Re: arc_usr_cmpxchg and preemption

> On Thu, Mar 15, 2018 at 12:03 PM, Alexey Brodkin > wrote: > Here's a brief analysis: > ARM: Looks like they got rid of that stuff in v4.4, see > commit db695c0509d6 ("ARM: remove user cmpxchg syscall"). > > M68K: That's even uglier implementation which is

Re: arc_usr_cmpxchg and preemption

> On Thu, Mar 15, 2018 at 12:03 PM, Alexey Brodkin > wrote: > Here's a brief analysis: > ARM: Looks like they got rid of that stuff in v4.4, see > commit db695c0509d6 ("ARM: remove user cmpxchg syscall"). > > M68K: That's even uglier implementation which is really asking for > a

Re: arc_usr_cmpxchg and preemption

On Fri, Mar 16, 2018 at 10:54:52AM -0700, Vineet Gupta wrote: > I'd say just bite the bullet, write the patch and we can refine it there ! Just be glad its not futex.c proper ;-) I'll try and have a look later..

Re: arc_usr_cmpxchg and preemption

On Fri, Mar 16, 2018 at 10:54:52AM -0700, Vineet Gupta wrote: > I'd say just bite the bullet, write the patch and we can refine it there ! Just be glad its not futex.c proper ;-) I'll try and have a look later..

Re: arc_usr_cmpxchg and preemption

On 03/16/2018 10:33 AM, Alexey Brodkin wrote: Hi Peter, Vineet, On Wed, 2018-03-14 at 18:53 +0100, Peter Zijlstra wrote: On Wed, Mar 14, 2018 at 09:58:19AM -0700, Vineet Gupta wrote: Well it is broken wrt the semantics the syscall is supposed to provide. Preemption disabling is what prevents

Re: arc_usr_cmpxchg and preemption

On 03/16/2018 10:33 AM, Alexey Brodkin wrote: Hi Peter, Vineet, On Wed, 2018-03-14 at 18:53 +0100, Peter Zijlstra wrote: On Wed, Mar 14, 2018 at 09:58:19AM -0700, Vineet Gupta wrote: Well it is broken wrt the semantics the syscall is supposed to provide. Preemption disabling is what prevents

Re: arc_usr_cmpxchg and preemption

Hi Peter, Vineet, On Wed, 2018-03-14 at 18:53 +0100, Peter Zijlstra wrote: > On Wed, Mar 14, 2018 at 09:58:19AM -0700, Vineet Gupta wrote: > > > Well it is broken wrt the semantics the syscall is supposed to provide. > > Preemption disabling is what prevents a concurrent thread from coming in

Re: arc_usr_cmpxchg and preemption

Hi Peter, Vineet, On Wed, 2018-03-14 at 18:53 +0100, Peter Zijlstra wrote: > On Wed, Mar 14, 2018 at 09:58:19AM -0700, Vineet Gupta wrote: > > > Well it is broken wrt the semantics the syscall is supposed to provide. > > Preemption disabling is what prevents a concurrent thread from coming in

Re: arc_usr_cmpxchg and preemption

On Thu, Mar 15, 2018 at 07:03:32PM +, Alexey Brodkin wrote: > > I think there's a bunch of architectures that are in the same boat. > > m68k, arm, mips was mentioned. Sure, the moment an arch has hardware > > support you don't need the syscall anymore. > > Here's a brief analysis: > ARM:

Re: arc_usr_cmpxchg and preemption

On Thu, Mar 15, 2018 at 07:03:32PM +, Alexey Brodkin wrote: > > I think there's a bunch of architectures that are in the same boat. > > m68k, arm, mips was mentioned. Sure, the moment an arch has hardware > > support you don't need the syscall anymore. > > Here's a brief analysis: > ARM:

Re: arc_usr_cmpxchg and preemption

Hi Peter, On Thu, 2018-03-15 at 12:28 +0100, Peter Zijlstra wrote: > On Thu, Mar 15, 2018 at 09:12:09AM +, Alexey Brodkin wrote: > > On Thu, 2018-03-15 at 09:18 +0100, Peter Zijlstra wrote: > > > Also, it might make sense to stuff this implementation in some lib/ file > > > somewhere and make

Re: arc_usr_cmpxchg and preemption

Hi Peter, On Thu, 2018-03-15 at 12:28 +0100, Peter Zijlstra wrote: > On Thu, Mar 15, 2018 at 09:12:09AM +, Alexey Brodkin wrote: > > On Thu, 2018-03-15 at 09:18 +0100, Peter Zijlstra wrote: > > > Also, it might make sense to stuff this implementation in some lib/ file > > > somewhere and make

Re: arc_usr_cmpxchg and preemption

On Thu, Mar 15, 2018 at 09:12:09AM +, Alexey Brodkin wrote: > On Thu, 2018-03-15 at 09:18 +0100, Peter Zijlstra wrote: > > Also, it might make sense to stuff this implementation in some lib/ file > > somewhere and make all platforms that need it use the same code, afaict > > there really

Re: arc_usr_cmpxchg and preemption

On Thu, Mar 15, 2018 at 09:12:09AM +, Alexey Brodkin wrote: > On Thu, 2018-03-15 at 09:18 +0100, Peter Zijlstra wrote: > > Also, it might make sense to stuff this implementation in some lib/ file > > somewhere and make all platforms that need it use the same code, afaict > > there really

Re: arc_usr_cmpxchg and preemption

Hi Peter, On Thu, 2018-03-15 at 09:18 +0100, Peter Zijlstra wrote: > On Wed, Mar 14, 2018 at 08:38:53PM +, Alexey Brodkin wrote: > > > int sys_cmpxchg(u32 __user *user_ptr, u32 old, u32 new) > > > { > > > u32 val; > > > int ret; > > > > > > again: > > > ret = 0; > > > > > >

Re: arc_usr_cmpxchg and preemption

Hi Peter, On Thu, 2018-03-15 at 09:18 +0100, Peter Zijlstra wrote: > On Wed, Mar 14, 2018 at 08:38:53PM +, Alexey Brodkin wrote: > > > int sys_cmpxchg(u32 __user *user_ptr, u32 old, u32 new) > > > { > > > u32 val; > > > int ret; > > > > > > again: > > > ret = 0; > > > > > >

Re: arc_usr_cmpxchg and preemption

On Wed, Mar 14, 2018 at 08:38:53PM +, Alexey Brodkin wrote: > > int sys_cmpxchg(u32 __user *user_ptr, u32 old, u32 new) > > { > > u32 val; > > int ret; > > > > again: > > ret = 0; > > > > preempt_disable(); > > val = get_user(user_ptr); > > if (val == old) > >

Re: arc_usr_cmpxchg and preemption

On Wed, Mar 14, 2018 at 08:38:53PM +, Alexey Brodkin wrote: > > int sys_cmpxchg(u32 __user *user_ptr, u32 old, u32 new) > > { > > u32 val; > > int ret; > > > > again: > > ret = 0; > > > > preempt_disable(); > > val = get_user(user_ptr); > > if (val == old) > >

Re: arc_usr_cmpxchg and preemption

On 03/14/2018 01:38 PM, Alexey Brodkin wrote: @Vineet, are you OK with proposed implementation? I couldn't agree any more ! -Vineet

Re: arc_usr_cmpxchg and preemption

On 03/14/2018 01:38 PM, Alexey Brodkin wrote: @Vineet, are you OK with proposed implementation? I couldn't agree any more ! -Vineet

Re: arc_usr_cmpxchg and preemption

Hi Peter, Vineet, On Wed, 2018-03-14 at 18:53 +0100, Peter Zijlstra wrote: > On Wed, Mar 14, 2018 at 09:58:19AM -0700, Vineet Gupta wrote: > > > Well it is broken wrt the semantics the syscall is supposed to provide. > > Preemption disabling is what prevents a concurrent thread from coming in

Re: arc_usr_cmpxchg and preemption

Hi Peter, Vineet, On Wed, 2018-03-14 at 18:53 +0100, Peter Zijlstra wrote: > On Wed, Mar 14, 2018 at 09:58:19AM -0700, Vineet Gupta wrote: > > > Well it is broken wrt the semantics the syscall is supposed to provide. > > Preemption disabling is what prevents a concurrent thread from coming in

Re: arc_usr_cmpxchg and preemption

On Wed, Mar 14, 2018 at 06:53:52PM +0100, Peter Zijlstra wrote: > On Wed, Mar 14, 2018 at 09:58:19AM -0700, Vineet Gupta wrote: > > > Well it is broken wrt the semantics the syscall is supposed to provide. > > Preemption disabling is what prevents a concurrent thread from coming in and > >

Re: arc_usr_cmpxchg and preemption

On Wed, Mar 14, 2018 at 06:53:52PM +0100, Peter Zijlstra wrote: > On Wed, Mar 14, 2018 at 09:58:19AM -0700, Vineet Gupta wrote: > > > Well it is broken wrt the semantics the syscall is supposed to provide. > > Preemption disabling is what prevents a concurrent thread from coming in and > >

Re: arc_usr_cmpxchg and preemption

On Wed, Mar 14, 2018 at 09:58:19AM -0700, Vineet Gupta wrote: > Well it is broken wrt the semantics the syscall is supposed to provide. > Preemption disabling is what prevents a concurrent thread from coming in and > modifying the same location (Imagine a variable which is being cmpxchg >

Re: arc_usr_cmpxchg and preemption

On Wed, Mar 14, 2018 at 09:58:19AM -0700, Vineet Gupta wrote: > Well it is broken wrt the semantics the syscall is supposed to provide. > Preemption disabling is what prevents a concurrent thread from coming in and > modifying the same location (Imagine a variable which is being cmpxchg >

Re: arc_usr_cmpxchg and preemption

+CC linux-arch, Peter for any preemption insights ! On 03/14/2018 09:36 AM, Alexey Brodkin wrote: Hi Vineet, While debugging a segfault of user-space app on system without atomic ops (I mean LLOCK/SCOND) I understood the root-cause is in implementation of kernel's __NR_arc_usr_cmpxchg syscall

Re: arc_usr_cmpxchg and preemption

+CC linux-arch, Peter for any preemption insights ! On 03/14/2018 09:36 AM, Alexey Brodkin wrote: Hi Vineet, While debugging a segfault of user-space app on system without atomic ops (I mean LLOCK/SCOND) I understood the root-cause is in implementation of kernel's __NR_arc_usr_cmpxchg syscall