http://groups.google.com/group/generic-abi Discuss modification and extension of generic System V Application Binary Interface. 2009-06-08T18:16:15Z Google Groups Roland McGrath rol...@redhat.com 2009-06-08T18:16:15Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/a1108b370ed03d2e?show_docid=a1108b370ed03d2e

You seem to have misunderstood the idea.

Roland McGrath rol...@redhat.com 2009-06-08T16:59:00Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/a46d35c74c0acb1b?show_docid=a46d35c74c0acb1b

You seem to have misunderstood the idea.

H.J. Lu hjl.to...@gmail.com 2009-06-08T13:57:25Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/357d3ba29ebfd887?show_docid=357d3ba29ebfd887

Please try the current binutils in CVS. I think I have fixed all <br> problems. If not, <br> please open a bug report with a testcase. <br> Thanks.

H.J. Lu hjl.to...@gmail.com 2009-06-06T15:21:44Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/64d1c0973bd9eac2?show_docid=64d1c0973bd9eac2

This patch implements #2. It may not solve all the problems. But it <br> works for my simple testcase.

Roland McGrath rol...@redhat.com 2009-06-06T12:55:21Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/c2c244515be1a219?show_docid=c2c244515be1a219

This does not seem like too much of a constraint, given that it also <br> unavoidably has to work before the DSO's initializers have run. <br> i.e., it is always special-purpose code that has to be written very <br> carefully. Being PIC code it should not be relying on much to have been <br> done before. It just has to avoid any PLT calls or other GOT use itself.

H.J. Lu hjl.to...@gmail.com 2009-06-06T12:15:57Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/75ac801cf19b5437?show_docid=75ac801cf19b5437

Hi, <br> There is an ordering issue on STT_GNU_IFUNC functions. In shared object, <br> we need dynamic relocation to get the real function address when there is a <br> non-GOT reference against a STT_GNU_IFUNC symbols. A STT_GNU_IFUNC <br> function may use some data which need to be relocated first. Possible solutions:

H.J. Lu hjl.to...@gmail.com 2009-06-06T12:11:09Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/9928d0492db5c8e9?show_docid=9928d0492db5c8e9

Hi, <br> There is an ordering issue on STT_GNU_IFUNC functions. In shared object, <br> we need dynamic relocation to get the real function address when there is a <br> non-GOT reference against a STT_GNU_IFUNC symbols. A STT_GNU_IFUNC <br> function may use some data which need to be relocated first. Possible solutions:

Ulrich Drepper drep...@redhat.com 2009-06-05T05:30:48Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/451b763a50a3aef5?show_docid=451b763a50a3aef5

-----BEGIN PGP SIGNED MESSAGE----- <br> Hash: SHA1 <br> What undefined STT_GNU_IFUNC symbol? The whole point of this symbol <br> type is that the caller doesn't have to care at all about the <br> optimizations. A library author can introduce them in DSOs and programs <br> at the next start can take advantage of this. <br> Linking against an STT_GNU_IFUNC symbol in a DSO should create a normal

Ulrich Drepper drep...@redhat.com 2009-06-05T05:26:17Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/a1dbedfe489d1d0d?show_docid=a1dbedfe489d1d0d

-----BEGIN PGP SIGNED MESSAGE----- <br> Hash: SHA1 <br> What undefined STT_GNU_IFUNC symbol? The whole point of this symbol <br> type is that the caller doesn't have to care at all about the <br> optimizations. A library author can introduce them in DSOs and programs <br> at the next start can take advantage of this. <br> Linking against an STT_GNU_IFUNC symbol in a DSO should create a normal

H.J. Lu hjl.to...@gmail.com 2009-06-04T20:46:43Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/97941e94aaa95d6d?show_docid=97941e94aaa95d6d

There is an issue with STT_GNU_IFUNC symbols. <br> For branch, it is easy to understand. But it is unclear what the value <br> of STT_GNU_IFUNC symbol should be. In a non-shared object, the value <br> of the undefined STT_GNU_IFUNC symbol is the address of the PLT entry <br> since it will only be resolved at run-time. In a shared object,

H.J. Lu hjl.to...@gmail.com 2009-06-04T20:34:04Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/59d48789a5e9ff6c?show_docid=59d48789a5e9ff6c

There is an issue with STT_GNU_IFUNC symbols. <br> For branch, it is easy to understand. But it is unclear what the value <br> of STT_GNU_IFUNC symbol should be. In a non-shared object, the value <br> of the undefined STT_GNU_IFUNC symbol is the address of the PLT entry <br> since it will only be resolved at run-time. In a shared object,

Ian Lance Taylor ianlancetay...@gmail.com 2009-06-03T22:02:42Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/bd70bccba08a83cc?show_docid=bd70bccba08a83cc

How is the &quot;corresponding R_*_RELATIVE relocation&quot; determined? <br> Ian

H.J. Lu hjl.to...@gmail.com 2009-06-03T22:02:44Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/1b8b133aca611ace?show_docid=1b8b133aca611ace

On Mon, Jun 1, 2009 at 1:07 PM, Ian Lance Taylor <br> I added &quot;as specified in the processor-specific ABI&quot;: <br> R_*_IRELATIVE <br> This relocation is similar to R_*_RELATIVE except that the <br> value used in this relocation is the program address returned <br> by the function, which takes no arguments, at the address of

H.J. Lu hjl.to...@gmail.com 2009-06-01T20:14:43Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/7519868fbae54003?show_docid=7519868fbae54003

How about this: <br> The calling convention of the STT_GNU_IFUNC function, which takes no <br> arguments and returns a function pointer, should follow the <br> processor-specific ABI. All rules for caller-saved and callee-saved <br> registers apply.

H.J. Lu hjl.to...@gmail.com 2009-06-01T20:12:44Z http://groups.google.com/group/generic-abi/browse_thread/thread/75ed47beafc5027f/f140de881c44deba?show_docid=f140de881c44deba

On Mon, Jun 1, 2009 at 1:07 PM, Ian Lance Taylor <br> I added &quot;as specified in the processor-specific ABI&quot;: <br> R_*_IRELATIVE <br> This relocation is similar to R_*_RELATIVE except that the <br> value used in this relocation is the program address returned <br> by the function, which takes no arguments, at the address of