On Thu, 20 Jul 2023 19:32:33 -0700 Andrew Hewus Fresh <and...@afresh1.com> wrote:
> On Mon, Jul 17, 2023 at 11:54:18PM -0400, George Koehler wrote: > > How to pread($fd, $buf, 4, 16)? > > > > 1. syscall(169, $fd, $buf, 4, 0, 16) > > 2. syscall(169, $fd, $buf, 4, 0, 0, 16) > > 3. syscall(169, $fd, $buf, 4, 16, 0) > > 4. syscall(169, $fd, $buf, 4, 0, 16, 0) > > 5. syscall(169, $fd, $buf, 4, 16) > > Which is the way it works without the this middleman? This is on macppc. Without your diff is syscall(2). With your diff is va_args(args, off_t). pread(2) without your diff needs line 2, $ perl 3</etc/passwd \ > -E 'syscall(169, 3, $_ = "c" x 7, 7, 0, 0, 11); say $_' Charlie pread(2) with your diff needs line 1, $ perl 3</etc/passwd \ > -E 'syscall(169, 3, $_ = "c" x 7, 7, 0, 11); say $_' Charlie I don't know how it worked before we removed __syscall(2). > > $ cd /usr/src/gnu/usr.bin/perl/obj > > $ MALLOC_OPTIONS=S perl t/op/syscall_emulator.t > > /usr/include/sys/syscall.h -> /usr/include/sys/syscall.ph > > 1..13 > > ok 1 - Opened test.txt for write/create > > ok 2 - Wrote out to test.txt > > ok 3 - closed test.txt > > perl(15511) in malloc(): write after free 0xc36d1efcddc0 > > Abort trap (core dumped) > > Hmm, I guess I need to get a powerpc64 machine. You didn't need a powerpc64. The crash was random; if you ran the test multiple times, then it would crash. Your updated diff from Thu 20 Jul doesn't crash.
