linux_dsm_epyc7002/drivers/gpu/drm/msm/msm_perf.c
Linus Torvalds be8454afc5 drm main pull request for v5.3-rc1 (sans mm changes)
-----BEGIN PGP SIGNATURE-----
 
 iQIcBAABAgAGBQJdLMSbAAoJEAx081l5xIa+udkP/iWr8mw44tWYb8Wuzc/aR91v
 02X/J4S9XTQttNn/1Gpq9ItTLMf0Gc08tk1wEBBHAWi/qGaGZS2al+rv0afeuuQa
 aFhQzioDi7K/YZt92iEJhdx7wVMyydICTg3INmYlSP7/FyzLp6gBQRGSJ1kX5mHZ
 qWsFZgUOH9V5evyB6fDMleDaqFOKfcwrD7XYwbOheL/HeYQSv5AYn3VBupBFQ76L
 0hclI5VzZQ5V0nnqRTNDQVA9Yl6NTl+2eXTn5vuBtwKXEI6JJw8eihZp2oZDXqfS
 L441w7wGbkRPzN5kjMZjs1ToPMTlMveR5kL6Sc+o3DT/HmIr1odeaSDXR/93UOLd
 z0CRJ6xMC8h1ThLNHp8UgbxCKqIwYPsY2wVqjsJt7lDY5jma7Yv2YJ9ocYGHN/sO
 DVHcU6ugbwvuC5wZZtVZl5J4hjnBZwNRGSVK+iM0tkjalgdEuSFehXT7eQ8SphF/
 yI5gD1xNEwGfZ4bvZ3u/QrDCcpUAgPIUYmxEa2tPJILQWOJ9O87yc0y9Z21k9Ef1
 9yDqrFV3sPqC2xj/0ufZG/18+Yt99Ykg1jQE3RGDwD/59KAeqPbOvqTKyVODV9jE
 qje6ScSIc2G0713uss2bcaD3k+rCB5YL2JkKrk5OWW/T2+n9T+JFaiNh7dnSFFcU
 gBKyeY24OyCDMwXrby0K
 =SI+Y
 -----END PGP SIGNATURE-----

Merge tag 'drm-next-2019-07-16' of git://anongit.freedesktop.org/drm/drm

Pull drm updates from Dave Airlie:
 "The biggest thing in this is the AMD Navi GPU support, this again
  contains a bunch of header files that are large. These are the new AMD
  RX5700 GPUs that just recently became available.

  New drivers:
   - ST-Ericsson MCDE driver
   - Ingenic JZ47xx SoC

  UAPI change:
   - HDR source metadata property

  Core:
   - HDR inforframes and EDID parsing
   - drm hdmi infoframe unpacking
   - remove prime sg_table caching into dma-buf
   - New gem vram helpers to reduce driver code
   - Lots of drmP.h removal
   - reservation fencing fix
   - documentation updates
   - drm_fb_helper_connector removed
   - mode name command handler rewrite

  fbcon:
   - Remove the fbcon notifiers

  ttm:
   - forward progress fixes

  dma-buf:
   - make mmap call optional
   - debugfs refcount fixes
   - dma-fence free with pending signals fix
   - each dma-buf gets an inode

  Panels:
   - Lots of additional panel bindings

  amdgpu:
   - initial navi10 support
   - avoid hw reset
   - HDR metadata support
   - new thermal sensors for vega asics
   - RAS fixes
   - use HMM rather than MMU notifier
   - xgmi topology via kfd
   - SR-IOV fixes
   - driver reload fixes
   - DC use a core bpc attribute
   - Aux fixes for DC
   - Bandwidth calc updates for DC
   - Clock handling refactor
   - kfd VEGAM support

  vmwgfx:
   - Coherent memory support changes

  i915:
   - HDR Support
   - HDMI i2c link
   - Icelake multi-segmented gamma support
   - GuC firmware update
   - Mule Creek Canyon PCH support for EHL
   - EHL platform updtes
   - move i915.alpha_support to i915.force_probe
   - runtime PM refactoring
   - VBT parsing refactoring
   - DSI fixes
   - struct mutex dependency reduction
   - GEM code reorg

  mali-dp:
   - Komeda driver features

  msm:
   - dsi vs EPROBE_DEFER fixes
   - msm8998 snapdragon 835 support
   - a540 gpu support
   - mdp5 and dpu interconnect support

  exynos:
   - drmP.h removal

  tegra:
   - misc fixes

  tda998x:
   - audio support improvements
   - pixel repeated mode support
   - quantisation range handling corrections
   - HDMI vendor info fix

  armada:
   - interlace support fix
   - overlay/video plane register handling refactor
   - add gamma support

  rockchip:
   - RX3328 support

  panfrost:
   - expose perf counters via hidden ioctls

  vkms:
   - enumerate CRC sources list

  ast:
   - rework BO handling

  mgag200:
   - rework BO handling

  dw-hdmi:
   - suspend/resume support

  rcar-du:
   - R8A774A1 Soc Support
   - LVDS dual-link mode support
   - Additional formats
   - Misc fixes

  omapdrm:
   - DSI command mode display support

  stm
   - fb modifier support
   - runtime PM support

  sun4i:
   - use vmap ops

  vc4:
   - binner bo binding rework

  v3d:
   - compute shader support
   - resync/sync fixes
   - job management refactoring

  lima:
   - NULL pointer in irq handler fix
   - scheduler default timeout

  virtio:
   - fence seqno support
   - trace events

  bochs:
   - misc fixes

  tc458767:
   - IRQ/HDP handling

  sii902x:
   - HDMI audio support

  atmel-hlcdc:
   - misc fixes

  meson:
   - zpos support"

* tag 'drm-next-2019-07-16' of git://anongit.freedesktop.org/drm/drm: (1815 commits)
  Revert "Merge branch 'vmwgfx-next' of git://people.freedesktop.org/~thomash/linux into drm-next"
  Revert "mm: adjust apply_to_pfn_range interface for dropped token."
  mm: adjust apply_to_pfn_range interface for dropped token.
  drm/amdgpu/navi10: add uclk activity sensor
  drm/amdgpu: properly guard the generic discovery code
  drm/amdgpu: add missing documentation on new module parameters
  drm/amdgpu: don't invalidate caches in RELEASE_MEM, only do the writeback
  drm/amd/display: avoid 64-bit division
  drm/amdgpu/psp11: simplify the ucode register logic
  drm/amdgpu: properly guard DC support in navi code
  drm/amd/powerplay: vega20: fix uninitialized variable use
  drm/amd/display: dcn20: include linux/delay.h
  amdgpu: make pmu support optional
  drm/amd/powerplay: Zero initialize current_rpm in vega20_get_fan_speed_percent
  drm/amd/powerplay: Zero initialize freq in smu_v11_0_get_current_clk_freq
  drm/amd/powerplay: Use memset to initialize metrics structs
  drm/amdgpu/mes10.1: Fix header guard
  drm/amd/powerplay: add temperature sensor support for navi10
  drm/amdgpu: fix scheduler timeout calc
  drm/amdgpu: Prepare for hmm_range_register API change (v2)
  ...
2019-07-15 19:04:27 -07:00

231 lines
4.5 KiB
C

// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright (C) 2013 Red Hat
* Author: Rob Clark <robdclark@gmail.com>
*/
/* For profiling, userspace can:
*
* tail -f /sys/kernel/debug/dri/<minor>/gpu
*
* This will enable performance counters/profiling to track the busy time
* and any gpu specific performance counters that are supported.
*/
#ifdef CONFIG_DEBUG_FS
#include <linux/debugfs.h>
#include "msm_drv.h"
#include "msm_gpu.h"
struct msm_perf_state {
struct drm_device *dev;
bool open;
int cnt;
struct mutex read_lock;
char buf[256];
int buftot, bufpos;
unsigned long next_jiffies;
};
#define SAMPLE_TIME (HZ/4)
/* wait for next sample time: */
static int wait_sample(struct msm_perf_state *perf)
{
unsigned long start_jiffies = jiffies;
if (time_after(perf->next_jiffies, start_jiffies)) {
unsigned long remaining_jiffies =
perf->next_jiffies - start_jiffies;
int ret = schedule_timeout_interruptible(remaining_jiffies);
if (ret > 0) {
/* interrupted */
return -ERESTARTSYS;
}
}
perf->next_jiffies += SAMPLE_TIME;
return 0;
}
static int refill_buf(struct msm_perf_state *perf)
{
struct msm_drm_private *priv = perf->dev->dev_private;
struct msm_gpu *gpu = priv->gpu;
char *ptr = perf->buf;
int rem = sizeof(perf->buf);
int i, n;
if ((perf->cnt++ % 32) == 0) {
/* Header line: */
n = snprintf(ptr, rem, "%%BUSY");
ptr += n;
rem -= n;
for (i = 0; i < gpu->num_perfcntrs; i++) {
const struct msm_gpu_perfcntr *perfcntr = &gpu->perfcntrs[i];
n = snprintf(ptr, rem, "\t%s", perfcntr->name);
ptr += n;
rem -= n;
}
} else {
/* Sample line: */
uint32_t activetime = 0, totaltime = 0;
uint32_t cntrs[5];
uint32_t val;
int ret;
/* sleep until next sample time: */
ret = wait_sample(perf);
if (ret)
return ret;
ret = msm_gpu_perfcntr_sample(gpu, &activetime, &totaltime,
ARRAY_SIZE(cntrs), cntrs);
if (ret < 0)
return ret;
val = totaltime ? 1000 * activetime / totaltime : 0;
n = snprintf(ptr, rem, "%3d.%d%%", val / 10, val % 10);
ptr += n;
rem -= n;
for (i = 0; i < ret; i++) {
/* cycle counters (I think).. convert to MHz.. */
val = cntrs[i] / 10000;
n = snprintf(ptr, rem, "\t%5d.%02d",
val / 100, val % 100);
ptr += n;
rem -= n;
}
}
n = snprintf(ptr, rem, "\n");
ptr += n;
rem -= n;
perf->bufpos = 0;
perf->buftot = ptr - perf->buf;
return 0;
}
static ssize_t perf_read(struct file *file, char __user *buf,
size_t sz, loff_t *ppos)
{
struct msm_perf_state *perf = file->private_data;
int n = 0, ret = 0;
mutex_lock(&perf->read_lock);
if (perf->bufpos >= perf->buftot) {
ret = refill_buf(perf);
if (ret)
goto out;
}
n = min((int)sz, perf->buftot - perf->bufpos);
if (copy_to_user(buf, &perf->buf[perf->bufpos], n)) {
ret = -EFAULT;
goto out;
}
perf->bufpos += n;
*ppos += n;
out:
mutex_unlock(&perf->read_lock);
if (ret)
return ret;
return n;
}
static int perf_open(struct inode *inode, struct file *file)
{
struct msm_perf_state *perf = inode->i_private;
struct drm_device *dev = perf->dev;
struct msm_drm_private *priv = dev->dev_private;
struct msm_gpu *gpu = priv->gpu;
int ret = 0;
mutex_lock(&dev->struct_mutex);
if (perf->open || !gpu) {
ret = -EBUSY;
goto out;
}
file->private_data = perf;
perf->open = true;
perf->cnt = 0;
perf->buftot = 0;
perf->bufpos = 0;
msm_gpu_perfcntr_start(gpu);
perf->next_jiffies = jiffies + SAMPLE_TIME;
out:
mutex_unlock(&dev->struct_mutex);
return ret;
}
static int perf_release(struct inode *inode, struct file *file)
{
struct msm_perf_state *perf = inode->i_private;
struct msm_drm_private *priv = perf->dev->dev_private;
msm_gpu_perfcntr_stop(priv->gpu);
perf->open = false;
return 0;
}
static const struct file_operations perf_debugfs_fops = {
.owner = THIS_MODULE,
.open = perf_open,
.read = perf_read,
.llseek = no_llseek,
.release = perf_release,
};
int msm_perf_debugfs_init(struct drm_minor *minor)
{
struct msm_drm_private *priv = minor->dev->dev_private;
struct msm_perf_state *perf;
/* only create on first minor: */
if (priv->perf)
return 0;
perf = kzalloc(sizeof(*perf), GFP_KERNEL);
if (!perf)
return -ENOMEM;
perf->dev = minor->dev;
mutex_init(&perf->read_lock);
priv->perf = perf;
debugfs_create_file("perf", S_IFREG | S_IRUGO, minor->debugfs_root,
perf, &perf_debugfs_fops);
return 0;
}
void msm_perf_debugfs_cleanup(struct msm_drm_private *priv)
{
struct msm_perf_state *perf = priv->perf;
if (!perf)
return;
priv->perf = NULL;
mutex_destroy(&perf->read_lock);
kfree(perf);
}
#endif