Re: [PATCH v5 1/2] x86/purgatory: Disable various profiling and sanitizing options

From: Hans de Goede
Date: Tue Mar 17 2020 - 09:08:48 EST


Hi,

On 3/13/20 7:28 PM, Borislav Petkov wrote:
On Thu, Mar 12, 2020 at 12:49:50PM +0100, Hans de Goede wrote:
Since the purgatory is a special stand-alone binary, we need to disable

Pls use passive voice in your commit message: no "we" or "I", etc, and
describe your changes in imperative mood.

Also, pls read section "2) Describe your changes" in
Documentation/process/submitting-patches.rst for more details.

Ok, fixed for v6 and I've also rebased the series on tip/master for v6.

Regards,

Hans




various profiling and sanitizing options. Having these options enabled
typically will cause dependency on various special symbols exported by
special libs / stubs used by these frameworks. Since the purgatory is
special we do not link against these stubs causing missing symbols in
the purgatory if we do not disable these options.

This commit syncs the set of disabled profiling and sanitizing options

Avoid having "This patch" or "This commit" in the commit message. It is
tautologically useless.

Also, do

$ git grep 'This patch' Documentation/process

for more details.

Those two review comments apply to patch 2's commit message too, pls fix
them there too.

with that from drivers/firmware/efi/libstub/Makefile, adding
-DDISABLE_BRANCH_PROFILING to the CFLAGS and setting:

GCOV_PROFILE := n
UBSAN_SANITIZE := n

This fixes broken references to ftrace_likely_update when
CONFIG_TRACE_BRANCH_PROFILING is enabled and to __gcov_init and
__gcov_exit when CONFIG_GCOV_KERNEL is enabled.

Signed-off-by: Hans de Goede <hdegoede@xxxxxxxxxx>
---
Changes in v5:
-Not only add -DDISABLE_BRANCH_PROFILING to the CFLAGS but also set:
GCOV_PROFILE := n
UBSAN_SANITIZE := n

Changes in v4:
-This is a new patch in v4 of this series
---
arch/x86/purgatory/Makefile | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)

$ test-apply.sh -g /tmp/01-x86-purgatory-disable_various_profiling_and_sanitizing_options.patch
checking file arch/x86/purgatory/Makefile
Hunk #1 FAILED at 17.
Hunk #2 succeeded at 27 (offset 2 lines).
1 out of 2 hunks FAILED

This happens because tip/master already has KCSAN merged in and it adds

KCSAN_SANITIZE := n

there.

Please redo the patches against current tip/master.

Thx.