# git rev-parse -q --verify dc8b1290efbbe974d9968106a03077823d0863d6^{commit}
dc8b1290efbbe974d9968106a03077823d0863d6
already have revision, skipping fetch
# git checkout -q -f -B kisskb dc8b1290efbbe974d9968106a03077823d0863d6
# git clean -qxdf
# < git log -1
# commit dc8b1290efbbe974d9968106a03077823d0863d6
# Author: Russell Currey <ruscur@russell.cc>
# Date:   Fri Mar 1 01:49:17 2019 +1100
# 
#     powerpc/64s: Implement KUAP for Radix MMU
#     
#     Kernel Userspace Access Prevention utilises a feature of the Radix MMU
#     which disallows read and write access to userspace addresses. By
#     utilising this, the kernel is prevented from accessing user data from
#     outside of trusted paths that perform proper safety checks, such as
#     copy_{to/from}_user() and friends.
#     
#     Userspace access is disabled from early boot and is only enabled when
#     performing an operation like copy_{to/from}_user(). The register that
#     controls this (AMR) does not prevent userspace from accessing itself,
#     so there is no need to save and restore when entering and exiting
#     userspace.
#     
#     This feature has a slight performance impact which I roughly measured
#     to be 3% slower in the worst case (performing 1GB of 1 byte
#     read()/write() syscalls), and is gated behind the CONFIG_PPC_KUAP
#     option for performance-critical builds.
#     
#     This feature can be tested by using the lkdtm driver (CONFIG_LKDTM=y)
#     and performing the following:
#     
#       # (echo ACCESS_USERSPACE) > [debugfs]/provoke-crash/DIRECT
#     
#     If enabled, this should send SIGSEGV to the thread.
#     
#     mpe:
#      - Drop the unused paca flags.
#      - Zero the UAMOR to be safe.
#      - Save the AMR when we enter the kernel from the kernel and then
#        block user access again if it's not already blocked.
#      - Restore on the way back to the kernel.
#      - This means we handle nesting of interrupts properly, ie. we are
#        protected inside the page fault handler caused by a user access.
#      - Add paranoid checking of AMR in switch and syscall return.
#      - Add isync()'s around AMR writes as per the ISA.
#      - Support selectively disabling read or write, with no support for
#        nesting.
#     
#     Co-authored-by: Michael Ellerman <mpe@ellerman.id.au>
#     Signed-off-by: Russell Currey <ruscur@russell.cc>
#     Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
#     ---
#     
#     v5:
#      - On kernel entry check if the AMR is already blocking user access
#        and if so don't do the mtspr again (pointed out by Nick).
#      - Rework the constants to make the asm a bit cleaner and avoid any
#        hard coded shifts.
#      - Selectively disable read or write, we don't support nesting and
#        shouldn't need to (famous last words).
# < /opt/cross/kisskb/korg/gcc-5.5.0-nolibc/powerpc64-linux/bin/powerpc64-linux-gcc --version
# < /opt/cross/kisskb/korg/gcc-5.5.0-nolibc/powerpc64-linux/bin/powerpc64-linux-ld --version
# < git log --format=%s --max-count=1 dc8b1290efbbe974d9968106a03077823d0863d6
# < make -s -j 48 ARCH=powerpc O=/kisskb/build/powerpc-next_adder875_defconfig_powerpc-gcc5 CROSS_COMPILE=/opt/cross/kisskb/korg/gcc-5.5.0-nolibc/powerpc64-linux/bin/powerpc64-linux-  adder875_defconfig
# make -s -j 48 ARCH=powerpc O=/kisskb/build/powerpc-next_adder875_defconfig_powerpc-gcc5 CROSS_COMPILE=/opt/cross/kisskb/korg/gcc-5.5.0-nolibc/powerpc64-linux/bin/powerpc64-linux-  
Image Name:   Linux-5.0.0-rc2-gdc8b1290efbb
Created:      Sat Mar  2 22:21:02 2019
Image Type:   PowerPC Linux Kernel Image (gzip compressed)
Data Size:    1936635 Bytes = 1891.25 KiB = 1.85 MiB
Load Address: 00400000
Entry Point:  00400204
Completed OK
# rm -rf /kisskb/build/powerpc-next_adder875_defconfig_powerpc-gcc5
# Build took: 0:00:36.793244