Re: [Intel-gfx] [PATCH i-g-t v2 01/13] lib/sw_sync: Add helper functions for managing synchronization primitives

2016-09-14 Thread Robert Foss 



On 2016-09-14 08:18 AM, Chris Wilson wrote:

On Tue, Sep 13, 2016 at 11:40:18AM -0400, Robert Foss wrote:



On 2016-09-13 07:03 AM, Chris Wilson wrote:

Try:

int __sw_sync_fence_create(int fd, int32_t seqno) /* int32_t not unsigned ? */
{

struct sw_sync_create_fence_data data;

memset(, 0, sizeof(data));
data.value = seqno;

if (igt_ioctl(fd, SW_SYNC_IOCT_CREATE_FENCE, ))
return -errno;

return data.fence;
}

int sw_sync_fence_create(int fd, int32_t seqno)
{
int fence = __sw_sync_fence_create(fd, seqno);
igt_assert(fence >= 0);
return fence;
}

Then only in the test code do you send garbage and check for the
expected errno.



What would the corresponding negative test code look like?
A call to __sw_sync_fence_create? Then __sw_sync_fence_create would
have to be made accessible outside of lib/sw_sync.

Or maybe creating a second user of __sw_sync_fence_create along the
lines of sw_sync_fence_create_fail with an inverted igt_assert check
is what you're suggesting.


Exactly. Make the raw unchecked version available to tests. We have been
using __gem_foo() and gem_foo() to identify the difference. __gem_foo()
reports the error to the caller (so that they can feed in different
values of garbage and check for different errno, or maybe used as a
probe to see if the kernel supports such a function) and gem_foo() for
everyone else where we want to just focus on writing a test and hide the
error handling clutter.

Try not to put the error handling tests in the library itself.


Thanks!
Coming up in v4.


Rob.
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH i-g-t v2 01/13] lib/sw_sync: Add helper functions for managing synchronization primitives

2016-09-14 Thread Chris Wilson 
On Tue, Sep 13, 2016 at 11:40:18AM -0400, Robert Foss wrote:
> 
> 
> On 2016-09-13 07:03 AM, Chris Wilson wrote:
> >Try:
> >
> >int __sw_sync_fence_create(int fd, int32_t seqno) /* int32_t not unsigned ? 
> >*/
> >{
> >
> > struct sw_sync_create_fence_data data;
> >
> > memset(, 0, sizeof(data));
> > data.value = seqno;
> >
> > if (igt_ioctl(fd, SW_SYNC_IOCT_CREATE_FENCE, ))
> > return -errno;
> > 
> > return data.fence;
> >}
> >
> >int sw_sync_fence_create(int fd, int32_t seqno)
> >{
> > int fence = __sw_sync_fence_create(fd, seqno);
> > igt_assert(fence >= 0);
> > return fence;
> >}
> >
> >Then only in the test code do you send garbage and check for the
> >expected errno.
> >
> 
> What would the corresponding negative test code look like?
> A call to __sw_sync_fence_create? Then __sw_sync_fence_create would
> have to be made accessible outside of lib/sw_sync.
> 
> Or maybe creating a second user of __sw_sync_fence_create along the
> lines of sw_sync_fence_create_fail with an inverted igt_assert check
> is what you're suggesting.

Exactly. Make the raw unchecked version available to tests. We have been
using __gem_foo() and gem_foo() to identify the difference. __gem_foo()
reports the error to the caller (so that they can feed in different
values of garbage and check for different errno, or maybe used as a
probe to see if the kernel supports such a function) and gem_foo() for
everyone else where we want to just focus on writing a test and hide the
error handling clutter.

Try not to put the error handling tests in the library itself.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH i-g-t v2 01/13] lib/sw_sync: Add helper functions for managing synchronization primitives

2016-09-13 Thread Robert Foss 



On 2016-09-13 07:03 AM, Chris Wilson wrote:

On Mon, Sep 12, 2016 at 06:08:30PM -0400, robert.f...@collabora.com wrote:

From: Robert Foss 

Base functions to help testing the Sync File Framework (explicit fencing
mechanism ported from Android).
These functions allow you to create, use and destroy timelines and fences.

Signed-off-by: Robert Foss 
Signed-off-by: Gustavo Padovan 
Reviewed-by: Eric Engestrom 
---
 lib/Makefile.sources |   2 +
 lib/sw_sync.c| 237 +++
 lib/sw_sync.h|  49 +++
 3 files changed, 288 insertions(+)
 create mode 100644 lib/sw_sync.c
 create mode 100644 lib/sw_sync.h

diff --git a/lib/Makefile.sources b/lib/Makefile.sources
index bac9a7f..3dc7c3c 100644
--- a/lib/Makefile.sources
+++ b/lib/Makefile.sources
@@ -61,6 +61,8 @@ lib_source_list = \
rendercopy_gen8.c   \
rendercopy_gen9.c   \
rendercopy.h\
+   sw_sync.c   \
+   sw_sync.h   \
intel_reg_map.c \
intel_iosf.c\
igt_kms.c   \
diff --git a/lib/sw_sync.c b/lib/sw_sync.c
new file mode 100644
index 000..179a473
--- /dev/null
+++ b/lib/sw_sync.c
@@ -0,0 +1,237 @@
+/*
+ * Copyright 2012 Google, Inc
+ * Copyright © 2016 Collabora, Ltd.
+ *
+ * Based on the implementation from the Android Open Source Project
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Authors:
+ *Robert Foss 
+ */
+
+#ifndef ANDROID
+#define _GNU_SOURCE
+#else
+#include 
+#endif
+#include 
+#include 
+#include 
+#include 
+#include 
+
+#include "sw_sync.h"
+#include "drmtest.h"
+#include "ioctl_wrappers.h"
+
+#ifndef SW_SYNC_IOC_INC
+struct sw_sync_create_fence_data {
+   __u32   value;
+   charname[32];
+   __s32   fence;
+};
+
+#define SW_SYNC_IOC_MAGIC  'W'
+#define SW_SYNC_IOC_CREATE_FENCE   _IOWR(SW_SYNC_IOC_MAGIC, 0,\
+   struct 
sw_sync_create_fence_data)
+#define SW_SYNC_IOC_INC_IOW(SW_SYNC_IOC_MAGIC, 1, 
__u32)
+#endif
+
+#define DEVFS_SW_SYNC   "/dev/sw_sync"
+#define DEBUGFS_SW_SYNC "/sys/kernel/debug/sync/sw_sync"
+
+
+int sw_sync_fd_is_valid(int fd)
+{
+   int status;
+
+   if (fd < 0)
+   return 0;
+
+   status = fcntl(fd, F_GETFD, 0);
+   return status >= 0;
+}
+
+static
+void sw_sync_fd_close(int fd)
+{
+   if (!sw_sync_fd_is_valid(fd))
+   return;
+
+   close(fd);


What bug are you trying to prevent? This is the same as just calling
close(fd) and seeing EINVAL.


It's not really trying to prevent any bugs, but rather the idea is have
low cognitive load on these functions. If an fd is valid it can always 
be closed.


Or are you suggesting that the above code is equivalent to close()?
If so it should be replaced by close().




+}
+
+int sw_sync_timeline_create(void)
+{
+   int fd = open(DEVFS_SW_SYNC, O_RDWR);


if (fd < 0)
fd = open(DEBUGFS_SW_SYNC, O_RDWR);

igt_require(fd >= 0);


Replacing if (!sw_sync_fd_is_valid(fd)) with fd<0 is seems to me like a 
good change, in this scenario they should be equivalent.


Adding igt_require() is a bit inflexbile, especially when it comes to 
actually testing timeline creation.





+
+   fd = open(DEBUGFS_SW_SYNC, O_RDWR);
+
+   igt_assert(sw_sync_fd_is_valid(fd));
+
+   return fd;
+}
+
+void sw_sync_timeline_destroy(int fd)
+{
+   return sw_sync_fd_close(fd);


Now just close(fd); and might as well not obfuscate that these are plain
fd.


You're definitely, right. I'll remove this function from lib/sw_sync and 
tests/sw_sync in v3.





+int sw_sync_fence_create(int fd, int32_t seqno)
+{
+   struct

Re: [Intel-gfx] [PATCH i-g-t v2 01/13] lib/sw_sync: Add helper functions for managing synchronization primitives

2016-09-13 Thread Chris Wilson 
On Mon, Sep 12, 2016 at 06:08:30PM -0400, robert.f...@collabora.com wrote:
> From: Robert Foss 
> 
> Base functions to help testing the Sync File Framework (explicit fencing
> mechanism ported from Android).
> These functions allow you to create, use and destroy timelines and fences.
> 
> Signed-off-by: Robert Foss 
> Signed-off-by: Gustavo Padovan 
> Reviewed-by: Eric Engestrom 
> ---
>  lib/Makefile.sources |   2 +
>  lib/sw_sync.c| 237 
> +++
>  lib/sw_sync.h|  49 +++
>  3 files changed, 288 insertions(+)
>  create mode 100644 lib/sw_sync.c
>  create mode 100644 lib/sw_sync.h
> 
> diff --git a/lib/Makefile.sources b/lib/Makefile.sources
> index bac9a7f..3dc7c3c 100644
> --- a/lib/Makefile.sources
> +++ b/lib/Makefile.sources
> @@ -61,6 +61,8 @@ lib_source_list =   \
>   rendercopy_gen8.c   \
>   rendercopy_gen9.c   \
>   rendercopy.h\
> + sw_sync.c   \
> + sw_sync.h   \
>   intel_reg_map.c \
>   intel_iosf.c\
>   igt_kms.c   \
> diff --git a/lib/sw_sync.c b/lib/sw_sync.c
> new file mode 100644
> index 000..179a473
> --- /dev/null
> +++ b/lib/sw_sync.c
> @@ -0,0 +1,237 @@
> +/*
> + * Copyright 2012 Google, Inc
> + * Copyright © 2016 Collabora, Ltd.
> + *
> + * Based on the implementation from the Android Open Source Project
> + *
> + * Permission is hereby granted, free of charge, to any person obtaining a
> + * copy of this software and associated documentation files (the "Software"),
> + * to deal in the Software without restriction, including without limitation
> + * the rights to use, copy, modify, merge, publish, distribute, sublicense,
> + * and/or sell copies of the Software, and to permit persons to whom the
> + * Software is furnished to do so, subject to the following conditions:
> + *
> + * The above copyright notice and this permission notice (including the next
> + * paragraph) shall be included in all copies or substantial portions of the
> + * Software.
> + *
> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
> + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
> + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
> + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER 
> DEALINGS
> + * IN THE SOFTWARE.
> + *
> + * Authors:
> + *Robert Foss 
> + */
> +
> +#ifndef ANDROID
> +#define _GNU_SOURCE
> +#else
> +#include 
> +#endif
> +#include 
> +#include 
> +#include 
> +#include 
> +#include 
> +
> +#include "sw_sync.h"
> +#include "drmtest.h"
> +#include "ioctl_wrappers.h"
> +
> +#ifndef SW_SYNC_IOC_INC
> +struct sw_sync_create_fence_data {
> + __u32   value;
> + charname[32];
> + __s32   fence;
> +};
> +
> +#define SW_SYNC_IOC_MAGIC'W'
> +#define SW_SYNC_IOC_CREATE_FENCE _IOWR(SW_SYNC_IOC_MAGIC, 0,\
> + struct 
> sw_sync_create_fence_data)
> +#define SW_SYNC_IOC_INC  _IOW(SW_SYNC_IOC_MAGIC, 1, 
> __u32)
> +#endif
> +
> +#define DEVFS_SW_SYNC   "/dev/sw_sync"
> +#define DEBUGFS_SW_SYNC "/sys/kernel/debug/sync/sw_sync"
> +
> +
> +int sw_sync_fd_is_valid(int fd)
> +{
> + int status;
> +
> + if (fd < 0)
> + return 0;
> +
> + status = fcntl(fd, F_GETFD, 0);
> + return status >= 0;
> +}
> +
> +static
> +void sw_sync_fd_close(int fd)
> +{
> + if (!sw_sync_fd_is_valid(fd))
> + return;
> +
> + close(fd);

What bug are you trying to prevent? This is the same as just calling
close(fd) and seeing EINVAL.

> +}
> +
> +int sw_sync_timeline_create(void)
> +{
> + int fd = open(DEVFS_SW_SYNC, O_RDWR);

if (fd < 0)
fd = open(DEBUGFS_SW_SYNC, O_RDWR);

igt_require(fd >= 0);

> +
> + fd = open(DEBUGFS_SW_SYNC, O_RDWR);
> +
> + igt_assert(sw_sync_fd_is_valid(fd));
> +
> + return fd;
> +}
> +
> +void sw_sync_timeline_destroy(int fd)
> +{
> + return sw_sync_fd_close(fd);

Now just close(fd); and might as well not obfuscate that these are plain
fd.

> +int sw_sync_fence_create(int fd, int32_t seqno)
> +{
> + struct sw_sync_create_fence_data data = {};
> + data.value = seqno;
> +
> + if (fd >= 0) {
> + do_ioctl(fd, SW_SYNC_IOC_CREATE_FENCE, );
> + return data.fence;
> + } else {
> + do_ioctl_err(fd, SW_SYNC_IOC_CREATE_FENCE, , EBADF);
> + return -1;

Try:

int __sw_sync_fence_create(int fd, int32_t seqno) /* int32_t not unsigned ? */
{

struct sw_sync_create_fence_data data;

memset(,