linux: implement pkey_alloc, pkey_free and pkey_mprotect

Bridge the Linux memory protection key syscalls to FreeBSD's native
MPK support instead of returning ENOSYS. Modern Linux software
probes the

linux: implement pkey_alloc, pkey_free and pkey_mprotect

Bridge the Linux memory protection key syscalls to FreeBSD's native
MPK support instead of returning ENOSYS. Modern Linux software
probes these at startup: Chromium-based browsers (found via
www/linux-brave) use protection keys for V8's heap and JIT
sandboxing, and glibc >= 2.27 exposes the full API.

pkey_alloc() allocates from a per-process bitmap kept in the process
emuldata (key 0 implicitly allocated, matching Linux's
mm_pkey_allocation_map; ENOSPC once keys 1..15 are exhausted or when
PKU is absent, as Linux returns on such hardware) and applies the
requested initial access rights to the calling thread's PKRU, located
in the XSAVE area via xsave_area_offset(). pkey_free() is
bookkeeping only: as on Linux, freeing neither untags pages nor
updates PKRU. pkey_mprotect() performs the protection change and
tags the range through amd64_pkru_update(), factored out of
sysarch(2)'s AMD64_SET_PKRU/AMD64_CLEAR_PKRU implementation so that
both share the same argument checking and map read lock
synchronization with a parallel pmap_vmspace_copy() on fork; tags die
with the mapping, matching Linux VMA semantics. A pkey of -1
degrades to plain mprotect.

The allocation map is inherited on fork and reset on exec. At exec
the Linux sysvecs initialize PKRU to 0x55555554, Linux's init_pkru
default (access disabled for keys 1..15), so memory tagged with a
not yet allocated key is inaccessible to threads that were never
granted rights -- the property V8's thread isolation relies on.
Setting PKRU at exec initializes the user FPU state slightly earlier
than the lazy first-use path; the state would be initialized moments
later in rtld/libc startup regardless. Protection key faults
already deliver SEGV_PKUERR through the existing siginfo
translation.

The common code carries no architecture ifdefs. Machine-dependent
state lives in struct linux_pemuldata_md, embedded in the process
emuldata in the manner of struct mdthread, and common code calls
per-arch lifecycle hooks (linux_pemuldata_init_md/_exec_md) and pkey
back ends after performing the parameter validation Linux applies
regardless of hardware support. On amd64 the implementation lives
in sys/amd64/linux/linux_pkru.c, compiled into linux_common and
serving both the 64-bit and 32-bit Linux ABIs. Elsewhere (arm64,
i386) linux_emul_md.c provides stubs returning what Linux returns on
hardware without protection keys (ENOSPC from pkey_alloc;
pkey_mprotect with a pkey of -1 acts as plain mprotect), so
applications take their normal no-PKU fallback instead of the ENOSYS
path.

PR: 297427
MFC after: 1 month
Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D58782

show more ...

linux: Add required symbol to EXPORT_SYMS list

PR: 291270

Revert "modules: Add missing opt_*.h files for stand-alone compile"

This reverts commit 1a37caeb076b9d31e13c54691d7f1eeb589798bb.

I'd overlooked 06c07e1203324bfe8ab4526c82dbb750864272d5 when I upda

Revert "modules: Add missing opt_*.h files for stand-alone compile"

This reverts commit 1a37caeb076b9d31e13c54691d7f1eeb589798bb.

I'd overlooked 06c07e1203324bfe8ab4526c82dbb750864272d5 when I updated a
branch I'd previously committed but not updated. My search for this in
my tree before the commit was to a stale file...

Sponsored by: Netflix

show more ...

modules: Add missing opt_*.h files for stand-alone compile

Standalone compile that we at least create these opt_*.h files.

Sponsored by: Netflix

Remove residual blank line at start of Makefile

This is a residual of the $FreeBSD$ removal.

MFC After: 3 days (though I'll just run the command on the branches)
Sponsored by: Netflix

linux: Make module standalone-buildable

Add opt_inet.h and opt_usb.h to make linux module buildable standalone.

Sponsored by: Netflix

sanitizers: Avoid building genassym.c and genoffset.c with sanitizers

Some, particularly KASAN, may insert redzones around global symbols,
resulting in incorrect offset definitions because genassym.

sanitizers: Avoid building genassym.c and genoffset.c with sanitizers

Some, particularly KASAN, may insert redzones around global symbols,
resulting in incorrect offset definitions because genassym.sh (ab)uses
symbol sizes to assign semantic meaning.

(Ideally I would be able to define this pattern in one place, but I
haven't found a way to define a GENSYM_CFLAGS that actually works for
all of the consumers (kern.post.mk, kmod.mk, sys/conf/files*).)

MFC after: 1 week
Sponsored by: Klara, Inc.
Sponsored by: Juniper Networks, Inc.

show more ...

sys: ${CFLAGS:N-flto} -> ${CFLAGS:N-flto*}

For the same reason as the original https://reviews.freebsd.org/D9659:
-flto=<N>, -flto=full, and -flto=thin also produce the GIMPLE/bitcode
which is not s

sys: ${CFLAGS:N-flto} -> ${CFLAGS:N-flto*}

For the same reason as the original https://reviews.freebsd.org/D9659:
-flto=<N>, -flto=full, and -flto=thin also produce the GIMPLE/bitcode
which is not supported by genassym, so filter those out as well.

Signed-off-by: Alex Xu (Hello71) <alex_y_xu@yahoo.ca>
Reviewed by: imp
Pull Request: https://github.com/freebsd/freebsd-src/pull/898

show more ...

netlink: move NETLINK define to opt_global.h

Move the NETLINK define into opt_global.h so we can rely on it being
set correctly, without having to remember to include opt_netlink.h.
This ensures tha

netlink: move NETLINK define to opt_global.h

Move the NETLINK define into opt_global.h so we can rely on it being
set correctly, without having to remember to include opt_netlink.h.
This ensures that the NETLINK define is correctly set. If not we
may end up with unloadable modules, due to missing symbols (such as
nlmsg_get_group_writer).

PR: 274306
Reviewed by: imp, markj
MFC after: 3 days
Differential Revision: https://reviews.freebsd.org/D42179

show more ...

sys: Remove $FreeBSD$: one-line sh pattern

Remove /^\s*#[#!]?\s*\$FreeBSD\$.*$\n/

linux(4): Implement xattr syscalls

Reviewed by:
Differential revision: https://reviews.freebsd.org/D35544
MFC after: 1 month

linux(4): Make linux_emul_path private

There is no need to refer to linux_emul_path from outside anymore.

MFC after: 2 month

linux(4): Fix opt_netlink.h inclusion

Add opt_netlink.h to the linux_common module, on i386, where we don't
uses linux_common module, move opt_netlink.h inclusion under
i386 condition.

MFC after:

linux(4): Fix opt_netlink.h inclusion

Add opt_netlink.h to the linux_common module, on i386, where we don't
uses linux_common module, move opt_netlink.h inclusion under
i386 condition.

MFC after: 2 weeks

show more ...

linux(4): Move inclusion of i386-specific files under common condition

netlink: fix linux module build w/ netlink.

Reported by: Marek Zarychta <zarychtam@plan-b.pwste.edu.pl>
MFC after: 2 weeks

linux(4): Trim unused opt_usb.h from modules Makefiles

MFC after: 2 weeks

linux(4): Get rid of the opt_compat.h include.

Since e013e369 COMPAT_LINUX, COMPAT_LINUX32 build options are removed,
so include of opt_compat.h is no more needed.

MFC after: 2 weeks

linux(4): Attach netlink on i386.

Discussed with: melifaro
MFC after: 3 days

linux(4): Add coredump support to i386.

MFC after: 1 week

linux64: improve linux_support.s make rules

Previously we relied on the .s.o rule in share/mk/bsd.suffixes.mk to
tell make that linux_support.o is built from linux_support.s, even
though we do not u

linux64: improve linux_support.s make rules

Previously we relied on the .s.o rule in share/mk/bsd.suffixes.mk to
tell make that linux_support.o is built from linux_support.s, even
though we do not use the .s.o rule to assemble it.

Reviewed by: sjg
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D35864

show more ...

sys: use .S for assembly language files that use the preprocessor

Reviewed by: imp
Pull Request: https://github.com/freebsd/freebsd-src/pull/609
Differential Revision: https://reviews.freebsd.org/D3

sys: use .S for assembly language files that use the preprocessor

Reviewed by: imp
Pull Request: https://github.com/freebsd/freebsd-src/pull/609
Differential Revision: https://reviews.freebsd.org/D35908

show more ...

modules: Add missing opt_*.h files for stand-alone compile

Standalone compile that we at least create these opt_*.h files.

Sponsored by: Netflix

linux, linux64: fix module load

The previous commit added references to to the syscallnames arrays, but
failed to add the relevant source files to the module build. Thus, the
modules failed to load

linux, linux64: fix module load

The previous commit added references to to the syscallnames arrays, but
failed to add the relevant source files to the module build. Thus, the
modules failed to load due to missing symbols.

Reported by: cy
Fixes: 1da65dcb1c57 ("linux: populate sv_syscallnames in each sysentvec")
Sponsored by: The FreeBSD Foundation

show more ...

linux, linux64: improve SRCS formatting

Sort the entries alphabetically, and list them with one entry per line.
This makes the diffs much cleaner when adding or removing a new entry,
as I will do in

linux, linux64: improve SRCS formatting

Sort the entries alphabetically, and list them with one entry per line.
This makes the diffs much cleaner when adding or removing a new entry,
as I will do in the next commit.

MFC after: 1 week
Sponsored by: The FreeBSD Foundation

show more ...

linux(4); Move vdso_selector_x86.c to the linux_common module

MFC after: 2 weeks