Displaying 20 results from an estimated 24 matches for "compiler_used".
2018 Feb 07
4
ThinLTO and linkonce_odr + unnamed_addr
...Teresa, we should probably do #2 to preserve behavior for now.
On Wed, Feb 7, 2018 at 9:34 AM, Teresa Johnson via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
> Hi Steven,
>
> I'd prefer not to inhibit importing. I am also concerned about putting
> these symbols in the llvm.compiler_used (I don't recall earlier discussion
> around this, but it seems like it could have effects on optimization as you
> mention).
>
> What are the downsides of #2 (adding visibility hidden)? We already do
> this when promoting internal linkage to external due to importing. I'm not...
2018 Feb 07
7
ThinLTO and linkonce_odr + unnamed_addr
...roperty. As a result, it introduces unnecessary weak external symbols and weak external are not good for performance on darwin platforms.
I have few proposed solutions for this issue but I don't know which one works the best for none macho platforms and other LTO clients like lld.
1. Use llvm.compiler_used.
As far as I know, the linkage promote are just there to keep the symbol through internalize and codegen so adding them to compiler used should solve this issue. I was told that there was some objections to do that in the first place. Is it because the globals added to compiler used is ignored by...
2018 Feb 07
0
ThinLTO and linkonce_odr + unnamed_addr
Isn't #2 actually changing the behavior by changing the visibility while
llvm.compiler_used seems actually closest to current behavior to me?
The visibility hidden would break comparison of function as you mentioned,
which is not the case with auto-hide if I didn't miss anything.
--
Mehdi
2018-02-07 11:29 GMT-08:00 Reid Kleckner via llvm-dev <
llvm-dev at lists.llvm.org>:...
2018 Feb 07
0
ThinLTO and linkonce_odr + unnamed_addr
Hi Steven,
I'd prefer not to inhibit importing. I am also concerned about putting
these symbols in the llvm.compiler_used (I don't recall earlier discussion
around this, but it seems like it could have effects on optimization as you
mention).
What are the downsides of #2 (adding visibility hidden)? We already do this
when promoting internal linkage to external due to importing. I'm not an
expert on how this w...
2018 Feb 07
0
ThinLTO and linkonce_odr + unnamed_addr
...es unnecessary weak
> external symbols and weak external are not good for performance on darwin
> platforms.
>
> I have few proposed solutions for this issue but I don't know which one
> works the best for none macho platforms and other LTO clients like lld.
>
> 1. Use llvm.compiler_used.
> As far as I know, the linkage promote are just there to keep the symbol
> through internalize and codegen so adding them to compiler used should
> solve this issue. I was told that there was some objections to do that in
> the first place. Is it because the globals added to compiler...
2018 Feb 07
0
ThinLTO and linkonce_odr + unnamed_addr
...ehavior for now.
>
> On Wed, Feb 7, 2018 at 9:34 AM, Teresa Johnson via llvm-dev <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote:
> Hi Steven,
>
> I'd prefer not to inhibit importing. I am also concerned about putting these symbols in the llvm.compiler_used (I don't recall earlier discussion around this, but it seems like it could have effects on optimization as you mention).
>
> What are the downsides of #2 (adding visibility hidden)? We already do this when promoting internal linkage to external due to importing. I'm not an expert on...
2018 Feb 07
2
ThinLTO and linkonce_odr + unnamed_addr
...o preserve behavior for now.
>
> On Wed, Feb 7, 2018 at 9:34 AM, Teresa Johnson via llvm-dev <
> llvm-dev at lists.llvm.org> wrote:
>
>> Hi Steven,
>>
>> I'd prefer not to inhibit importing. I am also concerned about putting
>> these symbols in the llvm.compiler_used (I don't recall earlier discussion
>> around this, but it seems like it could have effects on optimization as you
>> mention).
>>
>> What are the downsides of #2 (adding visibility hidden)? We already do
>> this when promoting internal linkage to external due to im...
2018 Feb 07
0
ThinLTO and linkonce_odr + unnamed_addr
...;>
>> On Wed, Feb 7, 2018 at 9:34 AM, Teresa Johnson via llvm-dev <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote:
>> Hi Steven,
>>
>> I'd prefer not to inhibit importing. I am also concerned about putting these symbols in the llvm.compiler_used (I don't recall earlier discussion around this, but it seems like it could have effects on optimization as you mention).
>>
>> What are the downsides of #2 (adding visibility hidden)? We already do this when promoting internal linkage to external due to importing. I'm not an ex...
2018 Feb 07
2
ThinLTO and linkonce_odr + unnamed_addr
...;>
>> On Wed, Feb 7, 2018 at 9:34 AM, Teresa Johnson via llvm-dev <
>> llvm-dev at lists.llvm.org> wrote:
>>
>>> Hi Steven,
>>>
>>> I'd prefer not to inhibit importing. I am also concerned about putting
>>> these symbols in the llvm.compiler_used (I don't recall earlier discussion
>>> around this, but it seems like it could have effects on optimization as you
>>> mention).
>>>
>>> What are the downsides of #2 (adding visibility hidden)? We already do
>>> this when promoting internal linkage t...
2018 Feb 08
2
ThinLTO and linkonce_odr + unnamed_addr
...eb 7, 2018 at 9:34 AM, Teresa Johnson via llvm-dev <
>>> llvm-dev at lists.llvm.org> wrote:
>>>
>>>> Hi Steven,
>>>>
>>>> I'd prefer not to inhibit importing. I am also concerned about putting
>>>> these symbols in the llvm.compiler_used (I don't recall earlier discussion
>>>> around this, but it seems like it could have effects on optimization as you
>>>> mention).
>>>>
>>>> What are the downsides of #2 (adding visibility hidden)? We already do
>>>> this when promotin...
2018 Feb 07
0
ThinLTO and linkonce_odr + unnamed_addr
...gt; On Wed, Feb 7, 2018 at 9:34 AM, Teresa Johnson via llvm-dev <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote:
>>> Hi Steven,
>>>
>>> I'd prefer not to inhibit importing. I am also concerned about putting these symbols in the llvm.compiler_used (I don't recall earlier discussion around this, but it seems like it could have effects on optimization as you mention).
>>>
>>> What are the downsides of #2 (adding visibility hidden)? We already do this when promoting internal linkage to external due to importing. I'm n...
2018 Feb 08
3
ThinLTO and linkonce_odr + unnamed_addr
...eresa Johnson via llvm-dev <
>>>> llvm-dev at lists.llvm.org> wrote:
>>>>
>>>>> Hi Steven,
>>>>>
>>>>> I'd prefer not to inhibit importing. I am also concerned about putting
>>>>> these symbols in the llvm.compiler_used (I don't recall earlier discussion
>>>>> around this, but it seems like it could have effects on optimization as you
>>>>> mention).
>>>>>
>>>>> What are the downsides of #2 (adding visibility hidden)? We already do
>>>>>...
2018 Feb 08
0
ThinLTO and linkonce_odr + unnamed_addr
...Feb 7, 2018 at 9:34 AM, Teresa Johnson via llvm-dev <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote:
>>>> Hi Steven,
>>>>
>>>> I'd prefer not to inhibit importing. I am also concerned about putting these symbols in the llvm.compiler_used (I don't recall earlier discussion around this, but it seems like it could have effects on optimization as you mention).
>>>>
>>>> What are the downsides of #2 (adding visibility hidden)? We already do this when promoting internal linkage to external due to importing. I...
2018 Feb 08
0
ThinLTO and linkonce_odr + unnamed_addr
...at 9:34 AM, Teresa Johnson via llvm-dev <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote:
>>>>> Hi Steven,
>>>>>
>>>>> I'd prefer not to inhibit importing. I am also concerned about putting these symbols in the llvm.compiler_used (I don't recall earlier discussion around this, but it seems like it could have effects on optimization as you mention).
>>>>>
>>>>> What are the downsides of #2 (adding visibility hidden)? We already do this when promoting internal linkage to external due to impo...
2018 Feb 08
2
ThinLTO and linkonce_odr + unnamed_addr
...ev <
>>>>> llvm-dev at lists.llvm.org> wrote:
>>>>>
>>>>>> Hi Steven,
>>>>>>
>>>>>> I'd prefer not to inhibit importing. I am also concerned about
>>>>>> putting these symbols in the llvm.compiler_used (I don't recall earlier
>>>>>> discussion around this, but it seems like it could have effects on
>>>>>> optimization as you mention).
>>>>>>
>>>>>> What are the downsides of #2 (adding visibility hidden)? We already
>&g...
2018 Feb 09
2
ThinLTO and linkonce_odr + unnamed_addr
...>> llvm-dev at lists.llvm.org> wrote:
>>>>>>
>>>>>>> Hi Steven,
>>>>>>>
>>>>>>> I'd prefer not to inhibit importing. I am also concerned about
>>>>>>> putting these symbols in the llvm.compiler_used (I don't recall earlier
>>>>>>> discussion around this, but it seems like it could have effects on
>>>>>>> optimization as you mention).
>>>>>>>
>>>>>>> What are the downsides of #2 (adding visibility hidden)? W...
2018 Feb 09
0
ThinLTO and linkonce_odr + unnamed_addr
...Teresa Johnson via llvm-dev <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote:
>>>>>> Hi Steven,
>>>>>>
>>>>>> I'd prefer not to inhibit importing. I am also concerned about putting these symbols in the llvm.compiler_used (I don't recall earlier discussion around this, but it seems like it could have effects on optimization as you mention).
>>>>>>
>>>>>> What are the downsides of #2 (adding visibility hidden)? We already do this when promoting internal linkage to external due...
2018 Feb 08
1
ThinLTO and linkonce_odr + unnamed_addr
...ev <
>>>>> llvm-dev at lists.llvm.org> wrote:
>>>>>
>>>>>> Hi Steven,
>>>>>>
>>>>>> I'd prefer not to inhibit importing. I am also concerned about
>>>>>> putting these symbols in the llvm.compiler_used (I don't recall earlier
>>>>>> discussion around this, but it seems like it could have effects on
>>>>>> optimization as you mention).
>>>>>>
>>>>>> What are the downsides of #2 (adding visibility hidden)? We already
>&g...
2018 Feb 09
0
ThinLTO and linkonce_odr + unnamed_addr
...on via llvm-dev <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote:
>>>>>>> Hi Steven,
>>>>>>>
>>>>>>> I'd prefer not to inhibit importing. I am also concerned about putting these symbols in the llvm.compiler_used (I don't recall earlier discussion around this, but it seems like it could have effects on optimization as you mention).
>>>>>>>
>>>>>>> What are the downsides of #2 (adding visibility hidden)? We already do this when promoting internal linkage to exte...
2018 Feb 07
3
ThinLTO and linkonce_odr + unnamed_addr
..., it introduces unnecessary weak external symbols and weak external are not good for performance on darwin platforms.
>
> I have few proposed solutions for this issue but I don't know which one works the best for none macho platforms and other LTO clients like lld.
>
> 1. Use llvm.compiler_used.
> As far as I know, the linkage promote are just there to keep the symbol through internalize and codegen so adding them to compiler used should solve this issue. I was told that there was some objections to do that in the first place. Is it because the globals added to compiler used is ignored...