Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
* [PATCH][gdb] Fix build breaker with gcc 4.8
@ 2019-06-19 11:04 Tom de Vries
  2019-06-19 12:43 ` Pedro Alves
  2019-06-19 14:56 ` Steve Ellcey
  0 siblings, 2 replies; 6+ messages in thread
From: Tom de Vries @ 2019-06-19 11:04 UTC (permalink / raw)
  To: gdb-patches

Hi,

When compiling with gcc 4.8, we run into:
...
/usr/include/c++/4.8/bits/unordered_map.h:100:18: required from \
  ‘class std::unordered_map<sect_offset, std::vector<sect_offset> >’
src/gdb/dwarf2read.h:260:5:   required from here
/usr/include/c++/4.8/bits/hashtable_policy.h:1070:12: error: invalid use of \
  incomplete type ‘struct std::hash<sect_offset>’
...

Fix this by adding std::hash<sect_offset>.

Build and reg-tested on x86_64-linux with gcc 4.8.

OK for trunk?

Thanks,
- Tom

[gdb] Fix build breaker with gcc 4.8

gdb/ChangeLog:

2019-06-19  Tom de Vries  <tdevries@suse.de>

	* dwarf2read.h (std::hash<sect_offset>): New specialization.

---
 gdb/dwarf2read.h | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/gdb/dwarf2read.h b/gdb/dwarf2read.h
index 776860e454..494dcb32c1 100644
--- a/gdb/dwarf2read.h
+++ b/gdb/dwarf2read.h
@@ -99,6 +99,26 @@ struct signatured_type;
 struct die_info;
 typedef struct die_info *die_info_ptr;
 
+/* An std::hash specialization, required for use of sect_offset in
+   std::unordered_map with gcc 4.8.  */
+
+namespace std
+{
+  template<> struct hash<sect_offset>
+  {
+    typedef sect_offset argument_type;
+    typedef std::size_t result_type;
+    using underlying_type
+    = typename std::underlying_type<argument_type>::type;
+
+    result_type operator() (const argument_type &o) const noexcept
+    {
+      underlying_type u = static_cast<underlying_type> (o);
+      return std::hash<underlying_type> {} (u);
+    }
+  };
+};
+
 /* Collection of data recorded per objfile.
    This hangs off of dwarf2_objfile_data_key.  */
 


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH][gdb] Fix build breaker with gcc 4.8
  2019-06-19 11:04 [PATCH][gdb] Fix build breaker with gcc 4.8 Tom de Vries
@ 2019-06-19 12:43 ` Pedro Alves
  2019-06-19 14:35   ` Tom de Vries
  2019-06-19 14:56 ` Steve Ellcey
  1 sibling, 1 reply; 6+ messages in thread
From: Pedro Alves @ 2019-06-19 12:43 UTC (permalink / raw)
  To: Tom de Vries, gdb-patches

On 6/19/19 12:04 PM, Tom de Vries wrote:
> Hi,
> 
> When compiling with gcc 4.8, we run into:
> ...
> /usr/include/c++/4.8/bits/unordered_map.h:100:18: required from \
>   ‘class std::unordered_map<sect_offset, std::vector<sect_offset> >’
> src/gdb/dwarf2read.h:260:5:   required from here
> /usr/include/c++/4.8/bits/hashtable_policy.h:1070:12: error: invalid use of \
>   incomplete type ‘struct std::hash<sect_offset>’
> ...
> 
> Fix this by adding std::hash<sect_offset>.
> 
> Build and reg-tested on x86_64-linux with gcc 4.8.
> 
> OK for trunk?
> 

hash_enum was added for this:

 https://sourceware.org/ml/gdb-patches/2017-12/msg00210.html

Can you use it here?

It's currently used in dwarf2read.c, here:

  std::unordered_map<sect_offset,
		     dwarf2_per_cu_data *,
		     gdb::hash_enum<sect_offset>>

Thanks,
Pedro Alves

> Thanks,
> - Tom
> 
> [gdb] Fix build breaker with gcc 4.8
> 
> gdb/ChangeLog:
> 
> 2019-06-19  Tom de Vries  <tdevries@suse.de>
> 
> 	* dwarf2read.h (std::hash<sect_offset>): New specialization.
> 
> ---
>  gdb/dwarf2read.h | 20 ++++++++++++++++++++
>  1 file changed, 20 insertions(+)
> 
> diff --git a/gdb/dwarf2read.h b/gdb/dwarf2read.h
> index 776860e454..494dcb32c1 100644
> --- a/gdb/dwarf2read.h
> +++ b/gdb/dwarf2read.h
> @@ -99,6 +99,26 @@ struct signatured_type;
>  struct die_info;
>  typedef struct die_info *die_info_ptr;
>  
> +/* An std::hash specialization, required for use of sect_offset in
> +   std::unordered_map with gcc 4.8.  */
> +
> +namespace std
> +{
> +  template<> struct hash<sect_offset>
> +  {
> +    typedef sect_offset argument_type;
> +    typedef std::size_t result_type;
> +    using underlying_type
> +    = typename std::underlying_type<argument_type>::type;
> +
> +    result_type operator() (const argument_type &o) const noexcept
> +    {
> +      underlying_type u = static_cast<underlying_type> (o);
> +      return std::hash<underlying_type> {} (u);
> +    }
> +  };
> +};
> +
>  /* Collection of data recorded per objfile.
>     This hangs off of dwarf2_objfile_data_key.  */
>  
> 


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH][gdb] Fix build breaker with gcc 4.8
  2019-06-19 12:43 ` Pedro Alves
@ 2019-06-19 14:35   ` Tom de Vries
  2019-06-19 14:49     ` Pedro Alves
  0 siblings, 1 reply; 6+ messages in thread
From: Tom de Vries @ 2019-06-19 14:35 UTC (permalink / raw)
  To: Pedro Alves, gdb-patches

[-- Attachment #1: Type: text/plain, Size: 969 bytes --]

On 19-06-19 14:43, Pedro Alves wrote:
> On 6/19/19 12:04 PM, Tom de Vries wrote:
>> Hi,
>>
>> When compiling with gcc 4.8, we run into:
>> ...
>> /usr/include/c++/4.8/bits/unordered_map.h:100:18: required from \
>>   ‘class std::unordered_map<sect_offset, std::vector<sect_offset> >’
>> src/gdb/dwarf2read.h:260:5:   required from here
>> /usr/include/c++/4.8/bits/hashtable_policy.h:1070:12: error: invalid use of \
>>   incomplete type ‘struct std::hash<sect_offset>’
>> ...
>>
>> Fix this by adding std::hash<sect_offset>.
>>
>> Build and reg-tested on x86_64-linux with gcc 4.8.
>>
>> OK for trunk?
>>
> 
> hash_enum was added for this:
> 
>  https://sourceware.org/ml/gdb-patches/2017-12/msg00210.html
> 
> Can you use it here?
> 
> It's currently used in dwarf2read.c, here:
> 
>   std::unordered_map<sect_offset,
> 		     dwarf2_per_cu_data *,
> 		     gdb::hash_enum<sect_offset>>
> 

Yes, that works for me.

OK for trunk?

Thanks,
- Tom



[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-gdb-Fix-build-breaker-with-gcc-4.8.patch --]
[-- Type: text/x-patch; name="0001-gdb-Fix-build-breaker-with-gcc-4.8.patch", Size: 1633 bytes --]

[gdb] Fix build breaker with gcc 4.8

When compiling with gcc 4.8, we run into:
...
/usr/include/c++/4.8/bits/unordered_map.h:100:18: required from \
  ‘class std::unordered_map<sect_offset, std::vector<sect_offset> >’
src/gdb/dwarf2read.h:260:5:   required from here
/usr/include/c++/4.8/bits/hashtable_policy.h:1070:12: error: invalid use of \
  incomplete type ‘struct std::hash<sect_offset>’
...

Fix this by setting the Hash template parameter of the unordered_map to
gdb::hash_enum<sect_offset>, rather than using the default
std::hash<sect_offset>.

Build and reg-tested on x86_64-linux with gcc 4.8.

gdb/ChangeLog:

2019-06-19  Tom de Vries  <tdevries@suse.de>

	* dwarf2read.h (abstract_to_concrete): Change type to
	std::unordered_map<sect_offset, std::vector<sect_offset>,
	gdb::hash_enum<sect_offset>>.

---
 gdb/dwarf2read.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/gdb/dwarf2read.h b/gdb/dwarf2read.h
index 776860e454..7113cfd384 100644
--- a/gdb/dwarf2read.h
+++ b/gdb/dwarf2read.h
@@ -24,6 +24,7 @@
 #include "dwarf-index-cache.h"
 #include "filename-seen-cache.h"
 #include "gdb_obstack.h"
+#include "common/hash_enum.h"
 
 /* Hold 'maintenance (set|show) dwarf' commands.  */
 extern struct cmd_list_element *set_dwarf_cmdlist;
@@ -256,7 +257,8 @@ public:
 
   /* Mapping from abstract origin DIE to concrete DIEs that reference it as
      DW_AT_abstract_origin.  */
-  std::unordered_map<sect_offset, std::vector<sect_offset>>
+  std::unordered_map<sect_offset, std::vector<sect_offset>, \
+		     gdb::hash_enum<sect_offset>> \
     abstract_to_concrete;
 };
 

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH][gdb] Fix build breaker with gcc 4.8
  2019-06-19 14:35   ` Tom de Vries
@ 2019-06-19 14:49     ` Pedro Alves
  0 siblings, 0 replies; 6+ messages in thread
From: Pedro Alves @ 2019-06-19 14:49 UTC (permalink / raw)
  To: Tom de Vries, gdb-patches

On 6/19/19 3:35 PM, Tom de Vries wrote:

> OK for trunk?
> 

OK.

Thanks,
Pedro Alves


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH][gdb] Fix build breaker with gcc 4.8
  2019-06-19 11:04 [PATCH][gdb] Fix build breaker with gcc 4.8 Tom de Vries
  2019-06-19 12:43 ` Pedro Alves
@ 2019-06-19 14:56 ` Steve Ellcey
       [not found]   ` <0cedb79e-1423-90d8-0c61-1386f398ab20@suse.de>
  1 sibling, 1 reply; 6+ messages in thread
From: Steve Ellcey @ 2019-06-19 14:56 UTC (permalink / raw)
  To: gdb-patches, tdevries

On Wed, 2019-06-19 at 13:04 +0200, Tom de Vries wrote:
> 
> [gdb] Fix build breaker with gcc 4.8
> 
> gdb/ChangeLog:
> 
> 2019-06-19  Tom de Vries  <tdevries@suse.de>
> 
> 	* dwarf2read.h (std::hash<sect_offset>): New specialization.

I think this might have broken the GDB build with GCC 5.4.  My
gdb/binutils build using the latest sources is dying with the
following message.  I am building on Ubuntu 18.04 and the installed GCC
5.4 compiler.  I haven't seen any errors from the buildbot, maybe that
is using GCC 4.8?

Error:


In file included from /usr/include/c++/5/bits/hashtable.h:35:0,
                 from /usr/include/c++/5/unordered_map:47,
                 from /home/sellcey/tot/src/binutils-
gdb/gdb/dwarf2read.h:23,
                 from /home/sellcey/tot/src/binutils-gdb/gdb/dwarf-
index-write.h:24,
                 from /home/sellcey/tot/src/binutils-gdb/gdb/dwarf-
index-cache.c:28:
/usr/include/c++/5/bits/hashtable_policy.h: In instantiation of ‘struct
std::__detail::__is_noexcept_hash<sect_offset, std::hash<sect_offset>
>’:
/usr/include/c++/5/type_traits:137:12:   required from ‘struct
std::__and_<std::__is_fast_hash<std::hash<sect_offset> >,
std::__detail::__is_noexcept_hash<sect_offset, std::hash<sect_offset> >
>’
/usr/include/c++/5/type_traits:148:38:   required from ‘struct
std::__not_<std::__and_<std::__is_fast_hash<std::hash<sect_offset> >,
std::__detail::__is_noexcept_hash<sect_offset, std::hash<sect_offset> >
> >’

(etc, etc, etc)

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [EXT] Re: [PATCH][gdb] Fix build breaker with gcc 4.8
       [not found]   ` <0cedb79e-1423-90d8-0c61-1386f398ab20@suse.de>
@ 2019-06-19 15:14     ` Steve Ellcey
  0 siblings, 0 replies; 6+ messages in thread
From: Steve Ellcey @ 2019-06-19 15:14 UTC (permalink / raw)
  To: gdb-patches, tdevries

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="utf-8", Size: 315 bytes --]

On Wed, 2019-06-19 at 17:00 +0200, Tom de Vries wrote:
> 
> 
> I think you're running into the same build breaker as me, for which I
> just pushed a fix (and not this one).
> 
> Thanks,
> - Tom

Yes, that patch fixes my build. Thanks.

Steve Ellcey
sellcey@marvell.com
\x16º&Öéj×!zÊÞ¶êçמ½ÓÙb²Ö«r\x18\x1dn–­r\x17¬

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2019-06-19 15:14 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-06-19 11:04 [PATCH][gdb] Fix build breaker with gcc 4.8 Tom de Vries
2019-06-19 12:43 ` Pedro Alves
2019-06-19 14:35   ` Tom de Vries
2019-06-19 14:49     ` Pedro Alves
2019-06-19 14:56 ` Steve Ellcey
     [not found]   ` <0cedb79e-1423-90d8-0c61-1386f398ab20@suse.de>
2019-06-19 15:14     ` [EXT] " Steve Ellcey

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox