DRAM_STR placed its backing string in a variable named __c in a section named
after DRAM_ATTR's __COUNTER__ value. Under LTO, when translation units are
merged, two such variables from different units could collide on the same name
and section, producing a "section type conflict" error.
Derive both the variable name and a dedicated .dram1.str subsection from a
single __COUNTER__ value so each DRAM_STR expansion is unique after merging.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Remove ~50 duplicate local definitions of ALIGN_UP/ALIGN_DOWN/ALIGN_UP_BY/
ALIGN_DOWN_BY across the codebase and replace them with canonical
ESP_ALIGN_UP/ESP_ALIGN_DOWN from esp_macros.h.
fix(mmap): fixed mmap read data wrong when flash being erased/written and cache not disabled
Closes IDFGH-14084
See merge request espressif/esp-idf!29804
ESP_FAULT_ASSERT(C) was silently deleted by the optimizer when C is a cached
flag/status already proven by a preceding `if (!C) return/goto`: the compiler
folds C to a constant and drops all three checks, removing the fault-injection
protection with no warning.
Before:
The cache won't be disabled when XIP on psram. But during flash
erasing/programming, read data will be courrupt.
When XIP in psram is enabled, the image is not mapped to the cache so
usually there will be no flash access. The only way to read from flash
is via the driver or use mmap. The driver has protection during erasing,
while th mmap region not.
Now:
Mmap APIs provide a flag to make mmap->unmap region mutually exclusive
to flash erase/programming when XIP from psram. SPI Flash write APIs
will benefit from this. When the flag is used, no concurrent access to
mapped region will happen while writing; otherwise the cache will be
disable to avoid data corruption.
Most ESP-IDF APIs calls mmap with this flag. As for users calling
mmap-like APIs directly, they can choose whether to enable this by a
flag.
Closes https://github.com/espressif/esp-idf/issues/14897
Refactor the esp_err_to_name() system to decouple esp_common from
higher-level components. Instead of a monolithic generated table,
each component registers its error codes into a dedicated linker
section (.esp_err_msg_table) via idf_define_esp_err_codes() in its
CMakeLists.txt.
New files:
- tools/err_codes_extract.py: extract ESP_ERR_* defines from headers to CSV
- tools/err_codes_to_c.py: generate C source placing entries into linker section
- tools/err_codes_to_rst.py: generate RST documentation from error codes
- tools/cmake/err_codes.cmake: CMake module providing idf_define_esp_err_codes()
- components/esp_common/include/esp_err_codes.h: esp_err_msg_t typedef
- components/esp_common/src/esp_err_to_name_new.c: new lookup using link-time array
- tools/test_apps/build_system/err_codes_check/: CI test app
Changes:
- Remove all optional component dependencies from esp_common/CMakeLists.txt
- Add .esp_err_msg_table section to all 5 linker scripts
- Register error codes in 18 components via idf_define_esp_err_codes()
- Add new scripts to .gitlab/ci/rules.yml build_check patterns
- use new scripts to generate doc and add CI validation
- Update esp_err.rst to add description of composable code registration
esp_http_client_set_redirection() now rejects any redirect target whose
scheme is not https:// when the origin is HTTPS. This catches http, ftp,
ws and any other scheme before client state is mutated. Same-host /
https-to-https redirects are unaffected. Apps that intentionally want
mixed-scheme redirects can set disable_auto_redirect=true and handle
HTTP_EVENT_REDIRECT.
platform.c (compiled into the 'builtin' archive) calls
esp_mbedtls_mem_calloc and esp_mbedtls_mem_free directly by name via
the MBEDTLS_PLATFORM_STD_CALLOC/FREE macros in esp_config.h.
Previously, esp_mem.c was compiled into the 'tfpsacrypto' archive, which
appears after 'builtin' in the link command, causing the linker to fail
with undefined reference errors when 'builtin' was processed first.
Moving esp_mem.c into 'builtin' to assure that the function used by
esp_mbedtls_mem_calloc and esp_mbedtls_mem_free are available when
pltform.c is linked.
Add a test app that verifies PLACE_IN_SECTION, _SECTION_ATTR_SYMBOL_DECL_GENERIC,
_SECTION_START and _SECTION_END macros work correctly on Linux.
The test places 5 uint32_t values into a custom .test_data_table section from
two separate translation units, then iterates the section at runtime to verify
the correct count and content of all entries.
Includes:
- Custom linker script (ld/test_section.ld)
- Build-test-rules entry (linux only)
- pytest host_test marker
Add PLACE_IN_SECTION, _SECTION_ATTR_IMPL_GENERIC, _SECTION_ATTR_SYMBOL_DECL_GENERIC,
_SECTION_START and _SECTION_END macros that emit real section attributes on every
platform (embedded ELF, Linux ELF, macOS Mach-O).
Unlike _SECTION_ATTR_IMPL which is a no-op on Linux, these macros work uniformly
across all targets, enabling link-time arrays for error-code tables, init-function
arrays, and similar patterns.
Also moves _COUNTER_STRINGIFY definition before its first use.
IDF_BUILD_V2 does not have BUILD_COMPONENTS build property. Therefore,
when IDF_BUILD_V2 is defined, use idf_component_optional_requires and
generator expressions instead. I believe that the one-liner should also
work with IDF_BUILD_V1, but the change is kept separate to clearly show
the difference between IDF_BUILD_V1 and IDF_BUILD_V2.
Signed-off-by: Frantisek Hrbata <frantisek.hrbata@espressif.com>
