gcc: Resolve relative prefix-map filenames
Quoted 67 lines · Richard Purdie via lists.openembedded.org
Add a patch to gcc so that relative paths are handled by -fdebug-prefix-map and friends. In OE we use relative paths in autotools and removing that creates a lot of issues we'd have to fix. This alternative allows us to fix the paths within gcc and improve our debug file coverage (and SPDX manifests) accordingly.
Signed-off-by Richard Purdie <[email protected]>--- meta/recipes-devtools/gcc/gcc-12.1.inc | 1 + .../gcc/gcc/prefix-map-realpath.patch | 46 +++++++++++++++++++ 2 files changed, 47 insertions(+) create mode 100644 meta/recipes-devtools/gcc/gcc/prefix-map-realpath.patchmeta/recipes-devtools/gcc/gcc-12.1.inc +1
-0@@ -65,6 +65,7 @@ SRC_URI = "${BASEURI} \ file://0025-Move-sched.h-include-ahead-of-user-headers.patch \ file://0026-rust-recursion-limit.patch \ file://0001-libsanitizer-cherry-pick-9cf13067cb5088626ba7-from-u.patch \ + file://prefix-map-realpath.patch \ " SRC_URI[sha256sum] = "62fd634889f31c02b64af2c468f064b47ad1ca78411c45abe6ac4b5f8dd19c7b"meta/recipes-devtools/gcc/gcc/prefix-map-realpath.patch new file +36
-0@@ -0,0 +1,46 @@ +Relative paths don't work with -fdebug-prefix-map and friends. This +can lead to paths which the user wanted to be remapped being missed. +Setting -fdebug-prefix-map to work with a relative path isn't practical +either. + +Instead, call gcc's realpath function on the incomming path name before +comparing it with the remapping. This means other issues like symlinks +are also accounted for and leads to a more consistent remapping experience. + +Upstream-Status: Pending [need to see if gcc developers would accept this] +Signed-off-by: Richard Purdie <[email protected]> + + +Index: gcc-12.1.0/gcc/file-prefix-map.cc +=================================================================== +--- gcc-12.1.0.orig/gcc/file-prefix-map.cc ++++ gcc-12.1.0/gcc/file-prefix-map.cc +@@ -70,19 +70,25 @@ remap_filename (file_prefix_map *maps, c + file_prefix_map *map; + char *s; + const char *name; ++ char *realname; + size_t name_len; + ++ realname = lrealpath (filename); ++ + for (map = maps; map; map = map->next) +- if (filename_ncmp (filename, map->old_prefix, map->old_len) == 0) ++ if (filename_ncmp (realname, map->old_prefix, map->old_len) == 0) + break; +- if (!map) ++ if (!map) { ++ free (realname); + return filename; +- name = filename + map->old_len; ++ }
meta/recipes-devtools/gcc/gcc/prefix-map-realpath.patch line 46
Quoted 10 lines
meta/recipes-devtools/gcc/gcc/prefix-map-realpath.patch new file +10
-0++ name = realname + map->old_len; + name_len = strlen (name) + 1; + + s = (char *) ggc_alloc_atomic (name_len + map->new_len); + memcpy (s, map->new_prefix, map->new_len); + memcpy (s + map->new_len, name, name_len); ++ free (realname); + return s; + } +
Seems there is a downside to this, it breaks ruby and xen-tools:
https://autobuilder.yoctoproject.org/typhoon/#/builders/117/builds/1386/steps/12/logs/stdio
https://autobuilder.yoctoproject.org/typhoon/#/builders/128/builds/482/steps/12/logs/stdio
which both do:
#include __FILE__
I'm tempted to suggest we patch that to something different in both cases as the alternatives to debug handling are much worse.
Cheers,
Richard