binutils-gdb/libctf/ctf-link.c

912 lines
26 KiB
C
Raw Normal View History

libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
/* CTF linking.
Copyright (C) 2019-2020 Free Software Foundation, Inc.
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
This file is part of libctf.
libctf is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free
Software Foundation; either version 3, or (at your option) any later
version.
This program is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; see the file COPYING. If not see
<http://www.gnu.org/licenses/>. */
#include <ctf-impl.h>
#include <string.h>
libctf: map from old to corresponding newly-added types in ctf_add_type This lets you call ctf_type_mapping (dest_fp, src_fp, src_type_id) and get told what type ID the corresponding type has in the target ctf_file_t. This works even if it was added by a recursive call, and because it is stored in the target ctf_file_t it works even if we had to add one type to multiple ctf_file_t's as part of conflicting type handling. We empty out this mapping after every archive is linked: because it maps input to output fps, and we only visit each input fp once, its contents are rendered entirely useless every time the source fp changes. v3: add several missing mapping additions. Add ctf_dynhash_empty, and empty after every input archive. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_file_t): New field ctf_link_type_mapping. (struct ctf_link_type_mapping_key): New. (ctf_hash_type_mapping_key): Likewise. (ctf_hash_eq_type_mapping_key): Likewise. (ctf_add_type_mapping): Likewise. (ctf_type_mapping): Likewise. (ctf_dynhash_empty): Likewise. * ctf-open.c (ctf_file_close): Update accordingly. * ctf-create.c (ctf_update): Likewise. (ctf_add_type): Populate the mapping. * ctf-hash.c (ctf_hash_type_mapping_key): Hash a type mapping key. (ctf_hash_eq_type_mapping_key): Check the key for equality. (ctf_dynhash_insert): Fix comment typo. (ctf_dynhash_empty): New. * ctf-link.c (ctf_add_type_mapping): New. (ctf_type_mapping): Likewise. (empty_link_type_mapping): New. (ctf_link_one_input_archive): Call it.
2019-07-13 22:31:26 +02:00
/* Type tracking machinery. */
/* Record the correspondence between a source and ctf_add_type()-added
destination type: both types are translated into parent type IDs if need be,
so they relate to the actual container they are in. Outside controlled
circumstances (like linking) it is probably not useful to do more than
compare these pointers, since there is nothing stopping the user closing the
source container whenever they want to.
Our OOM handling here is just to not do anything, because this is called deep
enough in the call stack that doing anything useful is painfully difficult:
the worst consequence if we do OOM is a bit of type duplication anyway. */
void
ctf_add_type_mapping (ctf_file_t *src_fp, ctf_id_t src_type,
ctf_file_t *dst_fp, ctf_id_t dst_type)
{
if (LCTF_TYPE_ISPARENT (src_fp, src_type) && src_fp->ctf_parent)
src_fp = src_fp->ctf_parent;
src_type = LCTF_TYPE_TO_INDEX(src_fp, src_type);
if (LCTF_TYPE_ISPARENT (dst_fp, dst_type) && dst_fp->ctf_parent)
dst_fp = dst_fp->ctf_parent;
dst_type = LCTF_TYPE_TO_INDEX(dst_fp, dst_type);
/* This dynhash is a bit tricky: it has a multivalued (structural) key, so we
need to use the sized-hash machinery to generate key hashing and equality
functions. */
if (dst_fp->ctf_link_type_mapping == NULL)
{
ctf_hash_fun f = ctf_hash_type_mapping_key;
ctf_hash_eq_fun e = ctf_hash_eq_type_mapping_key;
if ((dst_fp->ctf_link_type_mapping = ctf_dynhash_create (f, e, free,
NULL)) == NULL)
return;
}
ctf_link_type_mapping_key_t *key;
key = calloc (1, sizeof (struct ctf_link_type_mapping_key));
if (!key)
return;
key->cltm_fp = src_fp;
key->cltm_idx = src_type;
ctf_dynhash_insert (dst_fp->ctf_link_type_mapping, key,
(void *) (uintptr_t) dst_type);
}
/* Look up a type mapping: return 0 if none. The DST_FP is modified to point to
the parent if need be. The ID returned is from the dst_fp's perspective. */
ctf_id_t
ctf_type_mapping (ctf_file_t *src_fp, ctf_id_t src_type, ctf_file_t **dst_fp)
{
ctf_link_type_mapping_key_t key;
ctf_file_t *target_fp = *dst_fp;
ctf_id_t dst_type = 0;
if (LCTF_TYPE_ISPARENT (src_fp, src_type) && src_fp->ctf_parent)
src_fp = src_fp->ctf_parent;
src_type = LCTF_TYPE_TO_INDEX(src_fp, src_type);
key.cltm_fp = src_fp;
key.cltm_idx = src_type;
if (target_fp->ctf_link_type_mapping)
dst_type = (uintptr_t) ctf_dynhash_lookup (target_fp->ctf_link_type_mapping,
&key);
if (dst_type != 0)
{
dst_type = LCTF_INDEX_TO_TYPE (target_fp, dst_type,
target_fp->ctf_parent != NULL);
*dst_fp = target_fp;
return dst_type;
}
if (target_fp->ctf_parent)
target_fp = target_fp->ctf_parent;
else
return 0;
if (target_fp->ctf_link_type_mapping)
dst_type = (uintptr_t) ctf_dynhash_lookup (target_fp->ctf_link_type_mapping,
&key);
if (dst_type)
dst_type = LCTF_INDEX_TO_TYPE (target_fp, dst_type,
target_fp->ctf_parent != NULL);
*dst_fp = target_fp;
return dst_type;
}
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
/* Linker machinery.
CTF linking consists of adding CTF archives full of content to be merged into
this one to the current file (which must be writable) by calling
ctf_link_add_ctf(). Once this is done, a call to ctf_link() will merge the
type tables together, generating new CTF files as needed, with this one as a
parent, to contain types from the inputs which conflict.
ctf_link_add_strtab() takes a callback which provides string/offset pairs to
be added to the external symbol table and deduplicated from all CTF string
tables in the output link; ctf_link_shuffle_syms() takes a callback which
provides symtab entries in ascending order, and shuffles the function and
data sections to match; and ctf_link_write() emits a CTF file (if there are
no conflicts requiring per-compilation-unit sub-CTF files) or CTF archives
(otherwise) and returns it, suitable for addition in the .ctf section of the
output. */
/* Add a file to a link. */
static void ctf_arc_close_thunk (void *arc)
{
ctf_arc_close ((ctf_archive_t *) arc);
}
static void ctf_file_close_thunk (void *file)
{
ctf_file_close ((ctf_file_t *) file);
}
int
ctf_link_add_ctf (ctf_file_t *fp, ctf_archive_t *ctf, const char *name)
{
char *dupname = NULL;
if (fp->ctf_link_outputs)
return (ctf_set_errno (fp, ECTF_LINKADDEDLATE));
if (fp->ctf_link_inputs == NULL)
fp->ctf_link_inputs = ctf_dynhash_create (ctf_hash_string,
ctf_hash_eq_string, free,
ctf_arc_close_thunk);
if (fp->ctf_link_inputs == NULL)
goto oom;
if ((dupname = strdup (name)) == NULL)
goto oom;
if (ctf_dynhash_insert (fp->ctf_link_inputs, dupname, ctf) < 0)
goto oom;
return 0;
oom:
free (fp->ctf_link_inputs);
fp->ctf_link_inputs = NULL;
free (dupname);
return (ctf_set_errno (fp, ENOMEM));
}
/* Return a per-CU output CTF dictionary suitable for the given CU, creating and
interning it if need be. */
static ctf_file_t *
ctf_create_per_cu (ctf_file_t *fp, const char *filename, const char *cuname)
{
ctf_file_t *cu_fp;
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
const char *ctf_name = NULL;
char *dynname = NULL;
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
/* First, check the mapping table and translate the per-CU name we use
accordingly. We check both the input filename and the CU name. Only if
neither are set do we fall back to the input filename as the per-CU
dictionary name. We prefer the filename because this is easier for likely
callers to determine. */
if (fp->ctf_link_cu_mapping)
{
if (((ctf_name = ctf_dynhash_lookup (fp->ctf_link_cu_mapping, filename)) == NULL) &&
((ctf_name = ctf_dynhash_lookup (fp->ctf_link_cu_mapping, cuname)) == NULL))
ctf_name = filename;
}
if (ctf_name == NULL)
ctf_name = filename;
if ((cu_fp = ctf_dynhash_lookup (fp->ctf_link_outputs, ctf_name)) == NULL)
{
int err;
if ((cu_fp = ctf_create (&err)) == NULL)
{
ctf_dprintf ("Cannot create per-CU CTF archive for CU %s from "
"input file %s: %s\n", cuname, filename,
ctf_errmsg (err));
ctf_set_errno (fp, err);
return NULL;
}
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
if ((dynname = strdup (ctf_name)) == NULL)
goto oom;
if (ctf_dynhash_insert (fp->ctf_link_outputs, dynname, cu_fp) < 0)
goto oom;
ctf_import (cu_fp, fp);
ctf_cuname_set (cu_fp, cuname);
ctf_parent_name_set (cu_fp, _CTF_SECTION);
}
return cu_fp;
oom:
free (dynname);
ctf_file_close (cu_fp);
ctf_set_errno (fp, ENOMEM);
return NULL;
}
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
/* Add a mapping directing that the CU named FROM should have its
conflicting/non-duplicate types (depending on link mode) go into a container
named TO. Many FROMs can share a TO: in this case, the effect on conflicting
types is not yet defined (but in time an auto-renaming algorithm will be
added: ugly, but there is really no right thing one can do in this
situation).
We forcibly add a container named TO in every case, even though it may well
wind up empty, because clients that use this facility usually expect to find
every TO container present, even if empty, and malfunction otherwise. */
int
ctf_link_add_cu_mapping (ctf_file_t *fp, const char *from, const char *to)
{
int err;
char *f, *t;
if (fp->ctf_link_cu_mapping == NULL)
fp->ctf_link_cu_mapping = ctf_dynhash_create (ctf_hash_string,
ctf_hash_eq_string, free,
free);
if (fp->ctf_link_cu_mapping == NULL)
return ctf_set_errno (fp, ENOMEM);
if (fp->ctf_link_outputs == NULL)
fp->ctf_link_outputs = ctf_dynhash_create (ctf_hash_string,
ctf_hash_eq_string, free,
ctf_file_close_thunk);
if (fp->ctf_link_outputs == NULL)
return ctf_set_errno (fp, ENOMEM);
f = strdup (from);
t = strdup (to);
if (!f || !t)
goto oom;
if (ctf_create_per_cu (fp, t, t) == NULL)
goto oom_noerrno; /* Errno is set for us. */
err = ctf_dynhash_insert (fp->ctf_link_cu_mapping, f, t);
if (err)
{
ctf_set_errno (fp, err);
goto oom_noerrno;
}
return 0;
oom:
ctf_set_errno (fp, errno);
oom_noerrno:
free (f);
free (t);
return -1;
}
/* Set a function which is called to transform the names of archive members.
This is useful for applying regular transformations to many names, where
ctf_link_add_cu_mapping applies arbitrarily irregular changes to single
names. The member name changer is applied at ctf_link_write time, so it
cannot conflate multiple CUs into one the way ctf_link_add_cu_mapping can.
The changer function accepts a name and should return a new
dynamically-allocated name, or NULL if the name should be left unchanged. */
void
ctf_link_set_memb_name_changer (ctf_file_t *fp,
ctf_link_memb_name_changer_f *changer,
void *arg)
{
fp->ctf_link_memb_name_changer = changer;
fp->ctf_link_memb_name_changer_arg = arg;
}
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
typedef struct ctf_link_in_member_cb_arg
{
ctf_file_t *out_fp;
const char *file_name;
ctf_file_t *in_fp;
ctf_file_t *main_input_fp;
const char *cu_name;
char *arcname;
int done_main_member;
int share_mode;
int in_input_cu_file;
} ctf_link_in_member_cb_arg_t;
/* Link one type into the link. We rely on ctf_add_type() to detect
duplicates. This is not terribly reliable yet (unnmamed types will be
mindlessly duplicated), but will improve shortly. */
static int
ctf_link_one_type (ctf_id_t type, int isroot _libctf_unused_, void *arg_)
{
ctf_link_in_member_cb_arg_t *arg = (ctf_link_in_member_cb_arg_t *) arg_;
ctf_file_t *per_cu_out_fp;
int err;
if (arg->share_mode != CTF_LINK_SHARE_UNCONFLICTED)
{
ctf_dprintf ("Share-duplicated mode not yet implemented.\n");
return ctf_set_errno (arg->out_fp, ECTF_NOTYET);
}
/* Simply call ctf_add_type: if it reports a conflict and we're adding to the
main CTF file, add to the per-CU archive member instead, creating it if
necessary. If we got this type from a per-CU archive member, add it
straight back to the corresponding member in the output. */
if (!arg->in_input_cu_file)
{
if (ctf_add_type (arg->out_fp, arg->in_fp, type) != CTF_ERR)
return 0;
err = ctf_errno (arg->out_fp);
if (err != ECTF_CONFLICT)
{
libctf: handle nonrepresentable types at link time GCC can emit references to type 0 to indicate that this type is one that is not representable in the version of CTF it emits (for instance, version 3 cannot encode vector types). Type 0 is already used in the function section to indicate padding inserted to skip functions we do not want to encode the type of, so using zero in this way is a good extension of the format: but libctf reports such types as ECTF_BADID, which is indistinguishable from file corruption via links to truly nonexistent types with IDs like 0xDEADBEEF etc, which we really do want to stop for. In particular, this stops all traversals of types dead at this point, preventing us from even dumping CTF files containing unrepresentable types to see what's going on! So add a new error, ECTF_NONREPRESENTABLE, which is returned by recursive type resolution when a reference to a zero type is found. (No zero type is ever emitted into the CTF file by GCC, only references to one). We can't do much with types that are ultimately nonrepresentable, but we can do enough to keep functioning. Adjust ctf_add_type to ensure that top-level types of type zero and structure and union members of ultimate type zero are simply skipped without reporting an error, so we can copy structures and unions that contain nonrepresentable members (skipping them and leaving a hole where they would be, so no consumers downstream of the linker need to worry about this): adjust the dumper so that we dump members of nonrepresentable types in a simple form that indicates nonrepresentability rather than terminating the dump, and do not falsely assume all errors to be -ENOMEM: adjust the linker so that types that fail to get added are simply skipped, so that both nonrepresentable types and outright errors do not terminate the type addition, which could skip many valid types and cause further errors when variables of those types are added. In future, when we gain the ability to call back to the linker to report link-time type resolution errors, we should report failures to add all but nonrepresentable types. But we can't do that yet. v5: Fix tabdamage. include/ * ctf-api.h (ECTF_NONREPRESENTABLE): New. libctf/ * ctf-types.c (ctf_type_resolve): Return ECTF_NONREPRESENTABLE on type zero. * ctf-create.c (ctf_add_type): Detect and skip nonrepresentable members and types. (ctf_add_variable): Likewise for variables pointing to them. * ctf-link.c (ctf_link_one_type): Do not warn for nonrepresentable type link failure, but do warn for others. * ctf-dump.c (ctf_dump_format_type): Likewise. Do not assume all errors to be ENOMEM. (ctf_dump_member): Likewise. (ctf_dump_type): Likewise. (ctf_dump_header_strfield): Do not assume all errors to be ENOMEM. (ctf_dump_header_sectfield): Do not assume all errors to be ENOMEM. (ctf_dump_header): Likewise. (ctf_dump_label): likewise. (ctf_dump_objts): likewise. (ctf_dump_funcs): likewise. (ctf_dump_var): likewise. (ctf_dump_str): Likewise.
2019-08-05 12:40:33 +02:00
if (err != ECTF_NONREPRESENTABLE)
ctf_dprintf ("Cannot link type %lx from archive member %s, input file %s "
"into output link: %s\n", type, arg->arcname, arg->file_name,
ctf_errmsg (err));
/* We must ignore this problem or we end up losing future types, then
trying to link the variables in, then exploding. Better to link as
much as possible. XXX when we add a proper link warning
infrastructure, we should report the error here! */
return 0;
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
}
ctf_set_errno (arg->out_fp, 0);
}
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
if ((per_cu_out_fp = ctf_create_per_cu (arg->out_fp, arg->file_name,
arg->cu_name)) == NULL)
return -1; /* Errno is set for us. */
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
if (ctf_add_type (per_cu_out_fp, arg->in_fp, type) != CTF_ERR)
return 0;
err = ctf_errno (per_cu_out_fp);
libctf: handle nonrepresentable types at link time GCC can emit references to type 0 to indicate that this type is one that is not representable in the version of CTF it emits (for instance, version 3 cannot encode vector types). Type 0 is already used in the function section to indicate padding inserted to skip functions we do not want to encode the type of, so using zero in this way is a good extension of the format: but libctf reports such types as ECTF_BADID, which is indistinguishable from file corruption via links to truly nonexistent types with IDs like 0xDEADBEEF etc, which we really do want to stop for. In particular, this stops all traversals of types dead at this point, preventing us from even dumping CTF files containing unrepresentable types to see what's going on! So add a new error, ECTF_NONREPRESENTABLE, which is returned by recursive type resolution when a reference to a zero type is found. (No zero type is ever emitted into the CTF file by GCC, only references to one). We can't do much with types that are ultimately nonrepresentable, but we can do enough to keep functioning. Adjust ctf_add_type to ensure that top-level types of type zero and structure and union members of ultimate type zero are simply skipped without reporting an error, so we can copy structures and unions that contain nonrepresentable members (skipping them and leaving a hole where they would be, so no consumers downstream of the linker need to worry about this): adjust the dumper so that we dump members of nonrepresentable types in a simple form that indicates nonrepresentability rather than terminating the dump, and do not falsely assume all errors to be -ENOMEM: adjust the linker so that types that fail to get added are simply skipped, so that both nonrepresentable types and outright errors do not terminate the type addition, which could skip many valid types and cause further errors when variables of those types are added. In future, when we gain the ability to call back to the linker to report link-time type resolution errors, we should report failures to add all but nonrepresentable types. But we can't do that yet. v5: Fix tabdamage. include/ * ctf-api.h (ECTF_NONREPRESENTABLE): New. libctf/ * ctf-types.c (ctf_type_resolve): Return ECTF_NONREPRESENTABLE on type zero. * ctf-create.c (ctf_add_type): Detect and skip nonrepresentable members and types. (ctf_add_variable): Likewise for variables pointing to them. * ctf-link.c (ctf_link_one_type): Do not warn for nonrepresentable type link failure, but do warn for others. * ctf-dump.c (ctf_dump_format_type): Likewise. Do not assume all errors to be ENOMEM. (ctf_dump_member): Likewise. (ctf_dump_type): Likewise. (ctf_dump_header_strfield): Do not assume all errors to be ENOMEM. (ctf_dump_header_sectfield): Do not assume all errors to be ENOMEM. (ctf_dump_header): Likewise. (ctf_dump_label): likewise. (ctf_dump_objts): likewise. (ctf_dump_funcs): likewise. (ctf_dump_var): likewise. (ctf_dump_str): Likewise.
2019-08-05 12:40:33 +02:00
if (err != ECTF_NONREPRESENTABLE)
ctf_dprintf ("Cannot link type %lx from CTF archive member %s, input file %s "
"into output per-CU CTF archive member %s: %s: skipped\n", type,
arg->arcname, arg->file_name, arg->arcname,
ctf_errmsg (err));
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
if (err == ECTF_CONFLICT)
/* Conflicts are possible at this stage only if a non-ld user has combined
multiple TUs into a single output dictionary. Even in this case we do not
want to stop the link or propagate the error. */
ctf_set_errno (arg->out_fp, 0);
return 0; /* As above: do not lose types. */
}
/* Check if we can safely add a variable with the given type to this container. */
static int
check_variable (const char *name, ctf_file_t *fp, ctf_id_t type,
ctf_dvdef_t **out_dvd)
{
ctf_dvdef_t *dvd;
dvd = ctf_dynhash_lookup (fp->ctf_dvhash, name);
*out_dvd = dvd;
if (!dvd)
return 1;
if (dvd->dvd_type != type)
{
/* Variable here. Wrong type: cannot add. Just skip it, because there is
no way to express this in CTF. (This might be the parent, in which
case we'll try adding in the child first, and only then give up.) */
ctf_dprintf ("Inexpressible duplicate variable %s skipped.\n", name);
}
return 0; /* Already exists. */
}
/* Link one variable in. */
static int
ctf_link_one_variable (const char *name, ctf_id_t type, void *arg_)
{
ctf_link_in_member_cb_arg_t *arg = (ctf_link_in_member_cb_arg_t *) arg_;
ctf_file_t *per_cu_out_fp;
ctf_id_t dst_type = 0;
ctf_file_t *check_fp;
ctf_dvdef_t *dvd;
/* In unconflicted link mode, if this type is mapped to a type in the parent
container, we want to try to add to that first: if it reports a duplicate,
or if the type is in a child already, add straight to the child. */
check_fp = arg->out_fp;
dst_type = ctf_type_mapping (arg->in_fp, type, &check_fp);
if (dst_type != 0)
{
if (check_fp == arg->out_fp)
{
if (check_variable (name, check_fp, dst_type, &dvd))
{
/* No variable here: we can add it. */
if (ctf_add_variable (check_fp, name, dst_type) < 0)
return (ctf_set_errno (arg->out_fp, ctf_errno (check_fp)));
return 0;
}
/* Already present? Nothing to do. */
if (dvd && dvd->dvd_type == type)
return 0;
}
}
/* Can't add to the parent due to a name clash, or because it references a
type only present in the child. Try adding to the child, creating if need
be. */
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
if ((per_cu_out_fp = ctf_create_per_cu (arg->out_fp, arg->file_name,
arg->cu_name)) == NULL)
return -1; /* Errno is set for us. */
/* If the type was not found, check for it in the child too. */
if (dst_type == 0)
{
check_fp = per_cu_out_fp;
dst_type = ctf_type_mapping (arg->in_fp, type, &check_fp);
if (dst_type == 0)
{
ctf_dprintf ("Type %lx for variable %s in input file %s not "
"found: skipped.\n", type, name, arg->file_name);
/* Do not terminate the link: just skip the variable. */
return 0;
}
}
if (check_variable (name, per_cu_out_fp, dst_type, &dvd))
if (ctf_add_variable (per_cu_out_fp, name, dst_type) < 0)
return (ctf_set_errno (arg->out_fp, ctf_errno (per_cu_out_fp)));
return 0;
}
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
/* Merge every type and variable in this archive member into the link, so we can
relink things that have already had ld run on them. We use the archive
member name, sans any leading '.ctf.', as the CU name for ambiguous types if
there is one and it's not the default: otherwise, we use the name of the
input file. */
static int
ctf_link_one_input_archive_member (ctf_file_t *in_fp, const char *name, void *arg_)
{
ctf_link_in_member_cb_arg_t *arg = (ctf_link_in_member_cb_arg_t *) arg_;
int err = 0;
if (strcmp (name, _CTF_SECTION) == 0)
{
/* This file is the default member of this archive, and has already been
explicitly processed.
In the default sharing mode of CTF_LINK_SHARE_UNCONFLICTED, it does no
harm to rescan an existing shared repo again: all the types will just
end up in the same place. But in CTF_LINK_SHARE_DUPLICATED mode, this
causes the system to erroneously conclude that all types are duplicated
and should be shared, even if they are not. */
if (arg->done_main_member)
return 0;
arg->arcname = strdup (".ctf.");
if (arg->arcname)
{
char *new_name;
new_name = ctf_str_append (arg->arcname, arg->file_name);
if (new_name)
arg->arcname = new_name;
else
free (arg->arcname);
}
}
else
{
arg->arcname = strdup (name);
/* Get ambiguous types from our parent. */
ctf_import (in_fp, arg->main_input_fp);
arg->in_input_cu_file = 1;
}
if (!arg->arcname)
return ctf_set_errno (in_fp, ENOMEM);
arg->cu_name = name;
if (strncmp (arg->cu_name, ".ctf.", strlen (".ctf.")) == 0)
arg->cu_name += strlen (".ctf.");
arg->in_fp = in_fp;
if ((err = ctf_type_iter_all (in_fp, ctf_link_one_type, arg)) > -1)
err = ctf_variable_iter (in_fp, ctf_link_one_variable, arg);
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
arg->in_input_cu_file = 0;
free (arg->arcname);
if (err < 0)
return -1; /* Errno is set for us. */
return 0;
}
libctf: map from old to corresponding newly-added types in ctf_add_type This lets you call ctf_type_mapping (dest_fp, src_fp, src_type_id) and get told what type ID the corresponding type has in the target ctf_file_t. This works even if it was added by a recursive call, and because it is stored in the target ctf_file_t it works even if we had to add one type to multiple ctf_file_t's as part of conflicting type handling. We empty out this mapping after every archive is linked: because it maps input to output fps, and we only visit each input fp once, its contents are rendered entirely useless every time the source fp changes. v3: add several missing mapping additions. Add ctf_dynhash_empty, and empty after every input archive. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_file_t): New field ctf_link_type_mapping. (struct ctf_link_type_mapping_key): New. (ctf_hash_type_mapping_key): Likewise. (ctf_hash_eq_type_mapping_key): Likewise. (ctf_add_type_mapping): Likewise. (ctf_type_mapping): Likewise. (ctf_dynhash_empty): Likewise. * ctf-open.c (ctf_file_close): Update accordingly. * ctf-create.c (ctf_update): Likewise. (ctf_add_type): Populate the mapping. * ctf-hash.c (ctf_hash_type_mapping_key): Hash a type mapping key. (ctf_hash_eq_type_mapping_key): Check the key for equality. (ctf_dynhash_insert): Fix comment typo. (ctf_dynhash_empty): New. * ctf-link.c (ctf_add_type_mapping): New. (ctf_type_mapping): Likewise. (empty_link_type_mapping): New. (ctf_link_one_input_archive): Call it.
2019-07-13 22:31:26 +02:00
/* Dump the unnecessary link type mapping after one input file is processed. */
static void
empty_link_type_mapping (void *key _libctf_unused_, void *value,
void *arg _libctf_unused_)
{
ctf_file_t *fp = (ctf_file_t *) value;
if (fp->ctf_link_type_mapping)
ctf_dynhash_empty (fp->ctf_link_type_mapping);
}
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
/* Link one input file's types into the output file. */
static void
ctf_link_one_input_archive (void *key, void *value, void *arg_)
{
const char *file_name = (const char *) key;
ctf_archive_t *arc = (ctf_archive_t *) value;
ctf_link_in_member_cb_arg_t *arg = (ctf_link_in_member_cb_arg_t *) arg_;
int err;
arg->file_name = file_name;
arg->done_main_member = 0;
if ((arg->main_input_fp = ctf_arc_open_by_name (arc, NULL, &err)) == NULL)
if (err != ECTF_ARNNAME)
{
ctf_dprintf ("Cannot open main archive member in input file %s in the "
"link: skipping: %s.\n", arg->file_name,
ctf_errmsg (err));
return;
}
if (ctf_link_one_input_archive_member (arg->main_input_fp,
_CTF_SECTION, arg) < 0)
{
ctf_file_close (arg->main_input_fp);
return;
}
arg->done_main_member = 1;
if (ctf_archive_iter (arc, ctf_link_one_input_archive_member, arg) < 0)
ctf_dprintf ("Cannot traverse archive in input file %s: link "
"cannot continue: %s.\n", arg->file_name,
ctf_errmsg (ctf_errno (arg->out_fp)));
else
{
/* The only error indication to the caller is the errno: so ensure that it
is zero if there was no actual error from the caller. */
ctf_set_errno (arg->out_fp, 0);
}
ctf_file_close (arg->main_input_fp);
libctf: map from old to corresponding newly-added types in ctf_add_type This lets you call ctf_type_mapping (dest_fp, src_fp, src_type_id) and get told what type ID the corresponding type has in the target ctf_file_t. This works even if it was added by a recursive call, and because it is stored in the target ctf_file_t it works even if we had to add one type to multiple ctf_file_t's as part of conflicting type handling. We empty out this mapping after every archive is linked: because it maps input to output fps, and we only visit each input fp once, its contents are rendered entirely useless every time the source fp changes. v3: add several missing mapping additions. Add ctf_dynhash_empty, and empty after every input archive. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_file_t): New field ctf_link_type_mapping. (struct ctf_link_type_mapping_key): New. (ctf_hash_type_mapping_key): Likewise. (ctf_hash_eq_type_mapping_key): Likewise. (ctf_add_type_mapping): Likewise. (ctf_type_mapping): Likewise. (ctf_dynhash_empty): Likewise. * ctf-open.c (ctf_file_close): Update accordingly. * ctf-create.c (ctf_update): Likewise. (ctf_add_type): Populate the mapping. * ctf-hash.c (ctf_hash_type_mapping_key): Hash a type mapping key. (ctf_hash_eq_type_mapping_key): Check the key for equality. (ctf_dynhash_insert): Fix comment typo. (ctf_dynhash_empty): New. * ctf-link.c (ctf_add_type_mapping): New. (ctf_type_mapping): Likewise. (empty_link_type_mapping): New. (ctf_link_one_input_archive): Call it.
2019-07-13 22:31:26 +02:00
/* Discard the now-unnecessary mapping table data. */
if (arg->out_fp->ctf_link_type_mapping)
ctf_dynhash_empty (arg->out_fp->ctf_link_type_mapping);
ctf_dynhash_iter (arg->out_fp->ctf_link_outputs, empty_link_type_mapping, NULL);
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
}
/* Merge types and variable sections in all files added to the link
together. */
int
ctf_link (ctf_file_t *fp, int share_mode)
{
ctf_link_in_member_cb_arg_t arg;
memset (&arg, 0, sizeof (struct ctf_link_in_member_cb_arg));
arg.out_fp = fp;
arg.share_mode = share_mode;
if (fp->ctf_link_inputs == NULL)
return 0; /* Nothing to do. */
if (fp->ctf_link_outputs == NULL)
fp->ctf_link_outputs = ctf_dynhash_create (ctf_hash_string,
ctf_hash_eq_string, free,
ctf_file_close_thunk);
if (fp->ctf_link_outputs == NULL)
return ctf_set_errno (fp, ENOMEM);
ctf_dynhash_iter (fp->ctf_link_inputs, ctf_link_one_input_archive,
&arg);
if (ctf_errno (fp) != 0)
return -1;
return 0;
}
typedef struct ctf_link_out_string_cb_arg
{
const char *str;
uint32_t offset;
int err;
} ctf_link_out_string_cb_arg_t;
/* Intern a string in the string table of an output per-CU CTF file. */
static void
ctf_link_intern_extern_string (void *key _libctf_unused_, void *value,
void *arg_)
{
ctf_file_t *fp = (ctf_file_t *) value;
ctf_link_out_string_cb_arg_t *arg = (ctf_link_out_string_cb_arg_t *) arg_;
fp->ctf_flags |= LCTF_DIRTY;
libctf: avoid the need to ever use ctf_update The method of operation of libctf when the dictionary is writable has before now been that types that are added land in the dynamic type section, which is a linked list and hash of IDs -> dynamic type definitions (and, recently a hash of names): the DTDs are a bit of CTF representing the ctf_type_t and ad hoc C structures representing the vlen. Historically, libctf was unable to do anything with these types, not even look them up by ID, let alone by name: if you wanted to do that say if you were adding a type that depended on one you just added) you called ctf_update, which serializes all the DTDs into a CTF file and reopens it, copying its guts over the fp it's called with. The ctf_updated types are then frozen in amber and unchangeable: all lookups will return the types in the static portion in preference to the dynamic portion, and we will refuse to re-add things that already exist in the static portion (and, of late, in the dynamic portion too). The libctf machinery remembers the boundary between static and dynamic types and looks in the right portion for each type. Lots of things still don't quite work with dynamic types (e.g. getting their size), but enough works to do a bunch of additions and then a ctf_update, most of the time. Except it doesn't, because ctf_add_type finds it necessary to walk the full dynamic type definition list looking for types with matching names, so it gets slower and slower with every type you add: fixing this requires calling ctf_update periodically for no other reason than to avoid massively slowing things down. This is all clunky and very slow but kind of works, until you consider that it is in fact possible and indeed necessary to modify one sort of type after it has been added: forwards. These are necessarily promoted to structs, unions or enums, and when they do so *their type ID does not change*. So all of a sudden we are changing types that already exist in the static portion. ctf_update gets massively confused by this and allocates space enough for the forward (with no members), but then emits the new dynamic type (with all the members) into it. You get an assertion failure after that, if you're lucky, or a coredump. So this commit rejigs things a bit and arranges to exclusively use the dynamic type definitions in writable dictionaries, and the static type definitions in readable dictionaries: we don't at any time have a mixture of static and dynamic types, and you don't need to call ctf_update to make things "appear". The ctf_dtbyname hash I introduced a few months ago, which maps things like "struct foo" to DTDs, is removed, replaced instead by a change of type of the four dictionaries which track names. Rather than just being (unresizable) ctf_hash_t's populated only at ctf_bufopen time, they are now a ctf_names_t structure, which is a pair of ctf_hash_t and ctf_dynhash_t, with the ctf_hash_t portion being used in readonly dictionaries, and the ctf_dynhash_t being used in writable ones. The decision as to which to use is centralized in the new functions ctf_lookup_by_rawname (which takes a type kind) and ctf_lookup_by_rawhash, which it calls (which takes a ctf_names_t *.) This change lets us switch from using static to dynamic name hashes on the fly across the entirety of libctf without complexifying anything: in fact, because we now centralize the knowledge about how to map from type kind to name hash, it actually simplifies things and lets us throw out quite a lot of now-unnecessary complexity, from ctf_dtnyname (replaced by the dynamic half of the name tables), through to ctf_dtnextid (now that a dictionary's static portion is never referenced if the dictionary is writable, we can just use ctf_typemax to indicate the maximum type: dynamic or non-dynamic does not matter, and we no longer need to track the boundary between the types). You can now ctf_rollback() as far as you like, even past a ctf_update or for that matter a full writeout; all the iteration functions work just as well on writable as on read-only dictionaries; ctf_add_type no longer needs expensive duplicated code to run over the dynamic types hunting for ones it might be interested in; and the linker no longer needs a hack to call ctf_update so that calling ctf_add_type is not impossibly expensive. There is still a bit more complexity: some new code paths in ctf-types.c need to know how to extract information from dynamic types. This complexity will go away again in a few months when libctf acquires a proper intermediate representation. You can still call ctf_update if you like (it's public API, after all), but its only effect now is to set the point to which ctf_discard rolls back. Obviously *something* still needs to serialize the CTF file before writeout, and this job is done by ctf_serialize, which does everything ctf_update used to except set the counter used by ctf_discard. It is automatically called by the various functions that do CTF writeout: nobody else ever needs to call it. With this in place, forwards that are promoted to non-forwards no longer crash the link, even if it happens tens of thousands of types later. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_names_t): New. (ctf_lookup_t) <ctf_hash>: Now a ctf_names_t, not a ctf_hash_t. (ctf_file_t) <ctf_structs>: Likewise. <ctf_unions>: Likewise. <ctf_enums>: Likewise. <ctf_names>: Likewise. <ctf_lookups>: Improve comment. <ctf_ptrtab_len>: New. <ctf_prov_strtab>: New. <ctf_str_prov_offset>: New. <ctf_dtbyname>: Remove, redundant to the names hashes. <ctf_dtnextid>: Remove, redundant to ctf_typemax. (ctf_dtdef_t) <dtd_name>: Remove. <dtd_data>: Note that the ctt_name is now populated. (ctf_str_atom_t) <csa_offset>: This is now the strtab offset for internal strings too. <csa_external_offset>: New, the external strtab offset. (CTF_INDEX_TO_TYPEPTR): Handle the LCTF_RDWR case. (ctf_name_table): New declaration. (ctf_lookup_by_rawname): Likewise. (ctf_lookup_by_rawhash): Likewise. (ctf_set_ctl_hashes): Likewise. (ctf_serialize): Likewise. (ctf_dtd_insert): Adjust. (ctf_simple_open_internal): Likewise. (ctf_bufopen_internal): Likewise. (ctf_list_empty_p): Likewise. (ctf_str_remove_ref): Likewise. (ctf_str_add): Returns uint32_t now. (ctf_str_add_ref): Likewise. (ctf_str_add_external): Now returns a boolean (int). * ctf-string.c (ctf_strraw_explicit): Check the ctf_prov_strtab for strings in the appropriate range. (ctf_str_create_atoms): Create the ctf_prov_strtab. Detect OOM when adding the null string to the new strtab. (ctf_str_free_atoms): Destroy the ctf_prov_strtab. (ctf_str_add_ref_internal): Add make_provisional argument. If make_provisional, populate the offset and fill in the ctf_prov_strtab accordingly. (ctf_str_add): Return the offset, not the string. (ctf_str_add_ref): Likewise. (ctf_str_add_external): Return a success integer. (ctf_str_remove_ref): New, remove a single ref. (ctf_str_count_strtab): Do not count the initial null string's length or the existence or length of any unreferenced internal atoms. (ctf_str_populate_sorttab): Skip atoms with no refs. (ctf_str_write_strtab): Populate the nullstr earlier. Add one to the cts_len for the null string, since it is no longer done in ctf_str_count_strtab. Adjust for csa_external_offset rename. Populate the csa_offset for both internal and external cases. Flush the ctf_prov_strtab afterwards, and reset the ctf_str_prov_offset. * ctf-create.c (ctf_grow_ptrtab): New. (ctf_create): Call it. Initialize new fields rather than old ones. Tell ctf_bufopen_internal that this is a writable dictionary. Set the ctl hashes and data model. (ctf_update): Rename to... (ctf_serialize): ... this. Leave a compatibility function behind. Tell ctf_simple_open_internal that this is a writable dictionary. Pass the new fields along from the old dictionary. Drop ctf_dtnextid and ctf_dtbyname. Use ctf_strraw, not dtd_name. Do not zero out the DTD's ctt_name. (ctf_prefixed_name): Rename to... (ctf_name_table): ... this. No longer return a prefixed name: return the applicable name table instead. (ctf_dtd_insert): Use it, and use the right name table. Pass in the kind we're adding. Migrate away from dtd_name. (ctf_dtd_delete): Adjust similarly. Remove the ref to the deleted ctt_name. (ctf_dtd_lookup_type_by_name): Remove. (ctf_dynamic_type): Always return NULL on read-only dictionaries. No longer check ctf_dtnextid: check ctf_typemax instead. (ctf_snapshot): No longer use ctf_dtnextid: use ctf_typemax instead. (ctf_rollback): Likewise. No longer fail with ECTF_OVERROLLBACK. Use ctf_name_table and the right name table, and migrate away from dtd_name as in ctf_dtd_delete. (ctf_add_generic): Pass in the kind explicitly and pass it to ctf_dtd_insert. Use ctf_typemax, not ctf_dtnextid. Migrate away from dtd_name to using ctf_str_add_ref to populate the ctt_name. Grow the ptrtab if needed. (ctf_add_encoded): Pass in the kind. (ctf_add_slice): Likewise. (ctf_add_array): Likewise. (ctf_add_function): Likewise. (ctf_add_typedef): Likewise. (ctf_add_reftype): Likewise. Initialize the ctf_ptrtab, checking ctt_name rather than dtd_name. (ctf_add_struct_sized): Pass in the kind. Use ctf_lookup_by_rawname, not ctf_hash_lookup_type / ctf_dtd_lookup_type_by_name. (ctf_add_union_sized): Likewise. (ctf_add_enum): Likewise. (ctf_add_enum_encoded): Likewise. (ctf_add_forward): Likewise. (ctf_add_type): Likewise. (ctf_compress_write): Call ctf_serialize: adjust for ctf_size not being initialized until after the call. (ctf_write_mem): Likewise. (ctf_write): Likewise. * ctf-archive.c (arc_write_one_ctf): Likewise. * ctf-lookup.c (ctf_lookup_by_name): Use ctf_lookuup_by_rawhash, not ctf_hash_lookup_type. (ctf_lookup_by_id): No longer check the readonly types if the dictionary is writable. * ctf-open.c (init_types): Assert that this dictionary is not writable. Adjust to use the new name hashes, ctf_name_table, and ctf_ptrtab_len. GNU style fix for the final ptrtab scan. (ctf_bufopen_internal): New 'writable' parameter. Flip on LCTF_RDWR if set. Drop out early when dictionary is writable. Split the ctf_lookups initialization into... (ctf_set_cth_hashes): ... this new function. (ctf_simple_open_internal): Adjust. New 'writable' parameter. (ctf_simple_open): Adjust accordingly. (ctf_bufopen): Likewise. (ctf_file_close): Destroy the appropriate name hashes. No longer destroy ctf_dtbyname, which is gone. (ctf_getdatasect): Remove spurious "extern". * ctf-types.c (ctf_lookup_by_rawname): New, look up types in the specified name table, given a kind. (ctf_lookup_by_rawhash): Likewise, given a ctf_names_t *. (ctf_member_iter): Add support for iterating over the dynamic type list. (ctf_enum_iter): Likewise. (ctf_variable_iter): Likewise. (ctf_type_rvisit): Likewise. (ctf_member_info): Add support for types in the dynamic type list. (ctf_enum_name): Likewise. (ctf_enum_value): Likewise. (ctf_func_type_info): Likewise. (ctf_func_type_args): Likewise. * ctf-link.c (ctf_accumulate_archive_names): No longer call ctf_update. (ctf_link_write): Likewise. (ctf_link_intern_extern_string): Adjust for new ctf_str_add_external return value. (ctf_link_add_strtab): Likewise. * ctf-util.c (ctf_list_empty_p): New.
2019-08-07 18:55:09 +02:00
if (!ctf_str_add_external (fp, arg->str, arg->offset))
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
arg->err = ENOMEM;
}
/* Repeatedly call ADD_STRING to acquire strings from the external string table,
adding them to the atoms table for this CU and all subsidiary CUs.
If ctf_link() is also called, it must be called first if you want the new CTF
files ctf_link() can create to get their strings dedupped against the ELF
strtab properly. */
int
ctf_link_add_strtab (ctf_file_t *fp, ctf_link_strtab_string_f *add_string,
void *arg)
{
const char *str;
uint32_t offset;
int err = 0;
while ((str = add_string (&offset, arg)) != NULL)
{
ctf_link_out_string_cb_arg_t iter_arg = { str, offset, 0 };
fp->ctf_flags |= LCTF_DIRTY;
libctf: avoid the need to ever use ctf_update The method of operation of libctf when the dictionary is writable has before now been that types that are added land in the dynamic type section, which is a linked list and hash of IDs -> dynamic type definitions (and, recently a hash of names): the DTDs are a bit of CTF representing the ctf_type_t and ad hoc C structures representing the vlen. Historically, libctf was unable to do anything with these types, not even look them up by ID, let alone by name: if you wanted to do that say if you were adding a type that depended on one you just added) you called ctf_update, which serializes all the DTDs into a CTF file and reopens it, copying its guts over the fp it's called with. The ctf_updated types are then frozen in amber and unchangeable: all lookups will return the types in the static portion in preference to the dynamic portion, and we will refuse to re-add things that already exist in the static portion (and, of late, in the dynamic portion too). The libctf machinery remembers the boundary between static and dynamic types and looks in the right portion for each type. Lots of things still don't quite work with dynamic types (e.g. getting their size), but enough works to do a bunch of additions and then a ctf_update, most of the time. Except it doesn't, because ctf_add_type finds it necessary to walk the full dynamic type definition list looking for types with matching names, so it gets slower and slower with every type you add: fixing this requires calling ctf_update periodically for no other reason than to avoid massively slowing things down. This is all clunky and very slow but kind of works, until you consider that it is in fact possible and indeed necessary to modify one sort of type after it has been added: forwards. These are necessarily promoted to structs, unions or enums, and when they do so *their type ID does not change*. So all of a sudden we are changing types that already exist in the static portion. ctf_update gets massively confused by this and allocates space enough for the forward (with no members), but then emits the new dynamic type (with all the members) into it. You get an assertion failure after that, if you're lucky, or a coredump. So this commit rejigs things a bit and arranges to exclusively use the dynamic type definitions in writable dictionaries, and the static type definitions in readable dictionaries: we don't at any time have a mixture of static and dynamic types, and you don't need to call ctf_update to make things "appear". The ctf_dtbyname hash I introduced a few months ago, which maps things like "struct foo" to DTDs, is removed, replaced instead by a change of type of the four dictionaries which track names. Rather than just being (unresizable) ctf_hash_t's populated only at ctf_bufopen time, they are now a ctf_names_t structure, which is a pair of ctf_hash_t and ctf_dynhash_t, with the ctf_hash_t portion being used in readonly dictionaries, and the ctf_dynhash_t being used in writable ones. The decision as to which to use is centralized in the new functions ctf_lookup_by_rawname (which takes a type kind) and ctf_lookup_by_rawhash, which it calls (which takes a ctf_names_t *.) This change lets us switch from using static to dynamic name hashes on the fly across the entirety of libctf without complexifying anything: in fact, because we now centralize the knowledge about how to map from type kind to name hash, it actually simplifies things and lets us throw out quite a lot of now-unnecessary complexity, from ctf_dtnyname (replaced by the dynamic half of the name tables), through to ctf_dtnextid (now that a dictionary's static portion is never referenced if the dictionary is writable, we can just use ctf_typemax to indicate the maximum type: dynamic or non-dynamic does not matter, and we no longer need to track the boundary between the types). You can now ctf_rollback() as far as you like, even past a ctf_update or for that matter a full writeout; all the iteration functions work just as well on writable as on read-only dictionaries; ctf_add_type no longer needs expensive duplicated code to run over the dynamic types hunting for ones it might be interested in; and the linker no longer needs a hack to call ctf_update so that calling ctf_add_type is not impossibly expensive. There is still a bit more complexity: some new code paths in ctf-types.c need to know how to extract information from dynamic types. This complexity will go away again in a few months when libctf acquires a proper intermediate representation. You can still call ctf_update if you like (it's public API, after all), but its only effect now is to set the point to which ctf_discard rolls back. Obviously *something* still needs to serialize the CTF file before writeout, and this job is done by ctf_serialize, which does everything ctf_update used to except set the counter used by ctf_discard. It is automatically called by the various functions that do CTF writeout: nobody else ever needs to call it. With this in place, forwards that are promoted to non-forwards no longer crash the link, even if it happens tens of thousands of types later. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_names_t): New. (ctf_lookup_t) <ctf_hash>: Now a ctf_names_t, not a ctf_hash_t. (ctf_file_t) <ctf_structs>: Likewise. <ctf_unions>: Likewise. <ctf_enums>: Likewise. <ctf_names>: Likewise. <ctf_lookups>: Improve comment. <ctf_ptrtab_len>: New. <ctf_prov_strtab>: New. <ctf_str_prov_offset>: New. <ctf_dtbyname>: Remove, redundant to the names hashes. <ctf_dtnextid>: Remove, redundant to ctf_typemax. (ctf_dtdef_t) <dtd_name>: Remove. <dtd_data>: Note that the ctt_name is now populated. (ctf_str_atom_t) <csa_offset>: This is now the strtab offset for internal strings too. <csa_external_offset>: New, the external strtab offset. (CTF_INDEX_TO_TYPEPTR): Handle the LCTF_RDWR case. (ctf_name_table): New declaration. (ctf_lookup_by_rawname): Likewise. (ctf_lookup_by_rawhash): Likewise. (ctf_set_ctl_hashes): Likewise. (ctf_serialize): Likewise. (ctf_dtd_insert): Adjust. (ctf_simple_open_internal): Likewise. (ctf_bufopen_internal): Likewise. (ctf_list_empty_p): Likewise. (ctf_str_remove_ref): Likewise. (ctf_str_add): Returns uint32_t now. (ctf_str_add_ref): Likewise. (ctf_str_add_external): Now returns a boolean (int). * ctf-string.c (ctf_strraw_explicit): Check the ctf_prov_strtab for strings in the appropriate range. (ctf_str_create_atoms): Create the ctf_prov_strtab. Detect OOM when adding the null string to the new strtab. (ctf_str_free_atoms): Destroy the ctf_prov_strtab. (ctf_str_add_ref_internal): Add make_provisional argument. If make_provisional, populate the offset and fill in the ctf_prov_strtab accordingly. (ctf_str_add): Return the offset, not the string. (ctf_str_add_ref): Likewise. (ctf_str_add_external): Return a success integer. (ctf_str_remove_ref): New, remove a single ref. (ctf_str_count_strtab): Do not count the initial null string's length or the existence or length of any unreferenced internal atoms. (ctf_str_populate_sorttab): Skip atoms with no refs. (ctf_str_write_strtab): Populate the nullstr earlier. Add one to the cts_len for the null string, since it is no longer done in ctf_str_count_strtab. Adjust for csa_external_offset rename. Populate the csa_offset for both internal and external cases. Flush the ctf_prov_strtab afterwards, and reset the ctf_str_prov_offset. * ctf-create.c (ctf_grow_ptrtab): New. (ctf_create): Call it. Initialize new fields rather than old ones. Tell ctf_bufopen_internal that this is a writable dictionary. Set the ctl hashes and data model. (ctf_update): Rename to... (ctf_serialize): ... this. Leave a compatibility function behind. Tell ctf_simple_open_internal that this is a writable dictionary. Pass the new fields along from the old dictionary. Drop ctf_dtnextid and ctf_dtbyname. Use ctf_strraw, not dtd_name. Do not zero out the DTD's ctt_name. (ctf_prefixed_name): Rename to... (ctf_name_table): ... this. No longer return a prefixed name: return the applicable name table instead. (ctf_dtd_insert): Use it, and use the right name table. Pass in the kind we're adding. Migrate away from dtd_name. (ctf_dtd_delete): Adjust similarly. Remove the ref to the deleted ctt_name. (ctf_dtd_lookup_type_by_name): Remove. (ctf_dynamic_type): Always return NULL on read-only dictionaries. No longer check ctf_dtnextid: check ctf_typemax instead. (ctf_snapshot): No longer use ctf_dtnextid: use ctf_typemax instead. (ctf_rollback): Likewise. No longer fail with ECTF_OVERROLLBACK. Use ctf_name_table and the right name table, and migrate away from dtd_name as in ctf_dtd_delete. (ctf_add_generic): Pass in the kind explicitly and pass it to ctf_dtd_insert. Use ctf_typemax, not ctf_dtnextid. Migrate away from dtd_name to using ctf_str_add_ref to populate the ctt_name. Grow the ptrtab if needed. (ctf_add_encoded): Pass in the kind. (ctf_add_slice): Likewise. (ctf_add_array): Likewise. (ctf_add_function): Likewise. (ctf_add_typedef): Likewise. (ctf_add_reftype): Likewise. Initialize the ctf_ptrtab, checking ctt_name rather than dtd_name. (ctf_add_struct_sized): Pass in the kind. Use ctf_lookup_by_rawname, not ctf_hash_lookup_type / ctf_dtd_lookup_type_by_name. (ctf_add_union_sized): Likewise. (ctf_add_enum): Likewise. (ctf_add_enum_encoded): Likewise. (ctf_add_forward): Likewise. (ctf_add_type): Likewise. (ctf_compress_write): Call ctf_serialize: adjust for ctf_size not being initialized until after the call. (ctf_write_mem): Likewise. (ctf_write): Likewise. * ctf-archive.c (arc_write_one_ctf): Likewise. * ctf-lookup.c (ctf_lookup_by_name): Use ctf_lookuup_by_rawhash, not ctf_hash_lookup_type. (ctf_lookup_by_id): No longer check the readonly types if the dictionary is writable. * ctf-open.c (init_types): Assert that this dictionary is not writable. Adjust to use the new name hashes, ctf_name_table, and ctf_ptrtab_len. GNU style fix for the final ptrtab scan. (ctf_bufopen_internal): New 'writable' parameter. Flip on LCTF_RDWR if set. Drop out early when dictionary is writable. Split the ctf_lookups initialization into... (ctf_set_cth_hashes): ... this new function. (ctf_simple_open_internal): Adjust. New 'writable' parameter. (ctf_simple_open): Adjust accordingly. (ctf_bufopen): Likewise. (ctf_file_close): Destroy the appropriate name hashes. No longer destroy ctf_dtbyname, which is gone. (ctf_getdatasect): Remove spurious "extern". * ctf-types.c (ctf_lookup_by_rawname): New, look up types in the specified name table, given a kind. (ctf_lookup_by_rawhash): Likewise, given a ctf_names_t *. (ctf_member_iter): Add support for iterating over the dynamic type list. (ctf_enum_iter): Likewise. (ctf_variable_iter): Likewise. (ctf_type_rvisit): Likewise. (ctf_member_info): Add support for types in the dynamic type list. (ctf_enum_name): Likewise. (ctf_enum_value): Likewise. (ctf_func_type_info): Likewise. (ctf_func_type_args): Likewise. * ctf-link.c (ctf_accumulate_archive_names): No longer call ctf_update. (ctf_link_write): Likewise. (ctf_link_intern_extern_string): Adjust for new ctf_str_add_external return value. (ctf_link_add_strtab): Likewise. * ctf-util.c (ctf_list_empty_p): New.
2019-08-07 18:55:09 +02:00
if (!ctf_str_add_external (fp, str, offset))
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
err = ENOMEM;
ctf_dynhash_iter (fp->ctf_link_outputs, ctf_link_intern_extern_string,
&iter_arg);
if (iter_arg.err)
err = iter_arg.err;
}
return -err;
}
/* Not yet implemented. */
int
ctf_link_shuffle_syms (ctf_file_t *fp _libctf_unused_,
ctf_link_iter_symbol_f *add_sym _libctf_unused_,
void *arg _libctf_unused_)
{
return 0;
}
typedef struct ctf_name_list_accum_cb_arg
{
char **names;
ctf_file_t *fp;
ctf_file_t **files;
size_t i;
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
char **dynames;
size_t ndynames;
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
} ctf_name_list_accum_cb_arg_t;
libctf: avoid the need to ever use ctf_update The method of operation of libctf when the dictionary is writable has before now been that types that are added land in the dynamic type section, which is a linked list and hash of IDs -> dynamic type definitions (and, recently a hash of names): the DTDs are a bit of CTF representing the ctf_type_t and ad hoc C structures representing the vlen. Historically, libctf was unable to do anything with these types, not even look them up by ID, let alone by name: if you wanted to do that say if you were adding a type that depended on one you just added) you called ctf_update, which serializes all the DTDs into a CTF file and reopens it, copying its guts over the fp it's called with. The ctf_updated types are then frozen in amber and unchangeable: all lookups will return the types in the static portion in preference to the dynamic portion, and we will refuse to re-add things that already exist in the static portion (and, of late, in the dynamic portion too). The libctf machinery remembers the boundary between static and dynamic types and looks in the right portion for each type. Lots of things still don't quite work with dynamic types (e.g. getting their size), but enough works to do a bunch of additions and then a ctf_update, most of the time. Except it doesn't, because ctf_add_type finds it necessary to walk the full dynamic type definition list looking for types with matching names, so it gets slower and slower with every type you add: fixing this requires calling ctf_update periodically for no other reason than to avoid massively slowing things down. This is all clunky and very slow but kind of works, until you consider that it is in fact possible and indeed necessary to modify one sort of type after it has been added: forwards. These are necessarily promoted to structs, unions or enums, and when they do so *their type ID does not change*. So all of a sudden we are changing types that already exist in the static portion. ctf_update gets massively confused by this and allocates space enough for the forward (with no members), but then emits the new dynamic type (with all the members) into it. You get an assertion failure after that, if you're lucky, or a coredump. So this commit rejigs things a bit and arranges to exclusively use the dynamic type definitions in writable dictionaries, and the static type definitions in readable dictionaries: we don't at any time have a mixture of static and dynamic types, and you don't need to call ctf_update to make things "appear". The ctf_dtbyname hash I introduced a few months ago, which maps things like "struct foo" to DTDs, is removed, replaced instead by a change of type of the four dictionaries which track names. Rather than just being (unresizable) ctf_hash_t's populated only at ctf_bufopen time, they are now a ctf_names_t structure, which is a pair of ctf_hash_t and ctf_dynhash_t, with the ctf_hash_t portion being used in readonly dictionaries, and the ctf_dynhash_t being used in writable ones. The decision as to which to use is centralized in the new functions ctf_lookup_by_rawname (which takes a type kind) and ctf_lookup_by_rawhash, which it calls (which takes a ctf_names_t *.) This change lets us switch from using static to dynamic name hashes on the fly across the entirety of libctf without complexifying anything: in fact, because we now centralize the knowledge about how to map from type kind to name hash, it actually simplifies things and lets us throw out quite a lot of now-unnecessary complexity, from ctf_dtnyname (replaced by the dynamic half of the name tables), through to ctf_dtnextid (now that a dictionary's static portion is never referenced if the dictionary is writable, we can just use ctf_typemax to indicate the maximum type: dynamic or non-dynamic does not matter, and we no longer need to track the boundary between the types). You can now ctf_rollback() as far as you like, even past a ctf_update or for that matter a full writeout; all the iteration functions work just as well on writable as on read-only dictionaries; ctf_add_type no longer needs expensive duplicated code to run over the dynamic types hunting for ones it might be interested in; and the linker no longer needs a hack to call ctf_update so that calling ctf_add_type is not impossibly expensive. There is still a bit more complexity: some new code paths in ctf-types.c need to know how to extract information from dynamic types. This complexity will go away again in a few months when libctf acquires a proper intermediate representation. You can still call ctf_update if you like (it's public API, after all), but its only effect now is to set the point to which ctf_discard rolls back. Obviously *something* still needs to serialize the CTF file before writeout, and this job is done by ctf_serialize, which does everything ctf_update used to except set the counter used by ctf_discard. It is automatically called by the various functions that do CTF writeout: nobody else ever needs to call it. With this in place, forwards that are promoted to non-forwards no longer crash the link, even if it happens tens of thousands of types later. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_names_t): New. (ctf_lookup_t) <ctf_hash>: Now a ctf_names_t, not a ctf_hash_t. (ctf_file_t) <ctf_structs>: Likewise. <ctf_unions>: Likewise. <ctf_enums>: Likewise. <ctf_names>: Likewise. <ctf_lookups>: Improve comment. <ctf_ptrtab_len>: New. <ctf_prov_strtab>: New. <ctf_str_prov_offset>: New. <ctf_dtbyname>: Remove, redundant to the names hashes. <ctf_dtnextid>: Remove, redundant to ctf_typemax. (ctf_dtdef_t) <dtd_name>: Remove. <dtd_data>: Note that the ctt_name is now populated. (ctf_str_atom_t) <csa_offset>: This is now the strtab offset for internal strings too. <csa_external_offset>: New, the external strtab offset. (CTF_INDEX_TO_TYPEPTR): Handle the LCTF_RDWR case. (ctf_name_table): New declaration. (ctf_lookup_by_rawname): Likewise. (ctf_lookup_by_rawhash): Likewise. (ctf_set_ctl_hashes): Likewise. (ctf_serialize): Likewise. (ctf_dtd_insert): Adjust. (ctf_simple_open_internal): Likewise. (ctf_bufopen_internal): Likewise. (ctf_list_empty_p): Likewise. (ctf_str_remove_ref): Likewise. (ctf_str_add): Returns uint32_t now. (ctf_str_add_ref): Likewise. (ctf_str_add_external): Now returns a boolean (int). * ctf-string.c (ctf_strraw_explicit): Check the ctf_prov_strtab for strings in the appropriate range. (ctf_str_create_atoms): Create the ctf_prov_strtab. Detect OOM when adding the null string to the new strtab. (ctf_str_free_atoms): Destroy the ctf_prov_strtab. (ctf_str_add_ref_internal): Add make_provisional argument. If make_provisional, populate the offset and fill in the ctf_prov_strtab accordingly. (ctf_str_add): Return the offset, not the string. (ctf_str_add_ref): Likewise. (ctf_str_add_external): Return a success integer. (ctf_str_remove_ref): New, remove a single ref. (ctf_str_count_strtab): Do not count the initial null string's length or the existence or length of any unreferenced internal atoms. (ctf_str_populate_sorttab): Skip atoms with no refs. (ctf_str_write_strtab): Populate the nullstr earlier. Add one to the cts_len for the null string, since it is no longer done in ctf_str_count_strtab. Adjust for csa_external_offset rename. Populate the csa_offset for both internal and external cases. Flush the ctf_prov_strtab afterwards, and reset the ctf_str_prov_offset. * ctf-create.c (ctf_grow_ptrtab): New. (ctf_create): Call it. Initialize new fields rather than old ones. Tell ctf_bufopen_internal that this is a writable dictionary. Set the ctl hashes and data model. (ctf_update): Rename to... (ctf_serialize): ... this. Leave a compatibility function behind. Tell ctf_simple_open_internal that this is a writable dictionary. Pass the new fields along from the old dictionary. Drop ctf_dtnextid and ctf_dtbyname. Use ctf_strraw, not dtd_name. Do not zero out the DTD's ctt_name. (ctf_prefixed_name): Rename to... (ctf_name_table): ... this. No longer return a prefixed name: return the applicable name table instead. (ctf_dtd_insert): Use it, and use the right name table. Pass in the kind we're adding. Migrate away from dtd_name. (ctf_dtd_delete): Adjust similarly. Remove the ref to the deleted ctt_name. (ctf_dtd_lookup_type_by_name): Remove. (ctf_dynamic_type): Always return NULL on read-only dictionaries. No longer check ctf_dtnextid: check ctf_typemax instead. (ctf_snapshot): No longer use ctf_dtnextid: use ctf_typemax instead. (ctf_rollback): Likewise. No longer fail with ECTF_OVERROLLBACK. Use ctf_name_table and the right name table, and migrate away from dtd_name as in ctf_dtd_delete. (ctf_add_generic): Pass in the kind explicitly and pass it to ctf_dtd_insert. Use ctf_typemax, not ctf_dtnextid. Migrate away from dtd_name to using ctf_str_add_ref to populate the ctt_name. Grow the ptrtab if needed. (ctf_add_encoded): Pass in the kind. (ctf_add_slice): Likewise. (ctf_add_array): Likewise. (ctf_add_function): Likewise. (ctf_add_typedef): Likewise. (ctf_add_reftype): Likewise. Initialize the ctf_ptrtab, checking ctt_name rather than dtd_name. (ctf_add_struct_sized): Pass in the kind. Use ctf_lookup_by_rawname, not ctf_hash_lookup_type / ctf_dtd_lookup_type_by_name. (ctf_add_union_sized): Likewise. (ctf_add_enum): Likewise. (ctf_add_enum_encoded): Likewise. (ctf_add_forward): Likewise. (ctf_add_type): Likewise. (ctf_compress_write): Call ctf_serialize: adjust for ctf_size not being initialized until after the call. (ctf_write_mem): Likewise. (ctf_write): Likewise. * ctf-archive.c (arc_write_one_ctf): Likewise. * ctf-lookup.c (ctf_lookup_by_name): Use ctf_lookuup_by_rawhash, not ctf_hash_lookup_type. (ctf_lookup_by_id): No longer check the readonly types if the dictionary is writable. * ctf-open.c (init_types): Assert that this dictionary is not writable. Adjust to use the new name hashes, ctf_name_table, and ctf_ptrtab_len. GNU style fix for the final ptrtab scan. (ctf_bufopen_internal): New 'writable' parameter. Flip on LCTF_RDWR if set. Drop out early when dictionary is writable. Split the ctf_lookups initialization into... (ctf_set_cth_hashes): ... this new function. (ctf_simple_open_internal): Adjust. New 'writable' parameter. (ctf_simple_open): Adjust accordingly. (ctf_bufopen): Likewise. (ctf_file_close): Destroy the appropriate name hashes. No longer destroy ctf_dtbyname, which is gone. (ctf_getdatasect): Remove spurious "extern". * ctf-types.c (ctf_lookup_by_rawname): New, look up types in the specified name table, given a kind. (ctf_lookup_by_rawhash): Likewise, given a ctf_names_t *. (ctf_member_iter): Add support for iterating over the dynamic type list. (ctf_enum_iter): Likewise. (ctf_variable_iter): Likewise. (ctf_type_rvisit): Likewise. (ctf_member_info): Add support for types in the dynamic type list. (ctf_enum_name): Likewise. (ctf_enum_value): Likewise. (ctf_func_type_info): Likewise. (ctf_func_type_args): Likewise. * ctf-link.c (ctf_accumulate_archive_names): No longer call ctf_update. (ctf_link_write): Likewise. (ctf_link_intern_extern_string): Adjust for new ctf_str_add_external return value. (ctf_link_add_strtab): Likewise. * ctf-util.c (ctf_list_empty_p): New.
2019-08-07 18:55:09 +02:00
/* Accumulate the names and a count of the names in the link output hash. */
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
static void
ctf_accumulate_archive_names (void *key, void *value, void *arg_)
{
const char *name = (const char *) key;
ctf_file_t *fp = (ctf_file_t *) value;
char **names;
ctf_file_t **files;
ctf_name_list_accum_cb_arg_t *arg = (ctf_name_list_accum_cb_arg_t *) arg_;
if ((names = realloc (arg->names, sizeof (char *) * ++(arg->i))) == NULL)
{
(arg->i)--;
ctf_set_errno (arg->fp, ENOMEM);
return;
}
if ((files = realloc (arg->files, sizeof (ctf_file_t *) * arg->i)) == NULL)
{
(arg->i)--;
ctf_set_errno (arg->fp, ENOMEM);
return;
}
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
/* Allow the caller to get in and modify the name at the last minute. If the
caller *does* modify the name, we have to stash away the new name the
caller returned so we can free it later on. (The original name is the key
of the ctf_link_outputs hash and is freed by the dynhash machinery.) */
if (fp->ctf_link_memb_name_changer)
{
char **dynames;
char *dyname;
void *nc_arg = fp->ctf_link_memb_name_changer_arg;
dyname = fp->ctf_link_memb_name_changer (fp, name, nc_arg);
if (dyname != NULL)
{
if ((dynames = realloc (arg->dynames,
sizeof (char *) * ++(arg->ndynames))) == NULL)
{
(arg->ndynames)--;
ctf_set_errno (arg->fp, ENOMEM);
return;
}
arg->dynames = dynames;
name = (const char *) dyname;
}
}
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
arg->names = names;
arg->names[(arg->i) - 1] = (char *) name;
arg->files = files;
arg->files[(arg->i) - 1] = fp;
}
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
/* Change the name of the parent CTF section, if the name transformer has got to
it. */
static void
ctf_change_parent_name (void *key _libctf_unused_, void *value, void *arg)
{
ctf_file_t *fp = (ctf_file_t *) value;
const char *name = (const char *) arg;
ctf_parent_name_set (fp, name);
}
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
/* Write out a CTF archive (if there are per-CU CTF files) or a CTF file
(otherwise) into a new dynamically-allocated string, and return it.
Members with sizes above THRESHOLD are compressed. */
unsigned char *
ctf_link_write (ctf_file_t *fp, size_t *size, size_t threshold)
{
ctf_name_list_accum_cb_arg_t arg;
char **names;
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
char *transformed_name = NULL;
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
ctf_file_t **files;
FILE *f = NULL;
int err;
long fsize;
const char *errloc;
unsigned char *buf = NULL;
memset (&arg, 0, sizeof (ctf_name_list_accum_cb_arg_t));
arg.fp = fp;
if (fp->ctf_link_outputs)
{
ctf_dynhash_iter (fp->ctf_link_outputs, ctf_accumulate_archive_names, &arg);
if (ctf_errno (fp) < 0)
{
errloc = "hash creation";
goto err;
}
}
/* No extra outputs? Just write a simple ctf_file_t. */
if (arg.i == 0)
return ctf_write_mem (fp, size, threshold);
/* Writing an archive. Stick ourselves (the shared repository, parent of all
other archives) on the front of it with the default name. */
if ((names = realloc (arg.names, sizeof (char *) * (arg.i + 1))) == NULL)
{
errloc = "name reallocation";
goto err_no;
}
arg.names = names;
memmove (&(arg.names[1]), arg.names, sizeof (char *) * (arg.i));
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
arg.names[0] = (char *) _CTF_SECTION;
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
if (fp->ctf_link_memb_name_changer)
{
void *nc_arg = fp->ctf_link_memb_name_changer_arg;
transformed_name = fp->ctf_link_memb_name_changer (fp, _CTF_SECTION,
nc_arg);
if (transformed_name != NULL)
{
arg.names[0] = transformed_name;
ctf_dynhash_iter (fp->ctf_link_outputs, ctf_change_parent_name,
transformed_name);
}
}
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
if ((files = realloc (arg.files,
sizeof (struct ctf_file *) * (arg.i + 1))) == NULL)
{
errloc = "ctf_file reallocation";
goto err_no;
}
arg.files = files;
memmove (&(arg.files[1]), arg.files, sizeof (ctf_file_t *) * (arg.i));
arg.files[0] = fp;
if ((f = tmpfile ()) == NULL)
{
errloc = "tempfile creation";
goto err_no;
}
if ((err = ctf_arc_write_fd (fileno (f), arg.files, arg.i + 1,
(const char **) arg.names,
threshold)) < 0)
{
errloc = "archive writing";
ctf_set_errno (fp, err);
goto err;
}
if (fseek (f, 0, SEEK_END) < 0)
{
errloc = "seeking to end";
goto err_no;
}
if ((fsize = ftell (f)) < 0)
{
errloc = "filesize determination";
goto err_no;
}
if (fseek (f, 0, SEEK_SET) < 0)
{
errloc = "filepos resetting";
goto err_no;
}
if ((buf = malloc (fsize)) == NULL)
{
errloc = "CTF archive buffer allocation";
goto err_no;
}
while (!feof (f) && fread (buf, fsize, 1, f) == 0)
if (ferror (f))
{
errloc = "reading archive from temporary file";
goto err_no;
}
*size = fsize;
free (arg.names);
free (arg.files);
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
free (transformed_name);
if (arg.ndynames)
{
size_t i;
for (i = 0; i < arg.ndynames; i++)
free (arg.dynames[i]);
free (arg.dynames);
}
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
return buf;
err_no:
ctf_set_errno (fp, errno);
err:
free (buf);
if (f)
fclose (f);
free (arg.names);
free (arg.files);
libctf: add CU-mapping machinery Once the deduplicator is capable of actually detecting conflicting types with the same name (i.e., not yet) we will place such conflicting types, and types that depend on them, into CTF dictionaries that are the child of the main dictionary we usually emit: currently, this will lead to the .ctf section becoming a CTF archive rather than a single dictionary, with the default-named archive member (_CTF_SECTION, or NULL) being the main shared dictionary with most of the types in it. By default, the sections are named after the compilation unit they come from (complete path and all), with the cuname field in the CTF header providing further evidence of the name without requiring the caller to engage in tiresome parsing. But some callers may not wish the mapping from input CU to output sub-dictionary to be purely CU-based. The machinery here allows this to be freely changed, in two ways: - callers can call ctf_link_add_cu_mapping to specify that a single input compilation unit should have its types placed in some other CU if they conflict: the CU will always be created, even if empty, so the consuming program can depend on its existence. You can map multiple input CUs to one output CU to force all their types to be merged together: if some of *those* types conflict, the behaviour is currently unspecified (the new deduplicator will specify it). - callers can call ctf_link_set_memb_name_changer to provide a function which is passed every CTF sub-dictionary name in turn (including _CTF_SECTION) and can return a new name, or NULL if no change is desired. The mapping from input to output names should not map two input names to the same output name: if this happens, the two are not merged but will result in an archive with two members with the same name (technically valid, but it's hard to access the second same-named member: you have to do an iteration over archive members). This is used by the kernel's ctfarchive machinery (not yet upstream) to encode CTF under member names like {module name}.ctf rather than .ctf.CU, but it is anticipated that other large projects may wish to have their own storage for CTF outside of .ctf sections and may wish to have new naming schemes that suit their special-purpose consumers. New in v3. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (ctf_link_add_cu_mapping): New. (ctf_link_memb_name_changer_f): New. (ctf_link_set_memb_name_changer): New. libctf/ * ctf-impl.h (ctf_file_t) <ctf_link_cu_mappping>: New. <ctf_link_memb_name_changer>: Likewise. <ctf_link_memb_name_changer_arg>: Likewise. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-link.c (ctf_create_per_cu): Apply the cu mapping. (ctf_link_add_cu_mapping): New. (ctf_link_set_memb_name_changer): Likewise. (ctf_change_parent_name): New. (ctf_name_list_accum_cb_arg_t) <dynames>: New, storage for names allocated by the caller's ctf_link_memb_name_changer. <ndynames>: Likewise. (ctf_accumulate_archive_names): Call the ctf_link_memb_name_changer. (ctf_link_write): Likewise (for _CTF_SECTION only): also call ctf_change_parent_name. Free any resulting names.
2019-07-20 15:44:44 +02:00
free (transformed_name);
if (arg.ndynames)
{
size_t i;
for (i = 0; i < arg.ndynames; i++)
free (arg.dynames[i]);
free (arg.dynames);
}
libctf: add the ctf_link machinery This is the start of work on the core of the linking mechanism for CTF sections. This commit handles the type and string sections. The linker calls these functions in sequence: ctf_link_add_ctf: to add each CTF section in the input in turn to a newly-created ctf_file_t (which will appear in the output, and which itself will become the shared parent that contains types that all TUs have in common (in all link modes) and all types that do not have conflicting definitions between types (by default). Input files that are themselves products of ld -r are supported, though this is not heavily tested yet. ctf_link: called once all input files are added to merge the types in all the input containers into the output container, eliminating duplicates. ctf_link_add_strtab: called once the ELF string table is finalized and all its offsets are known, this calls a callback provided by the linker which returns the string content and offset of every string in the ELF strtab in turn: all these strings which appear in the input CTF strtab are eliminated from it in favour of the ELF strtab: equally, any strings that only appear in the input strtab will reappear in the internal CTF strtab of the output. ctf_link_shuffle_syms (not yet implemented): called once the ELF symtab is finalized, this calls a callback provided by the linker which returns information on every symbol in turn as a ctf_link_sym_t. This is then used to shuffle the function info and data object sections in the CTF section into symbol table order, eliminating the index sections which map those sections to symbol names before that point. Currently just returns ECTF_NOTYET. ctf_link_write: Returns a buffer containing either a serialized ctf_file_t (if there are no types with conflicting definitions in the object files in the link) or a ctf_archive_t containing a large ctf_file_t (the common types) and a bunch of small ones named after individual CUs in which conflicting types are found (containing the conflicting types, and all types that reference them). A threshold size above which compression takes place is passed as one parameter. (Currently, only gzip compression is supported, but I hope to add lzma as well.) Lifetime rules for this are simple: don't close the input CTF files until you've called ctf_link for the last time. We do not assume that symbols or strings passed in by the callback outlast the call to ctf_link_add_strtab or ctf_link_shuffle_syms. Right now, the duplicate elimination mechanism is the one already present as part of the ctf_add_type function, and is not particularly good: it misses numerous actual duplicates, and the conflicting-types detection hardly ever reports that types conflict, even when they do (one of them just tends to get silently dropped): it is also very slow. This will all be fixed in the next few weeks, but the fix hardly touches any of this code, and the linker does work without it, just not as well as it otherwise might. (And when no CTF section is present, there is no effect on performance, of course. So only people using a trunk GCC with not-yet-committed patches will even notice. By the time it gets upstream, things should be better.) v3: Fix error handling. v4: check for strdup failure. v5: fix tabdamage. include/ * ctf-api.h (struct ctf_link_sym): New, a symbol in flight to the libctf linking machinery. (CTF_LINK_SHARE_UNCONFLICTED): New. (CTF_LINK_SHARE_DUPLICATED): New. (ECTF_LINKADDEDLATE): New, replacing ECTF_UNUSED. (ECTF_NOTYET): New, a 'not yet implemented' message. (ctf_link_add_ctf): New, add an input file's CTF to the link. (ctf_link): New, merge the type and string sections. (ctf_link_strtab_string_f): New, callback for feeding strtab info. (ctf_link_iter_symbol_f): New, callback for feeding symtab info. (ctf_link_add_strtab): New, tell the CTF linker about the ELF strtab's strings. (ctf_link_shuffle_syms): New, ask the CTF linker to shuffle its symbols into symtab order. (ctf_link_write): New, ask the CTF linker to write the CTF out. libctf/ * ctf-link.c: New file, linking of the string and type sections. * Makefile.am (libctf_a_SOURCES): Add it. * Makefile.in: Regenerate. * ctf-impl.h (ctf_file_t): New fields ctf_link_inputs, ctf_link_outputs. * ctf-create.c (ctf_update): Update accordingly. * ctf-open.c (ctf_file_close): Likewise. * ctf-error.c (_ctf_errlist): Updated with new errors.
2019-07-13 22:06:55 +02:00
ctf_dprintf ("Cannot write archive in link: %s failure: %s\n", errloc,
ctf_errmsg (ctf_errno (fp)));
return NULL;
}