linux/drivers/gpu/drm/i915/i915_utils.h

425 lines
12 KiB
C
Raw Normal View History

/*
* Copyright © 2016 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.
*
*/
#ifndef __I915_UTILS_H
#define __I915_UTILS_H
#include <linux/list.h>
drm/i915: Allow a context to define its set of engines Over the last few years, we have debated how to extend the user API to support an increase in the number of engines, that may be sparse and even be heterogeneous within a class (not all video decoders created equal). We settled on using (class, instance) tuples to identify a specific engine, with an API for the user to construct a map of engines to capabilities. Into this picture, we then add a challenge of virtual engines; one user engine that maps behind the scenes to any number of physical engines. To keep it general, we want the user to have full control over that mapping. To that end, we allow the user to constrain a context to define the set of engines that it can access, order fully controlled by the user via (class, instance). With such precise control in context setup, we can continue to use the existing execbuf uABI of specifying a single index; only now it doesn't automagically map onto the engines, it uses the user defined engine map from the context. v2: Fixup freeing of local on success of get_engines() v3: Allow empty engines[] v4: s/nengine/num_engines/ v5: Replace 64 limit on num_engines with a note that execbuf is currently limited to only using the first 64 engines. v6: Actually use the engines_mutex to guard the ctx->engines. Testcase: igt/gem_ctx_engines 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/20190521211134.16117-2-chris@chris-wilson.co.uk
2019-05-21 21:11:26 +00:00
#include <linux/overflow.h>
#include <linux/sched.h>
#include <linux/types.h>
#include <linux/workqueue.h>
struct drm_i915_private;
#undef WARN_ON
/* Many gcc seem to no see through this and fall over :( */
#if 0
#define WARN_ON(x) ({ \
bool __i915_warn_cond = (x); \
if (__builtin_constant_p(__i915_warn_cond)) \
BUILD_BUG_ON(__i915_warn_cond); \
WARN(__i915_warn_cond, "WARN_ON(" #x ")"); })
#else
#define WARN_ON(x) WARN((x), "%s", "WARN_ON(" __stringify(x) ")")
#endif
#undef WARN_ON_ONCE
#define WARN_ON_ONCE(x) WARN_ONCE((x), "%s", "WARN_ON_ONCE(" __stringify(x) ")")
#define MISSING_CASE(x) WARN(1, "Missing case (%s == %ld)\n", \
__stringify(x), (long)(x))
void __printf(3, 4)
__i915_printk(struct drm_i915_private *dev_priv, const char *level,
const char *fmt, ...);
#define i915_report_error(dev_priv, fmt, ...) \
__i915_printk(dev_priv, KERN_ERR, fmt, ##__VA_ARGS__)
#if IS_ENABLED(CONFIG_DRM_I915_DEBUG)
int __i915_inject_load_error(struct drm_i915_private *i915, int err,
const char *func, int line);
#define i915_inject_load_error(_i915, _err) \
__i915_inject_load_error((_i915), (_err), __func__, __LINE__)
bool i915_error_injected(void);
#else
#define i915_inject_load_error(_i915, _err) 0
#define i915_error_injected() false
#endif
#define i915_inject_probe_failure(i915) i915_inject_load_error((i915), -ENODEV)
#define i915_probe_error(i915, fmt, ...) \
__i915_printk(i915, i915_error_injected() ? KERN_DEBUG : KERN_ERR, \
fmt, ##__VA_ARGS__)
#if defined(GCC_VERSION) && GCC_VERSION >= 70000
#define add_overflows_t(T, A, B) \
__builtin_add_overflow_p((A), (B), (T)0)
#else
#define add_overflows_t(T, A, B) ({ \
typeof(A) a = (A); \
typeof(B) b = (B); \
(T)(a + b) < a; \
})
#endif
#define add_overflows(A, B) \
add_overflows_t(typeof((A) + (B)), (A), (B))
#define range_overflows(start, size, max) ({ \
typeof(start) start__ = (start); \
typeof(size) size__ = (size); \
typeof(max) max__ = (max); \
(void)(&start__ == &size__); \
(void)(&start__ == &max__); \
start__ > max__ || size__ > max__ - start__; \
})
#define range_overflows_t(type, start, size, max) \
range_overflows((type)(start), (type)(size), (type)(max))
/* Note we don't consider signbits :| */
#define overflows_type(x, T) \
(sizeof(x) > sizeof(T) && (x) >> BITS_PER_TYPE(T))
drm/i915: Allow a context to define its set of engines Over the last few years, we have debated how to extend the user API to support an increase in the number of engines, that may be sparse and even be heterogeneous within a class (not all video decoders created equal). We settled on using (class, instance) tuples to identify a specific engine, with an API for the user to construct a map of engines to capabilities. Into this picture, we then add a challenge of virtual engines; one user engine that maps behind the scenes to any number of physical engines. To keep it general, we want the user to have full control over that mapping. To that end, we allow the user to constrain a context to define the set of engines that it can access, order fully controlled by the user via (class, instance). With such precise control in context setup, we can continue to use the existing execbuf uABI of specifying a single index; only now it doesn't automagically map onto the engines, it uses the user defined engine map from the context. v2: Fixup freeing of local on success of get_engines() v3: Allow empty engines[] v4: s/nengine/num_engines/ v5: Replace 64 limit on num_engines with a note that execbuf is currently limited to only using the first 64 engines. v6: Actually use the engines_mutex to guard the ctx->engines. Testcase: igt/gem_ctx_engines 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/20190521211134.16117-2-chris@chris-wilson.co.uk
2019-05-21 21:11:26 +00:00
static inline bool
__check_struct_size(size_t base, size_t arr, size_t count, size_t *size)
{
size_t sz;
if (check_mul_overflow(count, arr, &sz))
return false;
if (check_add_overflow(sz, base, &sz))
return false;
*size = sz;
return true;
}
/**
* check_struct_size() - Calculate size of structure with trailing array.
* @p: Pointer to the structure.
* @member: Name of the array member.
* @n: Number of elements in the array.
* @sz: Total size of structure and array
*
* Calculates size of memory needed for structure @p followed by an
* array of @n @member elements, like struct_size() but reports
* whether it overflowed, and the resultant size in @sz
*
* Return: false if the calculation overflowed.
*/
#define check_struct_size(p, member, n, sz) \
likely(__check_struct_size(sizeof(*(p)), \
sizeof(*(p)->member) + __must_be_array((p)->member), \
n, sz))
#define ptr_mask_bits(ptr, n) ({ \
unsigned long __v = (unsigned long)(ptr); \
(typeof(ptr))(__v & -BIT(n)); \
})
#define ptr_unmask_bits(ptr, n) ((unsigned long)(ptr) & (BIT(n) - 1))
#define ptr_unpack_bits(ptr, bits, n) ({ \
unsigned long __v = (unsigned long)(ptr); \
*(bits) = __v & (BIT(n) - 1); \
(typeof(ptr))(__v & -BIT(n)); \
})
#define ptr_pack_bits(ptr, bits, n) ({ \
unsigned long __bits = (bits); \
GEM_BUG_ON(__bits & -BIT(n)); \
((typeof(ptr))((unsigned long)(ptr) | __bits)); \
})
#define ptr_dec(ptr) ({ \
unsigned long __v = (unsigned long)(ptr); \
(typeof(ptr))(__v - 1); \
})
#define ptr_inc(ptr) ({ \
unsigned long __v = (unsigned long)(ptr); \
(typeof(ptr))(__v + 1); \
})
drm/i915/execlists: Preempt-to-busy When using a global seqno, we required a precise stop-the-workd event to handle preemption and unwind the global seqno counter. To accomplish this, we would preempt to a special out-of-band context and wait for the machine to report that it was idle. Given an idle machine, we could very precisely see which requests had completed and which we needed to feed back into the run queue. However, now that we have scrapped the global seqno, we no longer need to precisely unwind the global counter and only track requests by their per-context seqno. This allows us to loosely unwind inflight requests while scheduling a preemption, with the enormous caveat that the requests we put back on the run queue are still _inflight_ (until the preemption request is complete). This makes request tracking much more messy, as at any point then we can see a completed request that we believe is not currently scheduled for execution. We also have to be careful not to rewind RING_TAIL past RING_HEAD on preempting to the running context, and for this we use a semaphore to prevent completion of the request before continuing. To accomplish this feat, we change how we track requests scheduled to the HW. Instead of appending our requests onto a single list as we submit, we track each submission to ELSP as its own block. Then upon receiving the CS preemption event, we promote the pending block to the inflight block (discarding what was previously being tracked). As normal CS completion events arrive, we then remove stale entries from the inflight tracker. v2: Be a tinge paranoid and ensure we flush the write into the HWS page for the GPU semaphore to pick in a timely fashion. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190620142052.19311-1-chris@chris-wilson.co.uk
2019-06-20 14:20:51 +00:00
#define page_mask_bits(ptr) ptr_mask_bits(ptr, PAGE_SHIFT)
#define page_unmask_bits(ptr) ptr_unmask_bits(ptr, PAGE_SHIFT)
#define page_pack_bits(ptr, bits) ptr_pack_bits(ptr, bits, PAGE_SHIFT)
#define page_unpack_bits(ptr, bits) ptr_unpack_bits(ptr, bits, PAGE_SHIFT)
#define struct_member(T, member) (((T *)0)->member)
#define ptr_offset(ptr, member) offsetof(typeof(*(ptr)), member)
#define fetch_and_zero(ptr) ({ \
typeof(*ptr) __T = *(ptr); \
*(ptr) = (typeof(*ptr))0; \
__T; \
})
drm/i915: Introduce the i915_user_extension_method An idea for extending uABI inspired by Vulkan's extension chains. Instead of expanding the data struct for each ioctl every time we need to add a new feature, define an extension chain instead. As we add optional interfaces to control the ioctl, we define a new extension struct that can be linked into the ioctl data only when required by the user. The key advantage being able to ignore large control structs for optional interfaces/extensions, while being able to process them in a consistent manner. In comparison to other extensible ioctls, the key difference is the use of a linked chain of extension structs vs an array of tagged pointers. For example, struct drm_amdgpu_cs_chunk { __u32 chunk_id; __u32 length_dw; __u64 chunk_data; }; struct drm_amdgpu_cs_in { __u32 ctx_id; __u32 bo_list_handle; __u32 num_chunks; __u32 _pad; __u64 chunks; }; allows userspace to pass in array of pointers to extension structs, but must therefore keep constructing that array along side the command stream. In dynamic situations like that, a linked list is preferred and does not similar from extra cache line misses as the extension structs themselves must still be loaded separate to the chunks array. v2: Apply the tail call optimisation directly to nip the worry of stack overflow in the bud. v3: Defend against recursion. v4: Fixup local types to match new uabi Opens: - do we include the result as an out-field in each chain? struct i915_user_extension { __u64 next_extension; __u64 name; __s32 result; __u32 mbz; /* reserved for future use */ }; * Undecided, so provision some room for future expansion. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190322092325.5883-1-chris@chris-wilson.co.uk
2019-03-22 09:23:22 +00:00
/*
* container_of_user: Extract the superclass from a pointer to a member.
*
* Exactly like container_of() with the exception that it plays nicely
* with sparse for __user @ptr.
*/
#define container_of_user(ptr, type, member) ({ \
void __user *__mptr = (void __user *)(ptr); \
BUILD_BUG_ON_MSG(!__same_type(*(ptr), struct_member(type, member)) && \
drm/i915: Introduce the i915_user_extension_method An idea for extending uABI inspired by Vulkan's extension chains. Instead of expanding the data struct for each ioctl every time we need to add a new feature, define an extension chain instead. As we add optional interfaces to control the ioctl, we define a new extension struct that can be linked into the ioctl data only when required by the user. The key advantage being able to ignore large control structs for optional interfaces/extensions, while being able to process them in a consistent manner. In comparison to other extensible ioctls, the key difference is the use of a linked chain of extension structs vs an array of tagged pointers. For example, struct drm_amdgpu_cs_chunk { __u32 chunk_id; __u32 length_dw; __u64 chunk_data; }; struct drm_amdgpu_cs_in { __u32 ctx_id; __u32 bo_list_handle; __u32 num_chunks; __u32 _pad; __u64 chunks; }; allows userspace to pass in array of pointers to extension structs, but must therefore keep constructing that array along side the command stream. In dynamic situations like that, a linked list is preferred and does not similar from extra cache line misses as the extension structs themselves must still be loaded separate to the chunks array. v2: Apply the tail call optimisation directly to nip the worry of stack overflow in the bud. v3: Defend against recursion. v4: Fixup local types to match new uabi Opens: - do we include the result as an out-field in each chain? struct i915_user_extension { __u64 next_extension; __u64 name; __s32 result; __u32 mbz; /* reserved for future use */ }; * Undecided, so provision some room for future expansion. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190322092325.5883-1-chris@chris-wilson.co.uk
2019-03-22 09:23:22 +00:00
!__same_type(*(ptr), void), \
"pointer type mismatch in container_of()"); \
((type __user *)(__mptr - offsetof(type, member))); })
/*
* check_user_mbz: Check that a user value exists and is zero
*
* Frequently in our uABI we reserve space for future extensions, and
* two ensure that userspace is prepared we enforce that space must
* be zero. (Then any future extension can safely assume a default value
* of 0.)
*
* check_user_mbz() combines checking that the user pointer is accessible
* and that the contained value is zero.
*
* Returns: -EFAULT if not accessible, -EINVAL if !zero, or 0 on success.
*/
#define check_user_mbz(U) ({ \
typeof(*(U)) mbz__; \
get_user(mbz__, (U)) ? -EFAULT : mbz__ ? -EINVAL : 0; \
})
static inline u64 ptr_to_u64(const void *ptr)
{
return (uintptr_t)ptr;
}
#define u64_to_ptr(T, x) ({ \
typecheck(u64, x); \
(T *)(uintptr_t)(x); \
})
#define __mask_next_bit(mask) ({ \
int __idx = ffs(mask) - 1; \
mask &= ~BIT(__idx); \
__idx; \
})
drm/i915: Split execlist priority queue into rbtree + linked list All the requests at the same priority are executed in FIFO order. They do not need to be stored in the rbtree themselves, as they are a simple list within a level. If we move the requests at one priority into a list, we can then reduce the rbtree to the set of priorities. This should keep the height of the rbtree small, as the number of active priorities can not exceed the number of active requests and should be typically only a few. Currently, we have ~2k possible different priority levels, that may increase to allow even more fine grained selection. Allocating those in advance seems a waste (and may be impossible), so we opt for allocating upon first use, and freeing after its requests are depleted. To avoid the possibility of an allocation failure causing us to lose a request, we preallocate the default priority (0) and bump any request to that priority if we fail to allocate it the appropriate plist. Having a request (that is ready to run, so not leading to corruption) execute out-of-order is better than leaking the request (and its dependency tree) entirely. There should be a benefit to reducing execlists_dequeue() to principally using a simple list (and reducing the frequency of both rbtree iteration and balancing on erase) but for typical workloads, request coalescing should be small enough that we don't notice any change. The main gain is from improving PI calls to schedule, and the explicit list within a level should make request unwinding simpler (we just need to insert at the head of the list rather than the tail and not have to make the rbtree search more complicated). v2: Avoid use-after-free when deleting a depleted priolist v3: Michał found the solution to handling the allocation failure gracefully. If we disable all priority scheduling following the allocation failure, those requests will be executed in fifo and we will ensure that this request and its dependencies are in strict fifo (even when it doesn't realise it is only a single list). Normal scheduling is restored once we know the device is idle, until the next failure! Suggested-by: Michał Wajdeczko <michal.wajdeczko@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michał Winiarski <michal.winiarski@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Michał Winiarski <michal.winiarski@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20170517121007.27224-8-chris@chris-wilson.co.uk
2017-05-17 12:10:03 +00:00
static inline void __list_del_many(struct list_head *head,
struct list_head *first)
{
first->prev = head;
WRITE_ONCE(head->next, first);
}
/*
* Wait until the work is finally complete, even if it tries to postpone
* by requeueing itself. Note, that if the worker never cancels itself,
* we will spin forever.
*/
static inline void drain_delayed_work(struct delayed_work *dw)
{
do {
while (flush_delayed_work(dw))
;
} while (delayed_work_pending(dw));
}
static inline unsigned long msecs_to_jiffies_timeout(const unsigned int m)
{
unsigned long j = msecs_to_jiffies(m);
return min_t(unsigned long, MAX_JIFFY_OFFSET, j + 1);
}
/*
* If you need to wait X milliseconds between events A and B, but event B
* doesn't happen exactly after event A, you record the timestamp (jiffies) of
* when event A happened, then just before event B you call this function and
* pass the timestamp as the first argument, and X as the second argument.
*/
static inline void
wait_remaining_ms_from_jiffies(unsigned long timestamp_jiffies, int to_wait_ms)
{
unsigned long target_jiffies, tmp_jiffies, remaining_jiffies;
/*
* Don't re-read the value of "jiffies" every time since it may change
* behind our back and break the math.
*/
tmp_jiffies = jiffies;
target_jiffies = timestamp_jiffies +
msecs_to_jiffies_timeout(to_wait_ms);
if (time_after(target_jiffies, tmp_jiffies)) {
remaining_jiffies = target_jiffies - tmp_jiffies;
while (remaining_jiffies)
remaining_jiffies =
schedule_timeout_uninterruptible(remaining_jiffies);
}
}
/**
* __wait_for - magic wait macro
*
* Macro to help avoid open coding check/wait/timeout patterns. Note that it's
* important that we check the condition again after having timed out, since the
* timeout could be due to preemption or similar and we've never had a chance to
* check the condition before the timeout.
*/
#define __wait_for(OP, COND, US, Wmin, Wmax) ({ \
const ktime_t end__ = ktime_add_ns(ktime_get_raw(), 1000ll * (US)); \
long wait__ = (Wmin); /* recommended min for usleep is 10 us */ \
int ret__; \
might_sleep(); \
for (;;) { \
const bool expired__ = ktime_after(ktime_get_raw(), end__); \
OP; \
/* Guarantee COND check prior to timeout */ \
barrier(); \
if (COND) { \
ret__ = 0; \
break; \
} \
if (expired__) { \
ret__ = -ETIMEDOUT; \
break; \
} \
usleep_range(wait__, wait__ * 2); \
if (wait__ < (Wmax)) \
wait__ <<= 1; \
} \
ret__; \
})
#define _wait_for(COND, US, Wmin, Wmax) __wait_for(, (COND), (US), (Wmin), \
(Wmax))
#define wait_for(COND, MS) _wait_for((COND), (MS) * 1000, 10, 1000)
/* If CONFIG_PREEMPT_COUNT is disabled, in_atomic() always reports false. */
#if defined(CONFIG_DRM_I915_DEBUG) && defined(CONFIG_PREEMPT_COUNT)
# define _WAIT_FOR_ATOMIC_CHECK(ATOMIC) WARN_ON_ONCE((ATOMIC) && !in_atomic())
#else
# define _WAIT_FOR_ATOMIC_CHECK(ATOMIC) do { } while (0)
#endif
#define _wait_for_atomic(COND, US, ATOMIC) \
({ \
int cpu, ret, timeout = (US) * 1000; \
u64 base; \
_WAIT_FOR_ATOMIC_CHECK(ATOMIC); \
if (!(ATOMIC)) { \
preempt_disable(); \
cpu = smp_processor_id(); \
} \
base = local_clock(); \
for (;;) { \
u64 now = local_clock(); \
if (!(ATOMIC)) \
preempt_enable(); \
/* Guarantee COND check prior to timeout */ \
barrier(); \
if (COND) { \
ret = 0; \
break; \
} \
if (now - base >= timeout) { \
ret = -ETIMEDOUT; \
break; \
} \
cpu_relax(); \
if (!(ATOMIC)) { \
preempt_disable(); \
if (unlikely(cpu != smp_processor_id())) { \
timeout -= now - base; \
cpu = smp_processor_id(); \
base = local_clock(); \
} \
} \
} \
ret; \
})
#define wait_for_us(COND, US) \
({ \
int ret__; \
BUILD_BUG_ON(!__builtin_constant_p(US)); \
if ((US) > 10) \
ret__ = _wait_for((COND), (US), 10, 10); \
else \
ret__ = _wait_for_atomic((COND), (US), 0); \
ret__; \
})
#define wait_for_atomic_us(COND, US) \
({ \
BUILD_BUG_ON(!__builtin_constant_p(US)); \
BUILD_BUG_ON((US) > 50000); \
_wait_for_atomic((COND), (US), 1); \
})
#define wait_for_atomic(COND, MS) wait_for_atomic_us((COND), (MS) * 1000)
#define KHz(x) (1000 * (x))
#define MHz(x) KHz(1000 * (x))
#define KBps(x) (1000 * (x))
#define MBps(x) KBps(1000 * (x))
#define GBps(x) ((u64)1000 * MBps((x)))
static inline const char *yesno(bool v)
{
return v ? "yes" : "no";
}
static inline const char *onoff(bool v)
{
return v ? "on" : "off";
}
static inline const char *enableddisabled(bool v)
{
return v ? "enabled" : "disabled";
}
static inline void add_taint_for_CI(unsigned int taint)
{
/*
* The system is "ok", just about surviving for the user, but
* CI results are now unreliable as the HW is very suspect.
* CI checks the taint state after every test and will reboot
* the machine if the kernel is tainted.
*/
add_taint(taint, LOCKDEP_STILL_OK);
}
#endif /* !__I915_UTILS_H */