As per the discussion with Ingo, here is an updated patchset on top of tip/asm
to clean up the syscall stub naming convention. Changes to v1 include renames
(__do_ becomes __se_ for sign-extending, __in_ becomes __do_) and more
comments.
For the generic case, we now will have:
t kernel_waitid # common C function (see kernel/exit.c)
i __do_sys_waitid # inlined helper doing the actual work
# (takes original parameters as declared)
T __se_sys_waitid # sign-extending C function calling inlined
# helper (takes parameters of type long; casts
# them to the declared type)
i __do_compat_sys_waitid # inlined helper doing the actual work
# (takes parameters as declared)
T __se_compat_sys_waitid # sign-extending C function calling inlined
# helper (takes parameters of type long, casts
# them to unsigned long and then to the
# declared type)
T sys_waitid # alias to __se_sys_waitid() (taking
# parameters as declared), to be included
# in syscall table
T compat_sys_waitid # alias to __se_compat_sys_waitid()
# (taking parameters as declared), to
# be included in syscall table
For 64-bit x86, kernel_waitid, __do_sys_waitid and __se_sys_waitid are the
same. But instead of sys_waitid and compat_sys_waitid, there are:
T __x64_sys_waitid # x86 64-bit-ptregs -> C stub, calls
# __se_sys_waitid(); to be included in
# syscall table
T __ia32_sys_waitid # IA32_EMULATION 32-bit-ptregs -> C stub,
# calls __se_sys_waitid(); to be included
# in syscall table unless there is a
# compat syscall stub [in that case, it is
# unused]
T __ia32_compat_sys_waitid # IA32_EMULATION 32-bit-ptregs -> C stub,
# calls __se_compat_sys_waitid(); to be
# included in syscall table
T __x32_compat_sys_waitid # x32 64-bit-ptregs -> C stub, calls
# __se_compat_sys_waitid(); to be included
# in syscall table
In short (0xffffffff prefix removed, re-ordered):
810f0af0 t kernel_waitid # common (32/64) kernel helper
<inline> __do_sys_waitid # inlined helper doing actual work
810f0be0 t __se_sys_waitid # C func calling inlined helper
<inline> __do_compat_sys_waitid # inlined helper doing actual work
810f0d80 t __se_compat_sys_waitid # compat C func calling inlined helper
810f2080 T __x64_sys_waitid # x64 64-bit-ptregs -> C stub
810f20b0 T __ia32_sys_waitid # ia32 32-bit-ptregs -> C stub [unused]
810f2470 T __ia32_compat_sys_waitid # ia32 32-bit-ptregs -> compat C stub
810f2490 T __x32_compat_sys_waitid # x32 64-bit-ptregs -> compat C stub
The kbuild test robot barked at an alleged +20038 bytes kernel size regression
for i386-tinyconfig due to the first patch of this series. That seems to be a
false positive, as it likely doesn't take into account the change to
scripts/bloat-o-meter. Moreover, I could not reproduce such a size regression
on local i386 builds.
Thanks,
Dominik
Dominik Brodowski (4):
syscalls: clean up syscall stub naming convention
syscalls: clean up compat syscall stub naming convention
syscalls: rename struct pt_regs-based sys_*() to __x64_sys_*()
syscalls/x86: adapt syscall_wrapper.h to the new syscall stub naming
convention
Documentation/process/adding-syscalls.rst | 4 +-
arch/x86/entry/syscalls/syscall_32.tbl | 720 +++++++++++-----------
arch/x86/entry/syscalls/syscall_64.tbl | 710 ++++++++++-----------
arch/x86/entry/syscalls/syscalltbl.sh | 14 +-
arch/x86/entry/vsyscall/vsyscall_64.c | 6 +-
arch/x86/include/asm/syscall_wrapper.h | 156 +++--
arch/x86/include/asm/syscalls.h | 2 +-
include/linux/compat.h | 29 +-
include/linux/syscalls.h | 17 +-
scripts/bloat-o-meter | 4 +-
10 files changed, 858 insertions(+), 804 deletions(-)
--
2.17.0
