Commit Graph

111 Commits

Author SHA1 Message Date
Ed Maste
4b12fb6103 Remove duplicate copies of trivial getcontextx.c
Only i386 and amd64 provide a non-trivial __getcontextx(). Use a common
trivial implementation in gen/ for other architectures, rather than
copying the file to each MD subdirectory.

Reviewed by:	kib
Sponsored by:	The FreeBSD Foundation
Differential Revision:	https://reviews.freebsd.org/D1472
2015-01-12 18:13:38 +00:00
Dimitry Andric
18ecc14c78 Apparently more Makefiles use stuff from compiler-rt, so fix them up
too.  (This did not show during a make universe, strangely.)
2015-01-08 20:11:38 +00:00
Andrew Turner
a215cdfdb0 Switch to the ARM unified assembly language as the clang integrated as only
supports it. Binutils supports it when the ".syntax unified" directive
is set.

Sponsored by:	ABT Systems Ltd
2014-11-29 15:56:52 +00:00
Andrew Turner
cb8bcafa71 Add support for the __aeabi_c*cmp* functions. These are similar to the
existing functions with the exception they use the condition flags to
store the result.

Differential Revision:	https://reviews.freebsd.org/D872
Silence from:	current@ and numerics@
MFC after:	1 week
2014-10-14 14:27:51 +00:00
Andrew Turner
eabf853d8d Clean up detection of hard-float ABIs. As with big-endian in r272368 we
can check against arm*hf*.
2014-10-01 16:08:19 +00:00
Andrew Turner
6d4766c1b8 Remove MK_ARM_EABI, the armeb issues have been fixed. The code to support
the oabi is still in the tree, but it is expected this will be removed
as developers work on surrounding code.

With this commit the ARM EABI is the only supported supported ABI by
FreeBSD on ARMa 32-bit processors.

X-MFC after:	never
Relnotes:	yes
Differential Revision: https://reviews.freebsd.org/D876
2014-10-01 08:26:51 +00:00
Andrew Turner
954c837259 Add the fp{get,set}{mask,round} functions to the public symbols in the map.
These are only exported for armv6hf as the soft-float ABIs have these in
the softfloat Symbol.map file.
2014-09-05 18:11:36 +00:00
Ian Lepore
f2e71517e0 In ARM asm code, ensure that every ENTRY(foo) has a matching END(foo).
The EABI unwind info requires a .fnend for every .fnstart, and newer
binutils will complain about seeing two .fnstart in a row.  This change
allows newer tools to compile our code.

Reported by:	bapt
Reviewed by:	imp
2014-08-31 17:21:51 +00:00
Warner Losh
161fedb9cc From https://sourceware.org/ml/newlib/2014/msg00113.html
By Richard Earnshaw at ARM
>
>GCC has for a number of years provides a set of pre-defined macros for
>use with determining the ISA and features of the target during
>pre-processing.  However, the design was always somewhat cumbersome in
>that each new architecture revision created a new define and then
>removed the previous one.  This meant that it was necessary to keep
>updating the support code simply to recognise a new architecture being
>added.
>
>The ACLE specification (ARM C Language Extentions)
>(http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.set.swdev/index.html)
>provides a much more suitable interface and GCC has supported this
>since gcc-4.8.
>
>This patch makes use of the ACLE pre-defines to map to the internal
>feature definitions.  To support older versions of GCC a compatibility
>header is provided that maps the traditional pre-defines onto the new
>ACLE ones.

Stop using __FreeBSD_ARCH_armv6__ and switch to __ARM_ARCH >= 6 in the
couple of places in tree. clang already implements ACLE. Add a define
that says we implement version 1.1, even though the implementation
isn't quite complete.
2014-08-14 04:20:13 +00:00
Ian Lepore
677a88cec2 Add dl_unwind_find_exidx() for ARM EABI, required for C++ exception
handling.  For statically linked apps this uses the __exidx_start/end
symbols set up by the linker.  For dynamically linked apps it finds the
shared object that contains the given address and returns the location and
size of the exidx section in that shared object.

The dl_unwind_find_exidx() name is used by other BSD projects and Android,
and is mentioned in clang 3.5 comments as "the BSD interface" for finding
exidx data.  GCC (in libgcc_s) expects the exact same API and functionality
to be provided by a function named __gnu_Unwind_Find_exidx(), so we provide
that with an alias ("strong reference").

Reviewed by:	kib@
MFC after:	1 week
2014-07-19 22:13:27 +00:00
Andrew Turner
8258fd68ef Set the new floating point exception mask correctly
Submitted by:	Keith White <kwhite@site.uottawa.ca>
2014-04-28 18:54:12 +00:00
Andrew Turner
3f5730095b Rename the fp{get,set}* files so they no longer conflict with the softfloat
version of these files. Keep them within this directory so they can be used
to implement the armv6 version of these functions.
2014-04-21 09:43:22 +00:00
Andrew Turner
8772ff83ee Add the deprecated fp{get,set}* functions, a few ports use them. 2014-04-20 14:58:14 +00:00
Warner Losh
a5fc5b6223 Convert from WITHOUT_SYSCALL_COMPAT to MK_SYSCALL_COMPAT. 2014-04-05 17:54:43 +00:00
Andrew Turner
73279d4113 Add a new ARM TARGET_ARCH, armv6hf. This is considered experimental.
This targets the existing ARMv6 and ARMv7 SoCs that contain a VFP unit.
This is an optional coprocessors may not be present in all devices, however
it appears to be in all current SoCs we support.

armv6hf targets the VFP variant of the ARM EABI and our copy of gcc is too
old to support this. Because of this there are a number of WITH/WITHOUT
options that are unsupported and must be left as the default value. The
options and their required value are:
 * WITH_ARM_EABI
 * WITHOUT_GCC
 * WITHOUT_GNUCXX

In addition, without an external toolchain, the following need to be left
as their default:
 * WITH_CLANG
 * WITH_CLANG_IS_CC

As there is a different method of passing float and double values to
functions the ABI is incompatible with existing armv6 binaries. To use
this a full rebuild of world is required. Because no floating point values
are passed into the kernel an armv6 kernel with VFP enabled will work with
an armv6hf userland and vice versa.
2014-03-23 12:49:25 +00:00
Andrew Turner
cc0c9bba36 Implement __flt_rounds for ARMv6 hard-float. The fpscr register stores the
current rounding mode used by the VFP unit.
2014-03-22 12:28:21 +00:00
Andrew Turner
9b2d210438 Remove an extra 1 in an #if line
Reported by:	ian@
2014-03-16 14:21:09 +00:00
Andrew Turner
aac3f7c724 Fix the spelling of function.
Reported by:	ian@
2014-03-16 14:19:46 +00:00
Andrew Turner
a06918a349 Add an optimised version of the float and double helper functions. These
are only used on armv6 when the vfp unit is detected. They will also be
available for the upcoming armv6hf platform, however while not used by
default there will need to be defined for any software that calls them
directly.
2014-03-16 13:16:30 +00:00
Andrew Turner
2ab2eea1d9 Export _libc_arm_fpu_present as a private symbol to be used by other
system libraries, for example libm.
2014-03-11 00:21:49 +00:00
Marcel Moolenaar
8876613dc5 Replace use of ${.CURDIR} by ${LIBC_SRCTOP} and define ${LIBC_SRCTOP}
if not already defined. This allows building libc from outside of
lib/libc using a reach-over makefile.

A typical use-case is to build a standard ILP32 version and a COMPAT32
version in a single iteration by building the COMPAT32 version using a
reach-over makefile.

Obtained from:	Juniper Networks, Inc.
2014-03-04 02:19:39 +00:00
Andrew Turner
0a10f22a30 On ARM EABI double precision floating point values are stored in the
endian the CPU is in, i.e. little-endian on most ARM cores.

This allows ARMv4 and ARMv5 boards to boot with the ARM EABI.
2013-09-07 14:04:10 +00:00
Andrew Turner
f44fbb4c13 Teach siglongjmp about the VFP version of the setjmp buffer.
This fixes a bug where a call to sigsetjmp followed by siglongjmp may fail
when run on a machine with VFP enabled.
2013-07-21 12:47:33 +00:00
Andrew Turner
8e03dd5926 Finish pulling in the NetBSD setjmp/longjmp updates on ARM.
Store/restore the VFP registers in setjmp/longjmp on ARM EABI if VFP is
enabled in the kernel. It checks the hw.floatingpoint sysctl to see if
floating-point is available and uses this to determine if it should store
them. If it does it uses a different magic value so longjmp is able to know
if it should load them.
2013-06-07 22:01:06 +00:00
Andrew Turner
4e75169f43 Include machine/setjmp.h to get the definition of _JB_MAGIC__SETJMP. This
allows us to remove it from the ARM copy of machine/asm.h.
2013-06-07 21:13:28 +00:00
Andrew Turner
d2d3491d5a Remove part of the NetBSD longjmp code that was not ready to be merged. 2013-06-05 07:37:45 +00:00
Andrew Turner
8ed717de58 Start to merge the updated ARM NetBSD setjump/longjmp functions. To begin
with merge the functions but leave out the code to save/load the VFP
registers as that requires other changes to ensure the VFP is enabled
first.

This removes storing the old fpa registers. These were never fully
supported, and the only user of this code I can find have moved to newer
CPUs which use a VFP.
2013-06-04 19:47:26 +00:00
Konstantin Belousov
55a1911ef2 The getcontext() from the __fillcontextx() call in the
check_deferred_signal() returns twice, since handle_signal() emulates
the return from the normal signal handler by sigreturn(2)ing the
passed context.  Second return is performed on the destroyed stack
frame, because __fillcontextx() has already returned.  This causes
undefined and bad behaviour, usually the victim thread gets SIGSEGV.

Avoid nested frame and the need to return from it by doing direct call
to getcontext() in the check_deferred_signal() and using a new private
libc helper __fillcontextx2() to complement the context with the
extended CPU state if the deferred signal is still present.

The __fillcontextx() is now unused, but is kept to allow older
libthr.so to be used with the new libc.

Mark __fillcontextx() as returning twice [1].

Reported by:	pgj
Pointy hat to:	kib
Discussed with:	dim
Tested by:	pgj, dim
Suggested by:	jilles [1]
MFC after:	1 week
2013-05-28 04:54:16 +00:00
Andrew Turner
b27ca6313e When clang builds libc it may insert calls to __aeabi_* functions. Normally
this is not a problem as they are resolved by libgcc. The exception is for
the __aeabi_mem* functions. These call back into libc to the appropriate
function. This causes issues for static binaries as we only link against
libc once so there is no way for it to call into libgcc and back.

The fix for this is to include these symbols in libc but keep them hidden
so binaries use the libgcc version.
2013-02-12 06:04:51 +00:00
Konstantin Belousov
150facd256 Rework the __vdso_* symbols attributes to only make the symbols weak,
but use normal references instead of weak.  This makes the statically
linked binaries to use fast gettimeofday(2) by forcing the linker to
resolve references and providing the neccessary functions.

Reported by:	bde
Tested by:	marius (sparc64)
MFC after:	2 weeks
2013-01-30 12:48:16 +00:00
Andrew Turner
0129836894 Add the required __aeabi_* functions to libc.
The floating point functions are here rather than compiler-rt because the
libc softfloat code allows us to set the rounding mode.
2013-01-19 05:33:55 +00:00
Andrew Turner
bb41cbb27c Use the compiler-rt version __{u,}divsi3 and __{u,}modsi3 on ARM EABI 2013-01-19 04:11:45 +00:00
Andrew Turner
2aebb6cefe Update the syscall calling convention for ARM EABI. We store the syscall
in r7 and use ip to store the old version of r7 as it is not guaranteed to
be kept when calling a subroutine. The kernel will preserve the register
across system calls.
2013-01-19 04:03:18 +00:00
Andrew Turner
955c8cbb49 Silence a clang warning by telling it we are only interested in left
shifting the lower 32bits of the floating point value when we demangle it.
2013-01-06 00:49:06 +00:00
Oleksandr Tymoshenko
8e585021b5 Merging of projects/armv6, part 2
Handle TLS for ARMv6 and ARMv7
2012-08-15 03:09:00 +00:00
Tim Kientzle
061103347c Per kib, since __flt_rounds is being added to FreeBSD 10, it
belongs in FBSD_1.3.

MFC after:	1 week
2012-06-15 03:57:16 +00:00
Tim Kientzle
ed1589b9b7 __flt_rounds is a public symbol (expands from the FLT_ROUNDS macro),
so include it in the public namespace on arm just as with
other architectures.

This corrects r236816.

Submitted by:	Jan Sieka
MFC after:	1 week
2012-06-14 03:27:01 +00:00
Tim Kientzle
d849ce3315 __flt_rounds was omitted from the exported symbols here.
Submitted by:	Jan Sieka
Reviewed by:	arm@
MFC after:	1 week
2012-06-09 16:09:53 +00:00
Andrew Turner
0b898a9ef1 Replace the C implementation of __aeabi_read_tp with an assembly version.
This ensures we follow the ABI by preserving registers r1-r3.

Reviewed by:	jmallett, imp
2012-04-16 09:38:20 +00:00
Andrew Turner
fcb6e7147b Unlike other functions __aeabi_read_tp function must preserve r1-r3. The
currently generated code clobbers r3. Fix this by loading ARM_TP_ADDRESS
using inline assembly.

Approved by:	imp (mentor)
2012-04-08 04:36:27 +00:00
Stanislav Sedov
b30d62be27 - Revert part of r234005, which I did not intend to commit.
Sorry! :(
2012-04-07 23:51:16 +00:00
Stanislav Sedov
47db1ea0bf - Add kernel config file for QEMU-emulated gumstix board. 2012-04-07 23:48:51 +00:00
Oleksandr Tymoshenko
61ad92ef03 Unbreak trhe build by including proper header 2012-03-18 08:08:06 +00:00
Oleksandr Tymoshenko
7dc77cabb9 Fix TLS for statically linked binaries
Approved by:	cognet
2012-03-18 03:42:54 +00:00
Olivier Houchard
2dee026843 Add __aeabi_read_tp to the symbol list. 2012-03-04 15:25:11 +00:00
Oleksandr Tymoshenko
2643c5abff Add __aeabi_read_tp function required for thread-local storage
Reviewed by:	cognet
2012-02-14 00:12:15 +00:00
Konstantin Belousov
754f1c1e63 Make the sys/ucontext.h self-contained by changing the return type
of __getcontextx_size(3) from size_t to int.

PR:	ports/164654
MFC after:	1 month
2012-02-01 13:33:53 +00:00
Konstantin Belousov
2b1de0afd1 Add API for obtaining extended machine context states that cannot be
fit into existing mcontext_t.

On i386 and amd64 do return the extended FPU states using
getcontextx(3). For other architectures, getcontextx(3) returns the
same information as getcontext(2).

Tested by:  pho
MFC after:  1 month
2012-01-21 18:00:28 +00:00
David Schultz
7082a2cf72 Don't inline fenv.h functions on arm for now. Inlining makes sense:
the function bodies require only 2 to 10 instructions.  However, it
leads to application binaries that refer to a private ABI, namely, the
softfloat innards in libc.  This could complicate future changes in
the implementation of the floating-point emulation layer, so it seems
best to have programs refer to the official fe* entry points in libm.
2012-01-20 06:54:30 +00:00
David Schultz
86d131a04a Implement FLT_ROUNDS for arm. Some (all?) arm FPUs lack support for
dynamic rounding modes, but FPUless chips that use softfloat can support it
because everything is emulated anyway.  (We presently have incomplete
support for hardware FPUs.)

Submitted by:	Ian Lepore
2012-01-16 04:08:29 +00:00