llvm dev - Nov 2013 - [LLVMdev] bit code file incompatibility due to debug info changes

On Thu, Nov 21, 2013 at 10:38 AM, Manman Ren <manman.ren at gmail.com>
wrote:
>
>
>
> On Tue, Nov 19, 2013 at 8:47 PM, Manman Ren <manman.ren at gmail.com>
wrote:
>
>>
>>
>>
>> On Tue, Nov 19, 2013 at 5:26 PM, Manman Ren <manman.ren at
gmail.com> wrote:
>>
>>>
>>>
>>>
>>> On Mon, Nov 18, 2013 at 11:00 AM, Chandler Carruth <chandlerc at
google.com
>>> > wrote:
>>>
>>>>
>>>> On Mon, Nov 18, 2013 at 10:55 AM, David Blaikie <dblaikie at
gmail.com>wrote:
>>>>
>>>>> It depends a bit, also, on what kind of guarantees we need
to offer.
>>>>> If the guarantee when reading IR from disk is "will
not crash" then there's
>>>>> nothing for it but to run full debug info verification.
>>>>>
>>>>> On the other hand, if we can assume that some specific
metadata
>>>>> implies the correctness of some other metadata, then all we
need to do is
>>>>> check a known debug info version number. If it's the
current number, do
>>>>> nothing (assume the rest of the debug info is up to date
and well formed),
>>>>> otherwise do all the work to find the debug info and dump
it (no need to
>>>>> verify it - we just have to do the work to find it, so we
don't go
>>>>> following bad links later on - that should be as easy as
dropping the
>>>>> llvm.dbg.cu named node and removing all debug intrinsics
and the
>>>>> instruction metadata line references). But this latter
scheme isn't robust
>>>>> against arbitrary metadata (that could, coincidentally,
have the right
>>>>> version number and arbitrary metadata that breaks all our
debug info
>>>>> metadata assumptions)
>>>>>
>>>>> If the latter is sufficient for everyone's
needs/principles, great.
>>>>>
>>>>
>>>> This makes sense to me, but I see Eric's fundamental
concern with
>>>> upgrading test cases.
>>>>
>>>>
>>>> One other possible idea: we could artificially force coarseness
on
>>>> metadata while things are still iterating rapidly by just
having a version
>>>> number that rotates every "release". (Even
non-open-source-releases could
>>>> be handled by letting anyone, any time they need to rev it,
update what the
>>>> current version is and update every test case to reference that
version.)
>>>> If the version is nicely factored, it should at least be an
obvious diff if
>>>> still huge.
>>>>
>>>
>>> Adding a debug info version number and ignoring the debug info
MDNodes
>>> when version does not match makes sense. And changing the version
number
>>> with every release sounds reasonable.
>>>
>>> One implementation is
>>> 1> completely get rid of version number in the tag
>>>   LLVMDebugVersion is only used in two places and we should be able
to
>>> remove it.
>>>     include/llvm/DebugInfo.h:    return getUnsignedField(0) &
>>> ~LLVMDebugVersionMask;
>>>     lib/IR/DIBuilder.cpp:  assert((Tag & LLVMDebugVersionMask)
== 0 &&
>>>     lib/IR/DIBuilder.cpp:  return
>>> ConstantInt::get(Type::getInt32Ty(VMContext), Tag |
LLVMDebugVersion);
>>>   I remember there was something that blocks David from completely
>>> getting rid of version number, but it seems the blocker is gone
now.
>>>   This can be done separately as well.
>>>
>>> 2> introduce a debug info version in module flags similar to
"dwarf
>>> version" module flag
>>>      LTO can correctly handle the module flag during linking and we
only
>>> need to change one line in the debug info testing cases when we
update
>>> debug info version number.
>>>      One issue is that we don't have a mode that emits a
warning and
>>> outputs the largest value (we have a mode that emits a warning and
outputs
>>> the value from the first module, but we can definitely add another
mode if
>>> necessary)
>>>
>>> 3> when to drop the debug info MDNodes?
>>>      We could do this in the bit code loader but it seems a
violation of
>>> layers. One possibility is to run a module pass right after loading
the bit
>>> code, to remove debug intrinsics and debug tags on instructions.
>>>
>>
>> We have an existing pass StripSymbols that can strip debug info in the
>> module.
>> So it is just a matter of adding that pass after bit code loader if the
>> debug info version number does not match.
>>
>
> Can I assume no objection to the approach? :)
> Bill, are you okay with adding another mode to module flags? i.e. emits a
> warning and outputs the largest value.
>
How will this work as a module flag, though? If the flag gets merged and
maximized, how will we know which compile units have out of date debug info
metadata and drop them?

- David

>
> Thanks,
> Manman
>
>
>> 4> how to report warning?
>>     Is errs() << "WARNING: " good enough? BTW this is
how we emit warning
>> when linking module flags.
>>
>> Any objection to the above? Any other suggestion?
>>
>> Thanks,
>> Manman
>>
>>
>>> _______________________________________________
>>> LLVM Developers mailing list
>>> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
>>> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>>>
>>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20131121/66aad44d/attachment.html>

On Thu, Nov 21, 2013 at 10:57 AM, David Blaikie <dblaikie at gmail.com>
wrote:
>
>
>
> On Thu, Nov 21, 2013 at 10:38 AM, Manman Ren <manman.ren at
gmail.com> wrote:
>
>>
>>
>>
>> On Tue, Nov 19, 2013 at 8:47 PM, Manman Ren <manman.ren at
gmail.com> wrote:
>>
>>>
>>>
>>>
>>> On Tue, Nov 19, 2013 at 5:26 PM, Manman Ren <manman.ren at
gmail.com>wrote:
>>>
>>>>
>>>>
>>>>
>>>> On Mon, Nov 18, 2013 at 11:00 AM, Chandler Carruth <
>>>> chandlerc at google.com> wrote:
>>>>
>>>>>
>>>>> On Mon, Nov 18, 2013 at 10:55 AM, David Blaikie
<dblaikie at gmail.com>wrote:
>>>>>
>>>>>> It depends a bit, also, on what kind of guarantees we
need to offer.
>>>>>> If the guarantee when reading IR from disk is
"will not crash" then there's
>>>>>> nothing for it but to run full debug info verification.
>>>>>>
>>>>>> On the other hand, if we can assume that some specific
metadata
>>>>>> implies the correctness of some other metadata, then
all we need to do is
>>>>>> check a known debug info version number. If it's
the current number, do
>>>>>> nothing (assume the rest of the debug info is up to
date and well formed),
>>>>>> otherwise do all the work to find the debug info and
dump it (no need to
>>>>>> verify it - we just have to do the work to find it, so
we don't go
>>>>>> following bad links later on - that should be as easy
as dropping the
>>>>>> llvm.dbg.cu named node and removing all debug
intrinsics and the
>>>>>> instruction metadata line references). But this latter
scheme isn't robust
>>>>>> against arbitrary metadata (that could, coincidentally,
have the right
>>>>>> version number and arbitrary metadata that breaks all
our debug info
>>>>>> metadata assumptions)
>>>>>>
>>>>>> If the latter is sufficient for everyone's
needs/principles, great.
>>>>>>
>>>>>
>>>>> This makes sense to me, but I see Eric's fundamental
concern with
>>>>> upgrading test cases.
>>>>>
>>>>>
>>>>> One other possible idea: we could artificially force
coarseness on
>>>>> metadata while things are still iterating rapidly by just
having a version
>>>>> number that rotates every "release". (Even
non-open-source-releases could
>>>>> be handled by letting anyone, any time they need to rev it,
update what the
>>>>> current version is and update every test case to reference
that version.)
>>>>> If the version is nicely factored, it should at least be an
obvious diff if
>>>>> still huge.
>>>>>
>>>>
>>>> Adding a debug info version number and ignoring the debug info
MDNodes
>>>> when version does not match makes sense. And changing the
version number
>>>> with every release sounds reasonable.
>>>>
>>>> One implementation is
>>>> 1> completely get rid of version number in the tag
>>>>   LLVMDebugVersion is only used in two places and we should be
able to
>>>> remove it.
>>>>     include/llvm/DebugInfo.h:    return getUnsignedField(0)
&
>>>> ~LLVMDebugVersionMask;
>>>>     lib/IR/DIBuilder.cpp:  assert((Tag &
LLVMDebugVersionMask) == 0 &&
>>>>     lib/IR/DIBuilder.cpp:  return
>>>> ConstantInt::get(Type::getInt32Ty(VMContext), Tag |
LLVMDebugVersion);
>>>>   I remember there was something that blocks David from
completely
>>>> getting rid of version number, but it seems the blocker is gone
now.
>>>>   This can be done separately as well.
>>>>
>>>> 2> introduce a debug info version in module flags similar to
"dwarf
>>>> version" module flag
>>>>      LTO can correctly handle the module flag during linking
and we
>>>> only need to change one line in the debug info testing cases
when we update
>>>> debug info version number.
>>>>      One issue is that we don't have a mode that emits a
warning and
>>>> outputs the largest value (we have a mode that emits a warning
and outputs
>>>> the value from the first module, but we can definitely add
another mode if
>>>> necessary)
>>>>
>>>> 3> when to drop the debug info MDNodes?
>>>>      We could do this in the bit code loader but it seems a
violation
>>>> of layers. One possibility is to run a module pass right after
loading the
>>>> bit code, to remove debug intrinsics and debug tags on
instructions.
>>>>
>>>
>>> We have an existing pass StripSymbols that can strip debug info in
the
>>> module.
>>> So it is just a matter of adding that pass after bit code loader if
the
>>> debug info version number does not match.
>>>
>>
>> Can I assume no objection to the approach? :)
>> Bill, are you okay with adding another mode to module flags? i.e. emits
a
>> warning and outputs the largest value.
>>
>
> How will this work as a module flag, though? If the flag gets merged and
> maximized, how will we know which compile units have out of date debug info
> metadata and drop them?
>
The dropping part comes from item 3 above: adding StripSymbols pass right
after bit code loader if the debug info version number does not match.
Item 3 is shared between non-LTO and LTO. For LTO, we want the debug info
version number gets merged and maximized.

Thanks
Manman

>
> - David
>
>
>>
>> Thanks,
>> Manman
>>
>>
>>> 4> how to report warning?
>>>     Is errs() << "WARNING: " good enough? BTW this
is how we emit
>>> warning when linking module flags.
>>>
>>> Any objection to the above? Any other suggestion?
>>>
>>> Thanks,
>>> Manman
>>>
>>>
>>>> _______________________________________________
>>>> LLVM Developers mailing list
>>>> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
>>>> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>>>>
>>>>
>>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20131121/9ce852b9/attachment.html>

On Thu, Nov 21, 2013 at 11:06 AM, Manman Ren <manman.ren at gmail.com>
wrote:
>
>
>
> On Thu, Nov 21, 2013 at 10:57 AM, David Blaikie <dblaikie at
gmail.com> wrote:
>>
>>
>>
>>
>> On Thu, Nov 21, 2013 at 10:38 AM, Manman Ren <manman.ren at
gmail.com> wrote:
>>>
>>>
>>>
>>>
>>> On Tue, Nov 19, 2013 at 8:47 PM, Manman Ren <manman.ren at
gmail.com> wrote:
>>>>
>>>>
>>>>
>>>>
>>>> On Tue, Nov 19, 2013 at 5:26 PM, Manman Ren <manman.ren at
gmail.com>
>>>> wrote:
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Mon, Nov 18, 2013 at 11:00 AM, Chandler Carruth
>>>>> <chandlerc at google.com> wrote:
>>>>>>
>>>>>>
>>>>>> On Mon, Nov 18, 2013 at 10:55 AM, David Blaikie
<dblaikie at gmail.com>
>>>>>> wrote:
>>>>>>>
>>>>>>> It depends a bit, also, on what kind of guarantees
we need to offer.
>>>>>>> If the guarantee when reading IR from disk is
"will not crash" then there's
>>>>>>> nothing for it but to run full debug info
verification.
>>>>>>>
>>>>>>> On the other hand, if we can assume that some
specific metadata
>>>>>>> implies the correctness of some other metadata,
then all we need to do is
>>>>>>> check a known debug info version number. If
it's the current number, do
>>>>>>> nothing (assume the rest of the debug info is up to
date and well formed),
>>>>>>> otherwise do all the work to find the debug info
and dump it (no need to
>>>>>>> verify it - we just have to do the work to find it,
so we don't go following
>>>>>>> bad links later on - that should be as easy as
dropping the llvm.dbg.cu
>>>>>>> named node and removing all debug intrinsics and
the instruction metadata
>>>>>>> line references). But this latter scheme isn't
robust against arbitrary
>>>>>>> metadata (that could, coincidentally, have the
right version number and
>>>>>>> arbitrary metadata that breaks all our debug info
metadata assumptions)
>>>>>>>
>>>>>>> If the latter is sufficient for everyone's
needs/principles, great.
>>>>>>
>>>>>>
>>>>>> This makes sense to me, but I see Eric's
fundamental concern with
>>>>>> upgrading test cases.
>>>>>>
>>>>>>
>>>>>> One other possible idea: we could artificially force
coarseness on
>>>>>> metadata while things are still iterating rapidly by
just having a version
>>>>>> number that rotates every "release". (Even
non-open-source-releases could be
>>>>>> handled by letting anyone, any time they need to rev
it, update what the
>>>>>> current version is and update every test case to
reference that version.) If
>>>>>> the version is nicely factored, it should at least be
an obvious diff if
>>>>>> still huge.
>>>>>
>>>>>
>>>>> Adding a debug info version number and ignoring the debug
info MDNodes
>>>>> when version does not match makes sense. And changing the
version number
>>>>> with every release sounds reasonable.
>>>>>
>>>>> One implementation is
>>>>> 1> completely get rid of version number in the tag
>>>>>   LLVMDebugVersion is only used in two places and we should
be able to
>>>>> remove it.
>>>>>     include/llvm/DebugInfo.h:    return getUnsignedField(0)
&
>>>>> ~LLVMDebugVersionMask;
>>>>>     lib/IR/DIBuilder.cpp:  assert((Tag &
LLVMDebugVersionMask) == 0 &&
>>>>>     lib/IR/DIBuilder.cpp:  return
>>>>> ConstantInt::get(Type::getInt32Ty(VMContext), Tag |
LLVMDebugVersion);
>>>>>   I remember there was something that blocks David from
completely
>>>>> getting rid of version number, but it seems the blocker is
gone now.
>>>>>   This can be done separately as well.
>>>>>
>>>>> 2> introduce a debug info version in module flags
similar to "dwarf
>>>>> version" module flag
>>>>>      LTO can correctly handle the module flag during
linking and we
>>>>> only need to change one line in the debug info testing
cases when we update
>>>>> debug info version number.
>>>>>      One issue is that we don't have a mode that emits
a warning and
>>>>> outputs the largest value (we have a mode that emits a
warning and outputs
>>>>> the value from the first module, but we can definitely add
another mode if
>>>>> necessary)
>>>>>
>>>>> 3> when to drop the debug info MDNodes?
>>>>>      We could do this in the bit code loader but it seems a
violation
>>>>> of layers. One possibility is to run a module pass right
after loading the
>>>>> bit code, to remove debug intrinsics and debug tags on
instructions.
>>>>
>>>>
>>>> We have an existing pass StripSymbols that can strip debug info
in the
>>>> module.
>>>> So it is just a matter of adding that pass after bit code
loader if the
>>>> debug info version number does not match.
>>>
>>>
>>> Can I assume no objection to the approach? :)
>>> Bill, are you okay with adding another mode to module flags? i.e.
emits a
>>> warning and outputs the largest value.
>>
>>
>> How will this work as a module flag, though? If the flag gets merged
and
>> maximized, how will we know which compile units have out of date debug
info
>> metadata and drop them?
>
>
> The dropping part comes from item 3 above: adding StripSymbols pass right
> after bit code loader if the debug info version number does not match.
> Item 3 is shared between non-LTO and LTO. For LTO, we want the debug info
> version number gets merged and maximized.
>
You'll want to do it at loading time and not worry about merging at
all. I think this will need to go just after reading the module and
check the version against the hard coded version in the bitcode
reader/somewhere. I think we should do it there rather than as a later
pass as we really don't want to do anything to the metadata other than
strip it.

-eric
> Thanks
> Manman
>
>>
>>
>> - David
>>
>>>
>>>
>>> Thanks,
>>> Manman
>>>
>>>>
>>>> 4> how to report warning?
>>>>     Is errs() << "WARNING: " good enough? BTW
this is how we emit
>>>> warning when linking module flags.
>>>>
>>>> Any objection to the above? Any other suggestion?
>>>>
>>>> Thanks,
>>>> Manman
>>>>
>>>>>
>>>>> _______________________________________________
>>>>> LLVM Developers mailing list
>>>>> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
>>>>> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>>>>>
>>>>
>>>
>>>
>>
>

On Thu, Nov 21, 2013 at 11:06 AM, Manman Ren <manman.ren at gmail.com>
wrote:
>
>
>
> On Thu, Nov 21, 2013 at 10:57 AM, David Blaikie <dblaikie at
gmail.com>wrote:
>
>>
>>
>>
>> On Thu, Nov 21, 2013 at 10:38 AM, Manman Ren <manman.ren at
gmail.com>wrote:
>>
>>>
>>>
>>>
>>> On Tue, Nov 19, 2013 at 8:47 PM, Manman Ren <manman.ren at
gmail.com>wrote:
>>>
>>>>
>>>>
>>>>
>>>> On Tue, Nov 19, 2013 at 5:26 PM, Manman Ren <manman.ren at
gmail.com>wrote:
>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Mon, Nov 18, 2013 at 11:00 AM, Chandler Carruth <
>>>>> chandlerc at google.com> wrote:
>>>>>
>>>>>>
>>>>>> On Mon, Nov 18, 2013 at 10:55 AM, David Blaikie
<dblaikie at gmail.com>wrote:
>>>>>>
>>>>>>> It depends a bit, also, on what kind of guarantees
we need to offer.
>>>>>>> If the guarantee when reading IR from disk is
"will not crash" then there's
>>>>>>> nothing for it but to run full debug info
verification.
>>>>>>>
>>>>>>> On the other hand, if we can assume that some
specific metadata
>>>>>>> implies the correctness of some other metadata,
then all we need to do is
>>>>>>> check a known debug info version number. If
it's the current number, do
>>>>>>> nothing (assume the rest of the debug info is up to
date and well formed),
>>>>>>> otherwise do all the work to find the debug info
and dump it (no need to
>>>>>>> verify it - we just have to do the work to find it,
so we don't go
>>>>>>> following bad links later on - that should be as
easy as dropping the
>>>>>>> llvm.dbg.cu named node and removing all debug
intrinsics and the
>>>>>>> instruction metadata line references). But this
latter scheme isn't robust
>>>>>>> against arbitrary metadata (that could,
coincidentally, have the right
>>>>>>> version number and arbitrary metadata that breaks
all our debug info
>>>>>>> metadata assumptions)
>>>>>>>
>>>>>>> If the latter is sufficient for everyone's
needs/principles, great.
>>>>>>>
>>>>>>
>>>>>> This makes sense to me, but I see Eric's
fundamental concern with
>>>>>> upgrading test cases.
>>>>>>
>>>>>>
>>>>>> One other possible idea: we could artificially force
coarseness on
>>>>>> metadata while things are still iterating rapidly by
just having a version
>>>>>> number that rotates every "release". (Even
non-open-source-releases could
>>>>>> be handled by letting anyone, any time they need to rev
it, update what the
>>>>>> current version is and update every test case to
reference that version.)
>>>>>> If the version is nicely factored, it should at least
be an obvious diff if
>>>>>> still huge.
>>>>>>
>>>>>
>>>>> Adding a debug info version number and ignoring the debug
info MDNodes
>>>>> when version does not match makes sense. And changing the
version number
>>>>> with every release sounds reasonable.
>>>>>
>>>>> One implementation is
>>>>> 1> completely get rid of version number in the tag
>>>>>   LLVMDebugVersion is only used in two places and we should
be able to
>>>>> remove it.
>>>>>     include/llvm/DebugInfo.h:    return getUnsignedField(0)
&
>>>>> ~LLVMDebugVersionMask;
>>>>>     lib/IR/DIBuilder.cpp:  assert((Tag &
LLVMDebugVersionMask) == 0 &&
>>>>>     lib/IR/DIBuilder.cpp:  return
>>>>> ConstantInt::get(Type::getInt32Ty(VMContext), Tag |
LLVMDebugVersion);
>>>>>   I remember there was something that blocks David from
completely
>>>>> getting rid of version number, but it seems the blocker is
gone now.
>>>>>   This can be done separately as well.
>>>>>
>>>>> 2> introduce a debug info version in module flags
similar to "dwarf
>>>>> version" module flag
>>>>>      LTO can correctly handle the module flag during
linking and we
>>>>> only need to change one line in the debug info testing
cases when we update
>>>>> debug info version number.
>>>>>      One issue is that we don't have a mode that emits
a warning and
>>>>> outputs the largest value (we have a mode that emits a
warning and outputs
>>>>> the value from the first module, but we can definitely add
another mode if
>>>>> necessary)
>>>>>
>>>>> 3> when to drop the debug info MDNodes?
>>>>>      We could do this in the bit code loader but it seems a
violation
>>>>> of layers. One possibility is to run a module pass right
after loading the
>>>>> bit code, to remove debug intrinsics and debug tags on
instructions.
>>>>>
>>>>
>>>> We have an existing pass StripSymbols that can strip debug info
in the
>>>> module.
>>>> So it is just a matter of adding that pass after bit code
loader if the
>>>> debug info version number does not match.
>>>>
>>>
>>> Can I assume no objection to the approach? :)
>>> Bill, are you okay with adding another mode to module flags? i.e.
emits
>>> a warning and outputs the largest value.
>>>
>>
>> How will this work as a module flag, though? If the flag gets merged
and
>> maximized, how will we know which compile units have out of date debug
info
>> metadata and drop them?
>>
>
> The dropping part comes from item 3 above: adding StripSymbols pass right
> after bit code loader if the debug info version number does not match.
>
(what Eric said, but also)

I'm just not really understanding how you choose which things to strip in
(3). If you've already merged IR modules together and the debug info
version is bumped to the maximum - if we had an old and a new module, the
resulting version would be 'new' and we wouldn't know we still had
some old
data in there we needed to remove, would we? (and even if we did, we
wouldn't know which of the compile units we needed to drop and which
intrinsics we needed to strip, etc - since some was new and good and some
was old and busted)

> Item 3 is shared between non-LTO and LTO. For LTO, we want the debug info
> version number gets merged and maximized.
>
> Thanks
> Manman
>
>
>>
>> - David
>>
>>
>>>
>>> Thanks,
>>> Manman
>>>
>>>
>>>> 4> how to report warning?
>>>>     Is errs() << "WARNING: " good enough? BTW
this is how we emit
>>>> warning when linking module flags.
>>>>
>>>> Any objection to the above? Any other suggestion?
>>>>
>>>> Thanks,
>>>> Manman
>>>>
>>>>
>>>>> _______________________________________________
>>>>> LLVM Developers mailing list
>>>>> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
>>>>> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>>>>>
>>>>>
>>>>
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20131121/ef481334/attachment.html>