Re: [PATCH 1/2] arch: mark syscall number 435 reserved for clone3
On Fri, Jul 19, 2019 at 09:13:16PM +1000, Michael Ellerman wrote:
> Christian Brauner writes:
> > On Fri, Jul 19, 2019 at 08:18:02PM +1000, Michael Ellerman wrote:
> >> Christian Brauner writes:
> >> > On Mon, Jul 15, 2019 at 03:56:04PM +0200, Christian Borntraeger wrote:
> >> >> I think Vasily already has a clone3 patch for s390x with 435.
> >> >
> >> > A quick follow-up on this. Helge and Michael have asked whether there
> >> > are any tests for clone3. Yes, there will be and I try to have them
> >> > ready by the end of the this or next week for review. In the meantime I
> >> > hope the following minimalistic test program that just verifies very
> >> > very basic functionality (It's not pretty.) will help you test:
> >>
> >> Hi Christian,
> >>
> >> Thanks for the test.
> >>
> >> This actually oopses on powerpc, it hits the BUG_ON in CHECK_FULL_REGS
> >> in process.c around line 1633:
> >>
> >>} else {
> >>/* user thread */
> >>struct pt_regs *regs = current_pt_regs();
> >>CHECK_FULL_REGS(regs);
> >>*childregs = *regs;
> >>if (usp)
> >>
> >>
> >> So I'll have to dig into how we fix that before we wire up clone3.
> >>
> >> Turns out testing is good! :)
> >
> > Indeed. I have a test-suite for clone3 in mind and I hope to have it
> > ready by the end of next week. It's just always the finding the time
> > part that is annoying. :)
>
> I know the feeling!
>
> > Thanks for digging into this, Michael!
>
> No worries, happy to help where I can.
>
> In the intervening five minutes I remembered how we handle this, we just
> need a little wrapper to save the non-volatile regs:
>
> _GLOBAL(ppc_clone3)
> bl save_nvgprs
> bl sys_clone3
> b .Lsyscall_exit
Sounds good.
>
>
> A while back I meant to make it generate those automatically based on a
> flag in the syscall.tbl but of course haven't got around to it :)
>
> So with the above it seems all good:
>
> $ ./clone3 ; echo $?
> Parent process received child's pid 4204 as return value
> Parent process received child's pidfd 3
> Parent process received child's pid 4204 as return argument
> Child process with pid 4204
> 0
>
> I'll send a patch to wire it up on Monday.
Excellent! Thank you!
Christian
Re: [PATCH 1/2] arch: mark syscall number 435 reserved for clone3
Christian Brauner writes:
> On Fri, Jul 19, 2019 at 08:18:02PM +1000, Michael Ellerman wrote:
>> Christian Brauner writes:
>> > On Mon, Jul 15, 2019 at 03:56:04PM +0200, Christian Borntraeger wrote:
>> >> I think Vasily already has a clone3 patch for s390x with 435.
>> >
>> > A quick follow-up on this. Helge and Michael have asked whether there
>> > are any tests for clone3. Yes, there will be and I try to have them
>> > ready by the end of the this or next week for review. In the meantime I
>> > hope the following minimalistic test program that just verifies very
>> > very basic functionality (It's not pretty.) will help you test:
>>
>> Hi Christian,
>>
>> Thanks for the test.
>>
>> This actually oopses on powerpc, it hits the BUG_ON in CHECK_FULL_REGS
>> in process.c around line 1633:
>>
>> } else {
>> /* user thread */
>> struct pt_regs *regs = current_pt_regs();
>> CHECK_FULL_REGS(regs);
>> *childregs = *regs;
>> if (usp)
>>
>>
>> So I'll have to dig into how we fix that before we wire up clone3.
>>
>> Turns out testing is good! :)
>
> Indeed. I have a test-suite for clone3 in mind and I hope to have it
> ready by the end of next week. It's just always the finding the time
> part that is annoying. :)
I know the feeling!
> Thanks for digging into this, Michael!
No worries, happy to help where I can.
In the intervening five minutes I remembered how we handle this, we just
need a little wrapper to save the non-volatile regs:
_GLOBAL(ppc_clone3)
bl save_nvgprs
bl sys_clone3
b .Lsyscall_exit
A while back I meant to make it generate those automatically based on a
flag in the syscall.tbl but of course haven't got around to it :)
So with the above it seems all good:
$ ./clone3 ; echo $?
Parent process received child's pid 4204 as return value
Parent process received child's pidfd 3
Parent process received child's pid 4204 as return argument
Child process with pid 4204
0
I'll send a patch to wire it up on Monday.
cheers
Re: [PATCH 1/2] arch: mark syscall number 435 reserved for clone3
On Fri, Jul 19, 2019 at 08:18:02PM +1000, Michael Ellerman wrote:
> Christian Brauner writes:
> > On Mon, Jul 15, 2019 at 03:56:04PM +0200, Christian Borntraeger wrote:
> >> I think Vasily already has a clone3 patch for s390x with 435.
> >
> > A quick follow-up on this. Helge and Michael have asked whether there
> > are any tests for clone3. Yes, there will be and I try to have them
> > ready by the end of the this or next week for review. In the meantime I
> > hope the following minimalistic test program that just verifies very
> > very basic functionality (It's not pretty.) will help you test:
>
> Hi Christian,
>
> Thanks for the test.
>
> This actually oopses on powerpc, it hits the BUG_ON in CHECK_FULL_REGS
> in process.c around line 1633:
>
> } else {
> /* user thread */
> struct pt_regs *regs = current_pt_regs();
> CHECK_FULL_REGS(regs);
> *childregs = *regs;
> if (usp)
>
>
> So I'll have to dig into how we fix that before we wire up clone3.
>
> Turns out testing is good! :)
Indeed. I have a test-suite for clone3 in mind and I hope to have it
ready by the end of next week. It's just always the finding the time
part that is annoying. :)
Thanks for digging into this, Michael!
Christian
Re: [PATCH 1/2] arch: mark syscall number 435 reserved for clone3
Christian Brauner writes:
> On Mon, Jul 15, 2019 at 03:56:04PM +0200, Christian Borntraeger wrote:
>> I think Vasily already has a clone3 patch for s390x with 435.
>
> A quick follow-up on this. Helge and Michael have asked whether there
> are any tests for clone3. Yes, there will be and I try to have them
> ready by the end of the this or next week for review. In the meantime I
> hope the following minimalistic test program that just verifies very
> very basic functionality (It's not pretty.) will help you test:
Hi Christian,
Thanks for the test.
This actually oopses on powerpc, it hits the BUG_ON in CHECK_FULL_REGS
in process.c around line 1633:
} else {
/* user thread */
struct pt_regs *regs = current_pt_regs();
CHECK_FULL_REGS(regs);
*childregs = *regs;
if (usp)
So I'll have to dig into how we fix that before we wire up clone3.
Turns out testing is good! :)
cheers
Re: [PATCH 1/2] arch: mark syscall number 435 reserved for clone3
On 16.07.19 20:55, Christian Brauner wrote: On Tue, Jul 16, 2019 at 08:53:10PM +0200, Sven Schnelle wrote: Hi, [Adding Helge to CC list] On Tue, Jul 16, 2019 at 03:06:33PM +0200, Christian Brauner wrote: On Mon, Jul 15, 2019 at 03:56:04PM +0200, Christian Borntraeger wrote: I think Vasily already has a clone3 patch for s390x with 435. A quick follow-up on this. Helge and Michael have asked whether there are any tests for clone3. Yes, there will be and I try to have them ready by the end of the this or next week for review. In the meantime I hope the following minimalistic test program that just verifies very very basic functionality (It's not pretty.) will help you test: [..] On PA-RISC this seems to work fine with Helge's patch to wire up the clone3 syscall. [...] In any case, I saw Helge's patch and I think I might've missed to add an Acked-by but feel free to add it. Thanks! I've added the patch to the parisc-linux for-next tree. Helge
Re: [PATCH 1/2] arch: mark syscall number 435 reserved for clone3
Hi, [Adding Helge to CC list] On Tue, Jul 16, 2019 at 03:06:33PM +0200, Christian Brauner wrote: > On Mon, Jul 15, 2019 at 03:56:04PM +0200, Christian Borntraeger wrote: > > I think Vasily already has a clone3 patch for s390x with 435. > > A quick follow-up on this. Helge and Michael have asked whether there > are any tests for clone3. Yes, there will be and I try to have them > ready by the end of the this or next week for review. In the meantime I > hope the following minimalistic test program that just verifies very > very basic functionality (It's not pretty.) will help you test: > [..] On PA-RISC this seems to work fine with Helge's patch to wire up the clone3 syscall. root@c3750:/# clonetest Parent process received child's pid 84 as return value Parent process received child's pidfd 3 Parent process received child's pid 84 as return argument Child process with pid 84 root@c3750:/# echo $? 0 Regards Sven
Re: [PATCH 1/2] arch: mark syscall number 435 reserved for clone3
On Tue, Jul 16, 2019 at 08:53:10PM +0200, Sven Schnelle wrote: > Hi, > > [Adding Helge to CC list] > > On Tue, Jul 16, 2019 at 03:06:33PM +0200, Christian Brauner wrote: > > On Mon, Jul 15, 2019 at 03:56:04PM +0200, Christian Borntraeger wrote: > > > I think Vasily already has a clone3 patch for s390x with 435. > > > > A quick follow-up on this. Helge and Michael have asked whether there > > are any tests for clone3. Yes, there will be and I try to have them > > ready by the end of the this or next week for review. In the meantime I > > hope the following minimalistic test program that just verifies very > > very basic functionality (It's not pretty.) will help you test: > > [..] > > On PA-RISC this seems to work fine with Helge's patch to wire up the > clone3 syscall. I think I already responded to Helge before and yes, I think that parisc doesn't do anything special for fork, vfork, clone, and by extension also probably doesn't need to for clone3. It should only be a problem for arches that require mucking explicitly with arguments of clone-like syscalls. In any case, I saw Helge's patch and I think I might've missed to add an Acked-by but feel free to add it. Thanks for testing it and sorry that I couldn't test! Christian
Re: [PATCH 1/2] arch: mark syscall number 435 reserved for clone3
On Mon, Jul 15, 2019 at 03:56:04PM +0200, Christian Borntraeger wrote:
> I think Vasily already has a clone3 patch for s390x with 435.
A quick follow-up on this. Helge and Michael have asked whether there
are any tests for clone3. Yes, there will be and I try to have them
ready by the end of the this or next week for review. In the meantime I
hope the following minimalistic test program that just verifies very
very basic functionality (It's not pretty.) will help you test:
#define _GNU_SOURCE
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#ifndef CLONE_PIDFD
#define CLONE_PIDFD 0x1000
#endif
#ifndef __NR_clone3
#define __NR_clone3 -1
#endif
static pid_t sys_clone3(struct clone_args *args)
{
return syscall(__NR_clone3, args, sizeof(struct clone_args));
}
static int wait_for_pid(pid_t pid)
{
int status, ret;
again:
ret = waitpid(pid, , 0);
if (ret == -1) {
if (errno == EINTR)
goto again;
return -1;
}
if (ret != pid)
goto again;
if (!WIFEXITED(status) || WEXITSTATUS(status) != 0)
return -1;
return 0;
}
#define ptr_to_u64(ptr) ((__u64)((uintptr_t)(ptr)))
int main(int argc, char *argv[])
{
int pidfd = -1;
pid_t parent_tid = -1, pid = -1;
struct clone_args args = {0};
args.parent_tid = ptr_to_u64(_tid); /* CLONE_PARENT_SETTID */
args.pidfd = ptr_to_u64(); /* CLONE_PIDFD */
args.flags = CLONE_PIDFD | CLONE_PARENT_SETTID;
args.exit_signal = SIGCHLD;
pid = sys_clone3();
if (pid < 0) {
fprintf(stderr, "%s - Failed to create new process\n",
strerror(errno));
exit(EXIT_FAILURE);
}
if (pid == 0) {
printf("Child process with pid %d\n", getpid());
exit(EXIT_SUCCESS);
}
printf("Parent process received child's pid %d as return value\n", pid);
printf("Parent process received child's pidfd %d\n", *(int
*)args.pidfd);
printf("Parent process received child's pid %d as return argument\n",
*(pid_t *)args.parent_tid);
if (wait_for_pid(pid))
exit(EXIT_FAILURE);
if (pid != *(pid_t *)args.parent_tid)
exit(EXIT_FAILURE);
close(pidfd);
return 0;
}
Re: [PATCH 1/2] arch: mark syscall number 435 reserved for clone3
On Mon, Jul 15, 2019 at 03:56:04PM +0200, Christian Borntraeger wrote: > I think Vasily already has a clone3 patch for s390x with 435. Excellent. I'll leave the # 435 reserved for clone3 on s390x in until this patch has landed. It shouldn't be a merge conflict and if so it should be trivial. Christian
Re: [PATCH 1/2] arch: mark syscall number 435 reserved for clone3
I think Vasily already has a clone3 patch for s390x with 435. On 14.07.19 21:22, Christian Brauner wrote: > A while ago Arnd made it possible to give new system calls the same > syscall number on all architectures (except alpha). To not break this > nice new feature let's mark 435 for clone3 as reserved on all > architectures that do not yet implement it. > Even if an architecture does not plan to implement it this ensures that > new system calls coming after clone3 will have the same number on all > architectures. > > Signed-off-by: Christian Brauner > Cc: Arnd Bergmann > Cc: linux-a...@vger.kernel.org > Cc: linux-alpha@vger.kernel.org > Cc: linux-ker...@vger.kernel.org > Cc: linux-i...@vger.kernel.org > Cc: linux-m...@lists.linux-m68k.org > Cc: linux-m...@vger.kernel.org > Cc: linux-par...@vger.kernel.org > Cc: linuxppc-...@lists.ozlabs.org > Cc: linux-s...@vger.kernel.org > Cc: linux...@vger.kernel.org > Cc: sparcli...@vger.kernel.org > --- > arch/alpha/kernel/syscalls/syscall.tbl| 1 + > arch/ia64/kernel/syscalls/syscall.tbl | 1 + > arch/m68k/kernel/syscalls/syscall.tbl | 1 + > arch/mips/kernel/syscalls/syscall_n32.tbl | 1 + > arch/mips/kernel/syscalls/syscall_n64.tbl | 1 + > arch/mips/kernel/syscalls/syscall_o32.tbl | 1 + > arch/parisc/kernel/syscalls/syscall.tbl | 1 + > arch/powerpc/kernel/syscalls/syscall.tbl | 1 + > arch/s390/kernel/syscalls/syscall.tbl | 1 + > arch/sh/kernel/syscalls/syscall.tbl | 1 + > arch/sparc/kernel/syscalls/syscall.tbl| 1 + > 11 files changed, 11 insertions(+) > > diff --git a/arch/alpha/kernel/syscalls/syscall.tbl > b/arch/alpha/kernel/syscalls/syscall.tbl > index 1db9bbcfb84e..728fe028c02c 100644 > --- a/arch/alpha/kernel/syscalls/syscall.tbl > +++ b/arch/alpha/kernel/syscalls/syscall.tbl > @@ -474,3 +474,4 @@ > 542 common fsmount sys_fsmount > 543 common fspick sys_fspick > 544 common pidfd_open sys_pidfd_open > +# 545 reserved for clone3 > diff --git a/arch/ia64/kernel/syscalls/syscall.tbl > b/arch/ia64/kernel/syscalls/syscall.tbl > index ecc44926737b..36d5faf4c86c 100644 > --- a/arch/ia64/kernel/syscalls/syscall.tbl > +++ b/arch/ia64/kernel/syscalls/syscall.tbl > @@ -355,3 +355,4 @@ > 432 common fsmount sys_fsmount > 433 common fspick sys_fspick > 434 common pidfd_open sys_pidfd_open > +# 435 reserved for clone3 > diff --git a/arch/m68k/kernel/syscalls/syscall.tbl > b/arch/m68k/kernel/syscalls/syscall.tbl > index 9a3eb2558568..a88a285a0e5f 100644 > --- a/arch/m68k/kernel/syscalls/syscall.tbl > +++ b/arch/m68k/kernel/syscalls/syscall.tbl > @@ -434,3 +434,4 @@ > 432 common fsmount sys_fsmount > 433 common fspick sys_fspick > 434 common pidfd_open sys_pidfd_open > +# 435 reserved for clone3 > diff --git a/arch/mips/kernel/syscalls/syscall_n32.tbl > b/arch/mips/kernel/syscalls/syscall_n32.tbl > index 97035e19ad03..c9c879ec9b6d 100644 > --- a/arch/mips/kernel/syscalls/syscall_n32.tbl > +++ b/arch/mips/kernel/syscalls/syscall_n32.tbl > @@ -373,3 +373,4 @@ > 432 n32 fsmount sys_fsmount > 433 n32 fspick sys_fspick > 434 n32 pidfd_open sys_pidfd_open > +# 435 reserved for clone3 > diff --git a/arch/mips/kernel/syscalls/syscall_n64.tbl > b/arch/mips/kernel/syscalls/syscall_n64.tbl > index d7292722d3b0..bbce9159caa1 100644 > --- a/arch/mips/kernel/syscalls/syscall_n64.tbl > +++ b/arch/mips/kernel/syscalls/syscall_n64.tbl > @@ -349,3 +349,4 @@ > 432 n64 fsmount sys_fsmount > 433 n64 fspick sys_fspick > 434 n64 pidfd_open sys_pidfd_open > +# 435 reserved for clone3 > diff --git a/arch/mips/kernel/syscalls/syscall_o32.tbl > b/arch/mips/kernel/syscalls/syscall_o32.tbl > index dba084c92f14..9653591428ec 100644 > --- a/arch/mips/kernel/syscalls/syscall_o32.tbl > +++ b/arch/mips/kernel/syscalls/syscall_o32.tbl > @@ -422,3 +422,4 @@ > 432 o32 fsmount sys_fsmount > 433 o32 fspick sys_fspick > 434 o32 pidfd_open sys_pidfd_open > +# 435 reserved for clone3 > diff --git a/arch/parisc/kernel/syscalls/syscall.tbl > b/arch/parisc/kernel/syscalls/syscall.tbl > index 5022b9e179c2..c7aadfef5386 100644 > --- a/arch/parisc/kernel/syscalls/syscall.tbl > +++ b/arch/parisc/kernel/syscalls/syscall.tbl > @@ -431,3 +431,4 @@ > 432 common fsmount sys_fsmount > 433 common fspick sys_fspick > 434 common pidfd_open sys_pidfd_open > +# 435 reserved for clone3 > diff --git a/arch/powerpc/kernel/syscalls/syscall.tbl > b/arch/powerpc/kernel/syscalls/syscall.tbl > index
