* [SPDK] Posix framework on client
@ 2019-09-27 20:56 Tom Rainey
0 siblings, 0 replies; 3+ messages in thread
From: Tom Rainey @ 2019-09-27 20:56 UTC (permalink / raw)
To: spdk
[-- Attachment #1: Type: text/plain, Size: 432 bytes --]
Hi,
We are attempting to use TCP to connect to a remote target, however, we noticed that our posix socket layer is not being loaded. When spdk_sock_connect is called, there does not seem to be any items in g_net_impls.
How does g_net_impls get populated on the client side? We've looked at the hello_sock sample and there doesn't seem to be anything tricky happening before spdk_sock_connect is called.
Thanks,
-Tom
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [SPDK] Posix framework on client
@ 2019-09-27 21:04 Harris, James R
0 siblings, 0 replies; 3+ messages in thread
From: Harris, James R @ 2019-09-27 21:04 UTC (permalink / raw)
To: spdk
[-- Attachment #1: Type: text/plain, Size: 1390 bytes --]
Hi Tom,
This is most likely a link issue with your client application Makefile. net_impls are activated via C constructor functions, so you need to use --whole_archive directives to ensure the linker doesn't optimize out the constructor function (since the constructor function isn't explicitly called anywhere).
See mk/spdk.app_vars.mk. SPDK uses constructor functions for registration in a lot of places, so by default we wrap all SPDK libraries with --whole_archive directives to ensure constructors don't get optimized out during link phase.
Hope that helps - please report back how it goes!
Regards,
-Jim
On 9/27/19, 1:56 PM, "SPDK on behalf of Tom Rainey" <spdk-bounces(a)lists.01.org on behalf of tom.rainey(a)samsung.com> wrote:
Hi,
We are attempting to use TCP to connect to a remote target, however, we noticed that our posix socket layer is not being loaded. When spdk_sock_connect is called, there does not seem to be any items in g_net_impls.
How does g_net_impls get populated on the client side? We've looked at the hello_sock sample and there doesn't seem to be anything tricky happening before spdk_sock_connect is called.
Thanks,
-Tom
_______________________________________________
SPDK mailing list
SPDK(a)lists.01.org
https://lists.01.org/mailman/listinfo/spdk
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [SPDK] Posix framework on client
@ 2019-09-27 21:44 Tom Rainey
0 siblings, 0 replies; 3+ messages in thread
From: Tom Rainey @ 2019-09-27 21:44 UTC (permalink / raw)
To: spdk
[-- Attachment #1: Type: text/plain, Size: 2380 bytes --]
Hi Jim,
Thanks for the pointer. That's what we needed.
-Tom
-----Original Message-----
From: SPDK [mailto:spdk-bounces(a)lists.01.org] On Behalf Of Harris, James R
Sent: Friday, September 27, 2019 2:04 PM
To: Storage Performance Development Kit <spdk(a)lists.01.org>
Subject: Re: [SPDK] Posix framework on client
Hi Tom,
This is most likely a link issue with your client application Makefile. net_impls are activated via C constructor functions, so you need to use --whole_archive directives to ensure the linker doesn't optimize out the constructor function (since the constructor function isn't explicitly called anywhere).
See mk/spdk.app_vars.mk. SPDK uses constructor functions for registration in a lot of places, so by default we wrap all SPDK libraries with --whole_archive directives to ensure constructors don't get optimized out during link phase.
Hope that helps - please report back how it goes!
Regards,
-Jim
On 9/27/19, 1:56 PM, "SPDK on behalf of Tom Rainey" <spdk-bounces(a)lists.01.org on behalf of tom.rainey(a)samsung.com> wrote:
Hi,
We are attempting to use TCP to connect to a remote target, however, we noticed that our posix socket layer is not being loaded. When spdk_sock_connect is called, there does not seem to be any items in g_net_impls.
How does g_net_impls get populated on the client side? We've looked at the hello_sock sample and there doesn't seem to be anything tricky happening before spdk_sock_connect is called.
Thanks,
-Tom
_______________________________________________
SPDK mailing list
SPDK(a)lists.01.org
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.01.org_mailman_listinfo_spdk&d=DwIGaQ&c=JfeWlBa6VbDyTXraMENjy_b_0yKWuqQ4qY-FPhxK4x8w-TfgRBDyeV4hVQQBEgL2&r=TGM0bakVlcBrMpq-MEzqmroD2CrqAR3mlx-NLf9U4TU&m=1udY7g72tqf3g96ZtY29p0ErrT-e7MUYjloMBY5tzuM&s=YnnAuawXMkM9AxcBTumxiT-cRCz45B4XmVmtptDWZqg&e=
_______________________________________________
SPDK mailing list
SPDK(a)lists.01.org
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.01.org_mailman_listinfo_spdk&d=DwIGaQ&c=JfeWlBa6VbDyTXraMENjy_b_0yKWuqQ4qY-FPhxK4x8w-TfgRBDyeV4hVQQBEgL2&r=TGM0bakVlcBrMpq-MEzqmroD2CrqAR3mlx-NLf9U4TU&m=1udY7g72tqf3g96ZtY29p0ErrT-e7MUYjloMBY5tzuM&s=YnnAuawXMkM9AxcBTumxiT-cRCz45B4XmVmtptDWZqg&e=
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2019-09-27 21:44 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-09-27 21:44 [SPDK] Posix framework on client Tom Rainey
-- strict thread matches above, loose matches on Subject: below --
2019-09-27 21:04 Harris, James R
2019-09-27 20:56 Tom Rainey
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox