[LTP] [PATCH v4 1/4] lib/tst_kconfig.c: Add any kconfig to match the expected value function

[LTP] [PATCH v4 1/4] lib/tst_kconfig.c: Add any kconfig to match the expected value function

[Pengfei Xu]

Example: CONFIG_X86_INTEL_UMIP=y for umip kconfig before and v5.4
           mainline kernel.
         CONFIG_X86_UMIP=y for umip kconfig after v5.5 mainline kernel.
Format: CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y

Signed-off-by: Pengfei Xu <pengfei.xu@intel.com>
---
 lib/tst_kconfig.c | 36 +++++++++++++++++++++++-------------
 1 file changed, 23 insertions(+), 13 deletions(-)

diff --git a/lib/tst_kconfig.c b/lib/tst_kconfig.c
index 4b51413e5..74c46ebec 100644
--- a/lib/tst_kconfig.c
+++ b/lib/tst_kconfig.c
@@ -167,7 +167,8 @@ void tst_kconfig_read(const char *const *kconfigs,
 	struct match matches[cnt];
 	FILE *fp;
 	unsigned int i, j;
-	char buf[1024];
+	char buf[1024], kconfig_multi[100];
+	char *kconfig_token = NULL, *p_left = NULL;
 
 	for (i = 0; i < cnt; i++) {
 		const char *val = strchr(kconfigs[i], '=');
@@ -176,12 +177,9 @@ void tst_kconfig_read(const char *const *kconfigs,
 			tst_brk(TBROK, "Invalid config string '%s'", kconfigs[i]);
 
 		matches[i].match = 0;
-		matches[i].len = strlen(kconfigs[i]);
 
-		if (val) {
+		if (val)
 			matches[i].val = val + 1;
-			matches[i].len -= strlen(val);
-		}
 
 		results[i].match = 0;
 		results[i].value = NULL;
@@ -193,17 +191,29 @@ void tst_kconfig_read(const char *const *kconfigs,
 
 	while (fgets(buf, sizeof(buf), fp)) {
 		for (i = 0; i < cnt; i++) {
-			if (match(&matches[i], kconfigs[i], &results[i], buf)) {
-				for (j = 0; j < cnt; j++) {
-					if (matches[j].match)
-						break;
+			memset(kconfig_multi, 0, sizeof(kconfig_multi));
+			/* strtok_r will split kconfigs[i] to multi string, so copy it */
+			memcpy(kconfig_multi, kconfigs[i], strlen(kconfigs[i]));
+			kconfig_token = strtok_r(kconfig_multi, "|=", &p_left);
+
+			while (kconfig_token != NULL) {
+				if (strncmp("CONFIG_", kconfig_token, 7))
+					tst_brk(TBROK, "Invalid config string '%s'", kconfig_token);
+				matches[i].len = strlen(kconfig_token);
+				if (match(&matches[i], kconfig_token, &results[i], buf)) {
+					for (j = 0; j < cnt; j++) {
+						if (matches[j].match)
+							break;
+					}
 				}
-
-				if (j == cnt)
-					goto exit;
+				kconfig_token = strtok_r(NULL, "|=", &p_left);
+				/* avoid to use after "=" string */
+				if (strlen(p_left) == 0)
+					break;
 			}
+			if (j == cnt)
+				goto exit;
 		}
-
 	}
 
 exit:
-- 
2.14.1

[LTP] [PATCH v4 2/4] umip_basic_test.c: improve kconfig check to avoid umip wrong abort case

[Pengfei Xu]

From v5.5 main line, umip kconfig changed from "CONFIG_X86_INTEL_UMIP=y"
to "CONFIG_X86_UMIP=y".
We could use "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y" to check kconfig
CONFIG_X86_INTEL_UMIP=y(old kernel) or CONFIG_X86_UMIP=y(new kernel) for umip.

Signed-off-by: Pengfei Xu <pengfei.xu@intel.com>
---
 testcases/kernel/security/umip/umip_basic_test.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/testcases/kernel/security/umip/umip_basic_test.c b/testcases/kernel/security/umip/umip_basic_test.c
index 1baa26c52..24eca8890 100644
--- a/testcases/kernel/security/umip/umip_basic_test.c
+++ b/testcases/kernel/security/umip/umip_basic_test.c
@@ -171,7 +171,7 @@ static struct tst_test test = {
 	.forks_child = 1,
 	.test = verify_umip_instruction,
 	.needs_kconfigs = (const char *[]){
-		"CONFIG_X86_INTEL_UMIP=y",
+		"CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y",
 		NULL
 	},
 	.needs_root = 1,
-- 
2.14.1

[LTP] [PATCH v4 3/4] lib: add any kconfig to match the expected value function into kconfig test

[Pengfei Xu]

config01/02/03/04 should be passed for UMIP kconfig
All cases in config05 should be failed.

Signed-off-by: Pengfei Xu <pengfei.xu@intel.com>
---
 lib/newlib_tests/config01       | 1 +
 lib/newlib_tests/config02       | 1 +
 lib/newlib_tests/config03       | 1 +
 lib/newlib_tests/config04       | 1 +
 lib/newlib_tests/config05       | 2 ++
 lib/newlib_tests/test_kconfig.c | 1 +
 6 files changed, 7 insertions(+)

diff --git a/lib/newlib_tests/config01 b/lib/newlib_tests/config01
index 96d68d836..085c9368c 100644
--- a/lib/newlib_tests/config01
+++ b/lib/newlib_tests/config01
@@ -2,3 +2,4 @@
 CONFIG_MMU=y
 CONFIG_EXT4_FS=m
 CONFIG_PGTABLE_LEVELS=4
+CONFIG_X86_UMIP=y
diff --git a/lib/newlib_tests/config02 b/lib/newlib_tests/config02
index 2de45cff8..ca71d26c1 100644
--- a/lib/newlib_tests/config02
+++ b/lib/newlib_tests/config02
@@ -2,3 +2,4 @@
 # CONFIG_MMU is not set
 CONFIG_EXT4_FS=m
 CONFIG_PGTABLE_LEVELS=4
+CONFIG_X86_INTEL_UMIP=y
diff --git a/lib/newlib_tests/config03 b/lib/newlib_tests/config03
index 1a3b9e648..8a92def74 100644
--- a/lib/newlib_tests/config03
+++ b/lib/newlib_tests/config03
@@ -2,3 +2,4 @@
 CONFIG_MMU=y
 CONFIG_EXT4_FS=m
 CONFIG_PGTABLE_LEVELS=44
+CONFIG_X86_UMIP=y
diff --git a/lib/newlib_tests/config04 b/lib/newlib_tests/config04
index cce7051ae..424157fec 100644
--- a/lib/newlib_tests/config04
+++ b/lib/newlib_tests/config04
@@ -1,4 +1,5 @@
 # Unexpected CONFIG_EXT4_FS compiled in
 CONFIG_MMU=y
 CONFIG_EXT4_FS=y
+CONFIG_X86_INTEL_UMIP=y
 CONFIG_PGTABLE_LEVELS=4
diff --git a/lib/newlib_tests/config05 b/lib/newlib_tests/config05
index a9d7bab4d..e68fa8913 100644
--- a/lib/newlib_tests/config05
+++ b/lib/newlib_tests/config05
@@ -1,3 +1,5 @@
 # Everything is wrong
 CONFIG_EXT4_FS=y
 CONFIG_PGTABLE_LEVELS=44
+CONFIG_X86_UMI=y
+CONFIG_X86_INT_UMIP=y
diff --git a/lib/newlib_tests/test_kconfig.c b/lib/newlib_tests/test_kconfig.c
index d9c662fc5..9515b60e2 100644
--- a/lib/newlib_tests/test_kconfig.c
+++ b/lib/newlib_tests/test_kconfig.c
@@ -14,6 +14,7 @@ static const char *kconfigs[] = {
 	"CONFIG_MMU",
 	"CONFIG_EXT4_FS=m",
 	"CONFIG_PGTABLE_LEVELS=4",
+	"CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y",
 	NULL
 };
 
-- 
2.14.1

[LTP] [PATCH v4 4/4] lib: add any kconfig to match the expected value function document

[Pengfei Xu]

Signed-off-by: Pengfei Xu <pengfei.xu@intel.com>
---
 doc/test-writing-guidelines.txt | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/doc/test-writing-guidelines.txt b/doc/test-writing-guidelines.txt
index 79d857fea..e64ff8716 100644
--- a/doc/test-writing-guidelines.txt
+++ b/doc/test-writing-guidelines.txt
@@ -1590,7 +1590,9 @@ aborted with 'TCONF' if any of the required options were not set.
 #include "tst_test.h"
 
 static const char *kconfigs[] = {
-	"CONFIG_X86_INTEL_UMIP",
+	"CONFIG_EXT4_FS=y",
+	"CONFIG_MMU",
+	"CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y",
 	NULL
 };
 
-- 
2.14.1

[LTP] [PATCH v4 1/4] lib/tst_kconfig.c: Add any kconfig to match the expected value function

[Yang Xu]

Hi Pengfei

This patch set looks good to me, only a samll comment.
Reviewed-by: Yang Xu <xuyang2018.jy@cn.fujitsu.com>

on 2019/12/19 21:18, Pengfei Xu wrote:
> Example: CONFIG_X86_INTEL_UMIP=y for umip kconfig before and v5.4
>             mainline kernel.
>           CONFIG_X86_UMIP=y for umip kconfig after v5.5 mainline kernel.
> Format: CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y
> 
> Signed-off-by: Pengfei Xu <pengfei.xu@intel.com>
> ---
>   lib/tst_kconfig.c | 36 +++++++++++++++++++++++-------------
>   1 file changed, 23 insertions(+), 13 deletions(-)
> 
> diff --git a/lib/tst_kconfig.c b/lib/tst_kconfig.c
> index 4b51413e5..74c46ebec 100644
> --- a/lib/tst_kconfig.c
> +++ b/lib/tst_kconfig.c
> @@ -167,7 +167,8 @@ void tst_kconfig_read(const char *const *kconfigs,
>   	struct match matches[cnt];
>   	FILE *fp;
>   	unsigned int i, j;
> -	char buf[1024];
> +	char buf[1024], kconfig_multi[100];
> +	char *kconfig_token = NULL, *p_left = NULL;
>   
>   	for (i = 0; i < cnt; i++) {
>   		const char *val = strchr(kconfigs[i], '=');
> @@ -176,12 +177,9 @@ void tst_kconfig_read(const char *const *kconfigs,
>   			tst_brk(TBROK, "Invalid config string '%s'", kconfigs[i]);
>   
>   		matches[i].match = 0;
> -		matches[i].len = strlen(kconfigs[i]);
>   
> -		if (val) {
> +		if (val)
>   			matches[i].val = val + 1;
> -			matches[i].len -= strlen(val);
> -		}
>   
>   		results[i].match = 0;
>   		results[i].value = NULL;
> @@ -193,17 +191,29 @@ void tst_kconfig_read(const char *const *kconfigs,
>   
>   	while (fgets(buf, sizeof(buf), fp)) {
>   		for (i = 0; i < cnt; i++) {
> -			if (match(&matches[i], kconfigs[i], &results[i], buf)) {
> -				for (j = 0; j < cnt; j++) {
> -					if (matches[j].match)
> -						break;
> +			memset(kconfig_multi, 0, sizeof(kconfig_multi));
> +			/* strtok_r will split kconfigs[i] to multi string, so copy it */
> +			memcpy(kconfig_multi, kconfigs[i], strlen(kconfigs[i]));
> +			kconfig_token = strtok_r(kconfig_multi, "|=", &p_left);
> +
> +			while (kconfig_token != NULL) {
> +				if (strncmp("CONFIG_", kconfig_token, 7))
> +					tst_brk(TBROK, "Invalid config string '%s'", kconfig_token);
> +				matches[i].len = strlen(kconfig_token);
> +				if (match(&matches[i], kconfig_token, &results[i], buf)) {
> +					for (j = 0; j < cnt; j++) {
> +						if (matches[j].match)
> +							break;
> +					}
>   				}
> -
> -				if (j == cnt)
It will report ?j? may be used uninitialized in this function when 
compile. But I think it can be modified by maintainer when it is merged.
> -					goto exit;
> +				kconfig_token = strtok_r(NULL, "|=", &p_left);
> +				/* avoid to use after "=" string */
> +				if (strlen(p_left) == 0)
> +					break;
>   			}
> +			if (j == cnt)
> +				goto exit;
>   		}
> -
>   	}
>   
>   exit:
> 

[LTP] [PATCH v4 4/4] lib: add any kconfig to match the expected value function document

[Yang Xu]

Hi Pengfei

on 2019/12/19 21:18, Pengfei Xu wrote:
> Signed-off-by: Pengfei Xu <pengfei.xu@intel.com>
> ---
>   doc/test-writing-guidelines.txt | 4 +++-
>   1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/doc/test-writing-guidelines.txt b/doc/test-writing-guidelines.txt
> index 79d857fea..e64ff8716 100644
> --- a/doc/test-writing-guidelines.txt
> +++ b/doc/test-writing-guidelines.txt
> @@ -1590,7 +1590,9 @@ aborted with 'TCONF' if any of the required options were not set.
Before your patch, I know we can use the following two formats kconfigs
CONFIG_A
CONFIG_A=y/m/v
after your patch set, we can use the following three formats kconfigs
CONFIG_A
CONFIG_A=y/m/v
CONFIG_A|CONFIGB=y/m/v
As the usual extend logic,  we think  CONFIGA|CONFIGB is also ok. But in 
fact, we use "|" or "=" to delim string. So  we can't parse 
CONFIGA|CONFIGB correctly. So, if we can tell user or developer about 
this in here, it will be better.

ps: we can add configA| config B test in your third patch.

Kind Regards
Yang Xu
>   #include "tst_test.h"
>   
>   static const char *kconfigs[] = {
> -	"CONFIG_X86_INTEL_UMIP",
> +	"CONFIG_EXT4_FS=y",
> +	"CONFIG_MMU",
> +	"CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y",
>   	NULL
>   };
>   
> 

[LTP] [PATCH v4 1/4] lib/tst_kconfig.c: Add any kconfig to match the expected value function

[Pengfei Xu]

Hi Xu,
  Thanks for your advice!
  I will send the new patch to solve the warning.

  Thanks!
  BR.

On 2019-12-20 at 13:12:17 +0800, Yang Xu wrote:
> Hi Pengfei
> 
> This patch set looks good to me, only a samll comment.
> Reviewed-by: Yang Xu <xuyang2018.jy@cn.fujitsu.com>
> 
> on 2019/12/19 21:18, Pengfei Xu wrote:
> > Example: CONFIG_X86_INTEL_UMIP=y for umip kconfig before and v5.4
> >             mainline kernel.
> >           CONFIG_X86_UMIP=y for umip kconfig after v5.5 mainline kernel.
> > Format: CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y
> > 
> > Signed-off-by: Pengfei Xu <pengfei.xu@intel.com>
> > ---
> >   lib/tst_kconfig.c | 36 +++++++++++++++++++++++-------------
> >   1 file changed, 23 insertions(+), 13 deletions(-)
> > 
> > diff --git a/lib/tst_kconfig.c b/lib/tst_kconfig.c
> > index 4b51413e5..74c46ebec 100644
> > --- a/lib/tst_kconfig.c
> > +++ b/lib/tst_kconfig.c
> > @@ -167,7 +167,8 @@ void tst_kconfig_read(const char *const *kconfigs,
> >   	struct match matches[cnt];
> > +							break;
> > +					}
> >   				}
> > -
> > -				if (j == cnt)
> It will report ?j? may be used uninitialized in this function when compile.
> But I think it can be modified by maintainer when it is merged.
> > -					goto exit;
> > +				kconfig_token = strtok_r(NULL, "|=", &p_left);
> > +				/* avoid to use after "=" string */
> > +				if (strlen(p_left) == 0)
> > +					break;
> >   			}
> > +			if (j == cnt)
> > +				goto exit;
> >   		}
> > -
> >   	}
> >   exit:
> > 
> 
> 

[LTP] [PATCH v4 4/4] lib: add any kconfig to match the expected value function document

[Yang Xu]

Hi Pengfei
> Hi Xu,
> 
> 
> On 2019-12-20 at 13:37:21 +0800, Yang Xu wrote:
>>
>> Hi Pengfei
>>
>> on 2019/12/19 21:18, Pengfei Xu wrote:
>>> Signed-off-by: Pengfei Xu <pengfei.xu@intel.com>
>>> ---
>>>    doc/test-writing-guidelines.txt | 4 +++-
>>>    1 file changed, 3 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/doc/test-writing-guidelines.txt b/doc/test-writing-guidelines.txt
>>> index 79d857fea..e64ff8716 100644
>>> --- a/doc/test-writing-guidelines.txt
>>> +++ b/doc/test-writing-guidelines.txt
>>> @@ -1590,7 +1590,9 @@ aborted with 'TCONF' if any of the required options were not set.
>> Before your patch, I know we can use the following two formats kconfigs
>> CONFIG_A
>> CONFIG_A=y/m/v
>> after your patch set, we can use the following three formats kconfigs
>> CONFIG_A
>> CONFIG_A=y/m/v
>> CONFIG_A|CONFIGB=y/m/v
>> As the usual extend logic,  we think  CONFIGA|CONFIGB is also ok. But in
>> fact, we use "|" or "=" to delim string. So  we can't parse CONFIGA|CONFIGB
>> correctly. So, if we can tell user or developer about this in here, it will
>> be better.
>>
>> ps: we can add configA| config B test in your third patch.
>>
>    Actually present patch could support CONFIG_A|CONFIG_B style, and it works
> well, you could change "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y" to
> "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP" in test_kconfig.c and have a try. : )
"CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP" is useful because 
CONFIG_X86_INTEL_UMIP is in our kernel configs . If you use 
"CONFIG_X86_UMIP|CONFIG_X86_INTEL_UMIP", it will report error.

Kind Regards
Yang Xu
> 
> Thanks!
>> Kind Regards
>> Yang Xu
>>>    #include "tst_test.h"
>>>    static const char *kconfigs[] = {
>>> -	"CONFIG_X86_INTEL_UMIP",
>>> +	"CONFIG_EXT4_FS=y",
>>> +	"CONFIG_MMU",
>>> +	"CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y",
>>>    	NULL
>>>    };
>>>
>>
>>
> 
> 

[LTP] [PATCH v4 4/4] lib: add any kconfig to match the expected value function document

[Pengfei Xu]

Hi Xu,


On 2019-12-20 at 13:37:21 +0800, Yang Xu wrote:
> 
> Hi Pengfei
> 
> on 2019/12/19 21:18, Pengfei Xu wrote:
> > Signed-off-by: Pengfei Xu <pengfei.xu@intel.com>
> > ---
> >   doc/test-writing-guidelines.txt | 4 +++-
> >   1 file changed, 3 insertions(+), 1 deletion(-)
> > 
> > diff --git a/doc/test-writing-guidelines.txt b/doc/test-writing-guidelines.txt
> > index 79d857fea..e64ff8716 100644
> > --- a/doc/test-writing-guidelines.txt
> > +++ b/doc/test-writing-guidelines.txt
> > @@ -1590,7 +1590,9 @@ aborted with 'TCONF' if any of the required options were not set.
> Before your patch, I know we can use the following two formats kconfigs
> CONFIG_A
> CONFIG_A=y/m/v
> after your patch set, we can use the following three formats kconfigs
> CONFIG_A
> CONFIG_A=y/m/v
> CONFIG_A|CONFIGB=y/m/v
> As the usual extend logic,  we think  CONFIGA|CONFIGB is also ok. But in
> fact, we use "|" or "=" to delim string. So  we can't parse CONFIGA|CONFIGB
> correctly. So, if we can tell user or developer about this in here, it will
> be better.
> 
> ps: we can add configA| config B test in your third patch.
> 
  Actually present patch could support CONFIG_A|CONFIG_B style, and it works
well, you could change "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y" to
"CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP" in test_kconfig.c and have a try. : )

Thanks!
> Kind Regards
> Yang Xu
> >   #include "tst_test.h"
> >   static const char *kconfigs[] = {
> > -	"CONFIG_X86_INTEL_UMIP",
> > +	"CONFIG_EXT4_FS=y",
> > +	"CONFIG_MMU",
> > +	"CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y",
> >   	NULL
> >   };
> > 
> 
> 

[LTP] [PATCH v4 4/4] lib: add any kconfig to match the expected value function document

[Pengfei Xu]

Hi Xu,
  Thanks for your comment.
  I will try to fix it and send the patch.

  Thanks!
  BR.

On 2019-12-20 at 14:00:00 +0800, Yang Xu wrote:
> Hi Pengfei
> > Hi Xu,
> > 
> > 
> > On 2019-12-20 at 13:37:21 +0800, Yang Xu wrote:
> > > 
> > > Hi Pengfei
> > > 
> > > on 2019/12/19 21:18, Pengfei Xu wrote:
> > > > Signed-off-by: Pengfei Xu <pengfei.xu@intel.com>
> > > > ---
> > > >    doc/test-writing-guidelines.txt | 4 +++-
> > > >    1 file changed, 3 insertions(+), 1 deletion(-)
> > > > 
> > > > diff --git a/doc/test-writing-guidelines.txt b/doc/test-writing-guidelines.txt
> > > > index 79d857fea..e64ff8716 100644
> > > > --- a/doc/test-writing-guidelines.txt
> > > > +++ b/doc/test-writing-guidelines.txt
> > > > @@ -1590,7 +1590,9 @@ aborted with 'TCONF' if any of the required options were not set.
> > > Before your patch, I know we can use the following two formats kconfigs
> > > CONFIG_A
> > > CONFIG_A=y/m/v
> > > after your patch set, we can use the following three formats kconfigs
> > > CONFIG_A
> > > CONFIG_A=y/m/v
> > > CONFIG_A|CONFIGB=y/m/v
> > > As the usual extend logic,  we think  CONFIGA|CONFIGB is also ok. But in
> > > fact, we use "|" or "=" to delim string. So  we can't parse CONFIGA|CONFIGB
> > > correctly. So, if we can tell user or developer about this in here, it will
> > > be better.
> > > 
> > > ps: we can add configA| config B test in your third patch.
> > > 
> >    Actually present patch could support CONFIG_A|CONFIG_B style, and it works
> > well, you could change "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y" to
> > "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP" in test_kconfig.c and have a try. : )
> "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP" is useful because
> CONFIG_X86_INTEL_UMIP is in our kernel configs . If you use
> "CONFIG_X86_UMIP|CONFIG_X86_INTEL_UMIP", it will report error.
> 
> Kind Regards
> Yang Xu
> > 
> > Thanks!
> > > Kind Regards
> > > Yang Xu
> > > >    #include "tst_test.h"
> > > >    static const char *kconfigs[] = {
> > > > -	"CONFIG_X86_INTEL_UMIP",
> > > > +	"CONFIG_EXT4_FS=y",
> > > > +	"CONFIG_MMU",
> > > > +	"CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y",
> > > >    	NULL
> > > >    };
> > > > 
> > > 
> > > 
> > 
> > 
> 
> 

[LTP] [PATCH v4 4/4] lib: add any kconfig to match the expected value function document

[Pengfei Xu]

Hi Xu,
  Due to memory copy, there is some bit and display issue in last string.

  So for your question, for CONFIG_A|CONFIG_B without expect value, maybe y
  or m.
  We could add it with below style:
  "CONFIG_A|CONFIG_B|NA", and add it into guidelines, NA will not be solved,
  actually you could fill with any string after last '|'.
  Is it ok?

  Thanks!

On 2019-12-20 at 14:00:00 +0800, Yang Xu wrote:
> Hi Pengfei
> > Hi Xu,
> > 
> > 
> > > Before your patch, I know we can use the following two formats kconfigs
> > > CONFIG_A
> > > CONFIG_A=y/m/v
> > > after your patch set, we can use the following three formats kconfigs
> > > CONFIG_A
> > > CONFIG_A=y/m/v
> > > CONFIG_A|CONFIGB=y/m/v
> > > As the usual extend logic,  we think  CONFIGA|CONFIGB is also ok. But in
> > > fact, we use "|" or "=" to delim string. So  we can't parse CONFIGA|CONFIGB
> > > correctly. So, if we can tell user or developer about this in here, it will
> > > be better.
> > > 
> > > ps: we can add configA| config B test in your third patch.
> > > 
> >    Actually present patch could support CONFIG_A|CONFIG_B style, and it works
> > well, you could change "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y" to
> > "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP" in test_kconfig.c and have a try. : )
> "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP" is useful because
> CONFIG_X86_INTEL_UMIP is in our kernel configs . If you use
> "CONFIG_X86_UMIP|CONFIG_X86_INTEL_UMIP", it will report error.
> 
> Kind Regards
> Yang Xu
> > 
> > Thanks!
> > > Kind Regards
> > > Yang Xu
> > > >    #include "tst_test.h"
> > > >    static const char *kconfigs[] = {
> > > > -	"CONFIG_X86_INTEL_UMIP",
> > > > +	"CONFIG_EXT4_FS=y",
> > > > +	"CONFIG_MMU",
> > > > +	"CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y",
> > > >    	NULL
> > > >    };
> > > > 
> > > 
> > > 
> > 
> > 
> 
> 

[LTP] [PATCH v4 4/4] lib: add any kconfig to match the expected value function document

[Yang Xu]

Hi Pengfei
> Hi Xu,
>    Due to memory copy, there is some bit and display issue in last string.
> 
>    So for your question, for CONFIG_A|CONFIG_B without expect value, maybe y
>    or m.
>    We could add it with below style:
>    "CONFIG_A|CONFIG_B|NA", and add it into guidelines, NA will not be solved,
>    actually you could fill with any string after last '|'.
>    Is it ok?
It sounds reasonable. I think it is ok.

Kind Regards
Yang Xu
> 
>    Thanks!
> 
> On 2019-12-20 at 14:00:00 +0800, Yang Xu wrote:
>> Hi Pengfei
>>> Hi Xu,
>>>
>>>
>>>> Before your patch, I know we can use the following two formats kconfigs
>>>> CONFIG_A
>>>> CONFIG_A=y/m/v
>>>> after your patch set, we can use the following three formats kconfigs
>>>> CONFIG_A
>>>> CONFIG_A=y/m/v
>>>> CONFIG_A|CONFIGB=y/m/v
>>>> As the usual extend logic,  we think  CONFIGA|CONFIGB is also ok. But in
>>>> fact, we use "|" or "=" to delim string. So  we can't parse CONFIGA|CONFIGB
>>>> correctly. So, if we can tell user or developer about this in here, it will
>>>> be better.
>>>>
>>>> ps: we can add configA| config B test in your third patch.
>>>>
>>>     Actually present patch could support CONFIG_A|CONFIG_B style, and it works
>>> well, you could change "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y" to
>>> "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP" in test_kconfig.c and have a try. : )
>> "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP" is useful because
>> CONFIG_X86_INTEL_UMIP is in our kernel configs . If you use
>> "CONFIG_X86_UMIP|CONFIG_X86_INTEL_UMIP", it will report error.
>>
>> Kind Regards
>> Yang Xu
>>>
>>> Thanks!
>>>> Kind Regards
>>>> Yang Xu
>>>>>     #include "tst_test.h"
>>>>>     static const char *kconfigs[] = {
>>>>> -	"CONFIG_X86_INTEL_UMIP",
>>>>> +	"CONFIG_EXT4_FS=y",
>>>>> +	"CONFIG_MMU",
>>>>> +	"CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y",
>>>>>     	NULL
>>>>>     };
>>>>>
>>>>
>>>>
>>>
>>>
>>
>>
> 
> 

[LTP] [PATCH v4 4/4] lib: add any kconfig to match the expected value function document

[Xu, Pengfei]

Thanks Xu! :)

BR.
> ? 2019?12?20??15:20?Yang Xu <xuyang2018.jy@cn.fujitsu.com> ???
> 
> ?Hi Pengfei
>> Hi Xu,
>>   Due to memory copy, there is some bit and display issue in last string.
>>   So for your question, for CONFIG_A|CONFIG_B without expect value, maybe y
>>   or m.
>>   We could add it with below style:
>>   "CONFIG_A|CONFIG_B|NA", and add it into guidelines, NA will not be solved,
>>   actually you could fill with any string after last '|'.
>>   Is it ok?
> It sounds reasonable. I think it is ok.
> 
> Kind Regards
> Yang Xu
>>   Thanks!
>>> On 2019-12-20 at 14:00:00 +0800, Yang Xu wrote:
>>> Hi Pengfei
>>>> Hi Xu,
>>>> 
>>>> 
>>>>> Before your patch, I know we can use the following two formats kconfigs
>>>>> CONFIG_A
>>>>> CONFIG_A=y/m/v
>>>>> after your patch set, we can use the following three formats kconfigs
>>>>> CONFIG_A
>>>>> CONFIG_A=y/m/v
>>>>> CONFIG_A|CONFIGB=y/m/v
>>>>> As the usual extend logic,  we think  CONFIGA|CONFIGB is also ok. But in
>>>>> fact, we use "|" or "=" to delim string. So  we can't parse CONFIGA|CONFIGB
>>>>> correctly. So, if we can tell user or developer about this in here, it will
>>>>> be better.
>>>>> 
>>>>> ps: we can add configA| config B test in your third patch.
>>>>> 
>>>>    Actually present patch could support CONFIG_A|CONFIG_B style, and it works
>>>> well, you could change "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y" to
>>>> "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP" in test_kconfig.c and have a try. : )
>>> "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP" is useful because
>>> CONFIG_X86_INTEL_UMIP is in our kernel configs . If you use
>>> "CONFIG_X86_UMIP|CONFIG_X86_INTEL_UMIP", it will report error.
>>> 
>>> Kind Regards
>>> Yang Xu
>>>> 
>>>> Thanks!
>>>>> Kind Regards
>>>>> Yang Xu
>>>>>>    #include "tst_test.h"
>>>>>>    static const char *kconfigs[] = {
>>>>>> -    "CONFIG_X86_INTEL_UMIP",
>>>>>> +    "CONFIG_EXT4_FS=y",
>>>>>> +    "CONFIG_MMU",
>>>>>> +    "CONFIG_X86_INTEL_UMIP|CONFIG_X86_UMIP=y",
>>>>>>        NULL
>>>>>>    };
>>>>>> 
>>>>> 
>>>>> 
>>>> 
>>>> 
>>> 
>>> 
> 
>