software/arm-trusted-firmware.git/make_helpers/cflags.mk, branch master

feat(build): enable -Wundef for C files if W >= 1

2026-09-07T11:41:30+00:00

Add -Wundef to BL_FLAGS. This will then be enabled for all TF-A C files through MAKE_C macro, but not for libs. This flag is enabled only when W is defined in the build command line and greater or equal to 1. Change-Id: I6c327584b1e1bb0e1eb59c774a6c6a6788139b99 Signed-off-by: Yann Gautier

fix(build): fix misapplication of errata A53 843419

2026-08-05T13:28:18+00:00

When `ERRATA_A53_843419` is disabled, the build system tries to counter-act the default toolchain configuration behaviour by querying the linker for support for a `--no-fix-cortex-a53-843419` option, and passing it if it is supported. This option, however, is completely fictional; neither LD nor LLD can be configured to enable this errata work-around by default. GCC, on the other hand, *can* be configured to enable it by default. When a GCC configured this way is used as the linker, it behaves as if it was passed the `-mfix-cortex-a53-843419` option, which then passes the `--fix-cortex-a53-843419` option to the linker. Consequently, in this situation the errata work-around is applied regardless of the value of the `ERRATA_A53_843419` build option. The veneers generated by the linker when this work-around is enabled can change the image layout, as was discovered when the Debian maintainers encountered this as a section size overflow when building RK3328: - https://review.trustedfirmware.org/c/TF-A/trusted-firmware-a/+/36981/comments/877be809_5abf10d8 To resolve this, add a separate toolchain helper for probing options accepted directly by the linker driver without going to the linker itself, and use it to pass `-mno-fix-cortex-a53-843419` when the work-around is disabled. Change-Id: Ifffc683c0a5253a1085e1a1c2a2203950a232460 Signed-off-by: Chris Kay LTS-candidate: yes

fix(errata-abi): update management of REPORT_ERRATA

2026-06-23T16:06:40+00:00

For BL2, or when compiling libc, REPORT_ERRATA will not be defined. To avoid warnings when enabling -Wundef flag, force REPORT_ERRATA to 0 for BL2. For libc, when compiling setjmp.S, the issue is that it cannot be defined the same way. As REPORT_ERRATA will only be enabled for BL1, BL31 or BL32, rework the cpu_ops.h file to always check one of those images is used before checking REPORT_ERRATA. Change-Id: Ie4b708cfbb7f8eee9db3254069f179b148c5db6d Signed-off-by: Yann Gautier

fix(build): pass -flto-partition to the linker and not the compiler

2026-02-02T09:29:51+00:00

It's irrelevant to the compiler, although most of the times it gets enabled when compiler==linker so this doesn't show up. It is possible that compiler!=linker (eg compiler=armclang, linker=gcc) with LTO enabled though. Then this blows up since the check was done on the linker and the argument passed to the compiler. Change-Id: Ibc740599c71e892cae873976fed35e1b57658ff4 Signed-off-by: Boyan Karatotev

fix(build): fix BL2_CPPFLAGS when ENABLE_RME is set

2026-01-27T16:38:53+00:00

Add IMAGE_AT_EL3 to BL2_CPPFLAGS as BL2 runs at EL3 when ENABLE_RME is set. And use IMAGE_AT_EL3 for misc_helpers.S that is common for all BLs. This fix enables cpu_reset ops in CPU libs when RME is enabled. Signed-off-by: Arunachalam Ganapathy Change-Id: I9d30b6f900a5fed1993b1c09156830290203ed33

fix(build): assign the ldflags-common variable before appending to it

2026-01-19T08:37:00+00:00

Otherwise the first value is lost, causing odd behaviours. Change-Id: I6df892bf8f706db71d0aa7e67f5076cc9583e477 Signed-off-by: Boyan Karatotev

feat(morello): add Morello capability enablement changes

2026-01-13T08:02:15+00:00

This patch adds a build macro ENABLE_FEAT_MORELLO which when set will compile BL31 firmware with changes required to boot capability aware software. It also adds helper function in c and assmbly to check if morello hardware is present and if morello capability is enabled or not. CE field, bits [23:20] in ID_AA64PFR1_EL1 defines whether morello architecture is present or not, 0b0000 indicates that it is absent and 0b0001 indicates that it is present. While whether capabilities are enabled or not is decided at runtime with ENABLE_FEAT_MORELLO build option. Reference: https://developer.arm.com/documentation/ddi0606/latest/ Signed-off-by: Manoj Kumar Signed-off-by: Varshit Pandya Change-Id: Ib16877acbfcb72c4bd8c08e97e44edc0a3e46089

fix(build): set linker to lld before evaluating ld_option

2025-11-10T08:36:29+00:00

This ensures ld_option is called with valid -fuse-ld option, therefore filtering options against ones supported by lld and not the system default linker Clang chooses without the option specified. This change should not affect toolchains other than llvm-clang, since it only moves a conditional concerning the said toolchain. Resolves the following linking error that manifests itself when using Clang versions 21 and higher for building TF-A: > ld.lld: error: unknown argument '--no-warn-rwx-segments' Change-Id: I466f733377327a7c38ee27899be6681debf61e11 Signed-off-by: Dmitrii Sharshakov

fix(build): put the -target definitions in toolchain.mk

2025-10-20T13:16:13+00:00

Knowing the target is quite important and must be known early - it is necessary for flag discovery and should be available much earlier than cflags.mk's inclusion. So put it in toolchain.mk with the rest of the toolchain configuration. Change-Id: Ia8b522376d27171ad1282d05c162cddc0fca69ab Signed-off-by: Boyan Karatotev

fix(build): align the cpu-ops flags with all others

2025-10-20T13:16:01+00:00

Since the cpu-ops file was created we've obtained the constraints.mk and cflags.mk files and we also have the defaults.mk. The cpu-ops file is not much different to these three, just much more complex. This patch keeps the complicated bit in cpu-ops.mk but it makes it behave like defaults.mk. The non-complicated bits (like cross referencing and compiler flags) go to their corresponding files. This centralises responsibilities and makes it nicer to keep track of. The reason for doing this untangling is that the order of defaulting, cross referencing, and compiler flag selection is significant and we can run into problems where seemingly identical lines of code produce different outcomes. Change-Id: Id4b7714e432a0d628d33412836fd5c93f0488970 Signed-off-by: Boyan Karatotev