are generated from Unicode data.
Linux and FreeBSD use int as well. In addition, this fixes an Ada
incompatiblity problem on 64-bit targets. See also GCC:
gcc/ada/libgnarl/s-osinte__rtems.ads
Signed-off-by: Sebastian Huber <sebastian.huber@embedded-brains.de>
Locale modifier @cjkwide makes Unicode "ambiguous width" characters
wide. So ambiguous width characters can be enforced to have width 2
even in non-CJK locales. This gives e.g. users of "Powerline symbols"
the opportunity to adjust their width to the desired behaviour (and the
behaviour apparently expected by some tools) without having to set a CJK
locale and without losing consistence of terminal character width with
wcwidth/wcswidth locale width.
At least with Binutils 2.30 and GCC 7.3 we need symbol definitions
without the leading underscore.
Signed-off-by: Sebastian Huber <sebastian.huber@embedded-brains.de>
This is a NetBSD-specific detail which does not apply to Newlib, causing
linking issues in certain scenarios:
https://cygwin.com/ml/cygwin/2018-01/msg00189.html
Signed-off-by: Yaakov Selkowitz <yselkowi@redhat.com>
New optimized powf, logf, log2f, expf and exp2f yield worse performance
on Arm targets with only single precision instructions because the
double precision arithmetic is then implemented via softfloat routines.
This patch uses the old implementation when double precision
instructions are not available on Arm targets.
Testing: Built newlib with GCC's rmprofile Arm multilibs and compared
before/after -> only the above functions are changed and calls to them
(name change from logf to __ieee754_logf and similar). Testing the
changed function on a panel of values yields the same result before the
original patches to improve them and after this one. Double checking the
performance by looping the same panel of values being tested on Arm
Cortex-M4 does show the performance regression is fixed.
This patch fixes a syntax error in exit.c that was introduced during the
ANSI-fication of newlib. The patch fixes a compile-time issue that arises when
newlib is configured with the --enable-lite-exit feature.
mmacy
sbruno
brooks
jeff
kib
jtl
pkelsey
ae@FreeBSD.org
pfg
glebius
sephe
des
delphij
imp
ed@FreeBSD.org
jhb
Sebastian Huber
Masamichi Hosoda
Wilco Dijkstra
Aditya Upadhyay
Corinna Vinschen
Jordi Sanfeliu
Yaakov Selkowitz
DJ Delorie
Kumar Gala
Siddhesh Poyarekar
Szabolcs Nagy
Takashi Yano
Jon Beniston
Thomas Kindler
Jeff Johnston
Matthias Kannwischer
Freddie Chopin
Thomas Wolff
Hakan Lindqvist
Joel Sherrill
Richard Earnshaw
Thomas Wolff
Jon Turney
Our Air Quality
Jaap de Wolff
Thomas Preudhomme
Orlando Arias
Kito Cheng