From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by gabe.freedesktop.org (Postfix) with ESMTPS id AA59C10E821 for ; Tue, 14 Mar 2023 14:31:30 +0000 (UTC) Message-ID: Date: Tue, 14 Mar 2023 20:00:45 +0530 References: <20230314105452.2169208-1-bhanuprakash.modem@intel.com> <20230314105452.2169208-5-bhanuprakash.modem@intel.com> <20230314130805.7276dcd9@maurocar-mobl2> <20230314122019.nmjrvhx7xloqdcky@zkempczy-mobl2> Content-Language: en-US From: "Modem, Bhanuprakash" In-Reply-To: <20230314122019.nmjrvhx7xloqdcky@zkempczy-mobl2> Content-Type: text/plain; charset="utf-8"; format="flowed" Content-Transfer-Encoding: 8bit MIME-Version: 1.0 Subject: Re: [igt-dev] [i-g-t 4/5] lib/intel_chipset: Add support to XE driver to get devid List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: igt-dev-bounces@lists.freedesktop.org Sender: "igt-dev" To: =?UTF-8?Q?Zbigniew_Kempczy=c5=84ski?= , Mauro Carvalho Chehab Cc: igt-dev@lists.freedesktop.org List-ID: On Tue-14-03-2023 05:50 pm, Zbigniew Kempczyński wrote: > On Tue, Mar 14, 2023 at 01:08:05PM +0100, Mauro Carvalho Chehab wrote: >> On Tue, 14 Mar 2023 16:24:51 +0530 >> Bhanuprakash Modem wrote: >> >>> As many tests are using intel devid, add a helper to get it for >>> XE driver. >>> >>> Signed-off-by: Bhanuprakash Modem >>> --- >>> lib/intel_chipset.c | 30 ++++++++++++++++++++---------- >>> 1 file changed, 20 insertions(+), 10 deletions(-) >>> >>> diff --git a/lib/intel_chipset.c b/lib/intel_chipset.c >>> index efb6f1771..73dbd813d 100644 >>> --- a/lib/intel_chipset.c >>> +++ b/lib/intel_chipset.c >>> @@ -41,6 +41,7 @@ >>> #include "drmtest.h" >>> #include "intel_chipset.h" >>> #include "igt_core.h" >>> +#include "xe/xe_query.h" >>> >>> /** >>> * SECTION:intel_chipset >>> @@ -112,9 +113,22 @@ intel_get_pci_device(void) >>> return pci_dev; >>> } >>> >>> +static uint32_t __i915_get_drm_devid(int fd) >>> +{ >>> + struct drm_i915_getparam gp; >>> + int devid = 0; >>> + >>> + 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; >>> +} >>> + >> >> Hmm... Shouldn't it be cached? Anyway: >> >> Reviewed-by: Mauro Carvalho Chehab >> >> >>> /** >>> * intel_get_drm_devid: >>> - * @fd: open i915 drm file descriptor >>> + * @fd: open i915/xe drm file descriptor >>> * >>> * Queries the kernel for the pci device id corresponding to the drm file >>> * descriptor. >>> @@ -125,22 +139,18 @@ intel_get_pci_device(void) >>> 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(is_intel_device(fd)); >>> >>> override = getenv("INTEL_DEVID_OVERRIDE"); >>> if (override) >>> return strtol(override, NULL, 0); >>> >>> - 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; >>> + if (is_i915_device(fd)) >>> + return __i915_get_drm_devid(fd); >>> + else >>> + return xe_dev_id(fd); > > This call is problematic imo, it strongly depends on previously > calling xe_device_get(). As this library part I would explicitly > call config query ioctl() to get device id to avoid this dependency. Every KMS test is supposed to call xe_device_get() before starting the execution (Please check patch [5/5] in this series). Still, having a separate helper to get the dev_id is a good option to me. I'll float a new rev by adding this. - Bhanu > > -- > Zbigniew > > >>> } >>> >>> /**