sync machine list with spec

Do not warn for ELF note for FDO packaging metadata
(https://systemd.io/COREDUMP_PACKAGE_METADATA/),
found in userland binaries of Fedora 36.

s/hidding/hiding/

Arrange so that no options COREDUMP and no options PTRACE work together.
Thanks to Paul Goyette for testing.

Rename:
- EM_ALPHA (41) -> EM_OLD_ALPHA
- EM_ALPHA_EXP (36902) -> EM_ALPHA

36902 has been the value in common usage for a long time, and this naming
aligns with binutils, etc.

PR port-alpha/51926

Add defines for MIPS_XHASH (.MIPS.xhash)

MIPS version of GNU_HASH as supported by GNU toolchain.

Register DT_GNU_HASH

GNU-style hash table is supported in GNU and LLVM toolchains.

https://sourceware.org/ml/binutils/2006-10/msg00377.html

branches: 1.164.2;
Sync the e_ident[EI_OSABI] magic values with the ELF specification

source: http://www.sco.com/developers/gabi/latest/ch4.eheader.html

Update e_machine magic values with the ELF specification

source: http://www.sco.com/developers/gabi/latest/ch4.eheader.html

Fix the EM_TILEGX number in the ELF header

source: http://www.sco.com/developers/gabi/latest/ch4.eheader.html

191 is confirmed in LLVM, GNU toolchain, FreeBSD.

deduplicate the elf auxv builder code, welcome to 8.99.43

branches: 1.160.2;
Complete the list of DF_1 flags.

There have never been binaries with a checksum tag for NetBSD.
These days the equivalent functionality is provided by the build
id tag. It was incorrect in the first place to overload tag #2
(which used to be the emulation name) with a different meaning, so
actually document the meaning that was once used (because binaries
with that tag exist).

Tag existance noticed by kre@

branches: 1.158.2;
Since _RUMPKERNEL is really userland code it needs ARCH_ELFSIZE,
not KERN_ELFSIZE, make it so!

Use KERN_ELFSIZE if we are in the kernel.

Add mips abiflags section

Add an ElfW() macro like linux.

Add a note for the auxv array so we can find our load location from a
core file of a PIE binary.

Intel MCU.

Don't do _BSD_SIZE_T_ dance here. Use <stddef.h> in userland.

This file is used in tools, where _BSD_SIZE_T_ is unreliable. In the
kernel, <sys/types.h> will define size_t anyway. Outside the kernel,
<stddef.h> is the standard way to get size_t.

Ignore the ancient 01.01 tag that gnuc used to put in old NetBSD binaries.

Add buildid Go note

Revamp PaX:
- don't confuse between ELF flags and proc flags. Introduce the proc-
specific P_PAX_ASLR, P_PAX_MPROTECT and P_PAX_GUARD flags.
- introduce pax_setup_elf_flags(), which takes as argument the PaX flag
of the ELF PaX note section, and which sets the proc flag as
appropriate. Also introduce a couple of other functions used for that
purpose.
- modify pax_aslr_active(), and all the other similar pieces of code, so
that it checks the proc flag directly, without extra ELF computation

In addition to making PaX clearer, the combination of these changes fixes
the following bug: if a non-PaX'ed process is launched, and then someone
sets security.pax.{aslr,mprotect,segvguard}.global=1, the process becomes
PaX'ed while its address space hasn't been randomized, which is not likely
to be a good idea.

Now, only the proc flag is checked at runtime, which means the process's
PaX status won't be altered during the execution.

Also:
- declare PAX_DPRINTF, makes it more readable
- fix a typo in exec_elf.h

Remove unofficial AVR32

Add a few more EM_*

add struct exec_package;

branches: 1.145.2;
Add standard size_t dance

Add EM_RISCV

Add basic support for indirect functions. It allows providing a public
function symbol with an implementation choosen at run time.
Refactor calls to functions by address in ld.elf_so to create temporary
function descriptors on the stack, if the address is not leaked outside.

Limitations:
- no support for initialising static storage with function pointers
- no support for unnamed resolver functions

Inspired by FreeBSD's r228435 by kib@freebsd.org.

EM_OPENRISC is now EM_OR1K but keep a define of EM_OPENRISC to EM_OR1K around.

branches: 1.141.2;
Define ELF_MAXNOTESIZE, ELF_MAXSHNUM and ELF_MAXPHNUM in <sys/exec_elf.h>, so
that it can be used externally.

branches: 1.140.2;
Add more definitions, needed by the new elf toolchain.

Remove the last argument of elf_check_header(). It is easier - and faster - to
check the e_type field in the calling function. Other BSD's already do this.

ok christos@

Fix typo in DF_STATIC_TLS. From Elliott Hughes on tech-userlevel.

Fix bug in previous (panic during process core dump).
Change the interface to ELFNAMEEND(coredump_savenote) so that the caller
doesn't need to know the type of the elf note header.
Simplifies the calling code somewhat.

Instead of generating all the 'note' sections twice (and hoping that the
'fast path' size on the first path matches the actual size on the second)
save all the notes (mostly the cpu registers for all the LWPs) in
malloced memory on the first pass.
Sanity check that the number of memory segments matches written matches
the count obtained earlier. If gcore() is used they could differ.
(Not sure that returning ENOMEM is ideal, but it is better than a crash.)

Remove the incorrect and unused Elf64_Shalf.
At one time it was used for a 32bit item - but that definition has long
since been changed to use a reasonable 32bit type.

Add more gnu sections

Change the type of the 'cookie' that holds the state of the core dump file
from 'void *' to the actual type 'struct coredump_iostate *'.
In most of the code the contents of the structure are still unknown.
This just stops the wrong type of pointer being passed to the 'void *'
parameter.
I hope I've found everything, amd64 GENERIC and i386 GENERIC & ALL compile.

Define a new NetBSD specific note to encode the code model used for a binary.
Allow MD code to check these notes in the exec path.

Add EM_AARCH64 from Elliott Hughes

Add DF_* values for DT_FLAGS

Add a NetBSD note variant for recording the machine_arch for which the
executable was built. (currently only used by earm variants)

branches: 1.128.2;
Add the ARM processor-specific sections:
http://infocenter.arm.com/help/topic/com.arm.doc.ihi0044e/IHI0044E_aaelf.pdf

Add DT_PREINIT_ARRAY/DT_PREINIT_ARRAYSZ and few other missing ones.

branches: 1.126.2;
back out elf note changes and use EF_ARM_ABIVERS to determine EABI usage.

define<tab> consistency

Add a elf note to describe the ARM ABI in use. If encounters on arm,
set EXEC_ARM_AAPCS bit in exec_package's ep_flags.
XXX kind of gross but it there's isn't a MD hook for notes so ...

- Recognize the SuSE ABI note.
- Restructure the code to do the checking in the appropriate note type,
and harmonize all the checks to be positive.
- Print only the tag data being careful not to overrun the allocated buffer.

Add AuxInfo record with the base address of the main thread.

Kill duplicate SHF_MASKOS/SHF_MASKPROC defines.

Add more SHF_ and whitespace align SHT_

whitespace diffs [and comments]

branches: 1.118.2;
Add SHT_GNU_HASH.

branches: 1.117.2;
Replace the word `NetBSD' with `SuSE' in the rest of comments
for the cpp macro constatnt ELF_NOTE_SUSE_*.

Add SuSE note

define<tab>

document all the notes we know about in a consistent fashion.
XXX: someone please document the checksum tags.

PR/45015: ld.elf_so: support ELF symbol versioning
Applied latest patch.

fix ELFxx_M_SIZE macro.

fix spelling.

Refactor ps_strings access. Based on PK_32, write either the normal
version or the 32bit compat layout in execve1. Introduce a new function
copyin_psstrings for reading it back from userland and converting it to
the native layout. Refactor procfs to share most of the code with the
kern.proc_args sysctl handler.

This material is based upon work partially supported by
The NetBSD Foundation under a contract with Joerg Sonnenberger.

add the rest of the gnu specific pt sections.

branches: 1.108.2; 1.108.4;
Add support for DF_1_BIND_NOW, DF_1_NODELETE and DF_1_NOOPEN marked
objects, and the RTLD_NODELETE and RTLD_NOLOAD flags to dlopen(3).

Mark libpthread as DF_1_NOOPEN and use it to test the functionality.

Somewhat taken from FreeBSD.

Fixes PR 42029.

OK from christos and joerg.

Add PT_GNU_EH_FRAME definition

Implement dl_iterate_phdr.

Somewhat taken from FreeBSD. Manual page from OpenBSD.

Define PT_TLS.

Introduce a new type Elf_Symindx for use in decoding the symbol hash table
section and allow this type to be overridden.

The ELF specification says it should always be uint32_t (Elf_Word), but
alpha decided to be different (not sure why). Define Elf_Symindx to be
uint64_t on alpha.

Alpha no longer uses non-standard definitions of Elf64_Sword and
Elf64_Word. Remove the ability to override these types.

Fixes ld.elf_so after Herculean effort from me and martin.

Replace the current usage of Elf64_Half with Elf64_Word and rename
NetBSD specific Elf64_Quarter to Elf64_Half. This restores compatibility
with the common ELF specifications.

branches: 1.102.2;
Use #define<tab> consistently.

DTrace: fix a few problems introduced by the recent ZFS update (expected
problems from the OpenSolaris merge).

branches: 1.100.2;
Add defns for Elf{32,64}_Syminfo.

Add some definitions, from FreeBSD.

fix comments: st_name is index in .strtab

Add a signed offset type

branches: 1.96.4;
Remove LKMs and switch to the module framework, pass 1.

Proposed on tech-kern@.

branches: 1.95.6; 1.95.8; 1.95.10; 1.95.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.94.6; 1.94.8; 1.94.10;
Add PaX ASLR (Address Space Layout Randomization) [from elad and myself]

For regular (non PIE) executables randomization is enabled for:
1. The data segment
2. The stack

For PIE executables(*) randomization is enabled for:
1. The program itself
2. All shared libraries
3. The data segment
4. The stack

(*) To generate a PIE executable:
- compile everything with -fPIC
- link with -shared-libgcc -Wl,-pie

This feature is experimental, and might change. To use selectively add
options PAX_ASLR=0
in your kernel.

Currently we are using 12 bits for the stack, program, and data segment and
16 or 24 bits for mmap, depending on __LP64__.

branches: 1.93.4;
rtld changes for PR bin/37347:

- Apply patch from J.T. Conklin to execute .init/.fini functions in order.
- Support DF_1_INITFIRST and mark libc with DF_1_INITFIRST. Shared libs
should be recording a dependency on libc, but it's too late to do that.

Ok christos@.

branches: 1.92.2;
- add an elf aux vector entry for implementing $ORIGIN.
- the code to convert from a vnode to a path is commented out now until
a better solution is implemented. Only absolute paths work for now
(which is most of the cases).

requested by core

branches: 1.91.2; 1.91.8;
Add EM_AVR32

branches: 1.90.2; 1.90.6;
Use an elf note to handle pax arguments. This is a temporary solution to
avoid wasting OS flag bits. In the future we'll probably use fileassoc to
achieve this (once there is a way to make fileassoc persistent) or in the
shorter term libelf, so that we can add and remove the note on demand instead
of burning bits on each binary. Of course since this is a tool, this means
that we'll need to think about how to handle libelf...

branches: 1.89.2; 1.89.8; 1.89.10; 1.89.12; 1.89.16;
Make sys/exec_elf.h and machine/elf_machdep.h available to host
tools, and allow paxctl to use them when compiled as a host
tool. This fixes the build on MacOS X (and perhaps other platforms)

Damn, missed in previous commit. Add PaX Segvguard flags.

branches: 1.87.2; 1.87.4;
Regen.

Introduce PaX MPROTECT -- mprotect(2) restrictions used to strengthen
W^X mappings.

Disabled by default.

First proposed in:

http://mail-index.netbsd.org/tech-security/2005/12/18/0000.html

More information in:

http://pax.grsecurity.net/docs/mprotect.txt

Read relevant parts of options(4) and sysctl(3) before using!

Lots of thanks to the PaX author and Matt Thomas.

branches: 1.85.4; 1.85.6; 1.85.8; 1.85.12;
there is no elf64_read_from.

merge ktrace-lwp.

Add NetBSD checksum notes.

branches: 1.82.2;
Rework the coredump code to have no explicit knownledge of how coredump
i/o is done. Instead, pass an opaque cookie which is then passed to a
new routine, coredump_write, which does the actual i/o. This allows the
method of doing i/o to change without affecting any future MD code.
Also, make netbsd32_core.c [re]use core_netbsd.c (in a similar manner that
core_elf64.c uses core_elf32.c) and eliminate that code duplication.
cpu_coredump{,32} is now called twice, first with a NULL iocookie to fill
the core structure and a second to actually write md parts of the coredump.
All i/o is nolonger random access and is suitable for shipping over a stream.

nuke trailing whitespace

de-__P

branches: 1.79.2; 1.79.4;
Use uint32_t casts, rather than Elf_Byte, in ELF_ST_BIND() and ELF_ST_TYPE()
so we don't get screwed by promotion rules (and makes nlist_elf32.ln build
again).

According to the ELF docs, ELF{32,64}_ST_{BIND,TYPE,INFO,VISIBILITY} are
parts of the public interface. Define them in terms of ELF_ST_*.
Add STN_UNDEF which is part of the public interface also.

Remove ELFn_ST_{BIND,TYPE,INFO}(), and instead add a generic ELF_ST_*() that
works with an Elf_Byte. Also add ELF_ST_VISIBILITY() and STV_* types.

add EI_NIDENT (16) as seen in other <elf.h> headers.

Uppercase CPU, plural is CPUs.

fix typo in comment: magic numbers are indexed EI_MAG0 through EI_MAG3,
not ELFMAG0 through ELFMAG3 (which are the contents)

introduce another "magic" ELF address to tell that the file has to be
mapped at the address it was linked for

- Add execsw es_setup_stack
- gc all the rest of the setup stack functions.

Make elf{32|64}_check_header public, as it will be used by irix_elf32_probe.
While we are there, cut to 80 chars, and ANSIfy prototypes

branches: 1.70.2;
Back out the lwp/ktrace changes. They contained a lot of colateral damage,
and need to be examined and discussed more.

Pass lwp pointers throughtout the kernel, as required, so that the lwpid can
be inserted into ktrace records. The general change has been to replace
"struct proc *" with "struct lwp *" in various function prototypes, pass
the lwp through and use l_proc to get the process pointer when needed.

Bump the kernel rev up to 1.6V

Add a field to netbsd_elfcore_procinfo that identifies the LWP targeted
by the killing signal.

Merge the nathanw_sa branch.

Fix typo in comment.

Program header types 0x60000000 ~ 0x6fffffff are reserved for
OS-specific semantics.

Update URLs again (serves me right for committing an old change
without checking it). Back to sco.com, but with a different path.

Update URLs for ELF spec (sco.com -> caldera.com).

- pass struct proc to copyargs
- rename aux vector entries AT_{R,E}{U,G}ID from AT_SYUN_{R,}P{U,G}ID
- update new aux vector size.

Add official machine numbers for PDP10 and PDP11.

branches: 1.60.8;
Define ELF32_EHDR_FLAGS_OK()/ELF64_EHDR_FLAGS_OK() and use it
as an added measure to make sure that we can execute a binary.
These default to (1) if elf_machdep.h does not override them.

On Sun2, ELF32_EHDR_FLAGS_OK() checks for the presense of EF_M68000,
since the 68010 cannot run binaries for the 68020-and-up.

Add support for dumping ELF-cormat core files.

Rather than using a NetBSD-specific header file <machine/int_types.h>
and non-standard inttype-like types, pull in <sys/types.h> if
_KERNEL or _STANDALONE and <inttypes.h> otherwise, and use standard
inttype types.

Discussed with and OK'd by Christos.

Use <machine/elf_machdep.h> to override Elf64_Word and Elf64_Sword
on the Alpha (which uses a non-standard definitions for historical
reasons), rather than #ifdef __alpha__ (which is not cross-tool safe).

Include <machine/elf_machdep.h> earlier, at the beginning of the file,
after the inttypes header.

Add EM_NS32K for the ns32k series CPUs.

branches: 1.54.4;
make elf64_copyargs() of type int also per the recent copyargs changes

adjust to the new copyargs footprint

- add exec_read_from and make exec_elf32 use it.
- add a macho probe function

branches: 1.51.2;
Fix up comment botch.
_Never_ fix a simple formatting thing _after_ a test compile.

Add some more e_machine values and a reference to the SCO website with
ELF ABI information.

X8664 -> X86_64

Elf64_Word should be 32 bits; alpha is the exception rather than
the rule, so change the ifdef to reflect this.

Fix botched commit; I forgot to bring this up to date to -current
before committing it.

Add IDs for x86-64.

ElfNN_RelA -> ElfNN_Rela, which I missed in the Great Renaming;
noticed by Juan Jose Garcia-Ripoll in PR standards/12274.

Fix misinformation in comment

Slight readability improvement to rev. 1.39.

Describe possible section types with a few words.

populate OSABI list

branches: 1.40.2;
Tighten up the ELF signature checks, and actually look for the ABI tag added
in newer glibc versions.

Introduce Elf64_Xword and Elf64_Sxword, and use them in various places instead
of Elf64_Word.
Make Elf64_Word be 32 bits on sparc64. (XXXX Need to determine what it's
supposed to be on other, but it's current used to size entries in DT_HASH,
which is 32 bits on sparc64 and 64 bits on alpha.)

Be less invasive and use private names for exact-width integer types.

branches: 1.37.4;
Make this file possible to use again.

* Add more OS ABI constants and dynamic array tags from D1998-04-29.
* Enhance readibility: define data structures first, followed member value
constants, and add more comments.

Add OS ABI constants from the 1998-04-29 draft.

Delint.

Make sure the elf_args strucure is appropriate for the size of the ELF binary.

Define ELF_ST_*() to ELF{32,64}_ST_* as appropriate. They evaluate to
the same thing, but this provides consistency with everything else.

Fix comment typo introduced in rev. 1.27.

Fix obvious typos.

Add a few comments from <mips/elf.h>.

Add ELF64 versions of st_info manipulation macros I missed in previous.

As discussed with Christos, rename the definitions in this file to match those
found in various SVR4 ABIs.

Move ELFNAME macros out of #ifdef _KERNEL.

branches: 1.25.2; 1.25.4; 1.25.6;
Merge in NetBSD/sh3 from cvs.kame.net repository.

Tree structure:
- sys/arch/sh3: sh3 generic code
As commented, in-chip device drivers are put into sys/arch/sh3/dev.
- sys/arch/evbsh3: sh3 evaluation boards (pure sh3 CPU, no fancy external HW)
- sys/arch/mmeye: Brains mmEye, www.brains.co.jp
MI source code includes couple of #ifdef for sh3-coff support.
(sh3 uses coff or elf)

Needs some more improvements, especialy in sys/arch/sh3/conf/files.sh3,
to compile the tree (due to last minute tree structure change).

Add the newly blessed magic number for VAX.

Add NetBSD note constants.

Add a few new machine numbers and aux vectors.

Move elf function name building macros here (ELFNAME, ELFNAME2, etc...) Add defines related to elf note sections for use in determining executable type/emulation.

Fix copyright typos...

Assign copyright to TNF.

Oops, change previous to use the registered name found in the System V ABI,
draft edition as of April 29, 1998.

Add machine identifier for the PA.

Add machine identifier for ARM. Add (commented out) real ident
for Alpha as well.

add Elf_em_sparc64 at 43 (from solaris 2.7)

Remove a trailing ',' in an enum declaration.

Addes support for sparc32plus ELF executables: v9 in v8's clothing.

Generate dependencies on the exec format options EXEC_AOUT, EXEC_ECOFF,
EXEC_ELF32, EXEC_ELF64, and EXEC_SCRIPT.

Add Section Attribute Flags (SHF_*) as defined in _System V Application
Binary Interface, Third Edition_, section 4, page 16.

define Elf{,32,64}_Note as appropriate, add a constant for NetBSD-specific
notes' 'name' fields, and add a couple of NetBSD-specific note types.

rename the ELF section type constants to be Elf_* rather than Elf32_*,
since they're shared between ELF32 and ELF64. (Thought i checked this in
earlier, but apparently didn't!)

fix typo in definition of Elf32_e_ident

PowerPC port

fix up section header type enum/values

add and use a machine-dependent header, which currently defines some
macros to use to remove #ifdefs from the machine ID case check.
Eventually, these headers will contain other information, e.g.
machine-dependent relocation information, etc.

add definitions for 64-bit ELF. Provided by Matt Thomas, then integrated
and cleaned up a fair bit.

- Add definitions for Elf section headers.
- Add prototype for elf_read_from.
- KNF.

Attempt #2 :-( (where did that tag come from..)

Generic ELF exec include file, like the other exec_xxx include files.

Do not warn for ELF note for FDO packaging metadata
(https://systemd.io/COREDUMP_PACKAGE_METADATA/),
found in userland binaries of Fedora 36.

s/hidding/hiding/

Arrange so that no options COREDUMP and no options PTRACE work together.
Thanks to Paul Goyette for testing.

Rename:
- EM_ALPHA (41) -> EM_OLD_ALPHA
- EM_ALPHA_EXP (36902) -> EM_ALPHA

36902 has been the value in common usage for a long time, and this naming
aligns with binutils, etc.

PR port-alpha/51926

Add defines for MIPS_XHASH (.MIPS.xhash)

MIPS version of GNU_HASH as supported by GNU toolchain.

Register DT_GNU_HASH

GNU-style hash table is supported in GNU and LLVM toolchains.

https://sourceware.org/ml/binutils/2006-10/msg00377.html

branches: 1.164.2;
Sync the e_ident[EI_OSABI] magic values with the ELF specification

source: http://www.sco.com/developers/gabi/latest/ch4.eheader.html

Update e_machine magic values with the ELF specification

source: http://www.sco.com/developers/gabi/latest/ch4.eheader.html

Fix the EM_TILEGX number in the ELF header

source: http://www.sco.com/developers/gabi/latest/ch4.eheader.html

191 is confirmed in LLVM, GNU toolchain, FreeBSD.

deduplicate the elf auxv builder code, welcome to 8.99.43

branches: 1.160.2;
Complete the list of DF_1 flags.

There have never been binaries with a checksum tag for NetBSD.
These days the equivalent functionality is provided by the build
id tag. It was incorrect in the first place to overload tag #2
(which used to be the emulation name) with a different meaning, so
actually document the meaning that was once used (because binaries
with that tag exist).

Tag existance noticed by kre@

branches: 1.158.2;
Since _RUMPKERNEL is really userland code it needs ARCH_ELFSIZE,
not KERN_ELFSIZE, make it so!

Use KERN_ELFSIZE if we are in the kernel.

Add mips abiflags section

Add an ElfW() macro like linux.

Add a note for the auxv array so we can find our load location from a
core file of a PIE binary.

Intel MCU.

Don't do _BSD_SIZE_T_ dance here. Use <stddef.h> in userland.

This file is used in tools, where _BSD_SIZE_T_ is unreliable. In the
kernel, <sys/types.h> will define size_t anyway. Outside the kernel,
<stddef.h> is the standard way to get size_t.

Ignore the ancient 01.01 tag that gnuc used to put in old NetBSD binaries.

Add buildid Go note

Revamp PaX:
- don't confuse between ELF flags and proc flags. Introduce the proc-
specific P_PAX_ASLR, P_PAX_MPROTECT and P_PAX_GUARD flags.
- introduce pax_setup_elf_flags(), which takes as argument the PaX flag
of the ELF PaX note section, and which sets the proc flag as
appropriate. Also introduce a couple of other functions used for that
purpose.
- modify pax_aslr_active(), and all the other similar pieces of code, so
that it checks the proc flag directly, without extra ELF computation

In addition to making PaX clearer, the combination of these changes fixes
the following bug: if a non-PaX'ed process is launched, and then someone
sets security.pax.{aslr,mprotect,segvguard}.global=1, the process becomes
PaX'ed while its address space hasn't been randomized, which is not likely
to be a good idea.

Now, only the proc flag is checked at runtime, which means the process's
PaX status won't be altered during the execution.

Also:
- declare PAX_DPRINTF, makes it more readable
- fix a typo in exec_elf.h

Remove unofficial AVR32

Add a few more EM_*

add struct exec_package;

branches: 1.145.2;
Add standard size_t dance

Add EM_RISCV

Add basic support for indirect functions. It allows providing a public
function symbol with an implementation choosen at run time.
Refactor calls to functions by address in ld.elf_so to create temporary
function descriptors on the stack, if the address is not leaked outside.

Limitations:
- no support for initialising static storage with function pointers
- no support for unnamed resolver functions

Inspired by FreeBSD's r228435 by kib@freebsd.org.

EM_OPENRISC is now EM_OR1K but keep a define of EM_OPENRISC to EM_OR1K around.

branches: 1.141.2;
Define ELF_MAXNOTESIZE, ELF_MAXSHNUM and ELF_MAXPHNUM in <sys/exec_elf.h>, so
that it can be used externally.

branches: 1.140.2;
Add more definitions, needed by the new elf toolchain.

Remove the last argument of elf_check_header(). It is easier - and faster - to
check the e_type field in the calling function. Other BSD's already do this.

ok christos@

Fix typo in DF_STATIC_TLS. From Elliott Hughes on tech-userlevel.

Fix bug in previous (panic during process core dump).
Change the interface to ELFNAMEEND(coredump_savenote) so that the caller
doesn't need to know the type of the elf note header.
Simplifies the calling code somewhat.

Instead of generating all the 'note' sections twice (and hoping that the
'fast path' size on the first path matches the actual size on the second)
save all the notes (mostly the cpu registers for all the LWPs) in
malloced memory on the first pass.
Sanity check that the number of memory segments matches written matches
the count obtained earlier. If gcore() is used they could differ.
(Not sure that returning ENOMEM is ideal, but it is better than a crash.)

Remove the incorrect and unused Elf64_Shalf.
At one time it was used for a 32bit item - but that definition has long
since been changed to use a reasonable 32bit type.

Add more gnu sections

Change the type of the 'cookie' that holds the state of the core dump file
from 'void *' to the actual type 'struct coredump_iostate *'.
In most of the code the contents of the structure are still unknown.
This just stops the wrong type of pointer being passed to the 'void *'
parameter.
I hope I've found everything, amd64 GENERIC and i386 GENERIC & ALL compile.

Define a new NetBSD specific note to encode the code model used for a binary.
Allow MD code to check these notes in the exec path.

Add EM_AARCH64 from Elliott Hughes

Add DF_* values for DT_FLAGS

Add a NetBSD note variant for recording the machine_arch for which the
executable was built. (currently only used by earm variants)

branches: 1.128.2;
Add the ARM processor-specific sections:
http://infocenter.arm.com/help/topic/com.arm.doc.ihi0044e/IHI0044E_aaelf.pdf

Add DT_PREINIT_ARRAY/DT_PREINIT_ARRAYSZ and few other missing ones.

branches: 1.126.2;
back out elf note changes and use EF_ARM_ABIVERS to determine EABI usage.

define<tab> consistency

Add a elf note to describe the ARM ABI in use. If encounters on arm,
set EXEC_ARM_AAPCS bit in exec_package's ep_flags.
XXX kind of gross but it there's isn't a MD hook for notes so ...

- Recognize the SuSE ABI note.
- Restructure the code to do the checking in the appropriate note type,
and harmonize all the checks to be positive.
- Print only the tag data being careful not to overrun the allocated buffer.

Add AuxInfo record with the base address of the main thread.

Kill duplicate SHF_MASKOS/SHF_MASKPROC defines.

Add more SHF_ and whitespace align SHT_

whitespace diffs [and comments]

branches: 1.118.2;
Add SHT_GNU_HASH.

branches: 1.117.2;
Replace the word `NetBSD' with `SuSE' in the rest of comments
for the cpp macro constatnt ELF_NOTE_SUSE_*.

Add SuSE note

define<tab>

document all the notes we know about in a consistent fashion.
XXX: someone please document the checksum tags.

PR/45015: ld.elf_so: support ELF symbol versioning
Applied latest patch.

fix ELFxx_M_SIZE macro.

fix spelling.

Refactor ps_strings access. Based on PK_32, write either the normal
version or the 32bit compat layout in execve1. Introduce a new function
copyin_psstrings for reading it back from userland and converting it to
the native layout. Refactor procfs to share most of the code with the
kern.proc_args sysctl handler.

This material is based upon work partially supported by
The NetBSD Foundation under a contract with Joerg Sonnenberger.

add the rest of the gnu specific pt sections.

branches: 1.108.2; 1.108.4;
Add support for DF_1_BIND_NOW, DF_1_NODELETE and DF_1_NOOPEN marked
objects, and the RTLD_NODELETE and RTLD_NOLOAD flags to dlopen(3).

Mark libpthread as DF_1_NOOPEN and use it to test the functionality.

Somewhat taken from FreeBSD.

Fixes PR 42029.

OK from christos and joerg.

Add PT_GNU_EH_FRAME definition

Implement dl_iterate_phdr.

Somewhat taken from FreeBSD. Manual page from OpenBSD.

Define PT_TLS.

Introduce a new type Elf_Symindx for use in decoding the symbol hash table
section and allow this type to be overridden.

The ELF specification says it should always be uint32_t (Elf_Word), but
alpha decided to be different (not sure why). Define Elf_Symindx to be
uint64_t on alpha.

Alpha no longer uses non-standard definitions of Elf64_Sword and
Elf64_Word. Remove the ability to override these types.

Fixes ld.elf_so after Herculean effort from me and martin.

Replace the current usage of Elf64_Half with Elf64_Word and rename
NetBSD specific Elf64_Quarter to Elf64_Half. This restores compatibility
with the common ELF specifications.

branches: 1.102.2;
Use #define<tab> consistently.

DTrace: fix a few problems introduced by the recent ZFS update (expected
problems from the OpenSolaris merge).

branches: 1.100.2;
Add defns for Elf{32,64}_Syminfo.

Add some definitions, from FreeBSD.

fix comments: st_name is index in .strtab

Add a signed offset type

branches: 1.96.4;
Remove LKMs and switch to the module framework, pass 1.

Proposed on tech-kern@.

branches: 1.95.6; 1.95.8; 1.95.10; 1.95.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.94.6; 1.94.8; 1.94.10;
Add PaX ASLR (Address Space Layout Randomization) [from elad and myself]

For regular (non PIE) executables randomization is enabled for:
1. The data segment
2. The stack

For PIE executables(*) randomization is enabled for:
1. The program itself
2. All shared libraries
3. The data segment
4. The stack

(*) To generate a PIE executable:
- compile everything with -fPIC
- link with -shared-libgcc -Wl,-pie

This feature is experimental, and might change. To use selectively add
options PAX_ASLR=0
in your kernel.

Currently we are using 12 bits for the stack, program, and data segment and
16 or 24 bits for mmap, depending on __LP64__.

branches: 1.93.4;
rtld changes for PR bin/37347:

- Apply patch from J.T. Conklin to execute .init/.fini functions in order.
- Support DF_1_INITFIRST and mark libc with DF_1_INITFIRST. Shared libs
should be recording a dependency on libc, but it's too late to do that.

Ok christos@.

branches: 1.92.2;
- add an elf aux vector entry for implementing $ORIGIN.
- the code to convert from a vnode to a path is commented out now until
a better solution is implemented. Only absolute paths work for now
(which is most of the cases).

requested by core

branches: 1.91.2; 1.91.8;
Add EM_AVR32

branches: 1.90.2; 1.90.6;
Use an elf note to handle pax arguments. This is a temporary solution to
avoid wasting OS flag bits. In the future we'll probably use fileassoc to
achieve this (once there is a way to make fileassoc persistent) or in the
shorter term libelf, so that we can add and remove the note on demand instead
of burning bits on each binary. Of course since this is a tool, this means
that we'll need to think about how to handle libelf...

branches: 1.89.2; 1.89.8; 1.89.10; 1.89.12; 1.89.16;
Make sys/exec_elf.h and machine/elf_machdep.h available to host
tools, and allow paxctl to use them when compiled as a host
tool. This fixes the build on MacOS X (and perhaps other platforms)

Damn, missed in previous commit. Add PaX Segvguard flags.

branches: 1.87.2; 1.87.4;
Regen.

Introduce PaX MPROTECT -- mprotect(2) restrictions used to strengthen
W^X mappings.

Disabled by default.

First proposed in:

http://mail-index.netbsd.org/tech-security/2005/12/18/0000.html

More information in:

http://pax.grsecurity.net/docs/mprotect.txt

Read relevant parts of options(4) and sysctl(3) before using!

Lots of thanks to the PaX author and Matt Thomas.

branches: 1.85.4; 1.85.6; 1.85.8; 1.85.12;
there is no elf64_read_from.

merge ktrace-lwp.

Add NetBSD checksum notes.

branches: 1.82.2;
Rework the coredump code to have no explicit knownledge of how coredump
i/o is done. Instead, pass an opaque cookie which is then passed to a
new routine, coredump_write, which does the actual i/o. This allows the
method of doing i/o to change without affecting any future MD code.
Also, make netbsd32_core.c [re]use core_netbsd.c (in a similar manner that
core_elf64.c uses core_elf32.c) and eliminate that code duplication.
cpu_coredump{,32} is now called twice, first with a NULL iocookie to fill
the core structure and a second to actually write md parts of the coredump.
All i/o is nolonger random access and is suitable for shipping over a stream.

nuke trailing whitespace

de-__P

branches: 1.79.2; 1.79.4;
Use uint32_t casts, rather than Elf_Byte, in ELF_ST_BIND() and ELF_ST_TYPE()
so we don't get screwed by promotion rules (and makes nlist_elf32.ln build
again).

According to the ELF docs, ELF{32,64}_ST_{BIND,TYPE,INFO,VISIBILITY} are
parts of the public interface. Define them in terms of ELF_ST_*.
Add STN_UNDEF which is part of the public interface also.

Remove ELFn_ST_{BIND,TYPE,INFO}(), and instead add a generic ELF_ST_*() that
works with an Elf_Byte. Also add ELF_ST_VISIBILITY() and STV_* types.

add EI_NIDENT (16) as seen in other <elf.h> headers.

Uppercase CPU, plural is CPUs.

fix typo in comment: magic numbers are indexed EI_MAG0 through EI_MAG3,
not ELFMAG0 through ELFMAG3 (which are the contents)

introduce another "magic" ELF address to tell that the file has to be
mapped at the address it was linked for

- Add execsw es_setup_stack
- gc all the rest of the setup stack functions.

Make elf{32|64}_check_header public, as it will be used by irix_elf32_probe.
While we are there, cut to 80 chars, and ANSIfy prototypes

branches: 1.70.2;
Back out the lwp/ktrace changes. They contained a lot of colateral damage,
and need to be examined and discussed more.

Pass lwp pointers throughtout the kernel, as required, so that the lwpid can
be inserted into ktrace records. The general change has been to replace
"struct proc *" with "struct lwp *" in various function prototypes, pass
the lwp through and use l_proc to get the process pointer when needed.

Bump the kernel rev up to 1.6V

Add a field to netbsd_elfcore_procinfo that identifies the LWP targeted
by the killing signal.

Merge the nathanw_sa branch.

Fix typo in comment.

Program header types 0x60000000 ~ 0x6fffffff are reserved for
OS-specific semantics.

Update URLs again (serves me right for committing an old change
without checking it). Back to sco.com, but with a different path.

Update URLs for ELF spec (sco.com -> caldera.com).

- pass struct proc to copyargs
- rename aux vector entries AT_{R,E}{U,G}ID from AT_SYUN_{R,}P{U,G}ID
- update new aux vector size.

Add official machine numbers for PDP10 and PDP11.

branches: 1.60.8;
Define ELF32_EHDR_FLAGS_OK()/ELF64_EHDR_FLAGS_OK() and use it
as an added measure to make sure that we can execute a binary.
These default to (1) if elf_machdep.h does not override them.

On Sun2, ELF32_EHDR_FLAGS_OK() checks for the presense of EF_M68000,
since the 68010 cannot run binaries for the 68020-and-up.

Add support for dumping ELF-cormat core files.

Rather than using a NetBSD-specific header file <machine/int_types.h>
and non-standard inttype-like types, pull in <sys/types.h> if
_KERNEL or _STANDALONE and <inttypes.h> otherwise, and use standard
inttype types.

Discussed with and OK'd by Christos.

Use <machine/elf_machdep.h> to override Elf64_Word and Elf64_Sword
on the Alpha (which uses a non-standard definitions for historical
reasons), rather than #ifdef __alpha__ (which is not cross-tool safe).

Include <machine/elf_machdep.h> earlier, at the beginning of the file,
after the inttypes header.

Add EM_NS32K for the ns32k series CPUs.

branches: 1.54.4;
make elf64_copyargs() of type int also per the recent copyargs changes

adjust to the new copyargs footprint

- add exec_read_from and make exec_elf32 use it.
- add a macho probe function

branches: 1.51.2;
Fix up comment botch.
_Never_ fix a simple formatting thing _after_ a test compile.

Add some more e_machine values and a reference to the SCO website with
ELF ABI information.

X8664 -> X86_64

Elf64_Word should be 32 bits; alpha is the exception rather than
the rule, so change the ifdef to reflect this.

Fix botched commit; I forgot to bring this up to date to -current
before committing it.

Add IDs for x86-64.

ElfNN_RelA -> ElfNN_Rela, which I missed in the Great Renaming;
noticed by Juan Jose Garcia-Ripoll in PR standards/12274.

Fix misinformation in comment

Slight readability improvement to rev. 1.39.

Describe possible section types with a few words.

populate OSABI list

branches: 1.40.2;
Tighten up the ELF signature checks, and actually look for the ABI tag added
in newer glibc versions.

Introduce Elf64_Xword and Elf64_Sxword, and use them in various places instead
of Elf64_Word.
Make Elf64_Word be 32 bits on sparc64. (XXXX Need to determine what it's
supposed to be on other, but it's current used to size entries in DT_HASH,
which is 32 bits on sparc64 and 64 bits on alpha.)

Be less invasive and use private names for exact-width integer types.

branches: 1.37.4;
Make this file possible to use again.

* Add more OS ABI constants and dynamic array tags from D1998-04-29.
* Enhance readibility: define data structures first, followed member value
constants, and add more comments.

Add OS ABI constants from the 1998-04-29 draft.

Delint.

Make sure the elf_args strucure is appropriate for the size of the ELF binary.

Define ELF_ST_*() to ELF{32,64}_ST_* as appropriate. They evaluate to
the same thing, but this provides consistency with everything else.

Fix comment typo introduced in rev. 1.27.

Fix obvious typos.

Add a few comments from <mips/elf.h>.

Add ELF64 versions of st_info manipulation macros I missed in previous.

As discussed with Christos, rename the definitions in this file to match those
found in various SVR4 ABIs.

Move ELFNAME macros out of #ifdef _KERNEL.

branches: 1.25.2; 1.25.4; 1.25.6;
Merge in NetBSD/sh3 from cvs.kame.net repository.

Tree structure:
- sys/arch/sh3: sh3 generic code
As commented, in-chip device drivers are put into sys/arch/sh3/dev.
- sys/arch/evbsh3: sh3 evaluation boards (pure sh3 CPU, no fancy external HW)
- sys/arch/mmeye: Brains mmEye, www.brains.co.jp
MI source code includes couple of #ifdef for sh3-coff support.
(sh3 uses coff or elf)

Needs some more improvements, especialy in sys/arch/sh3/conf/files.sh3,
to compile the tree (due to last minute tree structure change).

Add the newly blessed magic number for VAX.

Add NetBSD note constants.

Add a few new machine numbers and aux vectors.

Move elf function name building macros here (ELFNAME, ELFNAME2, etc...) Add defines related to elf note sections for use in determining executable type/emulation.

Fix copyright typos...

Assign copyright to TNF.

Oops, change previous to use the registered name found in the System V ABI,
draft edition as of April 29, 1998.

Add machine identifier for the PA.

Add machine identifier for ARM. Add (commented out) real ident
for Alpha as well.

add Elf_em_sparc64 at 43 (from solaris 2.7)

Remove a trailing ',' in an enum declaration.

Addes support for sparc32plus ELF executables: v9 in v8's clothing.

Generate dependencies on the exec format options EXEC_AOUT, EXEC_ECOFF,
EXEC_ELF32, EXEC_ELF64, and EXEC_SCRIPT.

Add Section Attribute Flags (SHF_*) as defined in _System V Application
Binary Interface, Third Edition_, section 4, page 16.

define Elf{,32,64}_Note as appropriate, add a constant for NetBSD-specific
notes' 'name' fields, and add a couple of NetBSD-specific note types.

rename the ELF section type constants to be Elf_* rather than Elf32_*,
since they're shared between ELF32 and ELF64. (Thought i checked this in
earlier, but apparently didn't!)

fix typo in definition of Elf32_e_ident

PowerPC port

fix up section header type enum/values

add and use a machine-dependent header, which currently defines some
macros to use to remove #ifdefs from the machine ID case check.
Eventually, these headers will contain other information, e.g.
machine-dependent relocation information, etc.

add definitions for 64-bit ELF. Provided by Matt Thomas, then integrated
and cleaned up a fair bit.

- Add definitions for Elf section headers.
- Add prototype for elf_read_from.
- KNF.

Attempt #2 :-( (where did that tag come from..)

Generic ELF exec include file, like the other exec_xxx include files.

s/hidding/hiding/

Arrange so that no options COREDUMP and no options PTRACE work together.
Thanks to Paul Goyette for testing.

Rename:
- EM_ALPHA (41) -> EM_OLD_ALPHA
- EM_ALPHA_EXP (36902) -> EM_ALPHA

36902 has been the value in common usage for a long time, and this naming
aligns with binutils, etc.

PR port-alpha/51926

Add defines for MIPS_XHASH (.MIPS.xhash)

MIPS version of GNU_HASH as supported by GNU toolchain.

Register DT_GNU_HASH

GNU-style hash table is supported in GNU and LLVM toolchains.

https://sourceware.org/ml/binutils/2006-10/msg00377.html

branches: 1.164.2;
Sync the e_ident[EI_OSABI] magic values with the ELF specification

source: http://www.sco.com/developers/gabi/latest/ch4.eheader.html

Update e_machine magic values with the ELF specification

source: http://www.sco.com/developers/gabi/latest/ch4.eheader.html

Fix the EM_TILEGX number in the ELF header

source: http://www.sco.com/developers/gabi/latest/ch4.eheader.html

191 is confirmed in LLVM, GNU toolchain, FreeBSD.

deduplicate the elf auxv builder code, welcome to 8.99.43

branches: 1.160.2;
Complete the list of DF_1 flags.

There have never been binaries with a checksum tag for NetBSD.
These days the equivalent functionality is provided by the build
id tag. It was incorrect in the first place to overload tag #2
(which used to be the emulation name) with a different meaning, so
actually document the meaning that was once used (because binaries
with that tag exist).

Tag existance noticed by kre@

branches: 1.158.2;
Since _RUMPKERNEL is really userland code it needs ARCH_ELFSIZE,
not KERN_ELFSIZE, make it so!

Use KERN_ELFSIZE if we are in the kernel.

Add mips abiflags section

Add an ElfW() macro like linux.

Add a note for the auxv array so we can find our load location from a
core file of a PIE binary.

Intel MCU.

Don't do _BSD_SIZE_T_ dance here. Use <stddef.h> in userland.

This file is used in tools, where _BSD_SIZE_T_ is unreliable. In the
kernel, <sys/types.h> will define size_t anyway. Outside the kernel,
<stddef.h> is the standard way to get size_t.

Ignore the ancient 01.01 tag that gnuc used to put in old NetBSD binaries.

Add buildid Go note

Revamp PaX:
- don't confuse between ELF flags and proc flags. Introduce the proc-
specific P_PAX_ASLR, P_PAX_MPROTECT and P_PAX_GUARD flags.
- introduce pax_setup_elf_flags(), which takes as argument the PaX flag
of the ELF PaX note section, and which sets the proc flag as
appropriate. Also introduce a couple of other functions used for that
purpose.
- modify pax_aslr_active(), and all the other similar pieces of code, so
that it checks the proc flag directly, without extra ELF computation

In addition to making PaX clearer, the combination of these changes fixes
the following bug: if a non-PaX'ed process is launched, and then someone
sets security.pax.{aslr,mprotect,segvguard}.global=1, the process becomes
PaX'ed while its address space hasn't been randomized, which is not likely
to be a good idea.

Now, only the proc flag is checked at runtime, which means the process's
PaX status won't be altered during the execution.

Also:
- declare PAX_DPRINTF, makes it more readable
- fix a typo in exec_elf.h

Remove unofficial AVR32

Add a few more EM_*

add struct exec_package;

branches: 1.145.2;
Add standard size_t dance

Add EM_RISCV

Add basic support for indirect functions. It allows providing a public
function symbol with an implementation choosen at run time.
Refactor calls to functions by address in ld.elf_so to create temporary
function descriptors on the stack, if the address is not leaked outside.

Limitations:
- no support for initialising static storage with function pointers
- no support for unnamed resolver functions

Inspired by FreeBSD's r228435 by kib@freebsd.org.

EM_OPENRISC is now EM_OR1K but keep a define of EM_OPENRISC to EM_OR1K around.

branches: 1.141.2;
Define ELF_MAXNOTESIZE, ELF_MAXSHNUM and ELF_MAXPHNUM in <sys/exec_elf.h>, so
that it can be used externally.

branches: 1.140.2;
Add more definitions, needed by the new elf toolchain.

Remove the last argument of elf_check_header(). It is easier - and faster - to
check the e_type field in the calling function. Other BSD's already do this.

ok christos@

Fix typo in DF_STATIC_TLS. From Elliott Hughes on tech-userlevel.

Fix bug in previous (panic during process core dump).
Change the interface to ELFNAMEEND(coredump_savenote) so that the caller
doesn't need to know the type of the elf note header.
Simplifies the calling code somewhat.

Instead of generating all the 'note' sections twice (and hoping that the
'fast path' size on the first path matches the actual size on the second)
save all the notes (mostly the cpu registers for all the LWPs) in
malloced memory on the first pass.
Sanity check that the number of memory segments matches written matches
the count obtained earlier. If gcore() is used they could differ.
(Not sure that returning ENOMEM is ideal, but it is better than a crash.)

Remove the incorrect and unused Elf64_Shalf.
At one time it was used for a 32bit item - but that definition has long
since been changed to use a reasonable 32bit type.

Add more gnu sections

Change the type of the 'cookie' that holds the state of the core dump file
from 'void *' to the actual type 'struct coredump_iostate *'.
In most of the code the contents of the structure are still unknown.
This just stops the wrong type of pointer being passed to the 'void *'
parameter.
I hope I've found everything, amd64 GENERIC and i386 GENERIC & ALL compile.

Define a new NetBSD specific note to encode the code model used for a binary.
Allow MD code to check these notes in the exec path.

Add EM_AARCH64 from Elliott Hughes

Add DF_* values for DT_FLAGS

Add a NetBSD note variant for recording the machine_arch for which the
executable was built. (currently only used by earm variants)

branches: 1.128.2;
Add the ARM processor-specific sections:
http://infocenter.arm.com/help/topic/com.arm.doc.ihi0044e/IHI0044E_aaelf.pdf

Add DT_PREINIT_ARRAY/DT_PREINIT_ARRAYSZ and few other missing ones.

branches: 1.126.2;
back out elf note changes and use EF_ARM_ABIVERS to determine EABI usage.

define<tab> consistency

Add a elf note to describe the ARM ABI in use. If encounters on arm,
set EXEC_ARM_AAPCS bit in exec_package's ep_flags.
XXX kind of gross but it there's isn't a MD hook for notes so ...

- Recognize the SuSE ABI note.
- Restructure the code to do the checking in the appropriate note type,
and harmonize all the checks to be positive.
- Print only the tag data being careful not to overrun the allocated buffer.

Add AuxInfo record with the base address of the main thread.

Kill duplicate SHF_MASKOS/SHF_MASKPROC defines.

Add more SHF_ and whitespace align SHT_

whitespace diffs [and comments]

branches: 1.118.2;
Add SHT_GNU_HASH.

branches: 1.117.2;
Replace the word `NetBSD' with `SuSE' in the rest of comments
for the cpp macro constatnt ELF_NOTE_SUSE_*.

Add SuSE note

define<tab>

document all the notes we know about in a consistent fashion.
XXX: someone please document the checksum tags.

PR/45015: ld.elf_so: support ELF symbol versioning
Applied latest patch.

fix ELFxx_M_SIZE macro.

fix spelling.

Refactor ps_strings access. Based on PK_32, write either the normal
version or the 32bit compat layout in execve1. Introduce a new function
copyin_psstrings for reading it back from userland and converting it to
the native layout. Refactor procfs to share most of the code with the
kern.proc_args sysctl handler.

This material is based upon work partially supported by
The NetBSD Foundation under a contract with Joerg Sonnenberger.

add the rest of the gnu specific pt sections.

branches: 1.108.2; 1.108.4;
Add support for DF_1_BIND_NOW, DF_1_NODELETE and DF_1_NOOPEN marked
objects, and the RTLD_NODELETE and RTLD_NOLOAD flags to dlopen(3).

Mark libpthread as DF_1_NOOPEN and use it to test the functionality.

Somewhat taken from FreeBSD.

Fixes PR 42029.

OK from christos and joerg.

Add PT_GNU_EH_FRAME definition

Implement dl_iterate_phdr.

Somewhat taken from FreeBSD. Manual page from OpenBSD.

Define PT_TLS.

Introduce a new type Elf_Symindx for use in decoding the symbol hash table
section and allow this type to be overridden.

The ELF specification says it should always be uint32_t (Elf_Word), but
alpha decided to be different (not sure why). Define Elf_Symindx to be
uint64_t on alpha.

Alpha no longer uses non-standard definitions of Elf64_Sword and
Elf64_Word. Remove the ability to override these types.

Fixes ld.elf_so after Herculean effort from me and martin.

Replace the current usage of Elf64_Half with Elf64_Word and rename
NetBSD specific Elf64_Quarter to Elf64_Half. This restores compatibility
with the common ELF specifications.

branches: 1.102.2;
Use #define<tab> consistently.

DTrace: fix a few problems introduced by the recent ZFS update (expected
problems from the OpenSolaris merge).

branches: 1.100.2;
Add defns for Elf{32,64}_Syminfo.

Add some definitions, from FreeBSD.

fix comments: st_name is index in .strtab

Add a signed offset type

branches: 1.96.4;
Remove LKMs and switch to the module framework, pass 1.

Proposed on tech-kern@.

branches: 1.95.6; 1.95.8; 1.95.10; 1.95.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.94.6; 1.94.8; 1.94.10;
Add PaX ASLR (Address Space Layout Randomization) [from elad and myself]

For regular (non PIE) executables randomization is enabled for:
1. The data segment
2. The stack

For PIE executables(*) randomization is enabled for:
1. The program itself
2. All shared libraries
3. The data segment
4. The stack

(*) To generate a PIE executable:
- compile everything with -fPIC
- link with -shared-libgcc -Wl,-pie

This feature is experimental, and might change. To use selectively add
options PAX_ASLR=0
in your kernel.

Currently we are using 12 bits for the stack, program, and data segment and
16 or 24 bits for mmap, depending on __LP64__.

branches: 1.93.4;
rtld changes for PR bin/37347:

- Apply patch from J.T. Conklin to execute .init/.fini functions in order.
- Support DF_1_INITFIRST and mark libc with DF_1_INITFIRST. Shared libs
should be recording a dependency on libc, but it's too late to do that.

Ok christos@.

branches: 1.92.2;
- add an elf aux vector entry for implementing $ORIGIN.
- the code to convert from a vnode to a path is commented out now until
a better solution is implemented. Only absolute paths work for now
(which is most of the cases).

requested by core

branches: 1.91.2; 1.91.8;
Add EM_AVR32

branches: 1.90.2; 1.90.6;
Use an elf note to handle pax arguments. This is a temporary solution to
avoid wasting OS flag bits. In the future we'll probably use fileassoc to
achieve this (once there is a way to make fileassoc persistent) or in the
shorter term libelf, so that we can add and remove the note on demand instead
of burning bits on each binary. Of course since this is a tool, this means
that we'll need to think about how to handle libelf...

branches: 1.89.2; 1.89.8; 1.89.10; 1.89.12; 1.89.16;
Make sys/exec_elf.h and machine/elf_machdep.h available to host
tools, and allow paxctl to use them when compiled as a host
tool. This fixes the build on MacOS X (and perhaps other platforms)

Damn, missed in previous commit. Add PaX Segvguard flags.

branches: 1.87.2; 1.87.4;
Regen.

Introduce PaX MPROTECT -- mprotect(2) restrictions used to strengthen
W^X mappings.

Disabled by default.

First proposed in:

http://mail-index.netbsd.org/tech-security/2005/12/18/0000.html

More information in:

http://pax.grsecurity.net/docs/mprotect.txt

Read relevant parts of options(4) and sysctl(3) before using!

Lots of thanks to the PaX author and Matt Thomas.

branches: 1.85.4; 1.85.6; 1.85.8; 1.85.12;
there is no elf64_read_from.