Re: ARM issue with old binutils

2011-06-26 Thread Roman Divacky
On Sat, Jun 25, 2011 at 09:16:03AM -0600, Gerald Pfeifer wrote:
> On Wed, 22 Jun 2011, Damjan Marion wrote:
> > I see 3 options to fix this:
> > 
> > 1. Ask clang folks to patch llvm to use old mnemonics  ("mov r0, r0, rrx" 
> > instead of "rrx r0,r0")
> > 2. Maintain same patch for freebsd only
> > 3. patch binutils to support this new mnemonics
> 
>   4. Finally upgrade to a modern version of binutils.
> 
> Yes, I know that is GPLv3.  Been there, done that, and it is not a problem 
> at all, just FUD.  IBM, SAP, Oracle, and all the others are not concerned
> about GPLv3 in the toolchain, nor should we.  Except for FUD.

actually, there's 5. Use llvm integrated assembler (once it's mature enough),
we wont get newer linker this way but thats less pressing issue imho.

roman
___
freebsd-toolchain@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain
To unsubscribe, send any mail to "freebsd-toolchain-unsubscr...@freebsd.org"


Re: ARM issue with old binutils

2011-06-25 Thread Warner Losh

On Jun 25, 2011, at 9:16 AM, Gerald Pfeifer wrote:

> On Wed, 22 Jun 2011, Damjan Marion wrote:
>> I see 3 options to fix this:
>> 
>> 1. Ask clang folks to patch llvm to use old mnemonics  ("mov r0, r0, rrx" 
>> instead of "rrx r0,r0")
>> 2. Maintain same patch for freebsd only
>> 3. patch binutils to support this new mnemonics
> 
>  4. Finally upgrade to a modern version of binutils.
> 
> Yes, I know that is GPLv3.  Been there, done that, and it is not a problem 
> at all, just FUD.  IBM, SAP, Oracle, and all the others are not concerned
> about GPLv3 in the toolchain, nor should we.  Except for FUD.

Except there *ARE* FreeBSD users that have said that it is a real problem for 
them.  It isn't FUD.  The project has adopted the policy in reaction to large 
commercial FreeBSD users that have very restrictive company policies driven by 
their legal department's evaluation of GPLv3.

Warner

___
freebsd-toolchain@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain
To unsubscribe, send any mail to "freebsd-toolchain-unsubscr...@freebsd.org"


Re: ARM issue with old binutils

2011-06-25 Thread Gerald Pfeifer
On Wed, 22 Jun 2011, Damjan Marion wrote:
> I see 3 options to fix this:
> 
> 1. Ask clang folks to patch llvm to use old mnemonics  ("mov r0, r0, rrx" 
> instead of "rrx r0,r0")
> 2. Maintain same patch for freebsd only
> 3. patch binutils to support this new mnemonics

  4. Finally upgrade to a modern version of binutils.

Yes, I know that is GPLv3.  Been there, done that, and it is not a problem 
at all, just FUD.  IBM, SAP, Oracle, and all the others are not concerned
about GPLv3 in the toolchain, nor should we.  Except for FUD.

Gerald

___
freebsd-toolchain@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain
To unsubscribe, send any mail to "freebsd-toolchain-unsubscr...@freebsd.org"


Re: ARM issue with old binutils

2011-06-25 Thread Damjan Marion

On Jun 25, 2011, at 5:16 PM, Gerald Pfeifer wrote:

> On Wed, 22 Jun 2011, Damjan Marion wrote:
>> I see 3 options to fix this:
>> 
>> 1. Ask clang folks to patch llvm to use old mnemonics  ("mov r0, r0, rrx" 
>> instead of "rrx r0,r0")
>> 2. Maintain same patch for freebsd only
>> 3. patch binutils to support this new mnemonics
> 
>  4. Finally upgrade to a modern version of binutils.
> 
> Yes, I know that is GPLv3.  Been there, done that, and it is not a problem 
> at all, just FUD.  IBM, SAP, Oracle, and all the others are not concerned
> about GPLv3 in the toolchain, nor should we.  Except for FUD.

Too late. Fix is already in SVN (r223484) :)

Damjan___
freebsd-toolchain@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain
To unsubscribe, send any mail to "freebsd-toolchain-unsubscr...@freebsd.org"


Re: ARM issue with old binutils

2011-06-23 Thread Damjan Marion

On Jun 22, 2011, at 5:39 PM, Warner Losh wrote:

> I'd be inclined to include a small patch to FreeBSD's binutils to implement 
> this.  You might be able to snag it from a newer version of binutils if you 
> can track down the author of that code and ask if you can include it in a 
> GPLv2 version of binutils.

I just realized that apple is developing own assembler which is actually based 
on very old binutils 1.38 from 1988.
It is licensed under GPLv2. I can see some arm related stuff there which I miss 
in our assembler.

http://opensource.apple.com/tarballs/cctools/

Can we use parts from Apple AS and merge it with our binutils?

Damjan



___
freebsd-toolchain@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain
To unsubscribe, send any mail to "freebsd-toolchain-unsubscr...@freebsd.org"


Re: ARM issue with old binutils

2011-06-22 Thread Damjan Marion

On Jun 22, 2011, at 7:30 PM, Warner Losh wrote:

> 
> On Jun 22, 2011, at 10:42 AM, Ryan Stone wrote:
> 
>> On Wed, Jun 22, 2011 at 11:39 AM, Warner Losh  wrote:
>>> I'd be inclined to include a small patch to FreeBSD's binutils to implement 
>>> this.  You might be able to snag it from a newer version of binutils if you 
>>> can track down the author of that code and ask if you can include it in a 
>>> GPLv2 version of binutils.
>> 
>> Wouldn't the FSF be the copyright owner?  I thought that they required
>> copyright assignment for contributions?
> 
> Not necessarily.  Only large things need to be assigned.  Simple patches not 
> so much...

Does this mean that we can import ARMv7 support patch which is published before 
upgrade to GPLv3 ?

http://sourceware.org/ml/binutils/2006-02/msg00305.html

___
freebsd-toolchain@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain
To unsubscribe, send any mail to "freebsd-toolchain-unsubscr...@freebsd.org"


Re: ARM issue with old binutils

2011-06-22 Thread Warner Losh

On Jun 22, 2011, at 10:42 AM, Ryan Stone wrote:

> On Wed, Jun 22, 2011 at 11:39 AM, Warner Losh  wrote:
>> I'd be inclined to include a small patch to FreeBSD's binutils to implement 
>> this.  You might be able to snag it from a newer version of binutils if you 
>> can track down the author of that code and ask if you can include it in a 
>> GPLv2 version of binutils.
> 
> Wouldn't the FSF be the copyright owner?  I thought that they required
> copyright assignment for contributions?

Not necessarily.  Only large things need to be assigned.  Simple patches not so 
much...

Warner

___
freebsd-toolchain@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain
To unsubscribe, send any mail to "freebsd-toolchain-unsubscr...@freebsd.org"


Re: ARM issue with old binutils

2011-06-22 Thread Ryan Stone
On Wed, Jun 22, 2011 at 11:39 AM, Warner Losh  wrote:
> I'd be inclined to include a small patch to FreeBSD's binutils to implement 
> this.  You might be able to snag it from a newer version of binutils if you 
> can track down the author of that code and ask if you can include it in a 
> GPLv2 version of binutils.

Wouldn't the FSF be the copyright owner?  I thought that they required
copyright assignment for contributions?
___
freebsd-toolchain@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain
To unsubscribe, send any mail to "freebsd-toolchain-unsubscr...@freebsd.org"


Re: ARM issue with old binutils

2011-06-22 Thread Warner Losh
I'd be inclined to include a small patch to FreeBSD's binutils to implement 
this.  You might be able to snag it from a newer version of binutils if you can 
track down the author of that code and ask if you can include it in a GPLv2 
version of binutils.

Warner

On Jun 22, 2011, at 7:09 AM, Damjan Marion wrote:

> 
> Hi,
> 
> On my ongoing effort to support clang cross-compiling of world for ARM 
> architecture I hit one weird issue:
> 
> Problem is that old ARM Architecture Reference Manual (ARMv5) specifies 
> mnemonics "mov r0, r0, rrx" 
> while new ARM ARM defines same instruction as "rrx r0,r0". 
> 
> Both have same opcode: 0xe1a00060.
> 
> Problem is that clang currently uses GAS to assemble files and GAS version we 
> have (2.17.50)  doesn't have a clue about new mnemonics "rrx r0,r0".
> 
> I see 3 options to fix this:
> 
> 1. Ask clang folks to patch llvm to use old mnemonics  ("mov r0, r0, rrx" 
> instead of "rrx r0,r0")
> 2. Maintain same patch for freebsd only
> 3. patch binutils to support this new mnemonics
> 
> Any thoughts how to proceed?
> 
> Thanks,
> 
> Damjan
> 
> 
> 
> 
> ___
> freebsd-toolchain@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain
> To unsubscribe, send any mail to "freebsd-toolchain-unsubscr...@freebsd.org"
> 
> 

___
freebsd-toolchain@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain
To unsubscribe, send any mail to "freebsd-toolchain-unsubscr...@freebsd.org"


ARM issue with old binutils

2011-06-22 Thread Damjan Marion

Hi,

On my ongoing effort to support clang cross-compiling of world for ARM 
architecture I hit one weird issue:

Problem is that old ARM Architecture Reference Manual (ARMv5) specifies 
mnemonics "mov r0, r0, rrx" 
while new ARM ARM defines same instruction as "rrx r0,r0". 

Both have same opcode: 0xe1a00060.

Problem is that clang currently uses GAS to assemble files and GAS version we 
have (2.17.50)  doesn't have a clue about new mnemonics "rrx r0,r0".

I see 3 options to fix this:

1. Ask clang folks to patch llvm to use old mnemonics  ("mov r0, r0, rrx" 
instead of "rrx r0,r0")
2. Maintain same patch for freebsd only
3. patch binutils to support this new mnemonics

Any thoughts how to proceed?

Thanks,

Damjan




___
freebsd-toolchain@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain
To unsubscribe, send any mail to "freebsd-toolchain-unsubscr...@freebsd.org"