From: Paul Burton <paul.burton@mips.com>
To: 徐成华 <xuchenghua@loongson.cn>
Cc: "Yunqiang Su" <ysu@wavecomp.com>,
"Paul Burton" <pburton@wavecomp.com>,
"linux-mips@vger.kernel.org" <linux-mips@vger.kernel.org>,
"chenhc@lemote.com" <chenhc@lemote.com>,
"zhangfx@lemote.com" <zhangfx@lemote.com>,
"wuzhangjin@gmail.com" <wuzhangjin@gmail.com>,
"linux-mips@linux-mips.org" <linux-mips@linux-mips.org>,
黄沛 <huangpei@loongson.cn>
Subject: Re: [PATCH 1/2] MIPS: Loongson, add sync before target of branch between llsc
Date: Fri, 11 Jan 2019 19:00:50 +0000 [thread overview]
Message-ID: <20190111190049.pba3243a5ln5fw56@pburton-laptop> (raw)
In-Reply-To: <37e1dca1.5987.1683cede2ff.Coremail.xuchenghua@loongson.cn>
Hello,
On Fri, Jan 11, 2019 at 08:40:49PM +0800, 徐成华 wrote:
> For Loongson 3A1000 and 3A3000, when a memory access instruction
> (load, store, or prefetch)'s executing occurs between the execution of
> LL and SC, the success or failure of SC is not predictable. Although
> programmer would not insert memory access instructions between LL and
> SC, the memory instructions before LL in program-order, may
> dynamically executed between the execution of LL/SC, so a memory
> fence(SYNC) is needed before LL/LLD to avoid this situation.
>
> Since 3A3000, we improved our hardware design to handle this case.
> But we later deduce a rarely circumstance that some speculatively
> executed memory instructions due to branch misprediction between LL/SC
> still fall into the above case, so a memory fence(SYNC) at
> branch-target(if its target is not between LL/SC) is needed for 3A1000
> and 3A3000.
Thank you - that description is really helpful.
I have a few follow-up questions if you don't mind:
1) Is it correct to say that the only consequence of the bug is that an
SC might fail when it ought to have succeeded?
2) Does that mean placing a sync before the LL is purely a performance
optimization? ie. if we don't have the sync & the SC fails then
we'll retry the LL/SC anyway, and this time not have the reordered
instruction from before the LL to cause a problem.
3) In the speculative execution case would it also work to place a sync
before the branch instruction, instead of at the branch target? In
some cases this might be nicer since the workaround would be
contained within the LL/SC loop, but I guess it could potentially
add more overhead if the branch is conditional & not taken.
4) When we talk about branches here, is it really just branch
instructions that are affected or will the CPU speculate past jump
instructions too?
I just want to be sure that we work around this properly, and document
it in the kernel so that it's clear to developers why the workaround
exists & how to avoid introducing bugs for these CPUs in future.
> Our processor is continually evolving and we aim to to remove all
> these workaround-SYNCs around LL/SC for new-come processor.
I'm very glad to hear that :)
I hope one day I can get my hands on a nice Loongson laptop to test
with.
Thanks,
Paul
next prev parent reply other threads:[~2019-01-11 19:00 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-11 12:40 [PATCH 1/2] MIPS: Loongson, add sync before target of branch between llsc 徐成华
2019-01-11 12:45 ` huangpei
2019-01-11 19:00 ` Paul Burton [this message]
2019-01-12 8:02 ` 徐成华
2019-01-12 8:19 ` huangpei
2019-01-12 3:25 ` huangpei
2019-01-12 3:41 ` Yunqiang Su
2019-01-12 6:21 ` huangpei
-- strict thread matches above, loose matches on Subject: below --
2019-01-05 15:00 YunQiang Su
2019-01-09 22:08 ` Paul Burton
2019-01-10 1:59 ` Yunqiang Su
2019-01-10 17:35 ` Paul Burton
2019-01-10 18:42 ` YunQiang Su
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190111190049.pba3243a5ln5fw56@pburton-laptop \
--to=paul.burton@mips.com \
--cc=chenhc@lemote.com \
--cc=huangpei@loongson.cn \
--cc=linux-mips@linux-mips.org \
--cc=linux-mips@vger.kernel.org \
--cc=pburton@wavecomp.com \
--cc=wuzhangjin@gmail.com \
--cc=xuchenghua@loongson.cn \
--cc=ysu@wavecomp.com \
--cc=zhangfx@lemote.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).