DPDK patches and discussions
 help / color / mirror / Atom feed
From: Chao CH Zhu <bjzhuc@cn.ibm.com>
To: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH 02/12] Add atomic operations for IBM Power	architecture
Date: Thu, 16 Oct 2014 11:14:02 +0800	[thread overview]
Message-ID: <OFCD69120A.E4F3D7D4-ON48257D73.0011507B-48257D73.0011E4AC@cn.ibm.com> (raw)
In-Reply-To: <2601191342CEEE43887BDE71AB97725821393C8F@IRSMSX105.ger.corp.intel.com>

Konstantin,

In my understanding, compiler barrier is a kind of software barrier which 
prevents the compiler from moving memory accesses across the barrier. This 
should be architecture-independent. And the "sync" instruction is a 
hardware barrier which depends on PowerPC architecture. So I think the 
compiler barrier should be the same on x86 and PowerPC. Any comments? 
Please correct me if I was wrong.

Thanks a lot! 
 
Best Regards!
------------------------------
Chao Zhu 




From:   "Ananyev, Konstantin" <konstantin.ananyev@intel.com>
To:     Chao CH Zhu/China/IBM@IBMCN, "dev@dpdk.org" <dev@dpdk.org>
Date:   2014/10/16 08:38
Subject:        RE: [dpdk-dev] [PATCH 02/12] Add atomic operations for IBM 
Power   architecture




Hi,

> -----Original Message-----
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Chao Zhu
> Sent: Friday, September 26, 2014 10:36 AM
> To: dev@dpdk.org
> Subject: [dpdk-dev] [PATCH 02/12] Add atomic operations for IBM Power 
architecture
> 
> The atomic operations implemented with assembly code in DPDK only
> support x86. This patch add architecture specific atomic operations for
> IBM Power architecture.
> 
> Signed-off-by: Chao Zhu <bjzhuc@cn.ibm.com>
> ---
>  .../common/include/powerpc/arch/rte_atomic.h       |  387 
++++++++++++++++++++
>  .../common/include/powerpc/arch/rte_atomic_arch.h  |  318 
++++++++++++++++
>  2 files changed, 705 insertions(+), 0 deletions(-)
>  create mode 100644 
lib/librte_eal/common/include/powerpc/arch/rte_atomic.h
>  create mode 100644 
lib/librte_eal/common/include/powerpc/arch/rte_atomic_arch.h
> 
...
> +
> diff --git 
a/lib/librte_eal/common/include/powerpc/arch/rte_atomic_arch.h
> b/lib/librte_eal/common/include/powerpc/arch/rte_atomic_arch.h
> new file mode 100644
> index 0000000..fe5666e
> --- /dev/null
> +
...
>+#define                rte_arch_rmb() asm volatile("sync" : : : 
"memory")
>+
> +#define               rte_arch_compiler_barrier() do {        \
> +              asm volatile ("" : : : "memory");               \
> +} while(0)

I don't know much about PPC architecture, but as I remember it uses a 
weakly-ordering memory model.
Is that correct?
If so, then you probably need rte_arch_compiler_barrier() to be "sync" 
instruction (like mb()s above) .
The reason is that IA has much stronger memory ordering model and there 
are a lot of places in the code where it implies that  ordering.
For example - ring enqueue/dequeue functions. 

Konstantin

  reply	other threads:[~2014-10-16  3:07 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-09-26  9:36 [dpdk-dev] [PATCH 00/12] Patches for DPDK to support " Chao Zhu
2014-09-26  9:36 ` [dpdk-dev] [PATCH 01/12] Add compiling definations for IBM " Chao Zhu
2014-09-26  9:36 ` [dpdk-dev] [PATCH 02/12] Add atomic operations " Chao Zhu
2014-09-29  6:16   ` Hemant
2014-09-29  6:41     ` Chao CH Zhu
2014-10-16  0:39   ` Ananyev, Konstantin
2014-10-16  3:14     ` Chao CH Zhu [this message]
2014-10-16  9:42       ` Richardson, Bruce
2014-10-16 11:04         ` Ananyev, Konstantin
     [not found]       ` <2601191342CEEE43887BDE71AB97725821393F5D@IRSMSX105.ger.corp.intel.com>
2014-10-16 10:59         ` Ananyev, Konstantin
2014-09-26  9:36 ` [dpdk-dev] [PATCH 03/12] Add byte order " Chao Zhu
2014-09-26  9:36 ` [dpdk-dev] [PATCH 04/12] Add CPU cycle " Chao Zhu
2014-09-26  9:36 ` [dpdk-dev] [PATCH 05/12] Add prefetch operation " Chao Zhu
2014-09-26  9:36 ` [dpdk-dev] [PATCH 06/12] Add spinlock " Chao Zhu
2014-09-26  9:36 ` [dpdk-dev] [PATCH 07/12] Add vector memcpy " Chao Zhu
2014-09-26  9:36 ` [dpdk-dev] [PATCH 08/12] Add CPU flag checking " Chao Zhu
2014-09-26  9:36 ` [dpdk-dev] [PATCH 09/12] Remove iopl operation " Chao Zhu
2014-10-06 22:03   ` Cyril Chemparathy
2014-10-07 14:46     ` Ananyev, Konstantin
2014-10-13  2:33       ` Chao CH Zhu
2014-09-26  9:36 ` [dpdk-dev] [PATCH 10/12] Add cache size define for IBM Power Architecture Chao Zhu
2014-09-29  6:21   ` Hemant
2014-09-26  9:36 ` [dpdk-dev] [PATCH 11/12] Add huge page sizes for IBM Power architecture Chao Zhu
2014-09-26  9:36 ` [dpdk-dev] [PATCH 12/12] Add memory support for IBM Power Architecture Chao Zhu
2014-11-13 10:24 ` [dpdk-dev] [PATCH 00/12] Patches for DPDK to support Power architecture Thomas Monjalon
2014-11-13 10:31   ` Chao Zhu

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=OFCD69120A.E4F3D7D4-ON48257D73.0011507B-48257D73.0011E4AC@cn.ibm.com \
    --to=bjzhuc@cn.ibm.com \
    --cc=dev@dpdk.org \
    --cc=konstantin.ananyev@intel.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).