From: Jerome Glisse <j.glisse@gmail.com>
To: Oded Gabbay <oded.gabbay@gmail.com>
Cc: Andrew Lewycky <Andrew.Lewycky@amd.com>,
Ben Goz <ben.goz@amd.com>,
linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org,
Evgeny Pinchuk <evgeny.pinchuk@amd.com>,
Alexey Skidanov <Alexey.Skidanov@amd.com>,
linux-api@vger.kernel.org,
Alex Deucher <alexander.deucher@amd.com>
Subject: Re: [PATCH 32/83] hsa/radeon: implementing IOCTL for clock counters
Date: Fri, 11 Jul 2014 16:34:56 -0400 [thread overview]
Message-ID: <20140711203455.GR1870@gmail.com> (raw)
In-Reply-To: <1405029279-6894-4-git-send-email-oded.gabbay@amd.com>
On Fri, Jul 11, 2014 at 12:53:48AM +0300, Oded Gabbay wrote:
> From: Evgeny Pinchuk <evgeny.pinchuk@amd.com>
>
> Implemented new IOCTL to query the CPU and GPU clock counters.
>
> Signed-off-by: Evgeny Pinchuk <evgeny.pinchuk@amd.com>
> Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
> ---
> drivers/gpu/hsa/radeon/kfd_chardev.c | 37 ++++++++++++++++++++++++++++++++++++
> include/uapi/linux/kfd_ioctl.h | 9 +++++++++
> 2 files changed, 46 insertions(+)
>
> diff --git a/drivers/gpu/hsa/radeon/kfd_chardev.c b/drivers/gpu/hsa/radeon/kfd_chardev.c
> index ddaf357..d6fa980 100644
> --- a/drivers/gpu/hsa/radeon/kfd_chardev.c
> +++ b/drivers/gpu/hsa/radeon/kfd_chardev.c
> @@ -28,6 +28,7 @@
> #include <linux/slab.h>
> #include <linux/uaccess.h>
> #include <uapi/linux/kfd_ioctl.h>
> +#include <linux/time.h>
> #include "kfd_priv.h"
> #include "kfd_scheduler.h"
>
> @@ -284,6 +285,38 @@ out:
> return err;
> }
>
> +static long
> +kfd_ioctl_get_clock_counters(struct file *filep, struct kfd_process *p, void __user *arg)
> +{
> + struct kfd_ioctl_get_clock_counters_args args;
> + struct kfd_dev *dev;
> + struct timespec time;
> +
> + if (copy_from_user(&args, arg, sizeof(args)))
> + return -EFAULT;
> +
> + dev = radeon_kfd_device_by_id(args.gpu_id);
> + if (dev == NULL)
> + return -EINVAL;
> +
> + /* Reading GPU clock counter from KGD */
> + args.gpu_clock_counter = kfd2kgd->get_gpu_clock_counter(dev->kgd);
> +
> + /* No access to rdtsc. Using raw monotonic time */
> + getrawmonotonic(&time);
> + args.cpu_clock_counter = time.tv_nsec;
Is the GPU clock counter monotonic too ? Even after GPU reset (hard reset
included) what could go wrong if it rolls back ?
> +
> + get_monotonic_boottime(&time);
> + args.system_clock_counter = time.tv_nsec;
> +
> + /* Since the counter is in nano-seconds we use 1GHz frequency */
> + args.system_clock_freq = 1000000000;
> +
> + if (copy_to_user(arg, &args, sizeof(args)))
> + return -EFAULT;
> +
> + return 0;
> +}
>
> static long
> kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
> @@ -312,6 +345,10 @@ kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
> err = kfd_ioctl_set_memory_policy(filep, process, (void __user *)arg);
> break;
>
> + case KFD_IOC_GET_CLOCK_COUNTERS:
> + err = kfd_ioctl_get_clock_counters(filep, process, (void __user *)arg);
> + break;
> +
> default:
> dev_err(kfd_device,
> "unknown ioctl cmd 0x%x, arg 0x%lx)\n",
> diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
> index 928e628..5b9517e 100644
> --- a/include/uapi/linux/kfd_ioctl.h
> +++ b/include/uapi/linux/kfd_ioctl.h
> @@ -70,12 +70,21 @@ struct kfd_ioctl_set_memory_policy_args {
> uint64_t alternate_aperture_size; /* to KFD */
> };
>
> +struct kfd_ioctl_get_clock_counters_args {
> + uint32_t gpu_id; /* to KFD */
> + uint64_t gpu_clock_counter; /* from KFD */
> + uint64_t cpu_clock_counter; /* from KFD */
> + uint64_t system_clock_counter; /* from KFD */
> + uint64_t system_clock_freq; /* from KFD */
> +};
> +
> #define KFD_IOC_MAGIC 'K'
>
> #define KFD_IOC_GET_VERSION _IOR(KFD_IOC_MAGIC, 1, struct kfd_ioctl_get_version_args)
> #define KFD_IOC_CREATE_QUEUE _IOWR(KFD_IOC_MAGIC, 2, struct kfd_ioctl_create_queue_args)
> #define KFD_IOC_DESTROY_QUEUE _IOWR(KFD_IOC_MAGIC, 3, struct kfd_ioctl_destroy_queue_args)
> #define KFD_IOC_SET_MEMORY_POLICY _IOW(KFD_IOC_MAGIC, 4, struct kfd_ioctl_set_memory_policy_args)
> +#define KFD_IOC_GET_CLOCK_COUNTERS _IOWR(KFD_IOC_MAGIC, 5, struct kfd_ioctl_get_clock_counters_args)
>
> #pragma pack(pop)
>
> --
> 1.9.1
>
WARNING: multiple messages have this Message-ID (diff)
From: Jerome Glisse <j.glisse@gmail.com>
To: Oded Gabbay <oded.gabbay@gmail.com>
Cc: David Airlie <airlied@linux.ie>,
Alex Deucher <alexander.deucher@amd.com>,
linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org,
John Bridgman <John.Bridgman@amd.com>,
Andrew Lewycky <Andrew.Lewycky@amd.com>,
Joerg Roedel <joro@8bytes.org>,
Evgeny Pinchuk <evgeny.pinchuk@amd.com>,
Oded Gabbay <oded.gabbay@amd.com>, Ben Goz <ben.goz@amd.com>,
Alexey Skidanov <Alexey.Skidanov@amd.com>,
linux-api@vger.kernel.org
Subject: Re: [PATCH 32/83] hsa/radeon: implementing IOCTL for clock counters
Date: Fri, 11 Jul 2014 16:34:56 -0400 [thread overview]
Message-ID: <20140711203455.GR1870@gmail.com> (raw)
In-Reply-To: <1405029279-6894-4-git-send-email-oded.gabbay@amd.com>
On Fri, Jul 11, 2014 at 12:53:48AM +0300, Oded Gabbay wrote:
> From: Evgeny Pinchuk <evgeny.pinchuk@amd.com>
>
> Implemented new IOCTL to query the CPU and GPU clock counters.
>
> Signed-off-by: Evgeny Pinchuk <evgeny.pinchuk@amd.com>
> Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
> ---
> drivers/gpu/hsa/radeon/kfd_chardev.c | 37 ++++++++++++++++++++++++++++++++++++
> include/uapi/linux/kfd_ioctl.h | 9 +++++++++
> 2 files changed, 46 insertions(+)
>
> diff --git a/drivers/gpu/hsa/radeon/kfd_chardev.c b/drivers/gpu/hsa/radeon/kfd_chardev.c
> index ddaf357..d6fa980 100644
> --- a/drivers/gpu/hsa/radeon/kfd_chardev.c
> +++ b/drivers/gpu/hsa/radeon/kfd_chardev.c
> @@ -28,6 +28,7 @@
> #include <linux/slab.h>
> #include <linux/uaccess.h>
> #include <uapi/linux/kfd_ioctl.h>
> +#include <linux/time.h>
> #include "kfd_priv.h"
> #include "kfd_scheduler.h"
>
> @@ -284,6 +285,38 @@ out:
> return err;
> }
>
> +static long
> +kfd_ioctl_get_clock_counters(struct file *filep, struct kfd_process *p, void __user *arg)
> +{
> + struct kfd_ioctl_get_clock_counters_args args;
> + struct kfd_dev *dev;
> + struct timespec time;
> +
> + if (copy_from_user(&args, arg, sizeof(args)))
> + return -EFAULT;
> +
> + dev = radeon_kfd_device_by_id(args.gpu_id);
> + if (dev == NULL)
> + return -EINVAL;
> +
> + /* Reading GPU clock counter from KGD */
> + args.gpu_clock_counter = kfd2kgd->get_gpu_clock_counter(dev->kgd);
> +
> + /* No access to rdtsc. Using raw monotonic time */
> + getrawmonotonic(&time);
> + args.cpu_clock_counter = time.tv_nsec;
Is the GPU clock counter monotonic too ? Even after GPU reset (hard reset
included) what could go wrong if it rolls back ?
> +
> + get_monotonic_boottime(&time);
> + args.system_clock_counter = time.tv_nsec;
> +
> + /* Since the counter is in nano-seconds we use 1GHz frequency */
> + args.system_clock_freq = 1000000000;
> +
> + if (copy_to_user(arg, &args, sizeof(args)))
> + return -EFAULT;
> +
> + return 0;
> +}
>
> static long
> kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
> @@ -312,6 +345,10 @@ kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
> err = kfd_ioctl_set_memory_policy(filep, process, (void __user *)arg);
> break;
>
> + case KFD_IOC_GET_CLOCK_COUNTERS:
> + err = kfd_ioctl_get_clock_counters(filep, process, (void __user *)arg);
> + break;
> +
> default:
> dev_err(kfd_device,
> "unknown ioctl cmd 0x%x, arg 0x%lx)\n",
> diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
> index 928e628..5b9517e 100644
> --- a/include/uapi/linux/kfd_ioctl.h
> +++ b/include/uapi/linux/kfd_ioctl.h
> @@ -70,12 +70,21 @@ struct kfd_ioctl_set_memory_policy_args {
> uint64_t alternate_aperture_size; /* to KFD */
> };
>
> +struct kfd_ioctl_get_clock_counters_args {
> + uint32_t gpu_id; /* to KFD */
> + uint64_t gpu_clock_counter; /* from KFD */
> + uint64_t cpu_clock_counter; /* from KFD */
> + uint64_t system_clock_counter; /* from KFD */
> + uint64_t system_clock_freq; /* from KFD */
> +};
> +
> #define KFD_IOC_MAGIC 'K'
>
> #define KFD_IOC_GET_VERSION _IOR(KFD_IOC_MAGIC, 1, struct kfd_ioctl_get_version_args)
> #define KFD_IOC_CREATE_QUEUE _IOWR(KFD_IOC_MAGIC, 2, struct kfd_ioctl_create_queue_args)
> #define KFD_IOC_DESTROY_QUEUE _IOWR(KFD_IOC_MAGIC, 3, struct kfd_ioctl_destroy_queue_args)
> #define KFD_IOC_SET_MEMORY_POLICY _IOW(KFD_IOC_MAGIC, 4, struct kfd_ioctl_set_memory_policy_args)
> +#define KFD_IOC_GET_CLOCK_COUNTERS _IOWR(KFD_IOC_MAGIC, 5, struct kfd_ioctl_get_clock_counters_args)
>
> #pragma pack(pop)
>
> --
> 1.9.1
>
next prev parent reply other threads:[~2014-07-11 20:34 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-07-10 21:53 [PATCH 29/83] hsa/radeon: Fix memory size allocated for HPD Oded Gabbay
2014-07-10 21:53 ` [PATCH 30/83] hsa/radeon: Fix list of supported devices Oded Gabbay
2014-07-10 21:53 ` [PATCH 31/83] drm/radeon: extending kfd-kgd interface Oded Gabbay
[not found] ` <1405029279-6894-1-git-send-email-oded.gabbay-5C7GfCeVMHo@public.gmane.org>
2014-07-10 21:53 ` [PATCH 32/83] hsa/radeon: implementing IOCTL for clock counters Oded Gabbay
2014-07-10 21:53 ` Oded Gabbay
2014-07-11 20:34 ` Jerome Glisse [this message]
2014-07-11 20:34 ` Jerome Glisse
2014-07-10 21:53 ` [PATCH 33/83] hsa/radeon: Fix coding style in cik_int.h Oded Gabbay
2014-07-10 21:53 ` [PATCH 34/83] drm/radeon: adding synchronization for GRBM GFX Oded Gabbay
2014-07-10 21:53 ` [PATCH 35/83] hsa/radeon: Print ioctl commnad only in debug mode Oded Gabbay
2014-07-10 21:53 ` [PATCH 36/83] hsa/radeon: fixing clock counters bug Oded Gabbay
2014-07-10 21:53 ` [PATCH 37/83] hsa/radeon: Print ISR info only in debug mode Oded Gabbay
2014-07-10 21:53 ` [PATCH 38/83] hsa/radeon: Workaround for a bug in amd_iommu Oded Gabbay
2014-07-11 8:42 ` Joerg Roedel
2014-07-11 8:42 ` Joerg Roedel
2014-07-10 21:53 ` [PATCH 39/83] drm/radeon: Extending kfd interface Oded Gabbay
2014-07-10 21:53 ` [PATCH 40/83] hsa/radeon: Adding max clock speeds to topology Oded Gabbay
2014-07-10 21:53 ` [PATCH 41/83] hsa/radeon: Alternating the source of max clock Oded Gabbay
2014-07-10 21:53 ` [PATCH 42/83] hsa/radeon: 32-bit processes support Oded Gabbay
2014-07-10 21:53 ` [PATCH 43/83] hsa/radeon: NULL pointer dereference bug workaround Oded Gabbay
2014-07-10 21:54 ` [PATCH 44/83] hsa/radeon: HSA64/HSA32 modes support Oded Gabbay
[not found] ` <1405029279-6894-16-git-send-email-oded.gabbay-5C7GfCeVMHo@public.gmane.org>
2014-07-11 20:41 ` Jerome Glisse
2014-07-11 20:41 ` Jerome Glisse
2014-07-10 21:54 ` [PATCH 45/83] hsa/radeon: debugging print statements Oded Gabbay
2014-07-10 21:54 ` [PATCH 46/83] hsa/radeon: Add queue and hw_pointer_store modules Oded Gabbay
2014-07-10 21:54 ` [PATCH 47/83] hsa/radeon: Add support allocating kernel doorbells Oded Gabbay
2014-07-10 21:54 ` [PATCH 48/83] hsa/radeon: Add mqd_manager module Oded Gabbay
2014-07-10 21:54 ` [PATCH 49/83] hsa/radeon: Add kernel queue support for KFD Oded Gabbay
2014-07-10 21:54 ` [PATCH 50/83] hsa/radeon: Add module parameter of scheduling policy Oded Gabbay
2014-07-10 21:54 ` [PATCH 51/83] hsa/radeon: Add packet manager module Oded Gabbay
2014-07-10 21:54 ` [PATCH 52/83] hsa/radeon: Add process queue " Oded Gabbay
2014-07-10 21:54 ` [PATCH 53/83] hsa/radeon: Add device " Oded Gabbay
2014-07-10 21:54 ` [PATCH 54/83] hsa/radeon: Switch to new queue scheduler Oded Gabbay
2014-07-10 21:54 ` [PATCH 55/83] hsa/radeon: Add IOCTL for update queue Oded Gabbay
2014-07-10 21:54 ` [PATCH 56/83] hsa/radeon: Queue Management integration with Memory Management Oded Gabbay
2014-07-10 21:54 ` [PATCH 57/83] hsa/radeon: Eliminate warnings in compilation Oded Gabbay
2014-07-10 21:54 ` [PATCH 58/83] hsa/radeon: Various kernel styling fixes Oded Gabbay
2014-07-10 21:54 ` [PATCH 59/83] hsa/radeon: Exclusive access for perf. counters Oded Gabbay
2014-07-10 21:54 ` [PATCH 60/83] hsa/radeon: Rearrange structures in kfd_ioctl.h Oded Gabbay
2014-07-10 21:54 ` [PATCH 61/83] hsa/radeon: change another pr_info to pr_debug Oded Gabbay
2014-07-10 21:54 ` [PATCH 62/83] hsa/radeon: Fix timeout calculation in sync_with_hw Oded Gabbay
2014-07-10 21:54 ` [PATCH 63/83] hsa/radeon: Update module information and version Oded Gabbay
2014-07-10 21:54 ` [PATCH 64/83] hsa/radeon: update queue fault handling Oded Gabbay
2014-07-10 21:54 ` [PATCH 65/83] hsa/radeon: fixing a bug to support 32b processes Oded Gabbay
2014-07-10 21:54 ` [PATCH 66/83] hsa/radeon: Fix number of pipes per ME Oded Gabbay
2014-07-10 21:54 ` [PATCH 67/83] hsa/radeon: Removing hw pointer store module Oded Gabbay
2014-07-10 21:54 ` [PATCH 68/83] hsa/radeon: Update module version to 0.6.0 Oded Gabbay
2014-07-10 21:54 ` [PATCH 69/83] hsa/radeon: Fix initialization of sh_mem registers Oded Gabbay
2014-07-10 21:54 ` [PATCH 70/83] hsa/radeon: Fix compilation warnings Oded Gabbay
2014-07-10 21:54 ` [PATCH 71/83] hsa/radeon: Remove old scheduler code Oded Gabbay
2014-07-10 21:54 ` [PATCH 72/83] hsa/radeon: Static analysis (smatch) fixes Oded Gabbay
2014-07-10 21:54 ` [PATCH 73/83] hsa/radeon: Adding qcm fence return status Oded Gabbay
2014-07-10 21:54 ` [PATCH 74/83] hsa/radeon: Adding some error messages Oded Gabbay
2014-07-10 21:54 ` [PATCH 75/83] hsa/radeon: Fixing minor issues with kernel queues (DIQ) Oded Gabbay
2014-07-10 21:54 ` [PATCH 76/83] hsa/radeon: Check oversubscription before destroying runlist Oded Gabbay
2014-07-10 21:54 ` [PATCH 77/83] hsa/radeon: Add local memory to topology Oded Gabbay
2014-07-10 21:54 ` [PATCH 78/83] hsa/radeon: Don't verify cksum when parsing CRAT table Oded Gabbay
2014-07-10 21:54 ` [PATCH 79/83] hsa/radeon: Update module version to 0.6.1 Oded Gabbay
2014-07-10 21:54 ` [PATCH 80/83] drm/radeon: Add register access functions to kfd2kgd interface Oded Gabbay
2014-07-10 21:54 ` [PATCH 81/83] hsa/radeon: Eliminating all direct register accesses Oded Gabbay
2014-07-10 21:54 ` [PATCH 82/83] drm/radeon: Remove lock functions from kfd2kgd interface Oded Gabbay
2014-07-10 21:54 ` [PATCH 83/83] hsa/radeon: Update module version to 0.6.2 Oded Gabbay
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20140711203455.GR1870@gmail.com \
--to=j.glisse@gmail.com \
--cc=Alexey.Skidanov@amd.com \
--cc=Andrew.Lewycky@amd.com \
--cc=alexander.deucher@amd.com \
--cc=ben.goz@amd.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=evgeny.pinchuk@amd.com \
--cc=linux-api@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=oded.gabbay@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.