linux/drivers/gpu/drm/i915/i915_gpu_error.c

1870 lines
47 KiB
C
Raw Normal View History

/*
* Copyright (c) 2008 Intel Corporation
*
* 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:
* Eric Anholt <eric@anholt.net>
* Keith Packard <keithp@keithp.com>
* Mika Kuoppala <mika.kuoppala@intel.com>
*
*/
#include <linux/ascii85.h>
#include <linux/nmi.h>
#include <linux/scatterlist.h>
2016-10-12 09:05:19 +00:00
#include <linux/stop_machine.h>
#include <linux/utsname.h>
#include <linux/zlib.h>
#include <drm/drm_print.h>
#include "i915_gpu_error.h"
#include "i915_drv.h"
static inline const struct intel_engine_cs *
engine_lookup(const struct drm_i915_private *i915, unsigned int id)
{
if (id >= I915_NUM_ENGINES)
return NULL;
return i915->engine[id];
}
static inline const char *
__engine_name(const struct intel_engine_cs *engine)
{
return engine ? engine->name : "";
}
static const char *
engine_name(const struct drm_i915_private *i915, unsigned int id)
{
return __engine_name(engine_lookup(i915, id));
}
static const char *tiling_flag(int tiling)
{
switch (tiling) {
default:
case I915_TILING_NONE: return "";
case I915_TILING_X: return " X";
case I915_TILING_Y: return " Y";
}
}
static const char *dirty_flag(int dirty)
{
return dirty ? " dirty" : "";
}
static const char *purgeable_flag(int purgeable)
{
return purgeable ? " purgeable" : "";
}
static void __sg_set_buf(struct scatterlist *sg,
void *addr, unsigned int len, loff_t it)
{
sg->page_link = (unsigned long)virt_to_page(addr);
sg->offset = offset_in_page(addr);
sg->length = len;
sg->dma_address = it;
}
static bool __i915_error_grow(struct drm_i915_error_state_buf *e, size_t len)
{
if (!len)
return false;
if (e->bytes + len + 1 <= e->size)
return true;
if (e->bytes) {
__sg_set_buf(e->cur++, e->buf, e->bytes, e->iter);
e->iter += e->bytes;
e->buf = NULL;
e->bytes = 0;
}
if (e->cur == e->end) {
struct scatterlist *sgl;
sgl = (typeof(sgl))__get_free_page(GFP_KERNEL);
if (!sgl) {
e->err = -ENOMEM;
return false;
}
if (e->cur) {
e->cur->offset = 0;
e->cur->length = 0;
e->cur->page_link =
(unsigned long)sgl | SG_CHAIN;
} else {
e->sgl = sgl;
}
e->cur = sgl;
e->end = sgl + SG_MAX_SINGLE_ALLOC - 1;
}
e->size = ALIGN(len + 1, SZ_64K);
e->buf = kmalloc(e->size, GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY);
if (!e->buf) {
e->size = PAGE_ALIGN(len + 1);
e->buf = kmalloc(e->size, GFP_KERNEL);
}
if (!e->buf) {
e->err = -ENOMEM;
return false;
}
return true;
}
__printf(2, 0)
static void i915_error_vprintf(struct drm_i915_error_state_buf *e,
const char *fmt, va_list args)
{
va_list ap;
int len;
if (e->err)
return;
va_copy(ap, args);
len = vsnprintf(NULL, 0, fmt, ap);
va_end(ap);
if (len <= 0) {
e->err = len;
return;
}
if (!__i915_error_grow(e, len))
return;
GEM_BUG_ON(e->bytes >= e->size);
len = vscnprintf(e->buf + e->bytes, e->size - e->bytes, fmt, args);
if (len < 0) {
e->err = len;
return;
}
e->bytes += len;
}
static void i915_error_puts(struct drm_i915_error_state_buf *e, const char *str)
{
unsigned len;
if (e->err || !str)
return;
len = strlen(str);
if (!__i915_error_grow(e, len))
return;
GEM_BUG_ON(e->bytes + len > e->size);
memcpy(e->buf + e->bytes, str, len);
e->bytes += len;
}
#define err_printf(e, ...) i915_error_printf(e, __VA_ARGS__)
#define err_puts(e, s) i915_error_puts(e, s)
static void __i915_printfn_error(struct drm_printer *p, struct va_format *vaf)
{
i915_error_vprintf(p->arg, vaf->fmt, *vaf->va);
}
static inline struct drm_printer
i915_error_printer(struct drm_i915_error_state_buf *e)
{
struct drm_printer p = {
.printfn = __i915_printfn_error,
.arg = e,
};
return p;
}
#ifdef CONFIG_DRM_I915_COMPRESS_ERROR
struct compress {
struct z_stream_s zstream;
void *tmp;
};
static bool compress_init(struct compress *c)
{
struct z_stream_s *zstream = memset(&c->zstream, 0, sizeof(c->zstream));
zstream->workspace =
kmalloc(zlib_deflate_workspacesize(MAX_WBITS, MAX_MEM_LEVEL),
GFP_ATOMIC | __GFP_NOWARN);
if (!zstream->workspace)
return false;
if (zlib_deflateInit(zstream, Z_DEFAULT_COMPRESSION) != Z_OK) {
kfree(zstream->workspace);
return false;
}
c->tmp = NULL;
if (i915_has_memcpy_from_wc())
c->tmp = (void *)__get_free_page(GFP_ATOMIC | __GFP_NOWARN);
return true;
}
static void *compress_next_page(struct drm_i915_error_object *dst)
{
unsigned long page;
if (dst->page_count >= dst->num_pages)
return ERR_PTR(-ENOSPC);
page = __get_free_page(GFP_ATOMIC | __GFP_NOWARN);
if (!page)
return ERR_PTR(-ENOMEM);
return dst->pages[dst->page_count++] = (void *)page;
}
static int compress_page(struct compress *c,
void *src,
struct drm_i915_error_object *dst)
{
struct z_stream_s *zstream = &c->zstream;
zstream->next_in = src;
if (c->tmp && i915_memcpy_from_wc(c->tmp, src, PAGE_SIZE))
zstream->next_in = c->tmp;
zstream->avail_in = PAGE_SIZE;
do {
if (zstream->avail_out == 0) {
zstream->next_out = compress_next_page(dst);
if (IS_ERR(zstream->next_out))
return PTR_ERR(zstream->next_out);
zstream->avail_out = PAGE_SIZE;
}
if (zlib_deflate(zstream, Z_NO_FLUSH) != Z_OK)
return -EIO;
touch_nmi_watchdog();
} while (zstream->avail_in);
/* Fallback to uncompressed if we increase size? */
if (0 && zstream->total_out > zstream->total_in)
return -E2BIG;
return 0;
}
static int compress_flush(struct compress *c,
struct drm_i915_error_object *dst)
{
struct z_stream_s *zstream = &c->zstream;
do {
switch (zlib_deflate(zstream, Z_FINISH)) {
case Z_OK: /* more space requested */
zstream->next_out = compress_next_page(dst);
if (IS_ERR(zstream->next_out))
return PTR_ERR(zstream->next_out);
zstream->avail_out = PAGE_SIZE;
break;
case Z_STREAM_END:
goto end;
default: /* any error */
return -EIO;
}
} while (1);
end:
memset(zstream->next_out, 0, zstream->avail_out);
dst->unused = zstream->avail_out;
return 0;
}
static void compress_fini(struct compress *c,
struct drm_i915_error_object *dst)
{
struct z_stream_s *zstream = &c->zstream;
zlib_deflateEnd(zstream);
kfree(zstream->workspace);
if (c->tmp)
free_page((unsigned long)c->tmp);
}
static void err_compression_marker(struct drm_i915_error_state_buf *m)
{
err_puts(m, ":");
}
#else
struct compress {
};
static bool compress_init(struct compress *c)
{
return true;
}
static int compress_page(struct compress *c,
void *src,
struct drm_i915_error_object *dst)
{
unsigned long page;
void *ptr;
page = __get_free_page(GFP_ATOMIC | __GFP_NOWARN);
if (!page)
return -ENOMEM;
ptr = (void *)page;
if (!i915_memcpy_from_wc(ptr, src, PAGE_SIZE))
memcpy(ptr, src, PAGE_SIZE);
dst->pages[dst->page_count++] = ptr;
return 0;
}
static int compress_flush(struct compress *c,
struct drm_i915_error_object *dst)
{
return 0;
}
static void compress_fini(struct compress *c,
struct drm_i915_error_object *dst)
{
}
static void err_compression_marker(struct drm_i915_error_state_buf *m)
{
err_puts(m, "~");
}
#endif
static void print_error_buffers(struct drm_i915_error_state_buf *m,
const char *name,
struct drm_i915_error_buffer *err,
int count)
{
err_printf(m, "%s [%d]:\n", name, count);
while (count--) {
err_printf(m, " %08x_%08x %8u %02x %02x",
upper_32_bits(err->gtt_offset),
lower_32_bits(err->gtt_offset),
err->size,
err->read_domains,
err->write_domain);
err_puts(m, tiling_flag(err->tiling));
err_puts(m, dirty_flag(err->dirty));
err_puts(m, purgeable_flag(err->purgeable));
drm/i915: Introduce mapping of user pages into video memory (userptr) ioctl By exporting the ability to map user address and inserting PTEs representing their backing pages into the GTT, we can exploit UMA in order to utilize normal application data as a texture source or even as a render target (depending upon the capabilities of the chipset). This has a number of uses, with zero-copy downloads to the GPU and efficient readback making the intermixed streaming of CPU and GPU operations fairly efficient. This ability has many widespread implications from faster rendering of client-side software rasterisers (chromium), mitigation of stalls due to read back (firefox) and to faster pipelining of texture data (such as pixel buffer objects in GL or data blobs in CL). v2: Compile with CONFIG_MMU_NOTIFIER v3: We can sleep while performing invalidate-range, which we can utilise to drop our page references prior to the kernel manipulating the vma (for either discard or cloning) and so protect normal users. v4: Only run the invalidate notifier if the range intercepts the bo. v5: Prevent userspace from attempting to GTT mmap non-page aligned buffers v6: Recheck after reacquire mutex for lost mmu. v7: Fix implicit padding of ioctl struct by rounding to next 64bit boundary. v8: Fix rebasing error after forwarding porting the back port. v9: Limit the userptr to page aligned entries. We now expect userspace to handle all the offset-in-page adjustments itself. v10: Prevent vma from being copied across fork to avoid issues with cow. v11: Drop vma behaviour changes -- locking is nigh on impossible. Use a worker to load user pages to avoid lock inversions. v12: Use get_task_mm()/mmput() for correct refcounting of mm. v13: Use a worker to release the mmu_notifier to avoid lock inversion v14: Decouple mmu_notifier from struct_mutex using a custom mmu_notifer with its own locking and tree of objects for each mm/mmu_notifier. v15: Prevent overlapping userptr objects, and invalidate all objects within the mmu_notifier range v16: Fix a typo for iterating over multiple objects in the range and rearrange error path to destroy the mmu_notifier locklessly. Also close a race between invalidate_range and the get_pages_worker. v17: Close a race between get_pages_worker/invalidate_range and fresh allocations of the same userptr range - and notice that struct_mutex was presumed to be held when during creation it wasn't. v18: Sigh. Fix the refactor of st_set_pages() to allocate enough memory for the struct sg_table and to clear it before reporting an error. v19: Always error out on read-only userptr requests as we don't have the hardware infrastructure to support them at the moment. v20: Refuse to implement read-only support until we have the required infrastructure - but reserve the bit in flags for future use. v21: use_mm() is not required for get_user_pages(). It is only meant to be used to fix up the kernel thread's current->mm for use with copy_user(). v22: Use sg_alloc_table_from_pages for that chunky feeling v23: Export a function for sanity checking dma-buf rather than encode userptr details elsewhere, and clean up comments based on suggestions by Bradley. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> Cc: "Gong, Zhipeng" <zhipeng.gong@intel.com> Cc: Akash Goel <akash.goel@intel.com> Cc: "Volkin, Bradley D" <bradley.d.volkin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> Reviewed-by: Brad Volkin <bradley.d.volkin@intel.com> [danvet: Frob ioctl allocation to pick the next one - will cause a bit of fuss with create2 apparently, but such are the rules.] [danvet2: oops, forgot to git add after manual patch application] [danvet3: Appease sparse.] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-05-16 13:22:37 +00:00
err_puts(m, err->userptr ? " userptr" : "");
err_puts(m, i915_cache_level_str(m->i915, err->cache_level));
if (err->name)
err_printf(m, " (name: %d)", err->name);
if (err->fence_reg != I915_FENCE_REG_NONE)
err_printf(m, " (fence: %d)", err->fence_reg);
err_puts(m, "\n");
err++;
}
}
static void error_print_instdone(struct drm_i915_error_state_buf *m,
const struct drm_i915_error_engine *ee)
{
int slice;
int subslice;
err_printf(m, " INSTDONE: 0x%08x\n",
ee->instdone.instdone);
if (ee->engine_id != RCS0 || INTEL_GEN(m->i915) <= 3)
return;
err_printf(m, " SC_INSTDONE: 0x%08x\n",
ee->instdone.slice_common);
if (INTEL_GEN(m->i915) <= 6)
return;
for_each_instdone_slice_subslice(m->i915, slice, subslice)
err_printf(m, " SAMPLER_INSTDONE[%d][%d]: 0x%08x\n",
slice, subslice,
ee->instdone.sampler[slice][subslice]);
for_each_instdone_slice_subslice(m->i915, slice, subslice)
err_printf(m, " ROW_INSTDONE[%d][%d]: 0x%08x\n",
slice, subslice,
ee->instdone.row[slice][subslice]);
}
static void error_print_request(struct drm_i915_error_state_buf *m,
const char *prefix,
const struct drm_i915_error_request *erq,
const unsigned long epoch)
{
if (!erq->seqno)
return;
drm/i915: Use time based guilty context banning Currently, we accumulate each time a context hangs the GPU, offset against the number of requests it submits, and if that score exceeds a certain threshold, we ban that context from submitting any more requests (cancelling any work in flight). In contrast, we use a simple timer on the file, that if we see more than a 9 hangs faster than 60s apart in total across all of its contexts, we will ban the client from creating any more contexts. This leads to a confusing situation where the file may be banned before the context, so lets use a simple timer scheme for each. If the context submits 3 hanging requests within a 120s period, declare it forbidden to ever send more requests. This has the advantage of not being easy to repair by simply sending empty requests, but has the disadvantage that if the context is idle then it is forgiven. However, if the context is idle, it is not disrupting the system, but a hog can evade the request counting and cause much more severe disruption to the system. Updating ban_score from request retirement is dubious as the retirement is purposely not in sync with request submission (i.e. we try and batch retirement to reduce overhead and avoid latency on submission), which leads to surprising situations where we can forgive a hang immediately due to a backlog of requests from before the hang being retired afterwards. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190219122215.8941-2-chris@chris-wilson.co.uk
2019-02-19 12:21:52 +00:00
err_printf(m, "%s pid %d, seqno %8x:%08x%s%s, prio %d, emitted %dms, start %08x, head %08x, tail %08x\n",
prefix, erq->pid, erq->context, erq->seqno,
drm/i915: Replace global breadcrumbs with per-context interrupt tracking A few years ago, see commit 688e6c725816 ("drm/i915: Slaughter the thundering i915_wait_request herd"), the issue of handling multiple clients waiting in parallel was brought to our attention. The requirement was that every client should be woken immediately upon its request being signaled, without incurring any cpu overhead. To handle certain fragility of our hw meant that we could not do a simple check inside the irq handler (some generations required almost unbounded delays before we could be sure of seqno coherency) and so request completion checking required delegation. Before commit 688e6c725816, the solution was simple. Every client waiting on a request would be woken on every interrupt and each would do a heavyweight check to see if their request was complete. Commit 688e6c725816 introduced an rbtree so that only the earliest waiter on the global timeline would woken, and would wake the next and so on. (Along with various complications to handle requests being reordered along the global timeline, and also a requirement for kthread to provide a delegate for fence signaling that had no process context.) The global rbtree depends on knowing the execution timeline (and global seqno). Without knowing that order, we must instead check all contexts queued to the HW to see which may have advanced. We trim that list by only checking queued contexts that are being waited on, but still we keep a list of all active contexts and their active signalers that we inspect from inside the irq handler. By moving the waiters onto the fence signal list, we can combine the client wakeup with the dma_fence signaling (a dramatic reduction in complexity, but does require the HW being coherent, the seqno must be visible from the cpu before the interrupt is raised - we keep a timer backup just in case). Having previously fixed all the issues with irq-seqno serialisation (by inserting delays onto the GPU after each request instead of random delays on the CPU after each interrupt), we can rely on the seqno state to perfom direct wakeups from the interrupt handler. This allows us to preserve our single context switch behaviour of the current routine, with the only downside that we lose the RT priority sorting of wakeups. In general, direct wakeup latency of multiple clients is about the same (about 10% better in most cases) with a reduction in total CPU time spent in the waiter (about 20-50% depending on gen). Average herd behaviour is improved, but at the cost of not delegating wakeups on task_prio. v2: Capture fence signaling state for error state and add comments to warm even the most cold of hearts. v3: Check if the request is still active before busywaiting v4: Reduce the amount of pointer misdirection with list_for_each_safe and using a local i915_request variable inside the loops v5: Add a missing pluralisation to a purely informative selftest message. References: 688e6c725816 ("drm/i915: Slaughter the thundering i915_wait_request herd") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190129205230.19056-2-chris@chris-wilson.co.uk
2019-01-29 20:52:29 +00:00
test_bit(DMA_FENCE_FLAG_SIGNALED_BIT,
&erq->flags) ? "!" : "",
test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT,
&erq->flags) ? "+" : "",
erq->sched_attr.priority,
jiffies_to_msecs(erq->jiffies - epoch),
erq->start, erq->head, erq->tail);
}
static void error_print_context(struct drm_i915_error_state_buf *m,
const char *header,
const struct drm_i915_error_context *ctx)
{
err_printf(m, "%s%s[%d] hw_id %d, prio %d, guilty %d active %d\n",
header, ctx->comm, ctx->pid, ctx->hw_id,
drm/i915: Use time based guilty context banning Currently, we accumulate each time a context hangs the GPU, offset against the number of requests it submits, and if that score exceeds a certain threshold, we ban that context from submitting any more requests (cancelling any work in flight). In contrast, we use a simple timer on the file, that if we see more than a 9 hangs faster than 60s apart in total across all of its contexts, we will ban the client from creating any more contexts. This leads to a confusing situation where the file may be banned before the context, so lets use a simple timer scheme for each. If the context submits 3 hanging requests within a 120s period, declare it forbidden to ever send more requests. This has the advantage of not being easy to repair by simply sending empty requests, but has the disadvantage that if the context is idle then it is forgiven. However, if the context is idle, it is not disrupting the system, but a hog can evade the request counting and cause much more severe disruption to the system. Updating ban_score from request retirement is dubious as the retirement is purposely not in sync with request submission (i.e. we try and batch retirement to reduce overhead and avoid latency on submission), which leads to surprising situations where we can forgive a hang immediately due to a backlog of requests from before the hang being retired afterwards. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190219122215.8941-2-chris@chris-wilson.co.uk
2019-02-19 12:21:52 +00:00
ctx->sched_attr.priority, ctx->guilty, ctx->active);
}
static void error_print_engine(struct drm_i915_error_state_buf *m,
const struct drm_i915_error_engine *ee,
const unsigned long epoch)
{
int n;
err_printf(m, "%s command stream:\n",
engine_name(m->i915, ee->engine_id));
err_printf(m, " IDLE?: %s\n", yesno(ee->idle));
err_printf(m, " START: 0x%08x\n", ee->start);
err_printf(m, " HEAD: 0x%08x [0x%08x]\n", ee->head, ee->rq_head);
err_printf(m, " TAIL: 0x%08x [0x%08x, 0x%08x]\n",
ee->tail, ee->rq_post, ee->rq_tail);
err_printf(m, " CTL: 0x%08x\n", ee->ctl);
err_printf(m, " MODE: 0x%08x\n", ee->mode);
err_printf(m, " HWS: 0x%08x\n", ee->hws);
err_printf(m, " ACTHD: 0x%08x %08x\n",
(u32)(ee->acthd>>32), (u32)ee->acthd);
err_printf(m, " IPEIR: 0x%08x\n", ee->ipeir);
err_printf(m, " IPEHR: 0x%08x\n", ee->ipehr);
error_print_instdone(m, ee);
if (ee->batchbuffer) {
u64 start = ee->batchbuffer->gtt_offset;
u64 end = start + ee->batchbuffer->gtt_size;
err_printf(m, " batch: [0x%08x_%08x, 0x%08x_%08x]\n",
upper_32_bits(start), lower_32_bits(start),
upper_32_bits(end), lower_32_bits(end));
}
if (INTEL_GEN(m->i915) >= 4) {
err_printf(m, " BBADDR: 0x%08x_%08x\n",
(u32)(ee->bbaddr>>32), (u32)ee->bbaddr);
err_printf(m, " BB_STATE: 0x%08x\n", ee->bbstate);
err_printf(m, " INSTPS: 0x%08x\n", ee->instps);
}
err_printf(m, " INSTPM: 0x%08x\n", ee->instpm);
err_printf(m, " FADDR: 0x%08x %08x\n", upper_32_bits(ee->faddr),
lower_32_bits(ee->faddr));
if (INTEL_GEN(m->i915) >= 6) {
err_printf(m, " RC PSMI: 0x%08x\n", ee->rc_psmi);
err_printf(m, " FAULT_REG: 0x%08x\n", ee->fault_reg);
}
if (HAS_PPGTT(m->i915)) {
err_printf(m, " GFX_MODE: 0x%08x\n", ee->vm_info.gfx_mode);
if (INTEL_GEN(m->i915) >= 8) {
int i;
for (i = 0; i < 4; i++)
err_printf(m, " PDP%d: 0x%016llx\n",
i, ee->vm_info.pdp[i]);
} else {
err_printf(m, " PP_DIR_BASE: 0x%08x\n",
ee->vm_info.pp_dir_base);
}
}
err_printf(m, " ring->head: 0x%08x\n", ee->cpu_ring_head);
err_printf(m, " ring->tail: 0x%08x\n", ee->cpu_ring_tail);
err_printf(m, " hangcheck timestamp: %dms (%lu%s)\n",
jiffies_to_msecs(ee->hangcheck_timestamp - epoch),
drm/i915: Decouple hang detection from hangcheck period Hangcheck state accumulation has gained more steps along the years, like head movement and more recently the subunit inactivity check. As the subunit sampling is only done if the previous state check showed inactivity, we have added more stages (and time) to reach a hang verdict. Asymmetric engine states led to different actual weight of 'one hangcheck unit' and it was demonstrated in some hangs that due to difference in stages, simpler engines were accused falsely of a hang as their scoring was much more quicker to accumulate above the hang treshold. To completely decouple the hangcheck guilty score from the hangcheck period, convert hangcheck score to a rough period of inactivity measurement. As these are tracked as jiffies, they are meaningful also across reset boundaries. This makes finding a guilty engine more accurate across multi engine activity scenarios, especially across asymmetric engines. We lose the ability to detect cross batch malicious attempts to hinder the progress. Plan is to move this functionality to be part of context banning which is more natural fit, later in the series. v2: use time_before macros (Chris) reinstate the pardoning of moving engine after hc (Chris) v3: avoid global state for per engine stall detection (Chris) v4: take timeline last retirement into account (Chris) v5: do debug print on pardoning, split out retirement timestamp (Chris) Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
2016-11-18 13:09:04 +00:00
ee->hangcheck_timestamp,
ee->hangcheck_timestamp == epoch ? "; epoch" : "");
err_printf(m, " engine reset count: %u\n", ee->reset_count);
drm/i915: Decouple hang detection from hangcheck period Hangcheck state accumulation has gained more steps along the years, like head movement and more recently the subunit inactivity check. As the subunit sampling is only done if the previous state check showed inactivity, we have added more stages (and time) to reach a hang verdict. Asymmetric engine states led to different actual weight of 'one hangcheck unit' and it was demonstrated in some hangs that due to difference in stages, simpler engines were accused falsely of a hang as their scoring was much more quicker to accumulate above the hang treshold. To completely decouple the hangcheck guilty score from the hangcheck period, convert hangcheck score to a rough period of inactivity measurement. As these are tracked as jiffies, they are meaningful also across reset boundaries. This makes finding a guilty engine more accurate across multi engine activity scenarios, especially across asymmetric engines. We lose the ability to detect cross batch malicious attempts to hinder the progress. Plan is to move this functionality to be part of context banning which is more natural fit, later in the series. v2: use time_before macros (Chris) reinstate the pardoning of moving engine after hc (Chris) v3: avoid global state for per engine stall detection (Chris) v4: take timeline last retirement into account (Chris) v5: do debug print on pardoning, split out retirement timestamp (Chris) Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
2016-11-18 13:09:04 +00:00
for (n = 0; n < ee->num_ports; n++) {
err_printf(m, " ELSP[%d]:", n);
error_print_request(m, " ", &ee->execlist[n], epoch);
}
error_print_context(m, " Active context: ", &ee->context);
}
void i915_error_printf(struct drm_i915_error_state_buf *e, const char *f, ...)
{
va_list args;
va_start(args, f);
i915_error_vprintf(e, f, args);
va_end(args);
}
static void print_error_obj(struct drm_i915_error_state_buf *m,
struct intel_engine_cs *engine,
const char *name,
struct drm_i915_error_object *obj)
{
char out[ASCII85_BUFSZ];
int page;
if (!obj)
return;
if (name) {
err_printf(m, "%s --- %s = 0x%08x %08x\n",
engine ? engine->name : "global", name,
upper_32_bits(obj->gtt_offset),
lower_32_bits(obj->gtt_offset));
}
err_compression_marker(m);
for (page = 0; page < obj->page_count; page++) {
int i, len;
len = PAGE_SIZE;
if (page == obj->page_count - 1)
len -= obj->unused;
len = ascii85_encode_len(len);
for (i = 0; i < len; i++)
err_puts(m, ascii85_encode(obj->pages[page][i], out));
}
err_puts(m, "\n");
}
static void err_print_capabilities(struct drm_i915_error_state_buf *m,
const struct intel_device_info *info,
const struct intel_runtime_info *runtime,
const struct intel_driver_caps *caps)
{
struct drm_printer p = i915_error_printer(m);
intel_device_info_dump_flags(info, &p);
intel_driver_caps_print(caps, &p);
intel_device_info_dump_topology(&runtime->sseu, &p);
}
static void err_print_params(struct drm_i915_error_state_buf *m,
const struct i915_params *params)
{
struct drm_printer p = i915_error_printer(m);
i915_params_dump(params, &p);
}
static void err_print_pciid(struct drm_i915_error_state_buf *m,
struct drm_i915_private *i915)
{
struct pci_dev *pdev = i915->drm.pdev;
err_printf(m, "PCI ID: 0x%04x\n", pdev->device);
err_printf(m, "PCI Revision: 0x%02x\n", pdev->revision);
err_printf(m, "PCI Subsystem: %04x:%04x\n",
pdev->subsystem_vendor,
pdev->subsystem_device);
}
static void err_print_uc(struct drm_i915_error_state_buf *m,
const struct i915_error_uc *error_uc)
{
struct drm_printer p = i915_error_printer(m);
const struct i915_gpu_state *error =
container_of(error_uc, typeof(*error), uc);
if (!error->device_info.has_guc)
return;
intel_uc_fw_dump(&error_uc->guc_fw, &p);
intel_uc_fw_dump(&error_uc->huc_fw, &p);
print_error_obj(m, NULL, "GuC log buffer", error_uc->guc_log);
}
static void err_free_sgl(struct scatterlist *sgl)
{
while (sgl) {
struct scatterlist *sg;
for (sg = sgl; !sg_is_chain(sg); sg++) {
kfree(sg_virt(sg));
if (sg_is_last(sg))
break;
}
sg = sg_is_last(sg) ? NULL : sg_chain_ptr(sg);
free_page((unsigned long)sgl);
sgl = sg;
}
}
static void __err_print_to_sgl(struct drm_i915_error_state_buf *m,
struct i915_gpu_state *error)
{
struct drm_i915_error_object *obj;
struct timespec64 ts;
int i, j;
if (*error->error_msg)
err_printf(m, "%s\n", error->error_msg);
err_printf(m, "Kernel: %s %s\n",
init_utsname()->release,
init_utsname()->machine);
ts = ktime_to_timespec64(error->time);
err_printf(m, "Time: %lld s %ld us\n",
(s64)ts.tv_sec, ts.tv_nsec / NSEC_PER_USEC);
ts = ktime_to_timespec64(error->boottime);
err_printf(m, "Boottime: %lld s %ld us\n",
(s64)ts.tv_sec, ts.tv_nsec / NSEC_PER_USEC);
ts = ktime_to_timespec64(error->uptime);
err_printf(m, "Uptime: %lld s %ld us\n",
(s64)ts.tv_sec, ts.tv_nsec / NSEC_PER_USEC);
err_printf(m, "Epoch: %lu jiffies (%u HZ)\n", error->epoch, HZ);
err_printf(m, "Capture: %lu jiffies; %d ms ago, %d ms after epoch\n",
error->capture,
jiffies_to_msecs(jiffies - error->capture),
jiffies_to_msecs(error->capture - error->epoch));
drm/i915: Decouple hang detection from hangcheck period Hangcheck state accumulation has gained more steps along the years, like head movement and more recently the subunit inactivity check. As the subunit sampling is only done if the previous state check showed inactivity, we have added more stages (and time) to reach a hang verdict. Asymmetric engine states led to different actual weight of 'one hangcheck unit' and it was demonstrated in some hangs that due to difference in stages, simpler engines were accused falsely of a hang as their scoring was much more quicker to accumulate above the hang treshold. To completely decouple the hangcheck guilty score from the hangcheck period, convert hangcheck score to a rough period of inactivity measurement. As these are tracked as jiffies, they are meaningful also across reset boundaries. This makes finding a guilty engine more accurate across multi engine activity scenarios, especially across asymmetric engines. We lose the ability to detect cross batch malicious attempts to hinder the progress. Plan is to move this functionality to be part of context banning which is more natural fit, later in the series. v2: use time_before macros (Chris) reinstate the pardoning of moving engine after hc (Chris) v3: avoid global state for per engine stall detection (Chris) v4: take timeline last retirement into account (Chris) v5: do debug print on pardoning, split out retirement timestamp (Chris) Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
2016-11-18 13:09:04 +00:00
for (i = 0; i < ARRAY_SIZE(error->engine); i++) {
if (!error->engine[i].context.pid)
continue;
drm/i915: Use time based guilty context banning Currently, we accumulate each time a context hangs the GPU, offset against the number of requests it submits, and if that score exceeds a certain threshold, we ban that context from submitting any more requests (cancelling any work in flight). In contrast, we use a simple timer on the file, that if we see more than a 9 hangs faster than 60s apart in total across all of its contexts, we will ban the client from creating any more contexts. This leads to a confusing situation where the file may be banned before the context, so lets use a simple timer scheme for each. If the context submits 3 hanging requests within a 120s period, declare it forbidden to ever send more requests. This has the advantage of not being easy to repair by simply sending empty requests, but has the disadvantage that if the context is idle then it is forgiven. However, if the context is idle, it is not disrupting the system, but a hog can evade the request counting and cause much more severe disruption to the system. Updating ban_score from request retirement is dubious as the retirement is purposely not in sync with request submission (i.e. we try and batch retirement to reduce overhead and avoid latency on submission), which leads to surprising situations where we can forgive a hang immediately due to a backlog of requests from before the hang being retired afterwards. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190219122215.8941-2-chris@chris-wilson.co.uk
2019-02-19 12:21:52 +00:00
err_printf(m, "Active process (on ring %s): %s [%d]\n",
engine_name(m->i915, i),
error->engine[i].context.comm,
drm/i915: Use time based guilty context banning Currently, we accumulate each time a context hangs the GPU, offset against the number of requests it submits, and if that score exceeds a certain threshold, we ban that context from submitting any more requests (cancelling any work in flight). In contrast, we use a simple timer on the file, that if we see more than a 9 hangs faster than 60s apart in total across all of its contexts, we will ban the client from creating any more contexts. This leads to a confusing situation where the file may be banned before the context, so lets use a simple timer scheme for each. If the context submits 3 hanging requests within a 120s period, declare it forbidden to ever send more requests. This has the advantage of not being easy to repair by simply sending empty requests, but has the disadvantage that if the context is idle then it is forgiven. However, if the context is idle, it is not disrupting the system, but a hog can evade the request counting and cause much more severe disruption to the system. Updating ban_score from request retirement is dubious as the retirement is purposely not in sync with request submission (i.e. we try and batch retirement to reduce overhead and avoid latency on submission), which leads to surprising situations where we can forgive a hang immediately due to a backlog of requests from before the hang being retired afterwards. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190219122215.8941-2-chris@chris-wilson.co.uk
2019-02-19 12:21:52 +00:00
error->engine[i].context.pid);
}
err_printf(m, "Reset count: %u\n", error->reset_count);
err_printf(m, "Suspend count: %u\n", error->suspend_count);
err_printf(m, "Platform: %s\n", intel_platform_name(error->device_info.platform));
err_print_pciid(m, m->i915);
err_printf(m, "IOMMU enabled?: %d\n", error->iommu);
if (HAS_CSR(m->i915)) {
struct intel_csr *csr = &m->i915->csr;
err_printf(m, "DMC loaded: %s\n",
yesno(csr->dmc_payload != NULL));
err_printf(m, "DMC fw version: %d.%d\n",
CSR_VERSION_MAJOR(csr->version),
CSR_VERSION_MINOR(csr->version));
}
err_printf(m, "GT awake: %s\n", yesno(error->awake));
err_printf(m, "RPM wakelock: %s\n", yesno(error->wakelock));
err_printf(m, "PM suspended: %s\n", yesno(error->suspended));
err_printf(m, "EIR: 0x%08x\n", error->eir);
err_printf(m, "IER: 0x%08x\n", error->ier);
for (i = 0; i < error->ngtier; i++)
err_printf(m, "GTIER[%d]: 0x%08x\n", i, error->gtier[i]);
err_printf(m, "PGTBL_ER: 0x%08x\n", error->pgtbl_er);
err_printf(m, "FORCEWAKE: 0x%08x\n", error->forcewake);
err_printf(m, "DERRMR: 0x%08x\n", error->derrmr);
err_printf(m, "CCID: 0x%08x\n", error->ccid);
for (i = 0; i < error->nfence; i++)
err_printf(m, " fence[%d] = %08llx\n", i, error->fence[i]);
if (INTEL_GEN(m->i915) >= 6) {
err_printf(m, "ERROR: 0x%08x\n", error->error);
if (INTEL_GEN(m->i915) >= 8)
err_printf(m, "FAULT_TLB_DATA: 0x%08x 0x%08x\n",
error->fault_data1, error->fault_data0);
err_printf(m, "DONE_REG: 0x%08x\n", error->done_reg);
}
if (IS_GEN(m->i915, 7))
err_printf(m, "ERR_INT: 0x%08x\n", error->err_int);
for (i = 0; i < ARRAY_SIZE(error->engine); i++) {
if (error->engine[i].engine_id != -1)
error_print_engine(m, &error->engine[i], error->epoch);
}
for (i = 0; i < ARRAY_SIZE(error->active_vm); i++) {
char buf[128];
int len, first = 1;
if (!error->active_vm[i])
break;
len = scnprintf(buf, sizeof(buf), "Active (");
for (j = 0; j < ARRAY_SIZE(error->engine); j++) {
if (error->engine[j].vm != error->active_vm[i])
continue;
len += scnprintf(buf + len, sizeof(buf), "%s%s",
first ? "" : ", ",
m->i915->engine[j]->name);
first = 0;
}
scnprintf(buf + len, sizeof(buf), ")");
print_error_buffers(m, buf,
error->active_bo[i],
error->active_bo_count[i]);
}
print_error_buffers(m, "Pinned (global)",
error->pinned_bo,
error->pinned_bo_count);
for (i = 0; i < ARRAY_SIZE(error->engine); i++) {
const struct drm_i915_error_engine *ee = &error->engine[i];
obj = ee->batchbuffer;
if (obj) {
err_puts(m, m->i915->engine[i]->name);
if (ee->context.pid)
err_printf(m, " (submitted by %s [%d])",
ee->context.comm,
ee->context.pid);
err_printf(m, " --- gtt_offset = 0x%08x %08x\n",
upper_32_bits(obj->gtt_offset),
lower_32_bits(obj->gtt_offset));
print_error_obj(m, m->i915->engine[i], NULL, obj);
}
for (j = 0; j < ee->user_bo_count; j++)
print_error_obj(m, m->i915->engine[i],
"user", ee->user_bo[j]);
if (ee->num_requests) {
err_printf(m, "%s --- %d requests\n",
m->i915->engine[i]->name,
ee->num_requests);
for (j = 0; j < ee->num_requests; j++)
error_print_request(m, " ",
&ee->requests[j],
error->epoch);
}
print_error_obj(m, m->i915->engine[i],
"ringbuffer", ee->ringbuffer);
print_error_obj(m, m->i915->engine[i],
"HW Status", ee->hws_page);
print_error_obj(m, m->i915->engine[i],
"HW context", ee->ctx);
print_error_obj(m, m->i915->engine[i],
"WA context", ee->wa_ctx);
print_error_obj(m, m->i915->engine[i],
"WA batchbuffer", ee->wa_batchbuffer);
print_error_obj(m, m->i915->engine[i],
"NULL context", ee->default_state);
}
if (error->overlay)
intel_overlay_print_error_state(m, error->overlay);
if (error->display)
intel_display_print_error_state(m, error->display);
err_print_capabilities(m, &error->device_info, &error->runtime_info,
&error->driver_caps);
err_print_params(m, &error->params);
err_print_uc(m, &error->uc);
}
static int err_print_to_sgl(struct i915_gpu_state *error)
{
struct drm_i915_error_state_buf m;
if (IS_ERR(error))
return PTR_ERR(error);
if (READ_ONCE(error->sgl))
return 0;
memset(&m, 0, sizeof(m));
m.i915 = error->i915;
__err_print_to_sgl(&m, error);
if (m.buf) {
__sg_set_buf(m.cur++, m.buf, m.bytes, m.iter);
m.bytes = 0;
m.buf = NULL;
}
if (m.cur) {
GEM_BUG_ON(m.end < m.cur);
sg_mark_end(m.cur - 1);
}
GEM_BUG_ON(m.sgl && !m.cur);
if (m.err) {
err_free_sgl(m.sgl);
return m.err;
}
if (cmpxchg(&error->sgl, NULL, m.sgl))
err_free_sgl(m.sgl);
return 0;
}
ssize_t i915_gpu_state_copy_to_buffer(struct i915_gpu_state *error,
char *buf, loff_t off, size_t rem)
{
struct scatterlist *sg;
size_t count;
loff_t pos;
int err;
if (!error || !rem)
return 0;
err = err_print_to_sgl(error);
if (err)
return err;
sg = READ_ONCE(error->fit);
if (!sg || off < sg->dma_address)
sg = error->sgl;
if (!sg)
return 0;
pos = sg->dma_address;
count = 0;
do {
size_t len, start;
if (sg_is_chain(sg)) {
sg = sg_chain_ptr(sg);
GEM_BUG_ON(sg_is_chain(sg));
}
len = sg->length;
if (pos + len <= off) {
pos += len;
continue;
}
start = sg->offset;
if (pos < off) {
GEM_BUG_ON(off - pos > len);
len -= off - pos;
start += off - pos;
pos = off;
}
len = min(len, rem);
GEM_BUG_ON(!len || len > sg->length);
memcpy(buf, page_address(sg_page(sg)) + start, len);
count += len;
pos += len;
buf += len;
rem -= len;
if (!rem) {
WRITE_ONCE(error->fit, sg);
break;
}
} while (!sg_is_last(sg++));
return count;
}
static void i915_error_object_free(struct drm_i915_error_object *obj)
{
int page;
if (obj == NULL)
return;
for (page = 0; page < obj->page_count; page++)
free_page((unsigned long)obj->pages[page]);
kfree(obj);
}
static void cleanup_params(struct i915_gpu_state *error)
{
i915_params_free(&error->params);
}
static void cleanup_uc_state(struct i915_gpu_state *error)
{
struct i915_error_uc *error_uc = &error->uc;
kfree(error_uc->guc_fw.path);
kfree(error_uc->huc_fw.path);
i915_error_object_free(error_uc->guc_log);
}
void __i915_gpu_state_free(struct kref *error_ref)
{
struct i915_gpu_state *error =
container_of(error_ref, typeof(*error), ref);
long i, j;
for (i = 0; i < ARRAY_SIZE(error->engine); i++) {
struct drm_i915_error_engine *ee = &error->engine[i];
for (j = 0; j < ee->user_bo_count; j++)
i915_error_object_free(ee->user_bo[j]);
kfree(ee->user_bo);
i915_error_object_free(ee->batchbuffer);
i915_error_object_free(ee->wa_batchbuffer);
i915_error_object_free(ee->ringbuffer);
i915_error_object_free(ee->hws_page);
i915_error_object_free(ee->ctx);
i915_error_object_free(ee->wa_ctx);
kfree(ee->requests);
}
for (i = 0; i < ARRAY_SIZE(error->active_bo); i++)
drm/i915: Do not leak objects after capturing error state While running kmemleak chasing a different memleak, I saw that the capture_error_state function was leaking some objects, for example: unreferenced object 0xffff8800a9b72148 (size 8192): comm "kworker/u16:0", pid 1499, jiffies 4295201243 (age 990.096s) hex dump (first 32 bytes): 00 00 04 00 00 00 00 00 5d f4 ff ff 00 00 00 00 ........]....... 00 30 b0 01 00 00 00 00 37 00 00 00 00 00 00 00 .0......7....... backtrace: [<ffffffff811e5ae4>] create_object+0x104/0x2c0 [<ffffffff8178f50a>] kmemleak_alloc+0x7a/0xc0 [<ffffffff811cde4b>] __kmalloc+0xeb/0x220 [<ffffffffa038f1d9>] kcalloc.constprop.12+0x2d/0x2f [i915] [<ffffffffa0316064>] i915_capture_error_state+0x3f4/0x1660 [i915] [<ffffffffa03207df>] i915_handle_error+0x7f/0x660 [i915] [<ffffffffa03210f7>] i915_hangcheck_elapsed+0x2e7/0x470 [i915] [<ffffffff8108d574>] process_one_work+0x144/0x490 [<ffffffff8108dfbd>] worker_thread+0x11d/0x530 [<ffffffff81094079>] kthread+0xc9/0xe0 [<ffffffff817a2398>] ret_from_fork+0x58/0x90 [<ffffffffffffffff>] 0xffffffffffffffff The following objects are allocated in i915_gem_capture_buffers, but not released in i915_error_state_free: - error->active_bo_count - error->pinned_bo - error->pinned_bo_count - error->active_bo[vm_count] (allocated in i915_gem_capture_vm). The leaks were introduced by commit 95f5301dd880da2dea2c9a9c29750064536d426a Author: Ben Widawsky <ben@bwidawsk.net> Date: Wed Jul 31 17:00:15 2013 -0700 drm/i915: Update error capture for VMs v2: Reuse iterator and add culprit commit details (Chris) Cc: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Michel Thierry <michel.thierry@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2015-03-20 09:41:03 +00:00
kfree(error->active_bo[i]);
kfree(error->pinned_bo);
kfree(error->overlay);
kfree(error->display);
cleanup_params(error);
cleanup_uc_state(error);
err_free_sgl(error->sgl);
kfree(error);
}
static struct drm_i915_error_object *
i915_error_object_create(struct drm_i915_private *i915,
struct i915_vma *vma)
{
struct i915_ggtt *ggtt = &i915->ggtt;
const u64 slot = ggtt->error_capture.start;
struct drm_i915_error_object *dst;
struct compress compress;
unsigned long num_pages;
struct sgt_iter iter;
dma_addr_t dma;
int ret;
if (!vma || !vma->pages)
return NULL;
num_pages = min_t(u64, vma->size, vma->obj->base.size) >> PAGE_SHIFT;
num_pages = DIV_ROUND_UP(10 * num_pages, 8); /* worstcase zlib growth */
dst = kmalloc(sizeof(*dst) + num_pages * sizeof(u32 *),
GFP_ATOMIC | __GFP_NOWARN);
if (!dst)
return NULL;
dst->gtt_offset = vma->node.start;
dst->gtt_size = vma->node.size;
dst->num_pages = num_pages;
dst->page_count = 0;
dst->unused = 0;
if (!compress_init(&compress)) {
kfree(dst);
return NULL;
}
ret = -EINVAL;
for_each_sgt_dma(dma, iter, vma->pages) {
void __iomem *s;
ggtt->vm.insert_page(&ggtt->vm, dma, slot, I915_CACHE_NONE, 0);
s = io_mapping_map_atomic_wc(&ggtt->iomap, slot);
ret = compress_page(&compress, (void __force *)s, dst);
io_mapping_unmap_atomic(s);
if (ret)
break;
}
if (ret || compress_flush(&compress, dst)) {
while (dst->page_count--)
free_page((unsigned long)dst->pages[dst->page_count]);
kfree(dst);
dst = NULL;
}
compress_fini(&compress, dst);
return dst;
}
static void capture_bo(struct drm_i915_error_buffer *err,
struct i915_vma *vma)
{
struct drm_i915_gem_object *obj = vma->obj;
err->size = obj->base.size;
err->name = obj->base.name;
err->gtt_offset = vma->node.start;
err->read_domains = obj->read_domains;
err->write_domain = obj->write_domain;
err->fence_reg = vma->fence ? vma->fence->id : -1;
err->tiling = i915_gem_object_get_tiling(obj);
err->dirty = obj->mm.dirty;
err->purgeable = obj->mm.madv != I915_MADV_WILLNEED;
drm/i915: Introduce mapping of user pages into video memory (userptr) ioctl By exporting the ability to map user address and inserting PTEs representing their backing pages into the GTT, we can exploit UMA in order to utilize normal application data as a texture source or even as a render target (depending upon the capabilities of the chipset). This has a number of uses, with zero-copy downloads to the GPU and efficient readback making the intermixed streaming of CPU and GPU operations fairly efficient. This ability has many widespread implications from faster rendering of client-side software rasterisers (chromium), mitigation of stalls due to read back (firefox) and to faster pipelining of texture data (such as pixel buffer objects in GL or data blobs in CL). v2: Compile with CONFIG_MMU_NOTIFIER v3: We can sleep while performing invalidate-range, which we can utilise to drop our page references prior to the kernel manipulating the vma (for either discard or cloning) and so protect normal users. v4: Only run the invalidate notifier if the range intercepts the bo. v5: Prevent userspace from attempting to GTT mmap non-page aligned buffers v6: Recheck after reacquire mutex for lost mmu. v7: Fix implicit padding of ioctl struct by rounding to next 64bit boundary. v8: Fix rebasing error after forwarding porting the back port. v9: Limit the userptr to page aligned entries. We now expect userspace to handle all the offset-in-page adjustments itself. v10: Prevent vma from being copied across fork to avoid issues with cow. v11: Drop vma behaviour changes -- locking is nigh on impossible. Use a worker to load user pages to avoid lock inversions. v12: Use get_task_mm()/mmput() for correct refcounting of mm. v13: Use a worker to release the mmu_notifier to avoid lock inversion v14: Decouple mmu_notifier from struct_mutex using a custom mmu_notifer with its own locking and tree of objects for each mm/mmu_notifier. v15: Prevent overlapping userptr objects, and invalidate all objects within the mmu_notifier range v16: Fix a typo for iterating over multiple objects in the range and rearrange error path to destroy the mmu_notifier locklessly. Also close a race between invalidate_range and the get_pages_worker. v17: Close a race between get_pages_worker/invalidate_range and fresh allocations of the same userptr range - and notice that struct_mutex was presumed to be held when during creation it wasn't. v18: Sigh. Fix the refactor of st_set_pages() to allocate enough memory for the struct sg_table and to clear it before reporting an error. v19: Always error out on read-only userptr requests as we don't have the hardware infrastructure to support them at the moment. v20: Refuse to implement read-only support until we have the required infrastructure - but reserve the bit in flags for future use. v21: use_mm() is not required for get_user_pages(). It is only meant to be used to fix up the kernel thread's current->mm for use with copy_user(). v22: Use sg_alloc_table_from_pages for that chunky feeling v23: Export a function for sanity checking dma-buf rather than encode userptr details elsewhere, and clean up comments based on suggestions by Bradley. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> Cc: "Gong, Zhipeng" <zhipeng.gong@intel.com> Cc: Akash Goel <akash.goel@intel.com> Cc: "Volkin, Bradley D" <bradley.d.volkin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> Reviewed-by: Brad Volkin <bradley.d.volkin@intel.com> [danvet: Frob ioctl allocation to pick the next one - will cause a bit of fuss with create2 apparently, but such are the rules.] [danvet2: oops, forgot to git add after manual patch application] [danvet3: Appease sparse.] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-05-16 13:22:37 +00:00
err->userptr = obj->userptr.mm != NULL;
err->cache_level = obj->cache_level;
}
static u32 capture_error_bo(struct drm_i915_error_buffer *err,
int count, struct list_head *head,
drm/i915: Stop tracking MRU activity on VMA Our goal is to remove struct_mutex and replace it with fine grained locking. One of the thorny issues is our eviction logic for reclaiming space for an execbuffer (or GTT mmaping, among a few other examples). While eviction itself is easy to move under a per-VM mutex, performing the activity tracking is less agreeable. One solution is not to do any MRU tracking and do a simple coarse evaluation during eviction of active/inactive, with a loose temporal ordering of last insertion/evaluation. That keeps all the locking constrained to when we are manipulating the VM itself, neatly avoiding the tricky handling of possible recursive locking during execbuf and elsewhere. Note that discarding the MRU (currently implemented as a pair of lists, to avoid scanning the active list for a NONBLOCKING search) is unlikely to impact upon our efficiency to reclaim VM space (where we think a LRU model is best) as our current strategy is to use random idle replacement first before doing a search, and over time the use of softpinned 48b per-ppGTT is growing (thereby eliminating any need to perform any eviction searches, in theory at least) with the remaining users being found on much older devices (gen2-gen6). v2: Changelog and commentary rewritten to elaborate on the duality of a single list being both an inactive and active list. v3: Consolidate bool parameters into a single set of flags; don't comment on the duality of a single variable being a multiplicity of bits. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190128102356.15037-1-chris@chris-wilson.co.uk
2019-01-28 10:23:52 +00:00
unsigned int flags)
#define ACTIVE_ONLY BIT(0)
#define PINNED_ONLY BIT(1)
{
2013-08-01 00:00:14 +00:00
struct i915_vma *vma;
int i = 0;
list_for_each_entry(vma, head, vm_link) {
if (!vma->obj)
continue;
drm/i915: Stop tracking MRU activity on VMA Our goal is to remove struct_mutex and replace it with fine grained locking. One of the thorny issues is our eviction logic for reclaiming space for an execbuffer (or GTT mmaping, among a few other examples). While eviction itself is easy to move under a per-VM mutex, performing the activity tracking is less agreeable. One solution is not to do any MRU tracking and do a simple coarse evaluation during eviction of active/inactive, with a loose temporal ordering of last insertion/evaluation. That keeps all the locking constrained to when we are manipulating the VM itself, neatly avoiding the tricky handling of possible recursive locking during execbuf and elsewhere. Note that discarding the MRU (currently implemented as a pair of lists, to avoid scanning the active list for a NONBLOCKING search) is unlikely to impact upon our efficiency to reclaim VM space (where we think a LRU model is best) as our current strategy is to use random idle replacement first before doing a search, and over time the use of softpinned 48b per-ppGTT is growing (thereby eliminating any need to perform any eviction searches, in theory at least) with the remaining users being found on much older devices (gen2-gen6). v2: Changelog and commentary rewritten to elaborate on the duality of a single list being both an inactive and active list. v3: Consolidate bool parameters into a single set of flags; don't comment on the duality of a single variable being a multiplicity of bits. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190128102356.15037-1-chris@chris-wilson.co.uk
2019-01-28 10:23:52 +00:00
if (flags & ACTIVE_ONLY && !i915_vma_is_active(vma))
continue;
if (flags & PINNED_ONLY && !i915_vma_is_pinned(vma))
continue;
capture_bo(err++, vma);
if (++i == count)
break;
}
return i;
}
/*
* Generate a semi-unique error code. The code is not meant to have meaning, The
* code's only purpose is to try to prevent false duplicated bug reports by
* grossly estimating a GPU error state.
*
* TODO Ideally, hashing the batchbuffer would be a very nice way to determine
* the hang if we could strip the GTT offset information from it.
*
* It's only a small step better than a random number in its current form.
*/
static u32 i915_error_generate_code(struct i915_gpu_state *error,
unsigned long engine_mask)
{
/*
* IPEHR would be an ideal way to detect errors, as it's the gross
* measure of "the command that hung." However, has some very common
* synchronization commands which almost always appear in the case
* strictly a client bug. Use instdone to differentiate those some.
*/
if (engine_mask) {
struct drm_i915_error_engine *ee =
&error->engine[ffs(engine_mask)];
return ee->ipehr ^ ee->instdone.instdone;
}
return 0;
}
static void gem_record_fences(struct i915_gpu_state *error)
{
struct drm_i915_private *dev_priv = error->i915;
int i;
if (INTEL_GEN(dev_priv) >= 6) {
for (i = 0; i < dev_priv->num_fence_regs; i++)
error->fence[i] = I915_READ64(FENCE_REG_GEN6_LO(i));
} else if (INTEL_GEN(dev_priv) >= 4) {
for (i = 0; i < dev_priv->num_fence_regs; i++)
error->fence[i] = I915_READ64(FENCE_REG_965_LO(i));
} else {
for (i = 0; i < dev_priv->num_fence_regs; i++)
error->fence[i] = I915_READ(FENCE_REG(i));
}
error->nfence = i;
}
static void error_record_engine_registers(struct i915_gpu_state *error,
struct intel_engine_cs *engine,
struct drm_i915_error_engine *ee)
{
struct drm_i915_private *dev_priv = engine->i915;
if (INTEL_GEN(dev_priv) >= 6) {
ee->rc_psmi = ENGINE_READ(engine, RING_PSMI_CTL);
if (INTEL_GEN(dev_priv) >= 8)
ee->fault_reg = I915_READ(GEN8_RING_FAULT_REG);
else
ee->fault_reg = I915_READ(RING_FAULT_REG(engine));
}
if (INTEL_GEN(dev_priv) >= 4) {
ee->faddr = ENGINE_READ(engine, RING_DMA_FADD);
ee->ipeir = ENGINE_READ(engine, RING_IPEIR);
ee->ipehr = ENGINE_READ(engine, RING_IPEHR);
ee->instps = ENGINE_READ(engine, RING_INSTPS);
ee->bbaddr = ENGINE_READ(engine, RING_BBADDR);
if (INTEL_GEN(dev_priv) >= 8) {
ee->faddr |= (u64)ENGINE_READ(engine, RING_DMA_FADD_UDW) << 32;
ee->bbaddr |= (u64)ENGINE_READ(engine, RING_BBADDR_UDW) << 32;
}
ee->bbstate = ENGINE_READ(engine, RING_BBSTATE);
} else {
ee->faddr = ENGINE_READ(engine, DMA_FADD_I8XX);
ee->ipeir = ENGINE_READ(engine, IPEIR);
ee->ipehr = ENGINE_READ(engine, IPEHR);
}
intel_engine_get_instdone(engine, &ee->instdone);
ee->instpm = ENGINE_READ(engine, RING_INSTPM);
ee->acthd = intel_engine_get_active_head(engine);
ee->start = ENGINE_READ(engine, RING_START);
ee->head = ENGINE_READ(engine, RING_HEAD);
ee->tail = ENGINE_READ(engine, RING_TAIL);
ee->ctl = ENGINE_READ(engine, RING_CTL);
if (INTEL_GEN(dev_priv) > 2)
ee->mode = ENGINE_READ(engine, RING_MI_MODE);
if (!HWS_NEEDS_PHYSICAL(dev_priv)) {
drm/i915: Type safe register read/write Make I915_READ and I915_WRITE more type safe by wrapping the register offset in a struct. This should eliminate most of the fumbles we've had with misplaced parens. This only takes care of normal mmio registers. We could extend the idea to other register types and define each with its own struct. That way you wouldn't be able to accidentally pass the wrong thing to a specific register access function. The gpio_reg setup is probably the ugliest thing left. But I figure I'd just leave it for now, and wait for some divine inspiration to strike before making it nice. As for the generated code, it's actually a bit better sometimes. Eg. looking at i915_irq_handler(), we can see the following change: lea 0x70024(%rdx,%rax,1),%r9d mov $0x1,%edx - movslq %r9d,%r9 - mov %r9,%rsi - mov %r9,-0x58(%rbp) - callq *0xd8(%rbx) + mov %r9d,%esi + mov %r9d,-0x48(%rbp) callq *0xd8(%rbx) So previously gcc thought the register offset might be signed and decided to sign extend it, just in case. The rest appears to be mostly just minor shuffling of instructions. v2: i915_mmio_reg_{offset,equal,valid}() helpers added s/_REG/_MMIO/ in the register defines mo more switch statements left to worry about ring_emit stuff got sorted in a prep patch cmd parser, lrc context and w/a batch buildup also in prep patch vgpu stuff cleaned up and moved to a prep patch all other unrelated changes split out v3: Rebased due to BXT DSI/BLC, MOCS, etc. v4: Rebased due to churn, s/i915_mmio_reg_t/i915_reg_t/ Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: http://patchwork.freedesktop.org/patch/msgid/1447853606-2751-1-git-send-email-ville.syrjala@linux.intel.com
2015-11-18 13:33:26 +00:00
i915_reg_t mmio;
if (IS_GEN(dev_priv, 7)) {
switch (engine->id) {
default:
MISSING_CASE(engine->id);
case RCS0:
mmio = RENDER_HWS_PGA_GEN7;
break;
case BCS0:
mmio = BLT_HWS_PGA_GEN7;
break;
case VCS0:
mmio = BSD_HWS_PGA_GEN7;
break;
case VECS0:
mmio = VEBOX_HWS_PGA_GEN7;
break;
}
} else if (IS_GEN(engine->i915, 6)) {
mmio = RING_HWS_PGA_GEN6(engine->mmio_base);
} else {
/* XXX: gen8 returns to sanity */
mmio = RING_HWS_PGA(engine->mmio_base);
}
ee->hws = I915_READ(mmio);
}
ee->idle = intel_engine_is_idle(engine);
if (!ee->idle)
ee->hangcheck_timestamp = engine->hangcheck.action_timestamp;
ee->reset_count = i915_reset_engine_count(&dev_priv->gpu_error,
engine);
if (HAS_PPGTT(dev_priv)) {
int i;
ee->vm_info.gfx_mode = I915_READ(RING_MODE_GEN7(engine));
if (IS_GEN(dev_priv, 6))
ee->vm_info.pp_dir_base =
ENGINE_READ(engine, RING_PP_DIR_BASE_READ);
else if (IS_GEN(dev_priv, 7))
ee->vm_info.pp_dir_base =
ENGINE_READ(engine, RING_PP_DIR_BASE);
else if (INTEL_GEN(dev_priv) >= 8)
for (i = 0; i < 4; i++) {
ee->vm_info.pdp[i] =
I915_READ(GEN8_RING_PDP_UDW(engine, i));
ee->vm_info.pdp[i] <<= 32;
ee->vm_info.pdp[i] |=
I915_READ(GEN8_RING_PDP_LDW(engine, i));
}
}
}
static void record_request(struct i915_request *request,
struct drm_i915_error_request *erq)
{
struct i915_gem_context *ctx = request->gem_context;
drm/i915: Replace global breadcrumbs with per-context interrupt tracking A few years ago, see commit 688e6c725816 ("drm/i915: Slaughter the thundering i915_wait_request herd"), the issue of handling multiple clients waiting in parallel was brought to our attention. The requirement was that every client should be woken immediately upon its request being signaled, without incurring any cpu overhead. To handle certain fragility of our hw meant that we could not do a simple check inside the irq handler (some generations required almost unbounded delays before we could be sure of seqno coherency) and so request completion checking required delegation. Before commit 688e6c725816, the solution was simple. Every client waiting on a request would be woken on every interrupt and each would do a heavyweight check to see if their request was complete. Commit 688e6c725816 introduced an rbtree so that only the earliest waiter on the global timeline would woken, and would wake the next and so on. (Along with various complications to handle requests being reordered along the global timeline, and also a requirement for kthread to provide a delegate for fence signaling that had no process context.) The global rbtree depends on knowing the execution timeline (and global seqno). Without knowing that order, we must instead check all contexts queued to the HW to see which may have advanced. We trim that list by only checking queued contexts that are being waited on, but still we keep a list of all active contexts and their active signalers that we inspect from inside the irq handler. By moving the waiters onto the fence signal list, we can combine the client wakeup with the dma_fence signaling (a dramatic reduction in complexity, but does require the HW being coherent, the seqno must be visible from the cpu before the interrupt is raised - we keep a timer backup just in case). Having previously fixed all the issues with irq-seqno serialisation (by inserting delays onto the GPU after each request instead of random delays on the CPU after each interrupt), we can rely on the seqno state to perfom direct wakeups from the interrupt handler. This allows us to preserve our single context switch behaviour of the current routine, with the only downside that we lose the RT priority sorting of wakeups. In general, direct wakeup latency of multiple clients is about the same (about 10% better in most cases) with a reduction in total CPU time spent in the waiter (about 20-50% depending on gen). Average herd behaviour is improved, but at the cost of not delegating wakeups on task_prio. v2: Capture fence signaling state for error state and add comments to warm even the most cold of hearts. v3: Check if the request is still active before busywaiting v4: Reduce the amount of pointer misdirection with list_for_each_safe and using a local i915_request variable inside the loops v5: Add a missing pluralisation to a purely informative selftest message. References: 688e6c725816 ("drm/i915: Slaughter the thundering i915_wait_request herd") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190129205230.19056-2-chris@chris-wilson.co.uk
2019-01-29 20:52:29 +00:00
erq->flags = request->fence.flags;
erq->context = request->fence.context;
erq->seqno = request->fence.seqno;
erq->sched_attr = request->sched.attr;
erq->jiffies = request->emitted_jiffies;
erq->start = i915_ggtt_offset(request->ring->vma);
erq->head = request->head;
erq->tail = request->tail;
rcu_read_lock();
erq->pid = ctx->pid ? pid_nr(ctx->pid) : 0;
rcu_read_unlock();
}
static void engine_record_requests(struct intel_engine_cs *engine,
struct i915_request *first,
struct drm_i915_error_engine *ee)
{
struct i915_request *request;
int count;
count = 0;
request = first;
list_for_each_entry_from(request, &engine->timeline.requests, link)
count++;
if (!count)
return;
ee->requests = kcalloc(count, sizeof(*ee->requests), GFP_ATOMIC);
if (!ee->requests)
return;
ee->num_requests = count;
count = 0;
request = first;
list_for_each_entry_from(request, &engine->timeline.requests, link) {
if (count >= ee->num_requests) {
/*
* If the ring request list was changed in
* between the point where the error request
* list was created and dimensioned and this
* point then just exit early to avoid crashes.
*
* We don't need to communicate that the
* request list changed state during error
* state capture and that the error state is
* slightly incorrect as a consequence since we
* are typically only interested in the request
* list state at the point of error state
* capture, not in any changes happening during
* the capture.
*/
break;
}
record_request(request, &ee->requests[count++]);
}
ee->num_requests = count;
}
static void error_record_engine_execlists(struct intel_engine_cs *engine,
struct drm_i915_error_engine *ee)
{
const struct intel_engine_execlists * const execlists = &engine->execlists;
unsigned int n;
for (n = 0; n < execlists_num_ports(execlists); n++) {
struct i915_request *rq = port_request(&execlists->port[n]);
if (!rq)
break;
record_request(rq, &ee->execlist[n]);
}
ee->num_ports = n;
}
static void record_context(struct drm_i915_error_context *e,
struct i915_gem_context *ctx)
{
if (ctx->pid) {
struct task_struct *task;
rcu_read_lock();
task = pid_task(ctx->pid, PIDTYPE_PID);
if (task) {
strcpy(e->comm, task->comm);
e->pid = task->pid;
}
rcu_read_unlock();
}
e->hw_id = ctx->hw_id;
e->sched_attr = ctx->sched;
e->guilty = atomic_read(&ctx->guilty_count);
e->active = atomic_read(&ctx->active_count);
}
static void request_record_user_bo(struct i915_request *request,
struct drm_i915_error_engine *ee)
{
struct i915_capture_list *c;
struct drm_i915_error_object **bo;
long count, max;
max = 0;
for (c = request->capture_list; c; c = c->next)
max++;
if (!max)
return;
bo = kmalloc_array(max, sizeof(*bo), GFP_ATOMIC);
if (!bo) {
/* If we can't capture everything, try to capture something. */
max = min_t(long, max, PAGE_SIZE / sizeof(*bo));
bo = kmalloc_array(max, sizeof(*bo), GFP_ATOMIC);
}
if (!bo)
return;
count = 0;
for (c = request->capture_list; c; c = c->next) {
bo[count] = i915_error_object_create(request->i915, c->vma);
if (!bo[count])
break;
if (++count == max)
break;
}
ee->user_bo = bo;
ee->user_bo_count = count;
}
static struct drm_i915_error_object *
capture_object(struct drm_i915_private *dev_priv,
struct drm_i915_gem_object *obj)
{
if (obj && i915_gem_object_has_pages(obj)) {
struct i915_vma fake = {
.node = { .start = U64_MAX, .size = obj->base.size },
.size = obj->base.size,
.pages = obj->mm.pages,
.obj = obj,
};
return i915_error_object_create(dev_priv, &fake);
} else {
return NULL;
}
}
static void gem_record_rings(struct i915_gpu_state *error)
{
struct drm_i915_private *i915 = error->i915;
struct i915_ggtt *ggtt = &i915->ggtt;
int i;
for (i = 0; i < I915_NUM_ENGINES; i++) {
struct intel_engine_cs *engine = i915->engine[i];
struct drm_i915_error_engine *ee = &error->engine[i];
struct i915_request *request;
ee->engine_id = -1;
drm/i915: Allocate intel_engine_cs structure only for the enabled engines With the possibility of addition of many more number of rings in future, the drm_i915_private structure could bloat as an array, of type intel_engine_cs, is embedded inside it. struct intel_engine_cs engine[I915_NUM_ENGINES]; Though this is still fine as generally there is only a single instance of drm_i915_private structure used, but not all of the possible rings would be enabled or active on most of the platforms. Some memory can be saved by allocating intel_engine_cs structure only for the enabled/active engines. Currently the engine/ring ID is kept static and dev_priv->engine[] is simply indexed using the enums defined in intel_engine_id. To save memory and continue using the static engine/ring IDs, 'engine' is defined as an array of pointers. struct intel_engine_cs *engine[I915_NUM_ENGINES]; dev_priv->engine[engine_ID] will be NULL for disabled engine instances. There is a text size reduction of 928 bytes, from 1028200 to 1027272, for i915.o file (but for i915.ko file text size remain same as 1193131 bytes). v2: - Remove the engine iterator field added in drm_i915_private structure, instead pass a local iterator variable to the for_each_engine** macros. (Chris) - Do away with intel_engine_initialized() and instead directly use the NULL pointer check on engine pointer. (Chris) v3: - Remove for_each_engine_id() macro, as the updated macro for_each_engine() can be used in place of it. (Chris) - Protect the access to Render engine Fault register with a NULL check, as engine specific init is done later in Driver load sequence. v4: - Use !!dev_priv->engine[VCS] style for the engine check in getparam. (Chris) - Kill the superfluous init_engine_lists(). v5: - Cleanup the intel_engines_init() & intel_engines_setup(), with respect to allocation of intel_engine_cs structure. (Chris) v6: - Rebase. v7: - Optimize the for_each_engine_masked() macro. (Chris) - Change the type of 'iter' local variable to enum intel_engine_id. (Chris) - Rebase. v8: Rebase. v9: Rebase. v10: - For index calculation use engine ID instead of pointer based arithmetic in intel_engine_sync_index() as engine pointers are not contiguous now (Chris) - For appropriateness, rename local enum variable 'iter' to 'id'. (Joonas) - Use for_each_engine macro for cleanup in intel_engines_init() and remove check for NULL engine pointer in cleanup() routines. (Joonas) v11: Rebase. Cc: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Akash Goel <akash.goel@intel.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1476378888-7372-1-git-send-email-akash.goel@intel.com
2016-10-13 17:14:48 +00:00
if (!engine)
continue;
ee->engine_id = i;
error_record_engine_registers(error, engine, ee);
error_record_engine_execlists(engine, ee);
request = intel_engine_find_active_request(engine);
if (request) {
struct i915_gem_context *ctx = request->gem_context;
struct intel_ring *ring;
ee->vm = ctx->ppgtt ? &ctx->ppgtt->vm : &ggtt->vm;
record_context(&ee->context, ctx);
/* We need to copy these to an anonymous buffer
* as the simplest method to avoid being overwritten
* by userspace.
*/
ee->batchbuffer =
i915_error_object_create(i915, request->batch);
if (HAS_BROKEN_CS_TLB(i915))
ee->wa_batchbuffer =
i915_error_object_create(i915,
i915->gt.scratch);
request_record_user_bo(request, ee);
ee->ctx =
i915_error_object_create(i915,
request->hw_context->state);
error->simulated |=
i915_gem_context_no_error_capture(ctx);
ee->rq_head = request->head;
ee->rq_post = request->postfix;
ee->rq_tail = request->tail;
ring = request->ring;
ee->cpu_ring_head = ring->head;
ee->cpu_ring_tail = ring->tail;
ee->ringbuffer =
i915_error_object_create(i915, ring->vma);
engine_record_requests(engine, request, ee);
}
ee->hws_page =
i915_error_object_create(i915,
engine->status_page.vma);
ee->wa_ctx = i915_error_object_create(i915, engine->wa_ctx.vma);
ee->default_state = capture_object(i915, engine->default_state);
}
}
static void gem_capture_vm(struct i915_gpu_state *error,
struct i915_address_space *vm,
int idx)
{
struct drm_i915_error_buffer *active_bo;
struct i915_vma *vma;
int count;
count = 0;
drm/i915: Stop tracking MRU activity on VMA Our goal is to remove struct_mutex and replace it with fine grained locking. One of the thorny issues is our eviction logic for reclaiming space for an execbuffer (or GTT mmaping, among a few other examples). While eviction itself is easy to move under a per-VM mutex, performing the activity tracking is less agreeable. One solution is not to do any MRU tracking and do a simple coarse evaluation during eviction of active/inactive, with a loose temporal ordering of last insertion/evaluation. That keeps all the locking constrained to when we are manipulating the VM itself, neatly avoiding the tricky handling of possible recursive locking during execbuf and elsewhere. Note that discarding the MRU (currently implemented as a pair of lists, to avoid scanning the active list for a NONBLOCKING search) is unlikely to impact upon our efficiency to reclaim VM space (where we think a LRU model is best) as our current strategy is to use random idle replacement first before doing a search, and over time the use of softpinned 48b per-ppGTT is growing (thereby eliminating any need to perform any eviction searches, in theory at least) with the remaining users being found on much older devices (gen2-gen6). v2: Changelog and commentary rewritten to elaborate on the duality of a single list being both an inactive and active list. v3: Consolidate bool parameters into a single set of flags; don't comment on the duality of a single variable being a multiplicity of bits. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190128102356.15037-1-chris@chris-wilson.co.uk
2019-01-28 10:23:52 +00:00
list_for_each_entry(vma, &vm->bound_list, vm_link)
if (i915_vma_is_active(vma))
count++;
active_bo = NULL;
if (count)
active_bo = kcalloc(count, sizeof(*active_bo), GFP_ATOMIC);
if (active_bo)
drm/i915: Stop tracking MRU activity on VMA Our goal is to remove struct_mutex and replace it with fine grained locking. One of the thorny issues is our eviction logic for reclaiming space for an execbuffer (or GTT mmaping, among a few other examples). While eviction itself is easy to move under a per-VM mutex, performing the activity tracking is less agreeable. One solution is not to do any MRU tracking and do a simple coarse evaluation during eviction of active/inactive, with a loose temporal ordering of last insertion/evaluation. That keeps all the locking constrained to when we are manipulating the VM itself, neatly avoiding the tricky handling of possible recursive locking during execbuf and elsewhere. Note that discarding the MRU (currently implemented as a pair of lists, to avoid scanning the active list for a NONBLOCKING search) is unlikely to impact upon our efficiency to reclaim VM space (where we think a LRU model is best) as our current strategy is to use random idle replacement first before doing a search, and over time the use of softpinned 48b per-ppGTT is growing (thereby eliminating any need to perform any eviction searches, in theory at least) with the remaining users being found on much older devices (gen2-gen6). v2: Changelog and commentary rewritten to elaborate on the duality of a single list being both an inactive and active list. v3: Consolidate bool parameters into a single set of flags; don't comment on the duality of a single variable being a multiplicity of bits. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190128102356.15037-1-chris@chris-wilson.co.uk
2019-01-28 10:23:52 +00:00
count = capture_error_bo(active_bo,
count, &vm->bound_list,
ACTIVE_ONLY);
else
count = 0;
error->active_vm[idx] = vm;
error->active_bo[idx] = active_bo;
error->active_bo_count[idx] = count;
}
static void capture_active_buffers(struct i915_gpu_state *error)
{
int cnt = 0, i, j;
BUILD_BUG_ON(ARRAY_SIZE(error->engine) > ARRAY_SIZE(error->active_bo));
BUILD_BUG_ON(ARRAY_SIZE(error->active_bo) != ARRAY_SIZE(error->active_vm));
BUILD_BUG_ON(ARRAY_SIZE(error->active_bo) != ARRAY_SIZE(error->active_bo_count));
/* Scan each engine looking for unique active contexts/vm */
for (i = 0; i < ARRAY_SIZE(error->engine); i++) {
struct drm_i915_error_engine *ee = &error->engine[i];
bool found;
if (!ee->vm)
continue;
found = false;
for (j = 0; j < i && !found; j++)
found = error->engine[j].vm == ee->vm;
if (!found)
gem_capture_vm(error, ee->vm, cnt++);
}
}
static void capture_pinned_buffers(struct i915_gpu_state *error)
{
struct i915_address_space *vm = &error->i915->ggtt.vm;
struct drm_i915_error_buffer *bo;
struct i915_vma *vma;
drm/i915: Stop tracking MRU activity on VMA Our goal is to remove struct_mutex and replace it with fine grained locking. One of the thorny issues is our eviction logic for reclaiming space for an execbuffer (or GTT mmaping, among a few other examples). While eviction itself is easy to move under a per-VM mutex, performing the activity tracking is less agreeable. One solution is not to do any MRU tracking and do a simple coarse evaluation during eviction of active/inactive, with a loose temporal ordering of last insertion/evaluation. That keeps all the locking constrained to when we are manipulating the VM itself, neatly avoiding the tricky handling of possible recursive locking during execbuf and elsewhere. Note that discarding the MRU (currently implemented as a pair of lists, to avoid scanning the active list for a NONBLOCKING search) is unlikely to impact upon our efficiency to reclaim VM space (where we think a LRU model is best) as our current strategy is to use random idle replacement first before doing a search, and over time the use of softpinned 48b per-ppGTT is growing (thereby eliminating any need to perform any eviction searches, in theory at least) with the remaining users being found on much older devices (gen2-gen6). v2: Changelog and commentary rewritten to elaborate on the duality of a single list being both an inactive and active list. v3: Consolidate bool parameters into a single set of flags; don't comment on the duality of a single variable being a multiplicity of bits. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190128102356.15037-1-chris@chris-wilson.co.uk
2019-01-28 10:23:52 +00:00
int count;
drm/i915: Stop tracking MRU activity on VMA Our goal is to remove struct_mutex and replace it with fine grained locking. One of the thorny issues is our eviction logic for reclaiming space for an execbuffer (or GTT mmaping, among a few other examples). While eviction itself is easy to move under a per-VM mutex, performing the activity tracking is less agreeable. One solution is not to do any MRU tracking and do a simple coarse evaluation during eviction of active/inactive, with a loose temporal ordering of last insertion/evaluation. That keeps all the locking constrained to when we are manipulating the VM itself, neatly avoiding the tricky handling of possible recursive locking during execbuf and elsewhere. Note that discarding the MRU (currently implemented as a pair of lists, to avoid scanning the active list for a NONBLOCKING search) is unlikely to impact upon our efficiency to reclaim VM space (where we think a LRU model is best) as our current strategy is to use random idle replacement first before doing a search, and over time the use of softpinned 48b per-ppGTT is growing (thereby eliminating any need to perform any eviction searches, in theory at least) with the remaining users being found on much older devices (gen2-gen6). v2: Changelog and commentary rewritten to elaborate on the duality of a single list being both an inactive and active list. v3: Consolidate bool parameters into a single set of flags; don't comment on the duality of a single variable being a multiplicity of bits. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190128102356.15037-1-chris@chris-wilson.co.uk
2019-01-28 10:23:52 +00:00
count = 0;
list_for_each_entry(vma, &vm->bound_list, vm_link)
count++;
bo = NULL;
drm/i915: Stop tracking MRU activity on VMA Our goal is to remove struct_mutex and replace it with fine grained locking. One of the thorny issues is our eviction logic for reclaiming space for an execbuffer (or GTT mmaping, among a few other examples). While eviction itself is easy to move under a per-VM mutex, performing the activity tracking is less agreeable. One solution is not to do any MRU tracking and do a simple coarse evaluation during eviction of active/inactive, with a loose temporal ordering of last insertion/evaluation. That keeps all the locking constrained to when we are manipulating the VM itself, neatly avoiding the tricky handling of possible recursive locking during execbuf and elsewhere. Note that discarding the MRU (currently implemented as a pair of lists, to avoid scanning the active list for a NONBLOCKING search) is unlikely to impact upon our efficiency to reclaim VM space (where we think a LRU model is best) as our current strategy is to use random idle replacement first before doing a search, and over time the use of softpinned 48b per-ppGTT is growing (thereby eliminating any need to perform any eviction searches, in theory at least) with the remaining users being found on much older devices (gen2-gen6). v2: Changelog and commentary rewritten to elaborate on the duality of a single list being both an inactive and active list. v3: Consolidate bool parameters into a single set of flags; don't comment on the duality of a single variable being a multiplicity of bits. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190128102356.15037-1-chris@chris-wilson.co.uk
2019-01-28 10:23:52 +00:00
if (count)
bo = kcalloc(count, sizeof(*bo), GFP_ATOMIC);
if (!bo)
return;
drm/i915: Stop tracking MRU activity on VMA Our goal is to remove struct_mutex and replace it with fine grained locking. One of the thorny issues is our eviction logic for reclaiming space for an execbuffer (or GTT mmaping, among a few other examples). While eviction itself is easy to move under a per-VM mutex, performing the activity tracking is less agreeable. One solution is not to do any MRU tracking and do a simple coarse evaluation during eviction of active/inactive, with a loose temporal ordering of last insertion/evaluation. That keeps all the locking constrained to when we are manipulating the VM itself, neatly avoiding the tricky handling of possible recursive locking during execbuf and elsewhere. Note that discarding the MRU (currently implemented as a pair of lists, to avoid scanning the active list for a NONBLOCKING search) is unlikely to impact upon our efficiency to reclaim VM space (where we think a LRU model is best) as our current strategy is to use random idle replacement first before doing a search, and over time the use of softpinned 48b per-ppGTT is growing (thereby eliminating any need to perform any eviction searches, in theory at least) with the remaining users being found on much older devices (gen2-gen6). v2: Changelog and commentary rewritten to elaborate on the duality of a single list being both an inactive and active list. v3: Consolidate bool parameters into a single set of flags; don't comment on the duality of a single variable being a multiplicity of bits. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190128102356.15037-1-chris@chris-wilson.co.uk
2019-01-28 10:23:52 +00:00
error->pinned_bo_count =
capture_error_bo(bo, count, &vm->bound_list, PINNED_ONLY);
error->pinned_bo = bo;
}
static void capture_uc_state(struct i915_gpu_state *error)
{
struct drm_i915_private *i915 = error->i915;
struct i915_error_uc *error_uc = &error->uc;
/* Capturing uC state won't be useful if there is no GuC */
if (!error->device_info.has_guc)
return;
error_uc->guc_fw = i915->guc.fw;
error_uc->huc_fw = i915->huc.fw;
/* Non-default firmware paths will be specified by the modparam.
* As modparams are generally accesible from the userspace make
* explicit copies of the firmware paths.
*/
error_uc->guc_fw.path = kstrdup(i915->guc.fw.path, GFP_ATOMIC);
error_uc->huc_fw.path = kstrdup(i915->huc.fw.path, GFP_ATOMIC);
error_uc->guc_log = i915_error_object_create(i915, i915->guc.log.vma);
}
/* Capture all registers which don't fit into another category. */
static void capture_reg_state(struct i915_gpu_state *error)
{
struct drm_i915_private *dev_priv = error->i915;
int i;
/* General organization
* 1. Registers specific to a single generation
* 2. Registers which belong to multiple generations
* 3. Feature specific registers.
* 4. Everything else
* Please try to follow the order.
*/
/* 1: Registers specific to a single generation */
if (IS_VALLEYVIEW(dev_priv)) {
error->gtier[0] = I915_READ(GTIER);
error->ier = I915_READ(VLV_IER);
error->forcewake = I915_READ_FW(FORCEWAKE_VLV);
}
if (IS_GEN(dev_priv, 7))
error->err_int = I915_READ(GEN7_ERR_INT);
if (INTEL_GEN(dev_priv) >= 8) {
error->fault_data0 = I915_READ(GEN8_FAULT_TLB_DATA0);
error->fault_data1 = I915_READ(GEN8_FAULT_TLB_DATA1);
}
if (IS_GEN(dev_priv, 6)) {
error->forcewake = I915_READ_FW(FORCEWAKE);
error->gab_ctl = I915_READ(GAB_CTL);
error->gfx_mode = I915_READ(GFX_MODE);
}
/* 2: Registers which belong to multiple generations */
if (INTEL_GEN(dev_priv) >= 7)
error->forcewake = I915_READ_FW(FORCEWAKE_MT);
if (INTEL_GEN(dev_priv) >= 6) {
error->derrmr = I915_READ(DERRMR);
error->error = I915_READ(ERROR_GEN6);
error->done_reg = I915_READ(DONE_REG);
}
if (INTEL_GEN(dev_priv) >= 5)
error->ccid = I915_READ(CCID(RENDER_RING_BASE));
/* 3: Feature specific registers */
if (IS_GEN_RANGE(dev_priv, 6, 7)) {
error->gam_ecochk = I915_READ(GAM_ECOCHK);
error->gac_eco = I915_READ(GAC_ECO_BITS);
}
/* 4: Everything else */
if (INTEL_GEN(dev_priv) >= 11) {
error->ier = I915_READ(GEN8_DE_MISC_IER);
error->gtier[0] = I915_READ(GEN11_RENDER_COPY_INTR_ENABLE);
error->gtier[1] = I915_READ(GEN11_VCS_VECS_INTR_ENABLE);
error->gtier[2] = I915_READ(GEN11_GUC_SG_INTR_ENABLE);
error->gtier[3] = I915_READ(GEN11_GPM_WGBOXPERF_INTR_ENABLE);
error->gtier[4] = I915_READ(GEN11_CRYPTO_RSVD_INTR_ENABLE);
error->gtier[5] = I915_READ(GEN11_GUNIT_CSME_INTR_ENABLE);
error->ngtier = 6;
} else if (INTEL_GEN(dev_priv) >= 8) {
error->ier = I915_READ(GEN8_DE_MISC_IER);
for (i = 0; i < 4; i++)
error->gtier[i] = I915_READ(GEN8_GT_IER(i));
error->ngtier = 4;
} else if (HAS_PCH_SPLIT(dev_priv)) {
error->ier = I915_READ(DEIER);
error->gtier[0] = I915_READ(GTIER);
error->ngtier = 1;
} else if (IS_GEN(dev_priv, 2)) {
error->ier = I915_READ16(IER);
} else if (!IS_VALLEYVIEW(dev_priv)) {
error->ier = I915_READ(IER);
}
error->eir = I915_READ(EIR);
error->pgtbl_er = I915_READ(PGTBL_ER);
}
static const char *
error_msg(struct i915_gpu_state *error, unsigned long engines, const char *msg)
{
int len;
int i;
for (i = 0; i < ARRAY_SIZE(error->engine); i++)
if (!error->engine[i].context.pid)
engines &= ~BIT(i);
len = scnprintf(error->error_msg, sizeof(error->error_msg),
"GPU HANG: ecode %d:%lx:0x%08x",
INTEL_GEN(error->i915), engines,
i915_error_generate_code(error, engines));
if (engines) {
/* Just show the first executing process, more is confusing */
i = __ffs(engines);
len += scnprintf(error->error_msg + len,
sizeof(error->error_msg) - len,
", in %s [%d]",
error->engine[i].context.comm,
error->engine[i].context.pid);
}
if (msg)
len += scnprintf(error->error_msg + len,
sizeof(error->error_msg) - len,
", %s", msg);
return error->error_msg;
}
static void capture_gen_state(struct i915_gpu_state *error)
{
struct drm_i915_private *i915 = error->i915;
error->awake = i915->gt.awake;
error->wakelock = atomic_read(&i915->runtime_pm.wakeref_count);
error->suspended = i915->runtime_pm.suspended;
error->iommu = -1;
#ifdef CONFIG_INTEL_IOMMU
error->iommu = intel_iommu_gfx_mapped;
#endif
error->reset_count = i915_reset_count(&i915->gpu_error);
error->suspend_count = i915->suspend_count;
memcpy(&error->device_info,
INTEL_INFO(i915),
sizeof(error->device_info));
memcpy(&error->runtime_info,
RUNTIME_INFO(i915),
sizeof(error->runtime_info));
error->driver_caps = i915->caps;
}
static void capture_params(struct i915_gpu_state *error)
{
i915_params_copy(&error->params, &i915_modparams);
}
static unsigned long capture_find_epoch(const struct i915_gpu_state *error)
{
unsigned long epoch = error->capture;
int i;
for (i = 0; i < ARRAY_SIZE(error->engine); i++) {
const struct drm_i915_error_engine *ee = &error->engine[i];
if (ee->hangcheck_timestamp &&
time_before(ee->hangcheck_timestamp, epoch))
epoch = ee->hangcheck_timestamp;
}
return epoch;
}
static void capture_finish(struct i915_gpu_state *error)
{
struct i915_ggtt *ggtt = &error->i915->ggtt;
const u64 slot = ggtt->error_capture.start;
ggtt->vm.clear_range(&ggtt->vm, slot, PAGE_SIZE);
}
2016-10-12 09:05:19 +00:00
static int capture(void *data)
{
struct i915_gpu_state *error = data;
2016-10-12 09:05:19 +00:00
error->time = ktime_get_real();
error->boottime = ktime_get_boottime();
error->uptime = ktime_sub(ktime_get(),
error->i915->gt.last_init_time);
error->capture = jiffies;
capture_params(error);
capture_gen_state(error);
capture_uc_state(error);
capture_reg_state(error);
gem_record_fences(error);
gem_record_rings(error);
capture_active_buffers(error);
capture_pinned_buffers(error);
2016-10-12 09:05:19 +00:00
error->overlay = intel_overlay_capture_error_state(error->i915);
error->display = intel_display_capture_error_state(error->i915);
error->epoch = capture_find_epoch(error);
capture_finish(error);
2016-10-12 09:05:19 +00:00
return 0;
}
#define DAY_AS_SECONDS(x) (24 * 60 * 60 * (x))
struct i915_gpu_state *
i915_capture_gpu_state(struct drm_i915_private *i915)
{
struct i915_gpu_state *error;
/* Check if GPU capture has been disabled */
error = READ_ONCE(i915->gpu_error.first_error);
if (IS_ERR(error))
return error;
error = kzalloc(sizeof(*error), GFP_ATOMIC);
if (!error) {
i915_disable_error_state(i915, -ENOMEM);
return ERR_PTR(-ENOMEM);
}
kref_init(&error->ref);
error->i915 = i915;
stop_machine(capture, error, NULL);
return error;
}
/**
* i915_capture_error_state - capture an error record for later analysis
* @i915: i915 device
* @engine_mask: the mask of engines triggering the hang
* @msg: a message to insert into the error capture header
*
* Should be called when an error is detected (either a hang or an error
* interrupt) to capture error state from the time of the error. Fills
* out a structure which becomes available in debugfs for user level tools
* to pick up.
*/
void i915_capture_error_state(struct drm_i915_private *i915,
unsigned long engine_mask,
const char *msg)
{
static bool warned;
struct i915_gpu_state *error;
unsigned long flags;
if (!i915_modparams.error_capture)
return;
if (READ_ONCE(i915->gpu_error.first_error))
return;
error = i915_capture_gpu_state(i915);
if (IS_ERR(error))
return;
dev_info(i915->drm.dev, "%s\n", error_msg(error, engine_mask, msg));
if (!error->simulated) {
spin_lock_irqsave(&i915->gpu_error.lock, flags);
if (!i915->gpu_error.first_error) {
i915->gpu_error.first_error = error;
error = NULL;
}
spin_unlock_irqrestore(&i915->gpu_error.lock, flags);
}
if (error) {
__i915_gpu_state_free(&error->ref);
return;
}
if (!warned &&
ktime_get_real_seconds() - DRIVER_TIMESTAMP < DAY_AS_SECONDS(180)) {
DRM_INFO("GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.\n");
DRM_INFO("Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel\n");
DRM_INFO("drm/i915 developers can then reassign to the right component if it's not a kernel issue.\n");
DRM_INFO("The gpu crash dump is required to analyze gpu hangs, so please always attach it.\n");
DRM_INFO("GPU crash dump saved to /sys/class/drm/card%d/error\n",
i915->drm.primary->index);
warned = true;
}
}
struct i915_gpu_state *
i915_first_error_state(struct drm_i915_private *i915)
{
struct i915_gpu_state *error;
spin_lock_irq(&i915->gpu_error.lock);
error = i915->gpu_error.first_error;
if (!IS_ERR_OR_NULL(error))
i915_gpu_state_get(error);
spin_unlock_irq(&i915->gpu_error.lock);
return error;
}
void i915_reset_error_state(struct drm_i915_private *i915)
{
struct i915_gpu_state *error;
spin_lock_irq(&i915->gpu_error.lock);
error = i915->gpu_error.first_error;
if (error != ERR_PTR(-ENODEV)) /* if disabled, always disabled */
i915->gpu_error.first_error = NULL;
spin_unlock_irq(&i915->gpu_error.lock);
if (!IS_ERR_OR_NULL(error))
i915_gpu_state_put(error);
}
void i915_disable_error_state(struct drm_i915_private *i915, int err)
{
spin_lock_irq(&i915->gpu_error.lock);
if (!i915->gpu_error.first_error)
i915->gpu_error.first_error = ERR_PTR(err);
spin_unlock_irq(&i915->gpu_error.lock);
}