* [PATCH i-g-t] lib: Stop caching __drm_device_id
@ 2018-08-31 12:18 Chris Wilson
2018-08-31 14:20 ` [igt-dev] " Rodrigo Vivi
0 siblings, 1 reply; 2+ messages in thread
From: Chris Wilson @ 2018-08-31 12:18 UTC (permalink / raw)
To: igt-dev; +Cc: intel-gfx
In a multi-device system there is no guarantee that the fd being probed
in intel_get_drm_devid() is the same as was opened earlier. Any cache
may outlive the fd, so is frought with lifetime issues. The primary
reason for caching the devid was to avoid extra ioctls in the
dmesg/strace, but hopefully all users now grab the id in their fixture
and not inside every function.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Katarzyna Dec <katarzyna.dec@intel.com>
---
lib/drmtest.c | 3 ---
lib/intel_chipset.c | 14 +++++++++-----
2 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/lib/drmtest.c b/lib/drmtest.c
index fae6f86f2..ecb535f5d 100644
--- a/lib/drmtest.c
+++ b/lib/drmtest.c
@@ -75,8 +75,6 @@
* and [batchbuffer](igt-gpu-tools-intel-batchbuffer.html) libraries as dependencies.
*/
-uint16_t __drm_device_id;
-
static int __get_drm_device_name(int fd, char *name)
{
drm_version_t version;
@@ -142,7 +140,6 @@ static bool has_known_intel_chipset(int fd)
if (!intel_gen(devid))
return false;
- __drm_device_id = devid;
return true;
}
diff --git a/lib/intel_chipset.c b/lib/intel_chipset.c
index ab35fa70c..4748a3fb8 100644
--- a/lib/intel_chipset.c
+++ b/lib/intel_chipset.c
@@ -112,8 +112,6 @@ intel_get_pci_device(void)
return pci_dev;
}
-extern uint16_t __drm_device_id;
-
/**
* intel_get_drm_devid:
* @fd: open i915 drm file descriptor
@@ -127,16 +125,22 @@ extern uint16_t __drm_device_id;
uint32_t
intel_get_drm_devid(int fd)
{
+ struct drm_i915_getparam gp;
const char *override;
+ int devid = 0;
igt_assert(is_i915_device(fd));
- igt_assert(__drm_device_id);
override = getenv("INTEL_DEVID_OVERRIDE");
if (override)
return strtol(override, NULL, 0);
- else
- return __drm_device_id;
+
+ memset(&gp, 0, sizeof(gp));
+ gp.param = I915_PARAM_CHIPSET_ID;
+ gp.value = &devid;
+ ioctl(fd, DRM_IOCTL_I915_GETPARAM, &gp, sizeof(gp));
+
+ return devid;
}
/**
--
2.19.0.rc1
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply related [flat|nested] 2+ messages in thread* Re: [igt-dev] [PATCH i-g-t] lib: Stop caching __drm_device_id
2018-08-31 12:18 [PATCH i-g-t] lib: Stop caching __drm_device_id Chris Wilson
@ 2018-08-31 14:20 ` Rodrigo Vivi
0 siblings, 0 replies; 2+ messages in thread
From: Rodrigo Vivi @ 2018-08-31 14:20 UTC (permalink / raw)
To: Chris Wilson; +Cc: igt-dev, intel-gfx
On Fri, Aug 31, 2018 at 01:18:53PM +0100, Chris Wilson wrote:
> In a multi-device system there is no guarantee that the fd being probed
> in intel_get_drm_devid() is the same as was opened earlier. Any cache
> may outlive the fd, so is frought with lifetime issues. The primary
> reason for caching the devid was to avoid extra ioctls in the
> dmesg/strace, but hopefully all users now grab the id in their fixture
> and not inside every function.
>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Katarzyna Dec <katarzyna.dec@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> ---
> lib/drmtest.c | 3 ---
> lib/intel_chipset.c | 14 +++++++++-----
> 2 files changed, 9 insertions(+), 8 deletions(-)
>
> diff --git a/lib/drmtest.c b/lib/drmtest.c
> index fae6f86f2..ecb535f5d 100644
> --- a/lib/drmtest.c
> +++ b/lib/drmtest.c
> @@ -75,8 +75,6 @@
> * and [batchbuffer](igt-gpu-tools-intel-batchbuffer.html) libraries as dependencies.
> */
>
> -uint16_t __drm_device_id;
> -
> static int __get_drm_device_name(int fd, char *name)
> {
> drm_version_t version;
> @@ -142,7 +140,6 @@ static bool has_known_intel_chipset(int fd)
> if (!intel_gen(devid))
> return false;
>
> - __drm_device_id = devid;
> return true;
> }
>
> diff --git a/lib/intel_chipset.c b/lib/intel_chipset.c
> index ab35fa70c..4748a3fb8 100644
> --- a/lib/intel_chipset.c
> +++ b/lib/intel_chipset.c
> @@ -112,8 +112,6 @@ intel_get_pci_device(void)
> return pci_dev;
> }
>
> -extern uint16_t __drm_device_id;
> -
> /**
> * intel_get_drm_devid:
> * @fd: open i915 drm file descriptor
> @@ -127,16 +125,22 @@ extern uint16_t __drm_device_id;
> uint32_t
> intel_get_drm_devid(int fd)
> {
> + struct drm_i915_getparam gp;
> const char *override;
> + int devid = 0;
>
> igt_assert(is_i915_device(fd));
> - igt_assert(__drm_device_id);
>
> override = getenv("INTEL_DEVID_OVERRIDE");
> if (override)
> return strtol(override, NULL, 0);
> - else
> - return __drm_device_id;
> +
> + memset(&gp, 0, sizeof(gp));
> + gp.param = I915_PARAM_CHIPSET_ID;
> + gp.value = &devid;
> + ioctl(fd, DRM_IOCTL_I915_GETPARAM, &gp, sizeof(gp));
> +
> + return devid;
> }
>
> /**
> --
> 2.19.0.rc1
>
> _______________________________________________
> igt-dev mailing list
> igt-dev@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/igt-dev
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2018-08-31 14:20 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2018-08-31 12:18 [PATCH i-g-t] lib: Stop caching __drm_device_id Chris Wilson
2018-08-31 14:20 ` [igt-dev] " Rodrigo Vivi
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox