Re: [PATCH, GCC/LTO, ping] Fix PR69866: LTO with def for weak alias in regular object file

[Jan Hubicka <hubicka at ucw dot cz>]

Hello,
I apologize for absurdly long time to check the patch. I misread the testcase
originally and got confused.  Now it makes sense to me.

It does not make sense to keep the alias in symtab when we know that it
was prevailed by external definition.  So lto-symtab should turn it into
normal external symbol in this case. There are multiple cases to consider
  alias+target is interposed: in this case we probably want to keep them as
    external alias to maintain the info that they might point to same address
    but optimizer should take it as a may information.
  alias is interposed, target is not: in this case we want to reset alias
    into undefined state
  alias is not interposed, target is: in this case we need to preserve target
    definition but we also want to keep the extenral symbol. This will be bit
    tricky to do becuase LTO lacks mechanizm to keep the body around. 

I will fix the middle case first and look into other two (which are LTO-symtab
problem I know of for a while, but never got around fixing them)

Thanks and sorry for the slowness!
Honza
> Ping?
> 
> Best regards,
> 
> Thomas
> 
> On 06/06/17 11:12, Thomas Preudhomme wrote:
> >On 09/05/17 23:36, Jan Hubicka wrote:
> >>>Ping?
> >>Sorry for late reply
> >
> >My turn to apologize now.
> >
> >>>>Hi,
> >>>>
> >>>>This patch fixes an assert failure when linking one LTOed object file
> >>>>having a weak alias with a regular object file containing a strong
> >>>>definition for that same symbol. The patch is twofold:
> >>>>
> >>>>+ do not add an alias to a partition if it is external
> >>>>+ do not declare (.globl) an alias if it is external
> >>
> >>Adding external alises to partitions is important to keep the information
> >>that two symbols are the same.
> >
> >So how about simply relaxing the assert then? Right now it trips for any
> >external symbol, even external aliases.
> >
> >How about the following:
> >
> >
> >ChangeLog entries are as follow:
> >
> >*** gcc/lto/ChangeLog ***
> >
> >2017-06-02  Thomas Preud'homme  <thomas.preudhomme@arm.com>
> >
> >    * lto/lto-partition.c (add_symbol_to_partition_1): Change assert to
> >    allow external aliases to be added.
> >
> >*** gcc/ChangeLog ***
> >
> >2017-03-01  Thomas Preud'homme  <thomas.preudhomme@arm.com>
> >
> >    * cgraphunit.c (cgraph_node::assemble_thunks_and_aliases): Do not
> >    declare external aliases.
> >
> >*** gcc/testsuite/ChangeLog ***
> >
> >2017-02-28  Thomas Preud'homme  <thomas.preudhomme@arm.com>
> >
> >    * gcc.dg/lto/pr69866_0.c: New test.
> >    * gcc.dg/lto/pr69866_1.c: Likewise.
> >
> >Bootstrapped with LTO on Aarch64 and ARM and testsuite on both of these
> >architectures do not show any regression.
> >
> >Is this ok for trunk?
> >
> >Best regards,
> >
> >Thomas
> >
> >
> >>The second part makes sense to me.  What breaks when you drop the first
> >>change?
> >>
> >>Honza
> >>>>
> >>>>ChangeLog entries are as follow:
> >>>>
> >>>>*** gcc/lto/ChangeLog ***
> >>>>
> >>>>2017-03-01  Thomas Preud'homme  <thomas.preudhomme@arm.com>
> >>>>
> >>>>   PR lto/69866
> >>>>   * lto/lto-partition.c (add_symbol_to_partition_1): Do not add external
> >>>>   aliases to partition.
> >>>>
> >>>>*** gcc/ChangeLog ***
> >>>>
> >>>>2017-03-01  Thomas Preud'homme  <thomas.preudhomme@arm.com>
> >>>>
> >>>>   PR lto/69866
> >>>>   * cgraphunit.c (cgraph_node::assemble_thunks_and_aliases): Do not
> >>>>   declare external aliases.
> >>>>
> >>>>*** gcc/testsuite/ChangeLog ***
> >>>>
> >>>>2017-02-28  Thomas Preud'homme  <thomas.preudhomme@arm.com>
> >>>>
> >>>>   PR lto/69866
> >>>>   * gcc.dg/lto/pr69866_0.c: New test.
> >>>>   * gcc.dg/lto/pr69866_1.c: Likewise.
> >>>>
> >>>>
> >>>>Testing: Testsuite shows no regression when targeting Cortex-M3 with an
> >>>>arm-none-eabi GCC cross-compiler, neither does it show any regression with
> >>>>native LTO-bootstrapped x86-64_linux-gnu and aarch64-linux-gnu compilers.
> >>>>
> >>>>Is this ok for stage4?
> >>>>
> >>>>Best regards,
> >>>>
> >>>>Thomas
> >>>>
> >>>>On 31/03/17 18:07, Richard Biener wrote:
> >>>>>On March 31, 2017 5:23:03 PM GMT+02:00, Jeff Law <law@redhat.com> wrote:
> >>>>>>On 03/16/2017 08:05 AM, Thomas Preudhomme wrote:
> >>>>>>>Ping?
> >>>>>>>
> >>>>>>>Is this ok for stage4?
> >>>>>>Given the lack of response from Richi, I'd suggest deferring to stage1.
> >>>>>
> >>>>>Honza needs to review this, i habe too little knowledge here.
> >>>>>
> >>>>>Richard.
> >>>>>
> >>>>>>jeff
> >>>>>
> >>
> >>>diff --git a/gcc/cgraphunit.c b/gcc/cgraphunit.c
> >>>index
> >>>c82a88a599ca61b068dd9783d2a6158163809b37..580500ff922b8546d33119261a2455235edbf16d
> >>>100644
> >>>--- a/gcc/cgraphunit.c
> >>>+++ b/gcc/cgraphunit.c
> >>>@@ -1972,7 +1972,7 @@ cgraph_node::assemble_thunks_and_aliases (void)
> >>>   FOR_EACH_ALIAS (this, ref)
> >>>     {
> >>>       cgraph_node *alias = dyn_cast <cgraph_node *> (ref->referring);
> >>>-      if (!alias->transparent_alias)
> >>>+      if (!alias->transparent_alias && !DECL_EXTERNAL (alias->decl))
> >>>     {
> >>>       bool saved_written = TREE_ASM_WRITTEN (decl);
> >>>
> >>>diff --git a/gcc/lto/lto-partition.c b/gcc/lto/lto-partition.c
> >>>index
> >>>e27d0d1690c1fcfb39e2fac03ce0f4154031fc7c..f44fd435ed075a27e373bdfdf0464eb06e1731ef
> >>>100644
> >>>--- a/gcc/lto/lto-partition.c
> >>>+++ b/gcc/lto/lto-partition.c
> >>>@@ -178,7 +178,8 @@ add_symbol_to_partition_1 (ltrans_partition part,
> >>>symtab_node *node)
> >>>   /* Add all aliases associated with the symbol.  */
> >>>
> >>>   FOR_EACH_ALIAS (node, ref)
> >>>-    if (!ref->referring->transparent_alias)
> >>>+    if (!ref->referring->transparent_alias
> >>>+    && ref->referring->get_partitioning_class () != SYMBOL_EXTERNAL)
> >>>       add_symbol_to_partition_1 (part, ref->referring);
> >>>     else
> >>>       {
> >>>@@ -189,7 +190,8 @@ add_symbol_to_partition_1 (ltrans_partition part,
> >>>symtab_node *node)
> >>>       {
> >>>         /* Nested transparent aliases are not permitted.  */
> >>>         gcc_checking_assert (!ref2->referring->transparent_alias);
> >>>-        add_symbol_to_partition_1 (part, ref2->referring);
> >>>+        if (ref2->referring->get_partitioning_class () != SYMBOL_EXTERNAL)
> >>>+          add_symbol_to_partition_1 (part, ref2->referring);
> >>>       }
> >>>       }
> >>>
> >>>diff --git a/gcc/testsuite/gcc.dg/lto/pr69866_0.c
> >>>b/gcc/testsuite/gcc.dg/lto/pr69866_0.c
> >>>new file mode 100644
> >>>index
> >>>0000000000000000000000000000000000000000..f49ef8d4c1da7a21d1bfb5409d647bd18141595b
> >>>
> >>>--- /dev/null
> >>>+++ b/gcc/testsuite/gcc.dg/lto/pr69866_0.c
> >>>@@ -0,0 +1,13 @@
> >>>+/* { dg-lto-do link } */
> >>>+
> >>>+int _umh(int i)
> >>>+{
> >>>+  return i+1;
> >>>+}
> >>>+
> >>>+int weaks(int i) __attribute__((weak, alias("_umh")));
> >>>+
> >>>+int main()
> >>>+{
> >>>+  return weaks(10);
> >>>+}
> >>>diff --git a/gcc/testsuite/gcc.dg/lto/pr69866_1.c
> >>>b/gcc/testsuite/gcc.dg/lto/pr69866_1.c
> >>>new file mode 100644
> >>>index
> >>>0000000000000000000000000000000000000000..3a14f850eefaffbf659ce4642adef7900330f4ed
> >>>
> >>>--- /dev/null
> >>>+++ b/gcc/testsuite/gcc.dg/lto/pr69866_1.c
> >>>@@ -0,0 +1,6 @@
> >>>+/* { dg-options { -fno-lto } } */
> >>>+
> >>>+int weaks(int i)
> >>>+{
> >>>+  return i+1;
> >>>+}
> >>

> diff --git a/gcc/cgraphunit.c b/gcc/cgraphunit.c
> index 7b4f47e6efb41e7c401e7347d86fffca6618c4e9..c2cc9df6419573bdc6223e1d2ee348f8af69ccca 100644
> --- a/gcc/cgraphunit.c
> +++ b/gcc/cgraphunit.c
> @@ -1984,7 +1984,7 @@ cgraph_node::assemble_thunks_and_aliases (void)
>    FOR_EACH_ALIAS (this, ref)
>      {
>        cgraph_node *alias = dyn_cast <cgraph_node *> (ref->referring);
> -      if (!alias->transparent_alias)
> +      if (!alias->transparent_alias && !DECL_EXTERNAL (alias->decl))
>  	{
>  	  bool saved_written = TREE_ASM_WRITTEN (decl);
>  
> diff --git a/gcc/lto/lto-partition.c b/gcc/lto/lto-partition.c
> index 3600ab23bd9157aad76751912306abd498acae92..620deac090b7e718f05d3f37e6ef68e759de0008 100644
> --- a/gcc/lto/lto-partition.c
> +++ b/gcc/lto/lto-partition.c
> @@ -132,7 +132,7 @@ add_symbol_to_partition_1 (ltrans_partition part, symtab_node *node)
>  
>    /* Be sure that we never try to duplicate partitioned symbol
>       or add external symbol.  */
> -  gcc_assert (c != SYMBOL_EXTERNAL
> +  gcc_assert ((c != SYMBOL_EXTERNAL || node->alias)
>  	      && (c == SYMBOL_DUPLICATE || !symbol_partitioned_p (node)));
>  
>    part->symbols++;
> diff --git a/gcc/testsuite/gcc.dg/lto/pr69866_0.c b/gcc/testsuite/gcc.dg/lto/pr69866_0.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..f49ef8d4c1da7a21d1bfb5409d647bd18141595b
> --- /dev/null
> +++ b/gcc/testsuite/gcc.dg/lto/pr69866_0.c
> @@ -0,0 +1,13 @@
> +/* { dg-lto-do link } */
> +
> +int _umh(int i)
> +{
> +  return i+1;
> +}
> +
> +int weaks(int i) __attribute__((weak, alias("_umh")));
> +
> +int main()
> +{
> +  return weaks(10);
> +}
> diff --git a/gcc/testsuite/gcc.dg/lto/pr69866_1.c b/gcc/testsuite/gcc.dg/lto/pr69866_1.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..3a14f850eefaffbf659ce4642adef7900330f4ed
> --- /dev/null
> +++ b/gcc/testsuite/gcc.dg/lto/pr69866_1.c
> @@ -0,0 +1,6 @@
> +/* { dg-options { -fno-lto } } */
> +
> +int weaks(int i)
> +{
> +  return i+1;
> +}