Hi Gregory,
From that screen you will see a list below of test nodes, some will show pass and some
fail. Click on the ones that show failure and you will get a dir listing. From there,
the file 'build.log' has the main logs in it. It can be a very long file and not
so intuitive sometimes. A good start would be to search for "backtrace start"
and see what happened just before that.
Thanks
Paul
-----Original Message-----
From: SPDK [mailto:spdk-bounces@lists.01.org] On Behalf Of Gregory Shapiro
Sent: Wednesday, January 2, 2019 6:56 AM
To: Storage Performance Development Kit <spdk(a)lists.01.org>
Cc: open-source-contrib(a)kaminario.com
Subject: Re: [SPDK] change model number to be custom (from make file) instead define
HI,
I see now my commit failed CI
https://ci.spdk.io/spdk/builds/review/95ebd99e9f7fd361255d6b39ddb1a2914fa...
Can someone assists me to find out what the errors are.
How do I find them in the log?
Thanks,
Gregory
On Sun, Dec 30, 2018 at 10:18 AM Gregory Shapiro <shapiro.gregory(a)gmail.com>
wrote:
Hi,
I added a commit to gerrithub with discussed change.
The code can be found in below link:
438562: NVMF: Add module number as parameter to
construct_nvmf_subsystem (-d option).
Regards,
Gregory.
On Tue, Dec 18, 2018 at 8:34 PM Harris, James R
<james.r.harris(a)intel.com> wrote:
>
>
>
> On 12/16/18, 6:34 AM, "SPDK on behalf of Gregory Shapiro"
<spdk-bounces(a)lists.01.org on behalf of shapiro.gregory(a)gmail.com> wrote:
>
> Hello, my name is Gregory and I am new to spdk project.
> I now work on some branding feature and need to change MODEL_NUMBER to be
> custom.
> What do you think of defining it in makefile and reading the info from
> there?
> Proposed change look something like this:
>
> Hi Gregory,
>
> I think adding the ability to specify the MODEL_NUMBER is fantastic, but I would
prefer the configuration to be more dynamic. For example, make it an optional parameter
when creating an nvmf subsystem. If the user does not specify a MODEL_NUMBER, SPDK will
just use the current default.
>
> Would you like to submit a patch to GerritHub for this?
>
> Regards,
>
> -Jim
>
>
> diff --git a/lib/nvmf/ctrlr.c b/lib/nvmf/ctrlr.c
> index ed5e68f0..bc80bf83 100644
> --- a/lib/nvmf/ctrlr.c
> +++ b/lib/nvmf/ctrlr.c
> @@ -49,8 +49,9 @@
>
> #define MIN_KEEP_ALIVE_TIMEOUT 10000
>
> -#define MODEL_NUMBER "SPDK bdev Controller"
> -
> +#ifndef MODEL_NUMBER
> + #define MODEL_NUMBER "SPDK bdev Controller"
> +#endif /* MODEL_NUMBER */
> /*
> * Report the SPDK version as the firmware revision.
> * SPDK_VERSION_STRING won't fit into FR (only 8 bytes), so try to fit
the
> most important parts.
> diff --git a/mk/spdk.common.mk b/mk/spdk.common.mk
> index 3612551c..ff3e69cd 100644
> --- a/mk/spdk.common.mk
> +++ b/mk/spdk.common.mk
> @@ -187,6 +187,7 @@ COMMON_CFLAGS += -pthread
> LDFLAGS += -pthread
>
> CFLAGS += $(COMMON_CFLAGS) -Wno-pointer-sign -Wstrict-prototypes
> -Wold-style-definition -std=gnu99
> +CFLAGS += -DMODEL_NUMBER=\"SPDK\ bdev\ Controller\"
> CXXFLAGS += $(COMMON_CFLAGS) -std=c++0x
>
> SYS_LIBS += -lrt
>
> I will be happy to hear what do you think about such modification.
>
> Best Regards,
> Gregory
> _______________________________________________
> SPDK mailing list
> SPDK(a)lists.01.org
>
https://lists.01.org/mailman/listinfo/spdk
>
>
> _______________________________________________
> SPDK mailing list
> SPDK(a)lists.01.org
>
https://lists.01.org/mailman/listinfo/spdk
_______________________________________________
SPDK mailing list
SPDK(a)lists.01.org
https://lists.01.org/mailman/listinfo/spdk