On 10/06/10 14:32, Eric Blake wrote:
> On 06/10/2010 07:22 AM, Callahan, Patrick M. wrote:
>> if ((errno == ENOSYS || errno == EOPNOTSUPP)
>> ...
>> You could get in a debugger and determine where
>> to add "|| errno == 252" to solve what appears to be
>> an HP-UX-and/or-cvfs-specific problem.
>>
>> However, such a change is not appropriate for others,
>> and I doubt it will be worthwhile to attempt an
>> upstream workaround.
>>
>> [>> Agree. Appreciate the response. Restores my faith in humanity when
>> [>> I have to spend a lot of times with vendors whose only interest
>> [>> seems to be to separate me from as much as my budget as they can
>> [>> get.
>
> Can you confirm Pádraig's google analysis that 252 is ENOTSUP on your
> platform? If we have a symbolic name, then we can patch the code quite
> easily. We can also patch the gnulib strerror and perror to detect the
> failure to handle ENOTSUP gracefully.
Perhaps just changing to the more general ACL_NOT_WELL_SUPPORTED()
for HPUX is appropriate? I.E. could you test the following coreutils
patch Patrick?
I left the solaris and cygwin code doing the explicit errno checks,
but suggest they also change to the more general check too.
diff ../gnulib/lib/copy-acl.c gnulib/lib/copy-acl.c
--- ../gnulib/lib/copy-acl.c 2010-03-03 11:23:22.000000000 +0000
+++ gnulib/lib/copy-acl.c 2010-06-10 14:10:19.000000000 +0000
@@ -420,7 +420,7 @@
if (count < 0)
{
- if (errno == ENOSYS || errno == EOPNOTSUPP)
+ if (ACL_NOT_WELL_SUPPORTED (errno))
{
count = 0;
break;
@@ -455,7 +455,7 @@
{
int saved_errno = errno;
- if (errno == ENOSYS || errno == EOPNOTSUPP)
+ if (ACL_NOT_WELL_SUPPORTED (errno))
{
struct stat source_statbuf;
