|
From: Pim N. <pn...@ai...> - 2006-11-27 13:09:26
|
Hi, For the linux powerpc the syscall settimeofday (79) is commented out. Does someone has any idea why? I guess it just was never tested. I decommented it and it seems to work fine, however I didn't sincerely test it. If it is still commented out just because it wasn't tested I'm willing to test it. Is there a testcase available? If there's another reason, please point it out. I need this to work so I'm willing to spend some time on it. Pim -- This message has been scanned for viruses and is believed to be clean |
|
From: Nicholas N. <nj...@cs...> - 2006-11-28 00:22:36
|
On Mon, 27 Nov 2006, Pim Nijdam wrote: > For the linux powerpc the syscall settimeofday (79) is commented out. > Does someone has any idea why? I guess it just was never tested. Probably. > I decommented it and it seems to work fine, however I didn't sincerely > test it. If it is still commented out just because it wasn't tested I'm > willing to test it. Is there a testcase available? > > If there's another reason, please point it out. I need this to work so > I'm willing to spend some time on it. No test case, please test it thoroughly and let us know if it works well in which case we can commit the change! Nick |
|
From: Pim N. <pn...@ai...> - 2006-12-18 14:14:23
|
Nicholas Nethercote wrote: > On Mon, 27 Nov 2006, Pim Nijdam wrote: > >> For the linux powerpc the syscall settimeofday (79) is commented out. >> Does someone has any idea why? I guess it just was never tested. > > > Probably. > >> I decommented it and it seems to work fine, however I didn't sincerely >> test it. If it is still commented out just because it wasn't tested I'm >> willing to test it. Is there a testcase available? >> >> If there's another reason, please point it out. I need this to work so >> I'm willing to spend some time on it. > > > No test case, please test it thoroughly and let us know if it works well > in which case we can commit the change! I tested it and it works well. However, PRE(sys_settimeofday) only checks if the second argument is NULL, but according to the Linux manual page of settimeofday both arguments can be NULL. Adding a check and Valgrind reported as expected. Unfortunetly I don't know if this specification of settimeofday is generic or Linux specific. Pim Nijdam -- This message has been scanned for viruses and is believed to be clean |