coreboot-gerrit April 2022

coreboot-gerrit@coreboot.org

1 participants
5879 discussions

Change in coreboot[master]: tests: Split Makefile to allow for making host-side test tools
by Jakub Czapiga (Code Review) 14 Apr '22

14 Apr '22

Attention is currently required from: Martin Roth. Hello build bot (Jenkins), Martin Roth, I'd like you to reexamine a change. Please visit https://review.coreboot.org/c/coreboot/+/63637 to look at the new patch set (#2). Change subject: tests: Split Makefile to allow for making host-side test tools ...................................................................... tests: Split Makefile to allow for making host-side test tools This patch is based on similar changes [1] done in Depthcharge projects, which aimed to provide unified way to build host-side programs for testing internal code. New test tools might benefit from it by having same base code as unit-tests. [1] https://crrev.com/c/3412108 TEST=make unit-tests TEST=COV=1 make unit-tests coverage-report Signed-off-by: Jakub Czapiga <jacz(a)semihalf.com> Change-Id: Iac4517ab6146fa3f2d2b7a20df54601ab2d04c3d --- A tests/Makefile.common M tests/Makefile.inc 2 files changed, 170 insertions(+), 161 deletions(-) git pull ssh://review.coreboot.org:29418/coreboot refs/changes/37/63637/2 -- To view, visit https://review.coreboot.org/c/coreboot/+/63637 To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings Gerrit-Project: coreboot Gerrit-Branch: master Gerrit-Change-Id: Iac4517ab6146fa3f2d2b7a20df54601ab2d04c3d Gerrit-Change-Number: 63637 Gerrit-PatchSet: 2 Gerrit-Owner: Jakub Czapiga <jacz(a)semihalf.com> Gerrit-Reviewer: Martin Roth <martinroth(a)google.com> Gerrit-Reviewer: build bot (Jenkins) <no-reply(a)coreboot.org> Gerrit-CC: Jan Dabros <jsd(a)semihalf.com> Gerrit-Attention: Martin Roth <martinroth(a)google.com> Gerrit-MessageType: newpatchset

1 0

Change in coreboot[master]: tests: update CMocka to stable-1.1
by Jakub Czapiga (Code Review) 14 Apr '22

14 Apr '22

Attention is currently required from: Martin Roth. Hello build bot (Jenkins), Martin Roth, I'd like you to reexamine a change. Please visit https://review.coreboot.org/c/coreboot/+/63636 to look at the new patch set (#3). Change subject: tests: update CMocka to stable-1.1 ...................................................................... tests: update CMocka to stable-1.1 CMocka stable-1.1 has some convenience bugfixes like vprint buffer increase or leftover values log fix (funtion names display correctly now. Signed-off-by: Jakub Czapiga <jacz(a)semihalf.com> Change-Id: I20ebd15324a21c17cccd2976ae9c3f86b040426d --- M .gitmodules M 3rdparty/cmocka M tests/lib/b64_decode-test.c 3 files changed, 3 insertions(+), 2 deletions(-) git pull ssh://review.coreboot.org:29418/coreboot refs/changes/36/63636/3 -- To view, visit https://review.coreboot.org/c/coreboot/+/63636 To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings Gerrit-Project: coreboot Gerrit-Branch: master Gerrit-Change-Id: I20ebd15324a21c17cccd2976ae9c3f86b040426d Gerrit-Change-Number: 63636 Gerrit-PatchSet: 3 Gerrit-Owner: Jakub Czapiga <jacz(a)semihalf.com> Gerrit-Reviewer: Martin Roth <martinroth(a)google.com> Gerrit-Reviewer: build bot (Jenkins) <no-reply(a)coreboot.org> Gerrit-CC: Jan Dabros <jsd(a)semihalf.com> Gerrit-Attention: Martin Roth <martinroth(a)google.com> Gerrit-MessageType: newpatchset

1 0

Change in coreboot[master]: mb/intel/adlrvp_n: Disable SATA controller
by Usha P (Code Review) 14 Apr '22

14 Apr '22

Attention is currently required from: Rizwan Qureshi, Krishna P Bhat D. Usha P has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/63443 ) Change subject: mb/intel/adlrvp_n: Disable SATA controller ...................................................................... Set Ready For Review -- To view, visit https://review.coreboot.org/c/coreboot/+/63443 To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings Gerrit-Project: coreboot Gerrit-Branch: master Gerrit-Change-Id: Ic9dce3a0b06e1a0d0d9fa495aa406eb12557d842 Gerrit-Change-Number: 63443 Gerrit-PatchSet: 1 Gerrit-Owner: Usha P <usha.p(a)intel.com> Gerrit-Reviewer: Krishna P Bhat D <krishna.p.bhat.d(a)intel.com> Gerrit-Reviewer: Rizwan Qureshi <rizwan.qureshi(a)intel.com> Gerrit-Reviewer: build bot (Jenkins) <no-reply(a)coreboot.org> Gerrit-Attention: Rizwan Qureshi <rizwan.qureshi(a)intel.com> Gerrit-Attention: Krishna P Bhat D <krishna.p.bhat.d(a)intel.com> Gerrit-Comment-Date: Thu, 14 Apr 2022 12:33:29 +0000 Gerrit-HasComments: No Gerrit-Has-Labels: No Gerrit-MessageType: comment

1 0

Change in coreboot[master]: mb/intel/adlrvp: Set half_populated true for ADL-N
by Usha P (Code Review) 14 Apr '22

14 Apr '22

Attention is currently required from: V Sowmya, Paul Menzel, Balaji Manigandan, Krishna P Bhat D. Usha P has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/62913 ) Change subject: mb/intel/adlrvp: Set half_populated true for ADL-N ...................................................................... Patch Set 13: (4 comments) Commit Message: https://review.coreboot.org/c/coreboot/+/62913/comment/25e980ab_c00b6daa PS10, Line 9: Alder Lake N > I believe, the official spelling is Alder Lake-N. (If true, also one occurrence below. […] Ack File src/mainboard/intel/adlrvp/romstage_fsp_params.c: https://review.coreboot.org/c/coreboot/+/62913/comment/1abd5f7c_57e013a4 PS10, Line 52: AlderLake > Alder Lake Ack https://review.coreboot.org/c/coreboot/+/62913/comment/da80f5de_05e99c1d PS10, Line 53: Alder Lake N > I think the official spelling is Alder Lake-N. Ack https://review.coreboot.org/c/coreboot/+/62913/comment/9b78a1fa_acce549d PS10, Line 58: (CONFIG(BOARD_INTEL_ADLRVP_N_EXT_EC) || : CONFIG(BOARD_INTEL_ADLRVP_N)) > Sure. Will check and update. Ack -- To view, visit https://review.coreboot.org/c/coreboot/+/62913 To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings Gerrit-Project: coreboot Gerrit-Branch: master Gerrit-Change-Id: I2ecc3018a1ab039990ba47898ff0e0e2ede695cc Gerrit-Change-Number: 62913 Gerrit-PatchSet: 13 Gerrit-Owner: Usha P <usha.p(a)intel.com> Gerrit-Reviewer: Balaji Manigandan <balaji.manigandan(a)intel.com> Gerrit-Reviewer: Krishna P Bhat D <krishna.p.bhat.d(a)intel.com> Gerrit-Reviewer: Reka Norman <rekanorman(a)chromium.org> Gerrit-Reviewer: Rizwan Qureshi <rizwan.qureshi(a)intel.com> Gerrit-Reviewer: V Sowmya <v.sowmya(a)intel.com> Gerrit-Reviewer: build bot (Jenkins) <no-reply(a)coreboot.org> Gerrit-CC: Felix Held <felix-coreboot(a)felixheld.de> Gerrit-CC: Kangheui Won <khwon(a)chromium.org> Gerrit-CC: Paul Menzel <paulepanter(a)mailbox.org> Gerrit-CC: Tim Wawrzynczak <twawrzynczak(a)chromium.org> Gerrit-Attention: V Sowmya <v.sowmya(a)intel.com> Gerrit-Attention: Paul Menzel <paulepanter(a)mailbox.org> Gerrit-Attention: Balaji Manigandan <balaji.manigandan(a)intel.com> Gerrit-Attention: Krishna P Bhat D <krishna.p.bhat.d(a)intel.com> Gerrit-Comment-Date: Thu, 14 Apr 2022 12:32:34 +0000 Gerrit-HasComments: Yes Gerrit-Has-Labels: No Comment-In-Reply-To: Paul Menzel <paulepanter(a)mailbox.org> Comment-In-Reply-To: Usha P <usha.p(a)intel.com> Gerrit-MessageType: comment

1 0

Change in coreboot[master]: mb/intel/adlrvp: Set half_populated true for ADL-N
by Usha P (Code Review) 14 Apr '22

14 Apr '22

Attention is currently required from: V Sowmya, Paul Menzel, Balaji Manigandan, Krishna P Bhat D. Hello V Sowmya, build bot (Jenkins), Rizwan Qureshi, Reka Norman, Balaji Manigandan, Krishna P Bhat D, I'd like you to reexamine a change. Please visit https://review.coreboot.org/c/coreboot/+/62913 to look at the new patch set (#13). Change subject: mb/intel/adlrvp: Set half_populated true for ADL-N ...................................................................... mb/intel/adlrvp: Set half_populated true for ADL-N Alder Lake-N has single memory controller with 64-bit bus width. Alder Lake common meminit block driver considers bus width to be 128-bit and populates the meminit data accordingly. By setting half_populated to true, only the bottom half is populated. Ideally, half_populated is used in platforms with multiple channels to enable only one half of the channel. Alder Lake N has single channel, and it would require for new structures to be defined in meminit block driver for LPx memory configurations. In order to avoid adding new structures, set half_populated to true. This has the same effect as having single channel with 64-bit width. BRANCH=NONE TEST=Build and boot ADL-N RVP. Signed-off-by: Usha P <usha.p(a)intel.com> Change-Id: I2ecc3018a1ab039990ba47898ff0e0e2ede695cc --- M src/mainboard/intel/adlrvp/romstage_fsp_params.c 1 file changed, 10 insertions(+), 1 deletion(-) git pull ssh://review.coreboot.org:29418/coreboot refs/changes/13/62913/13 -- To view, visit https://review.coreboot.org/c/coreboot/+/62913 To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings Gerrit-Project: coreboot Gerrit-Branch: master Gerrit-Change-Id: I2ecc3018a1ab039990ba47898ff0e0e2ede695cc Gerrit-Change-Number: 62913 Gerrit-PatchSet: 13 Gerrit-Owner: Usha P <usha.p(a)intel.com> Gerrit-Reviewer: Balaji Manigandan <balaji.manigandan(a)intel.com> Gerrit-Reviewer: Krishna P Bhat D <krishna.p.bhat.d(a)intel.com> Gerrit-Reviewer: Reka Norman <rekanorman(a)chromium.org> Gerrit-Reviewer: Rizwan Qureshi <rizwan.qureshi(a)intel.com> Gerrit-Reviewer: V Sowmya <v.sowmya(a)intel.com> Gerrit-Reviewer: build bot (Jenkins) <no-reply(a)coreboot.org> Gerrit-CC: Felix Held <felix-coreboot(a)felixheld.de> Gerrit-CC: Kangheui Won <khwon(a)chromium.org> Gerrit-CC: Paul Menzel <paulepanter(a)mailbox.org> Gerrit-CC: Tim Wawrzynczak <twawrzynczak(a)chromium.org> Gerrit-Attention: V Sowmya <v.sowmya(a)intel.com> Gerrit-Attention: Paul Menzel <paulepanter(a)mailbox.org> Gerrit-Attention: Balaji Manigandan <balaji.manigandan(a)intel.com> Gerrit-Attention: Krishna P Bhat D <krishna.p.bhat.d(a)intel.com> Gerrit-MessageType: newpatchset

1 0

Change in coreboot[master]: mb/intel/adlrvp: Set half_populated true for ADL-N
by Usha P (Code Review) 14 Apr '22

14 Apr '22

Attention is currently required from: V Sowmya, Paul Menzel, Balaji Manigandan, Krishna P Bhat D. Hello V Sowmya, build bot (Jenkins), Rizwan Qureshi, Reka Norman, Balaji Manigandan, Krishna P Bhat D, I'd like you to reexamine a change. Please visit https://review.coreboot.org/c/coreboot/+/62913 to look at the new patch set (#12). Change subject: mb/intel/adlrvp: Set half_populated true for ADL-N ...................................................................... mb/intel/adlrvp: Set half_populated true for ADL-N Alder Lake-N has single memory controller with 64-bit bus width. Alder Lake common meminit block driver considers bus width to be 128-bit and populates the meminit data accordingly. By setting half_populated to true, only the bottom half is populated. Ideally, half_populated is used in platforms with multiple channels to enable only one half of the channel. Alder Lake N has single channel, and it would require for new structures to be defined in meminit block driver for LPx memory configurations. In order to avoid adding new structures, set half_populated to true. This has the same effect as having single channel with 64-bit width. Signed-off-by: Usha P <usha.p(a)intel.com> Change-Id: I2ecc3018a1ab039990ba47898ff0e0e2ede695cc --- M src/mainboard/intel/adlrvp/romstage_fsp_params.c 1 file changed, 10 insertions(+), 1 deletion(-) git pull ssh://review.coreboot.org:29418/coreboot refs/changes/13/62913/12 -- To view, visit https://review.coreboot.org/c/coreboot/+/62913 To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings Gerrit-Project: coreboot Gerrit-Branch: master Gerrit-Change-Id: I2ecc3018a1ab039990ba47898ff0e0e2ede695cc Gerrit-Change-Number: 62913 Gerrit-PatchSet: 12 Gerrit-Owner: Usha P <usha.p(a)intel.com> Gerrit-Reviewer: Balaji Manigandan <balaji.manigandan(a)intel.com> Gerrit-Reviewer: Krishna P Bhat D <krishna.p.bhat.d(a)intel.com> Gerrit-Reviewer: Reka Norman <rekanorman(a)chromium.org> Gerrit-Reviewer: Rizwan Qureshi <rizwan.qureshi(a)intel.com> Gerrit-Reviewer: V Sowmya <v.sowmya(a)intel.com> Gerrit-Reviewer: build bot (Jenkins) <no-reply(a)coreboot.org> Gerrit-CC: Felix Held <felix-coreboot(a)felixheld.de> Gerrit-CC: Kangheui Won <khwon(a)chromium.org> Gerrit-CC: Paul Menzel <paulepanter(a)mailbox.org> Gerrit-CC: Tim Wawrzynczak <twawrzynczak(a)chromium.org> Gerrit-Attention: V Sowmya <v.sowmya(a)intel.com> Gerrit-Attention: Paul Menzel <paulepanter(a)mailbox.org> Gerrit-Attention: Balaji Manigandan <balaji.manigandan(a)intel.com> Gerrit-Attention: Krishna P Bhat D <krishna.p.bhat.d(a)intel.com> Gerrit-MessageType: newpatchset

1 0

Change in coreboot[master]: tests: update CMocka to stable-1.1
by Jakub Czapiga (Code Review) 14 Apr '22

14 Apr '22

Attention is currently required from: Martin Roth. Hello Martin Roth, I'd like you to reexamine a change. Please visit https://review.coreboot.org/c/coreboot/+/63636 to look at the new patch set (#2). Change subject: tests: update CMocka to stable-1.1 ...................................................................... tests: update CMocka to stable-1.1 CMocka stable-1.1 has some convenience bugfixes like vprint buffer increase or leftover values log fix (funtion names display correctly now. Signed-off-by: Jakub Czapiga <jacz(a)semihalf.com> Change-Id: I20ebd15324a21c17cccd2976ae9c3f86b040426d --- M .gitmodules M 3rdparty/cmocka M tests/lib/b64_decode-test.c 3 files changed, 3 insertions(+), 2 deletions(-) git pull ssh://review.coreboot.org:29418/coreboot refs/changes/36/63636/2 -- To view, visit https://review.coreboot.org/c/coreboot/+/63636 To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings Gerrit-Project: coreboot Gerrit-Branch: master Gerrit-Change-Id: I20ebd15324a21c17cccd2976ae9c3f86b040426d Gerrit-Change-Number: 63636 Gerrit-PatchSet: 2 Gerrit-Owner: Jakub Czapiga <jacz(a)semihalf.com> Gerrit-Reviewer: Martin Roth <martinroth(a)google.com> Gerrit-CC: Jan Dabros <jsd(a)semihalf.com> Gerrit-Attention: Martin Roth <martinroth(a)google.com> Gerrit-MessageType: newpatchset

1 0

Change in coreboot[master]: mb/google/brya/var/anahera{4es}: select DRIVERS_GENESYSLOGIC_GL9750
by Chen Wisley (Code Review) 14 Apr '22

14 Apr '22

Attention is currently required from: Wisley Chen. Hello Wisley Chen, I'd like you to do a code review. Please visit https://review.coreboot.org/c/coreboot/+/63638 to review the following change. Change subject: mb/google/brya/var/anahera{4es}: select DRIVERS_GENESYSLOGIC_GL9750 ...................................................................... mb/google/brya/var/anahera{4es}: select DRIVERS_GENESYSLOGIC_GL9750 select DRIVERS_GENESYSLOGIC_GL9750 to disable ASPM L0s. BUG=b:229213455 TEST=emerge-brya coreboot chromeos-bootimage Signed-off-by: Wisley Chen <wisley.chen(a)quanta.corp-partner.google.com> Change-Id: Ie89fa6c66974284063cd25ae8097db94a93326ca --- M src/mainboard/google/brya/Kconfig.name 1 file changed, 2 insertions(+), 0 deletions(-) git pull ssh://review.coreboot.org:29418/coreboot refs/changes/38/63638/1 diff --git a/src/mainboard/google/brya/Kconfig.name b/src/mainboard/google/brya/Kconfig.name index b79fce6..3028aa1 100644 --- a/src/mainboard/google/brya/Kconfig.name +++ b/src/mainboard/google/brya/Kconfig.name @@ -13,6 +13,7 @@ bool "-> Anahera" select BOARD_GOOGLE_BASEBOARD_BRYA select DRIVERS_GENESYSLOGIC_GL9763E + select DRIVERS_GENESYSLOGIC_GL9750 select DRIVERS_GFX_GENERIC select HAVE_PCIE_WWAN select HAVE_WWAN_POWER_SEQUENCE @@ -21,6 +22,7 @@ bool "-> Anahera4ES" select BOARD_GOOGLE_BASEBOARD_BRYA select DRIVERS_GENESYSLOGIC_GL9763E + select DRIVERS_GENESYSLOGIC_GL9750 select DRIVERS_GFX_GENERIC select HAVE_PCIE_WWAN select HAVE_WWAN_POWER_SEQUENCE -- To view, visit https://review.coreboot.org/c/coreboot/+/63638 To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings Gerrit-Project: coreboot Gerrit-Branch: master Gerrit-Change-Id: Ie89fa6c66974284063cd25ae8097db94a93326ca Gerrit-Change-Number: 63638 Gerrit-PatchSet: 1 Gerrit-Owner: Chen Wisley <wisley.chen(a)quantatw.com> Gerrit-Reviewer: Wisley Chen <wisley.chen(a)quanta.corp-partner.google.com> Gerrit-Attention: Wisley Chen <wisley.chen(a)quanta.corp-partner.google.com> Gerrit-MessageType: newchange

1 0

Change in coreboot[master]: tests: Split Makefile to allow for making host-side test tools
by Jakub Czapiga (Code Review) 14 Apr '22

14 Apr '22

Jakub Czapiga has uploaded this change for review. ( https://review.coreboot.org/c/coreboot/+/63637 ) Change subject: tests: Split Makefile to allow for making host-side test tools ...................................................................... tests: Split Makefile to allow for making host-side test tools This patch is based on similar changes [1] done in Depthcharge projects, which aimed to provide unified way to build host-side programs for testing internal code. New test tools might benefit from it by having same base code as unit-tests. [1] https://crrev.com/c/3412108 TEST=make unit-tests TEST=COV=1 make unit-tests coverage-report Signed-off-by: Jakub Czapiga <jacz(a)semihalf.com> Change-Id: Iac4517ab6146fa3f2d2b7a20df54601ab2d04c3d --- A tests/Makefile.common M tests/Makefile.inc 2 files changed, 168 insertions(+), 159 deletions(-) git pull ssh://review.coreboot.org:29418/coreboot refs/changes/37/63637/1 diff --git a/tests/Makefile.common b/tests/Makefile.common new file mode 100644 index 0000000..0fb425e --- /dev/null +++ b/tests/Makefile.common @@ -0,0 +1,167 @@ +# SPDX-License-Identifier: GPL-2.0-only + +# This file contains common definitions, macros and targets for depthcharge +# unit-tests and screenshot utility. It requires list of defined variables: +# - src - source directory +# - testobj - build directory of tests +# - objutil - utility programs/libraries output directory + +cmockasrc := 3rdparty/cmocka +cmockaobj := $(objutil)/cmocka +coverage_dir := coverage_reports + +CMOCKA_LIB := $(cmockaobj)/src/libcmocka.so + +CMAKE := cmake +OBJCOPY ?= objcopy +OBJDUMP ?= objdump + +TEST_DEFAULT_CONFIG := $(top)/configs/config.emulation_qemu_x86_i440fx +TEST_DOTCONFIG := $(testobj)/.config +TEST_KCONFIG_AUTOHEADER := $(testobj)/config.src.h +TEST_KCONFIG_AUTOCONFIG := $(testobj)/auto.conf +TEST_KCONFIG_DEPENDENCIES := $(testobj)/auto.conf.cmd +TEST_KCONFIG_SPLITCONFIG := $(testobj)/config/ +TEST_KCONFIG_TRISTATE := $(testobj)/tristate.conf + +TEST_CFLAGS := -include $(src)/include/kconfig.h \ + -include $(src)/commonlib/bsd/include/commonlib/bsd/compiler.h \ + -include $(src)/include/rules.h + +# Include generic test mock headers, before original ones +TEST_CFLAGS += -I$(testsrc)/include/mocks -I$(testsrc)/include + +TEST_CFLAGS += -I$(src) -I$(src)/include -I$(src)/commonlib/include \ + -I$(src)/commonlib/bsd/include -I$(src)/arch/x86/include \ + -I$(top)/3rdparty/vboot/firmware/include + +# Note: This is intentionally just a subset of the warnings in the toplevel +# Makefile.inc. We don't need to be as strict with test code, and things like +# -Wmissing-prototypes just make working with the test framework cumbersome. +# Only put conservative warnings here that really detect code that's obviously +# unintentional. +TEST_CFLAGS += -Wall -Werror -Wundef -Wstrict-prototypes -Wno-inline-asm +TEST_CFLAGS += -Wno-unknown-warning-option -Wno-source-mgr -Wno-main-return-type + +# Path for Kconfig autoheader +TEST_CFLAGS += -I$(dir $(TEST_KCONFIG_AUTOHEADER)) + +TEST_CFLAGS += -std=gnu11 -Os -ffunction-sections -fdata-sections -fno-builtin + +TEST_CFLAGS += -D__TEST__ + +TEST_CFLAGS += -I$(cmockasrc)/include + +ifneq ($(filter-out 0,$(TEST_PRINT)),) +TEST_CFLAGS += -DTEST_PRINT=1 +endif + +# Link against Cmocka +TEST_LDFLAGS := -L$(cmockaobj)/src -lcmocka -Wl,-rpath=$(cmockaobj)/src +TEST_LDFLAGS += -Wl,--gc-sections + +# Some memlayout symbols don't work with userspace relocation -- disable it. +TEST_CFLAGS += -fno-pie -fno-pic +TEST_LDFLAGS += -no-pie + +# Extra attributes for unit tests, declared per test +attributes := srcs cflags config mocks stage + +# Copy attributes of one test to another. +# $1 - input test name +# $2 - output test name +copy-test = $(foreach attr,$(attributes), \ + $(eval $(strip $(2))-$(attr) := $($(strip $(1))-$(attr)))) + +# Create actual targets for unit test binaries +# $1 - test name +define TEST_CC_template + +# Generate custom config.h redefining given config symbols, and declaring mocked +# functions weak. It is important that the compiler already sees that they are +# weak (and they aren't just turned weak at a later stage) to prevent certain +# optimizations that would break if the function gets replaced. (For clang this +# file needs to be marked `system_header` to prevent it from warning about +# #pragma weak entries without a matching function declaration, since there's no +# -Wno-xxx command line option for that.) +$(1)-config-file := $(testobj)/$(1)/config.h +$$($(1)-config-file): $(TEST_KCONFIG_AUTOHEADER) + mkdir -p $$(dir $$@) + printf '// File generated by tests/Makefile.inc\n// Do not change\n' > $$@ + printf '#include <%s>\n\n' "$(notdir $(TEST_KCONFIG_AUTOHEADER))" >> $$@ + for kv in $$($(1)-config); do \ + key="`echo $$$$kv | cut -d '=' -f -1`"; \ + value="`echo $$$$kv | cut -d '=' -f 2-`"; \ + printf '#undef %s\n' "$$$$key" >> $$@; \ + printf '#define %s %s\n\n' "$$$$key" "$$$$value" >> $$@; \ + done + printf '#ifdef __clang__\n' >> $$@; + printf '#pragma clang system_header\n' >> $$@; + printf '#endif\n' >> $$@; + printf '#ifdef __TEST_SRCOBJ__\n' >> $$@; + for m in $$($(1)-mocks); do \ + printf '#pragma weak %s\n' "$$$$m" >> $$@; \ + done + printf '#endif\n' >> $$@; + +$($(1)-objs): TEST_CFLAGS += -I$$(dir $$($(1)-config-file)) \ + -D__$$(shell echo $$($(1)-stage) | tr '[:lower:]' '[:upper:]')__ \ + -D__TEST_NAME__=\"$(subst /,_,$(1))\" \ + -D__TEST_DATA_DIR__=\"$(testsrc)/data\" + +# Give us a way to distinguish between coreboot source files and test files in code. +$($(1)-srcobjs): TEST_CFLAGS += -D__TEST_SRCOBJ__ + +# Compile sources and apply mocking/wrapping of selected symbols. +# For each listed mock add new symbol with prefix `__real_`, +# and pointing to the same section:address. +$($(1)-objs): $(testobj)/$(1)/%.o: $$$$*.c $$($(1)-config-file) + mkdir -p $$(dir $$@) + $(HOSTCC) $(HOSTCFLAGS) $$(TEST_CFLAGS) $($(1)-cflags) -MMD \ + -MF $$(basename $$(a)).d -MT $$@ -c $$< -o $$@.orig + objcopy_wrap_flags=''; \ + for sym in $$($(1)-mocks); do \ + sym_line="$$$$($(OBJDUMP) -t $$@.orig \ + | grep -E "[0-9a-fA-F]+\\s+w\\s+F\\s+.*\\s$$$$sym$$$$")"; \ + if [ ! -z "$$$$sym_line" ] ; then \ + addr="$$$$(echo "$$$$sym_line" | awk '{ print $$$$1 }')"; \ + section="$$$$(echo "$$$$sym_line" | awk '{ print $$$$(NF - 2) }')"; \ + objcopy_wrap_flags="$$$$objcopy_wrap_flags --add-symbol __real_$$$${sym}=$$$${section}:0x$$$${addr},function,global"; \ + fi \ + done ; \ + $(OBJCOPY) $$@.orig $$$$objcopy_wrap_flags $$@ + +$($(1)-bin): $($(1)-objs) $(CMOCKA_LIB) + $(HOSTCC) $$^ $($(1)-cflags) $$(TEST_LDFLAGS) -o $$@ + +endef + +# Build cmocka +$(CMOCKA_LIB): + echo "*** Building CMOCKA ***" + mkdir -p $(cmockaobj) + cd $(cmockaobj) && $(CMAKE) $(abspath $(cmockasrc)) + $(MAKE) -C $(cmockaobj) + +# Kconfig targets +$(TEST_DOTCONFIG): + mkdir -p $(dir $@) + cp $(TEST_DEFAULT_CONFIG) $(TEST_DOTCONFIG) + +# Don't override default Kconfig variables, since this will affect all +# Kconfig targets. Change them only when calling sub-make instead. +$(TEST_KCONFIG_AUTOHEADER): TEST_KCONFIG_FLAGS := DOTCONFIG=$(TEST_DOTCONFIG) \ + KCONFIG_AUTOHEADER=$(TEST_KCONFIG_AUTOHEADER) \ + KCONFIG_AUTOCONFIG=$(TEST_KCONFIG_AUTOCONFIG) \ + KCONFIG_DEPENDENCIES=$(TEST_KCONFIG_DEPENDENCIES) \ + KCONFIG_SPLITCONFIG=$(TEST_KCONFIG_SPLITCONFIG) \ + KCONFIG_TRISTATE=$(TEST_KCONFIG_TRISTATE) \ + KBUILD_DEFCONFIG=$(TEST_DEFAULT_CONFIG) + +$(TEST_KCONFIG_AUTOHEADER): $(TEST_DOTCONFIG) $(objutil)/kconfig/conf + mkdir -p $(dir $@) + $(MAKE) $(TEST_KCONFIG_FLAGS) olddefconfig + $(MAKE) $(TEST_KCONFIG_FLAGS) syncconfig + +$(TEST_KCONFIG_AUTOCONFIG): $(TEST_KCONFIG_AUTOHEADER) + true diff --git a/tests/Makefile.inc b/tests/Makefile.inc index 847beaf..994aa49 100644 --- a/tests/Makefile.inc +++ b/tests/Makefile.inc @@ -10,63 +10,7 @@ testobj := $(obj)/tests endif -cmockasrc := 3rdparty/cmocka -cmockaobj := $(objutil)/cmocka -coverage_dir := coverage_reports - -CMOCKA_LIB := $(cmockaobj)/src/libcmocka.so - -CMAKE := cmake -OBJCOPY ?= objcopy -OBJDUMP ?= objdump - -TEST_DEFAULT_CONFIG := $(top)/configs/config.emulation_qemu_x86_i440fx -TEST_DOTCONFIG := $(testobj)/.config -TEST_KCONFIG_AUTOHEADER := $(testobj)/config.src.h -TEST_KCONFIG_AUTOCONFIG := $(testobj)/auto.conf -TEST_KCONFIG_DEPENDENCIES := $(testobj)/auto.conf.cmd -TEST_KCONFIG_SPLITCONFIG := $(testobj)/config/ -TEST_KCONFIG_TRISTATE := $(testobj)/tristate.conf - -TEST_CFLAGS := -include $(src)/include/kconfig.h \ - -include $(src)/commonlib/bsd/include/commonlib/bsd/compiler.h \ - -include $(src)/include/rules.h - -# Include generic test mock headers, before original ones -TEST_CFLAGS += -I$(testsrc)/include/mocks -I$(testsrc)/include - -TEST_CFLAGS += -I$(src) -I$(src)/include -I$(src)/commonlib/include \ - -I$(src)/commonlib/bsd/include -I$(src)/arch/x86/include \ - -I$(top)/3rdparty/vboot/firmware/include - -# Note: This is intentionally just a subset of the warnings in the toplevel -# Makefile.inc. We don't need to be as strict with test code, and things like -# -Wmissing-prototypes just make working with the test framework cumbersome. -# Only put conservative warnings here that really detect code that's obviously -# unintentional. -TEST_CFLAGS += -Wall -Werror -Wundef -Wstrict-prototypes -Wno-inline-asm -TEST_CFLAGS += -Wno-unknown-warning-option -Wno-source-mgr -Wno-main-return-type - -# Path for Kconfig autoheader -TEST_CFLAGS += -I$(dir $(TEST_KCONFIG_AUTOHEADER)) - -TEST_CFLAGS += -std=gnu11 -Os -ffunction-sections -fdata-sections -fno-builtin - -TEST_CFLAGS += -D__TEST__ - -TEST_CFLAGS += -I$(cmockasrc)/include - -ifneq ($(filter-out 0,$(TEST_PRINT)),) -TEST_CFLAGS += -DTEST_PRINT=1 -endif - -# Link against Cmocka -TEST_LDFLAGS := -L$(cmockaobj)/src -lcmocka -Wl,-rpath=$(cmockaobj)/src -TEST_LDFLAGS += -Wl,--gc-sections - -# Some memlayout symbols don't work with userspace relocation -- disable it. -TEST_CFLAGS += -fno-pie -fno-pic -TEST_LDFLAGS += -no-pie +include $(top)/tests/Makefile.common # Enable code coverage if COV=1 ifeq ($(COV),1) @@ -74,9 +18,6 @@ TEST_LDFLAGS += --coverage endif -# Extra attributes for unit tests, declared per test -attributes := srcs cflags config mocks stage - stages := decompressor bootblock romstage smm verstage stages += ramstage rmodule postcar libagesa @@ -100,78 +41,9 @@ Check your $(dir $(1)$(2))Makefile.inc)) endef -# Copy attributes of one test to another. -# $1 - input test name -# $2 - output test name -copy-test = $(foreach attr,$(attributes), \ - $(eval $(strip $(2))-$(attr) := $($(strip $(1))-$(attr)))) - $(call add-special-class, tests) $(call evaluate_subdirs) -# Create actual targets for unit test binaries -# $1 - test name -define TEST_CC_template - -# Generate custom config.h redefining given config symbols, and declaring mocked -# functions weak. It is important that the compiler already sees that they are -# weak (and they aren't just turned weak at a later stage) to prevent certain -# optimizations that would break if the function gets replaced. (For clang this -# file needs to be marked `system_header` to prevent it from warning about -# #pragma weak entries without a matching function declaration, since there's no -# -Wno-xxx command line option for that.) -$(1)-config-file := $(testobj)/$(1)/config.h -$$($(1)-config-file): $(TEST_KCONFIG_AUTOHEADER) - mkdir -p $$(dir $$@) - printf '// File generated by tests/Makefile.inc\n// Do not change\n' > $$@ - printf '#include <%s>\n\n' "$(notdir $(TEST_KCONFIG_AUTOHEADER))" >> $$@ - for kv in $$($(1)-config); do \ - key="`echo $$$$kv | cut -d '=' -f -1`"; \ - value="`echo $$$$kv | cut -d '=' -f 2-`"; \ - printf '#undef %s\n' "$$$$key" >> $$@; \ - printf '#define %s %s\n\n' "$$$$key" "$$$$value" >> $$@; \ - done - printf '#ifdef __clang__\n' >> $$@; - printf '#pragma clang system_header\n' >> $$@; - printf '#endif\n' >> $$@; - printf '#ifdef __TEST_SRCOBJ__\n' >> $$@; - for m in $$($(1)-mocks); do \ - printf '#pragma weak %s\n' "$$$$m" >> $$@; \ - done - printf '#endif\n' >> $$@; - -$($(1)-objs): TEST_CFLAGS += -I$$(dir $$($(1)-config-file)) \ - -D__$$(shell echo $$($(1)-stage) | tr '[:lower:]' '[:upper:]')__ \ - -D__TEST_NAME__=\"$(subst /,_,$(1))\" \ - -D__TEST_DATA_DIR__=\"$(testsrc)/data\" - -# Give us a way to distinguish between coreboot source files and test files in code. -$($(1)-srcobjs): TEST_CFLAGS += -D__TEST_SRCOBJ__ - -# Compile sources and apply mocking/wrapping of selected symbols. -# For each listed mock add new symbol with prefix `__real_`, -# and pointing to the same section:address. -$($(1)-objs): $(testobj)/$(1)/%.o: $$$$*.c $$($(1)-config-file) - mkdir -p $$(dir $$@) - $(HOSTCC) $(HOSTCFLAGS) $$(TEST_CFLAGS) $($(1)-cflags) -MMD \ - -MF $$(basename $$(a)).d -MT $$@ -c $$< -o $$@.orig - objcopy_wrap_flags=''; \ - for sym in $$($(1)-mocks); do \ - sym_line="$$$$($(OBJDUMP) -t $$@.orig \ - | grep -E "[0-9a-fA-F]+\\s+w\\s+F\\s+.*\\s$$$$sym$$$$")"; \ - if [ ! -z "$$$$sym_line" ] ; then \ - addr="$$$$(echo "$$$$sym_line" | awk '{ print $$$$1 }')"; \ - section="$$$$(echo "$$$$sym_line" | awk '{ print $$$$(NF - 2) }')"; \ - objcopy_wrap_flags="$$$$objcopy_wrap_flags --add-symbol __real_$$$${sym}=$$$${section}:0x$$$${addr},function,global"; \ - fi \ - done ; \ - $(OBJCOPY) $$@.orig $$$$objcopy_wrap_flags $$@ - -$($(1)-bin): $($(1)-objs) $(CMOCKA_LIB) - $(HOSTCC) $$^ $($(1)-cflags) $$(TEST_LDFLAGS) -o $$@ - -endef - $(foreach test, $(alltests), \ $(eval $(test)-srcobjs := $(addprefix $(testobj)/$(test)/, \ $(patsubst %.c,%.o,$(filter src/%,$($(test)-srcs))))) \ @@ -190,36 +62,6 @@ DEPENDENCIES += $(addsuffix .d,$(basename $(all-test-objs))) -include $(DEPENDENCIES) -# Build cmocka -$(CMOCKA_LIB): - echo "*** Building CMOCKA ***" - mkdir -p $(cmockaobj) - cd $(cmockaobj) && $(CMAKE) $(abspath $(cmockasrc)) - $(MAKE) -C $(cmockaobj) - -# Kconfig targets -$(TEST_DOTCONFIG): - mkdir -p $(dir $@) - cp $(TEST_DEFAULT_CONFIG) $(TEST_DOTCONFIG) - -# Don't override default Kconfig variables, since this will affect all -# Kconfig targets. Change them only when calling sub-make instead. -$(TEST_KCONFIG_AUTOHEADER): TEST_KCONFIG_FLAGS := DOTCONFIG=$(TEST_DOTCONFIG) \ - KCONFIG_AUTOHEADER=$(TEST_KCONFIG_AUTOHEADER) \ - KCONFIG_AUTOCONFIG=$(TEST_KCONFIG_AUTOCONFIG) \ - KCONFIG_DEPENDENCIES=$(TEST_KCONFIG_DEPENDENCIES) \ - KCONFIG_SPLITCONFIG=$(TEST_KCONFIG_SPLITCONFIG) \ - KCONFIG_TRISTATE=$(TEST_KCONFIG_TRISTATE) \ - KBUILD_DEFCONFIG=$(TEST_DEFAULT_CONFIG) - -$(TEST_KCONFIG_AUTOHEADER): $(TEST_DOTCONFIG) $(objutil)/kconfig/conf - mkdir -p $(dir $@) - $(MAKE) $(TEST_KCONFIG_FLAGS) olddefconfig - $(MAKE) $(TEST_KCONFIG_FLAGS) syncconfig - -$(TEST_KCONFIG_AUTOCONFIG): $(TEST_KCONFIG_AUTOHEADER) - true - .PHONY: $(alltests) $(addprefix clean-,$(alltests)) .PHONY: unit-tests build-unit-tests run-unit-tests clean-unit-tests -- To view, visit https://review.coreboot.org/c/coreboot/+/63637 To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings Gerrit-Project: coreboot Gerrit-Branch: master Gerrit-Change-Id: Iac4517ab6146fa3f2d2b7a20df54601ab2d04c3d Gerrit-Change-Number: 63637 Gerrit-PatchSet: 1 Gerrit-Owner: Jakub Czapiga <jacz(a)semihalf.com> Gerrit-MessageType: newchange

1 0

Change in coreboot[master]: tests: update CMocka to stable-1.1
by Jakub Czapiga (Code Review) 14 Apr '22

14 Apr '22

Jakub Czapiga has uploaded this change for review. ( https://review.coreboot.org/c/coreboot/+/63636 ) Change subject: tests: update CMocka to stable-1.1 ...................................................................... tests: update CMocka to stable-1.1 CMocka stable-1.1 has some convinience bugfixes like vprint buffer increase or leftover values log fix (funtion napes display correctly now. Signed-off-by: Jakub Czapiga <jacz(a)semihalf.com> Change-Id: I20ebd15324a21c17cccd2976ae9c3f86b040426d --- M .gitmodules M 3rdparty/cmocka M tests/lib/b64_decode-test.c 3 files changed, 3 insertions(+), 2 deletions(-) git pull ssh://review.coreboot.org:29418/coreboot refs/changes/36/63636/1 diff --git a/.gitmodules b/.gitmodules index 68230b9..69aa470 100644 --- a/.gitmodules +++ b/.gitmodules @@ -48,6 +48,7 @@ path = 3rdparty/cmocka url = ../cmocka.git update = none + branch = stable-1.1 [submodule "3rdparty/qc_blobs"] path = 3rdparty/qc_blobs url = ../qc_blobs.git diff --git a/3rdparty/cmocka b/3rdparty/cmocka index 672c5ce..8931845 160000 --- a/3rdparty/cmocka +++ b/3rdparty/cmocka @@ -1 +1 @@ -Subproject commit 672c5cee79eb412025c3dd8b034e611c1f119055 +Subproject commit 8931845c35e78b5123d73430b071affd537d5935 diff --git a/tests/lib/b64_decode-test.c b/tests/lib/b64_decode-test.c index 631dc15..189ce96 100644 --- a/tests/lib/b64_decode-test.c +++ b/tests/lib/b64_decode-test.c @@ -35,7 +35,7 @@ res = b64_decode((uint8_t *)messages[i].enc, strlen(messages[i].enc), decoded); assert_int_equal(res, (strlen(messages[i].dec))); - assert_string_equal(decoded, messages[i].dec); + assert_string_equal((const char *)decoded, messages[i].dec); free(decoded); } -- To view, visit https://review.coreboot.org/c/coreboot/+/63636 To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings Gerrit-Project: coreboot Gerrit-Branch: master Gerrit-Change-Id: I20ebd15324a21c17cccd2976ae9c3f86b040426d Gerrit-Change-Number: 63636 Gerrit-PatchSet: 1 Gerrit-Owner: Jakub Czapiga <jacz(a)semihalf.com> Gerrit-MessageType: newchange

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

coreboot-gerrit April 2022