[i-g-t] igt/debugfs_test: Skip dummy reads for sysfs pipe crc data
diff mbox

Message ID 1498645996-8019-1-git-send-email-abdiel.janulgue@linux.intel.com
State New
Headers show

Commit Message

Abdiel Janulgue June 28, 2017, 10:33 a.m. UTC
Doing this tends to takes forever.

Signed-off-by: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
---
 tests/debugfs_test.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

Comments

Arkadiusz Hiler June 28, 2017, 11:06 a.m. UTC | #1
On Wed, Jun 28, 2017 at 01:33:16PM +0300, Abdiel Janulgue wrote:
> Doing this tends to takes forever.
> 
> Signed-off-by: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
> ---
>  tests/debugfs_test.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/tests/debugfs_test.c b/tests/debugfs_test.c
> index 3f0eaae..3d973d9 100644
> --- a/tests/debugfs_test.c
> +++ b/tests/debugfs_test.c
> @@ -40,7 +40,8 @@ static void read_and_discard_sysfs_entries(int path_fd)
>  
>  	while ((dirent = readdir(dir))) {
>  		if (!strcmp(dirent->d_name, ".") ||
> -		    !strcmp(dirent->d_name, ".."))
> +		    !strcmp(dirent->d_name, "..") ||
> +		    !strcmp(dirent->d_name, "crc"))

Do we have one that actually is named just "crc"?

IIRC 'i915_something_crc_something' was the format, and the check won't
catch them.
Abdiel Janulgue June 29, 2017, 8:05 a.m. UTC | #2
On 28.06.2017 14:06, Arkadiusz Hiler wrote:
> On Wed, Jun 28, 2017 at 01:33:16PM +0300, Abdiel Janulgue wrote:
>> Doing this tends to takes forever.
>>
>> Signed-off-by: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
>> ---
>>  tests/debugfs_test.c | 3 ++-
>>  1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/tests/debugfs_test.c b/tests/debugfs_test.c
>> index 3f0eaae..3d973d9 100644
>> --- a/tests/debugfs_test.c
>> +++ b/tests/debugfs_test.c
>> @@ -40,7 +40,8 @@ static void read_and_discard_sysfs_entries(int path_fd)
>>  
>>  	while ((dirent = readdir(dir))) {
>>  		if (!strcmp(dirent->d_name, ".") ||
>> -		    !strcmp(dirent->d_name, ".."))
>> +		    !strcmp(dirent->d_name, "..") ||
>> +		    !strcmp(dirent->d_name, "crc"))
> 
> Do we have one that actually is named just "crc"?
> 
> IIRC 'i915_something_crc_something' was the format, and the check won't
> catch them.

Hi,

I was referring to crc capture data at crtc-*/crc/data which is an
unlimited stream of bytes when read outright (tests for this are already
covered by kms_pipe_crc_basic).

How about just checking for "crtc-"?

Patch
diff mbox

diff --git a/tests/debugfs_test.c b/tests/debugfs_test.c
index 3f0eaae..3d973d9 100644
--- a/tests/debugfs_test.c
+++ b/tests/debugfs_test.c
@@ -40,7 +40,8 @@  static void read_and_discard_sysfs_entries(int path_fd)
 
 	while ((dirent = readdir(dir))) {
 		if (!strcmp(dirent->d_name, ".") ||
-		    !strcmp(dirent->d_name, ".."))
+		    !strcmp(dirent->d_name, "..") ||
+		    !strcmp(dirent->d_name, "crc"))
 			continue;
 		if (dirent->d_type == DT_DIR) {
 			int sub_fd = -1;