Re: [webkit-dev] Data Memory Barrier ARMv6 question

> On Jul 5, 2017, at 19:19, Caio Lima wrote: > > 2017-07-05 12:41 GMT-03:00 JF Bastien : >> On Linux you can do the following: >> >> ((void(*)())0x0fa0)(); >> >> >> That address contains a helper which does the “right” barrier, including if >>

Re: [webkit-dev] Data Memory Barrier ARMv6 question

2017-07-05 12:41 GMT-03:00 JF Bastien : > On Linux you can do the following: > > ((void(*)())0x0fa0)(); > > > That address contains a helper which does the “right” barrier, including if > you’re not on an SMP system it’ll do nothing. > > Details: >

Re: [webkit-dev] Data Memory Barrier ARMv6 question

Alternatively, the JIT can read the content of these addresses and write them back as-is  > On Jul 5, 2017, at 09:01, Filip Pizlo wrote: > > We should not use those helpers, especially in the JIT. It does not make > sense for the JIT to emit calls to system functions when

Re: [webkit-dev] Data Memory Barrier ARMv6 question

We should not use those helpers, especially in the JIT. It does not make sense for the JIT to emit calls to system functions when the user is expecting it to emit an instruction. If we cannot perfectly select the right barrier on a particular CPU, we should disable concurrency on that CPU.

Re: [webkit-dev] Data Memory Barrier ARMv6 question

On Linux you can do the following: ((void(*)())0x0fa0)(); That address contains a helper which does the “right” barrier, including if you’re not on an SMP system it’ll do nothing. Details: https://www.kernel.org/doc/Documentation/arm/kernel_user_helpers.txt

[webkit-dev] Data Memory Barrier ARMv6 question

Hi all. I'm working in this patch (https://bugs.webkit.org/show_bug.cgi?id=172767) and Mark Lam raised some questions about the data memory barrier (DMB for short) in ARMv6 using "mcr 15 ...". The point is that we are having divergences in ARM official reference manual about the semantics of this