On 7/27/21 6:36 AM, Tony Battersby wrote:
Backport ustream gcc patch that enables -fdebug-prefix-map to cover additional cases with LTO enabled to make LTO builds more reproducible. [YOCTO #14481] Signed-off-by: Tony Battersby <[email protected]> --- meta/recipes-devtools/gcc/gcc-11.1.inc | 1 + ...prefix-maps-before-checksumming-DIEs.patch | 95 +++++++++++++++++++ 2 files changed, 96 insertions(+) create mode 100644 meta/recipes-devtools/gcc/gcc/0041-apply-debug-prefix-maps-before-checksumming-DIEs.patch diff --git a/meta/recipes-devtools/gcc/gcc-11.1.inc b/meta/recipes-devtools/gcc/gcc-11.1.inc index c21242af58..167fddb65a 100644 --- a/meta/recipes-devtools/gcc/gcc-11.1.inc +++ b/meta/recipes-devtools/gcc/gcc-11.1.inc @@ -72,6 +72,7 @@ SRC_URI = "\ file://0038-arc-Update-64bit-move-split-patterns.patch \ file://0039-arc-Fix-u-maddhisi-patterns.patch \ file://0040-arc-Update-doloop_end-patterns.patch \ + file://0041-apply-debug-prefix-maps-before-checksumming-DIEs.patch \ " SRC_URI[sha256sum] = "4c4a6fb8a8396059241c2e674b85b351c26a5d678274007f076957afa1cc9ddf" SRC_URI[backports.sha256sum] = "69274bebd6c069a13443d4af61070e854740a639ec4d66eedf3e80070363587b" diff --git a/meta/recipes-devtools/gcc/gcc/0041-apply-debug-prefix-maps-before-checksumming-DIEs.patch b/meta/recipes-devtools/gcc/gcc/0041-apply-debug-prefix-maps-before-checksumming-DIEs.patch new file mode 100644 index 0000000000..c8dcd74b9d --- /dev/null +++ b/meta/recipes-devtools/gcc/gcc/0041-apply-debug-prefix-maps-before-checksumming-DIEs.patch @@ -0,0 +1,95 @@ +From 7cc2df084b7977653a9b59cbc34a9ad500ae619c Mon Sep 17 00:00:00 2001 +From: Richard Biener <[email protected]> +Date: Tue, 20 Jul 2021 11:00:33 +0200 +Subject: [PATCH] debug/101473 - apply debug prefix maps before checksumming DIEs + +The following makes sure to apply the debug prefix maps to filenames +before checksumming DIEs to create the global symbol for the CU DIE +used by LTO to link the late debug to the early debug. This avoids +binary differences (in said symbol) when compiling with toolchains +installed under a different path and that compensated with appropriate +-fdebug-prefix-map options. + +The easiest and most scalable way is to record both the unmapped +and the remapped filename in the dwarf_file_data so the remapping +process takes place at a single point and only once (otherwise it +creates GC garbage at each point doing that). + +2021-07-20 Richard Biener <[email protected]> + + PR debug/101473 + * dwarf2out.h (dwarf_file_data): Add key member. + * dwarf2out.c (dwarf_file_hasher::equal): Compare key. + (dwarf_file_hasher::hash): Hash key. + (lookup_filename): Remap the filename and store it in the + filename member of dwarf_file_data when creating a new + dwarf_file_data. + (file_name_acquire): Do not remap the filename again. + (maybe_emit_file): Likewise. + +[YOCTO #14481] + +Upstream-Status: Backport [https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=7cc2df084b7977653a9b59cbc34a9ad500ae619c]
looks ok.
+ +The upstream patch was modified to compensate for the definition of +"struct dwarf_file_data" being in dwarf2out.c rather than dwarf2out.h in +this version of gcc. + +Signed-off-by: Tony Battersby <[email protected]> +--- +diff -urpN a/gcc/dwarf2out.c b/gcc/dwarf2out.c +--- a/gcc/dwarf2out.c 2021-04-27 06:00:13.000000000 -0400 ++++ b/gcc/dwarf2out.c 2021-07-23 16:40:06.141886167 -0400 +@@ -1283,6 +1283,7 @@ dwarf2out_switch_text_section (void) + + /* Data about a single source file. */ + struct GTY((for_user)) dwarf_file_data { ++ const char * key; + const char * filename; + int emitted_number; + }; +@@ -12334,7 +12335,7 @@ file_name_acquire (dwarf_file_data **slo + + fi = fnad->files + fnad->used_files++; + +- f = remap_debug_filename (d->filename); ++ f = d->filename; + + /* Skip all leading "./". */ + while (f[0] == '.' && IS_DIR_SEPARATOR (f[1])) +@@ -27231,13 +27232,13 @@ dwarf2out_ignore_block (const_tree block + bool + dwarf_file_hasher::equal (dwarf_file_data *p1, const char *p2) + { +- return filename_cmp (p1->filename, p2) == 0; ++ return filename_cmp (p1->key, p2) == 0; + } + + hashval_t + dwarf_file_hasher::hash (dwarf_file_data *p) + { +- return htab_hash_string (p->filename); ++ return htab_hash_string (p->key); + } + + /* Lookup FILE_NAME (in the list of filenames that we know about here in +@@ -27267,7 +27268,8 @@ lookup_filename (const char *file_name) + return *slot; + + created = ggc_alloc<dwarf_file_data> (); +- created->filename = file_name; ++ created->key = file_name; ++ created->filename = remap_debug_filename (file_name); + created->emitted_number = 0; + *slot = created; + return created; +@@ -27293,8 +27295,7 @@ maybe_emit_file (struct dwarf_file_data + if (output_asm_line_debug_info ()) + { + fprintf (asm_out_file, "\t.file %u ", fd->emitted_number); +- output_quoted_string (asm_out_file, +- remap_debug_filename (fd->filename)); ++ output_quoted_string (asm_out_file, fd->filename); + fputc ('\n', asm_out_file); + } + }
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#154179): https://lists.openembedded.org/g/openembedded-core/message/154179 Mute This Topic: https://lists.openembedded.org/mt/84481685/21656 Group Owner: [email protected] Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]] -=-=-=-=-=-=-=-=-=-=-=-
