summaryrefslogtreecommitdiff
path: root/libgralloc-qsd8k/gpu.cpp
blob: 90ff445a2c1e88d929c1baa73d7df8625efa3ac1 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
/*
 * Copyright (C) 2010 The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#include <limits.h>
#include <unistd.h>
#include <fcntl.h>

#include <sys/mman.h>

#include "gr.h"
#include "gpu.h"

gpu_context_t::gpu_context_t(Deps& deps, PmemAllocator& pmemAllocator,
        PmemAllocator& pmemAdspAllocator, const private_module_t* module) :
    deps(deps),
    pmemAllocator(pmemAllocator),
    pmemAdspAllocator(pmemAdspAllocator)
{
    // Zero out the alloc_device_t
    memset(static_cast<alloc_device_t*>(this), 0, sizeof(alloc_device_t));

    // Initialize the procs
    common.tag     = HARDWARE_DEVICE_TAG;
    common.version = 0;
    common.module  = const_cast<hw_module_t*>(&module->base.common);
    common.close   = gralloc_close;
    alloc          = gralloc_alloc;
    free           = gralloc_free;
}

int gpu_context_t::gralloc_alloc_framebuffer_locked(size_t size, int usage,
        buffer_handle_t* pHandle)
{
    private_module_t* m = reinterpret_cast<private_module_t*>(common.module);

    // we don't support allocations with both the FB and PMEM_ADSP flags
    if (usage & GRALLOC_USAGE_PRIVATE_PMEM_ADSP) {
        return -EINVAL;
    }

    // allocate the framebuffer
    if (m->framebuffer == NULL) {
        // initialize the framebuffer, the framebuffer is mapped once
        // and forever.
        int err = deps.mapFrameBufferLocked(m);
        if (err < 0) {
            return err;
        }
    }

    const uint32_t bufferMask = m->bufferMask;
    const uint32_t numBuffers = m->numBuffers;
    const size_t bufferSize = m->finfo.line_length * m->info.yres;
    if (numBuffers == 1) {
        // If we have only one buffer, we never use page-flipping. Instead,
        // we return a regular buffer which will be memcpy'ed to the main
        // screen when post is called.
        int newUsage = (usage & ~GRALLOC_USAGE_HW_FB) | GRALLOC_USAGE_HW_2D;
        return gralloc_alloc_buffer(bufferSize, newUsage, pHandle);
    }

    if (bufferMask >= ((1LU<<numBuffers)-1)) {
        // We ran out of buffers.
        return -ENOMEM;
    }

    // create a "fake" handles for it
    intptr_t vaddr = intptr_t(m->framebuffer->base);
    private_handle_t* hnd = new private_handle_t(dup(m->framebuffer->fd), size,
                                                 private_handle_t::PRIV_FLAGS_USES_PMEM |
                                                 private_handle_t::PRIV_FLAGS_FRAMEBUFFER);

    // find a free slot
    for (uint32_t i=0 ; i<numBuffers ; i++) {
        if ((bufferMask & (1LU<<i)) == 0) {
            m->bufferMask |= (1LU<<i);
            break;
        }
        vaddr += bufferSize;
    }

    hnd->base = vaddr;
    hnd->offset = vaddr - intptr_t(m->framebuffer->base);
    *pHandle = hnd;

    return 0;
}


int gpu_context_t::gralloc_alloc_framebuffer(size_t size, int usage,
        buffer_handle_t* pHandle)
{
    private_module_t* m = reinterpret_cast<private_module_t*>(common.module);
    pthread_mutex_lock(&m->lock);
    int err = gralloc_alloc_framebuffer_locked(size, usage, pHandle);
    pthread_mutex_unlock(&m->lock);
    return err;
}


int gpu_context_t::gralloc_alloc_buffer(size_t size, int usage, buffer_handle_t* pHandle)
{
    int err = 0;
    int flags = 0;

    int fd = -1;
    void* base = 0; // XXX JMG: This should change to just get an address from
                    // the PmemAllocator rather than getting the base & offset separately
    int offset = 0;
    int lockState = 0;

    size = roundUpToPageSize(size);

    if (usage & GRALLOC_USAGE_HW_TEXTURE) {
        // enable pmem in that case, so our software GL can fallback to
        // the copybit module.
        flags |= private_handle_t::PRIV_FLAGS_USES_PMEM;
    }

    if (usage & GRALLOC_USAGE_HW_2D) {
        flags |= private_handle_t::PRIV_FLAGS_USES_PMEM;
    }

    if (usage & GRALLOC_USAGE_PRIVATE_PMEM_ADSP) {
        flags |= private_handle_t::PRIV_FLAGS_USES_PMEM_ADSP;
        flags &= ~private_handle_t::PRIV_FLAGS_USES_PMEM;
    }

    private_module_t* m = reinterpret_cast<private_module_t*>(common.module);

    if ((flags & private_handle_t::PRIV_FLAGS_USES_PMEM) != 0 ||
        (flags & private_handle_t::PRIV_FLAGS_USES_PMEM_ADSP) != 0) {

        PmemAllocator* pma = 0;

        if ((flags & private_handle_t::PRIV_FLAGS_USES_PMEM) != 0) {
          if ((flags & private_handle_t::PRIV_FLAGS_USES_PMEM_ADSP) != 0) {
              LOGE("attempting to allocate a gralloc buffer with both the "
                   "USES_PMEM and USES_PMEM_ADSP flags.  Unsetting the "
                   "USES_PMEM_ADSP flag.");
              flags &= ~private_handle_t::PRIV_FLAGS_USES_PMEM_ADSP;
          }
          pma = &pmemAllocator;
        } else { // (flags & private_handle_t::PRIV_FLAGS_USES_PMEM_ADSP) != 0
          pma = &pmemAdspAllocator;
        }

        // PMEM buffers are always mmapped
        lockState |= private_handle_t::LOCK_STATE_MAPPED;

        // Allocate the buffer from pmem
        err = pma->alloc_pmem_buffer(size, usage, &base, &offset, &fd);
        if (err < 0) {
            if (((usage & GRALLOC_USAGE_HW_MASK) == 0) &&
                ((usage & GRALLOC_USAGE_PRIVATE_PMEM_ADSP) == 0)) {
                // the caller didn't request PMEM, so we can try something else
                flags &= ~private_handle_t::PRIV_FLAGS_USES_PMEM;
                err = 0;
                goto try_ashmem;
            } else {
                LOGE("couldn't open pmem (%s)", strerror(errno));
            }
        }
    } else {
try_ashmem:
        fd = deps.ashmem_create_region("gralloc-buffer", size);
        if (fd < 0) {
            LOGE("couldn't create ashmem (%s)", strerror(errno));
            err = -errno;
        }
    }

    if (err == 0) {
        private_handle_t* hnd = new private_handle_t(fd, size, flags);
        hnd->offset = offset;
        hnd->base = int(base)+offset;
        hnd->lockState = lockState;
        *pHandle = hnd;
    }

    LOGE_IF(err, "gralloc failed err=%s", strerror(-err));

    return err;
}

static inline size_t ALIGN(size_t x, size_t align) {
    return (x + align-1) & ~(align-1);
}

int gpu_context_t::alloc_impl(int w, int h, int format, int usage,
        buffer_handle_t* pHandle, int* pStride) {
    if (!pHandle || !pStride)
        return -EINVAL;

    size_t size, alignedw, alignedh;

    alignedw = ALIGN(w, 32);
    alignedh = ALIGN(h, 32);
    switch (format) {
        case HAL_PIXEL_FORMAT_RGBA_8888:
        case HAL_PIXEL_FORMAT_RGBX_8888:
        case HAL_PIXEL_FORMAT_BGRA_8888:
            size = alignedw * alignedh * 4;
            break;
        case HAL_PIXEL_FORMAT_RGB_888:
            size = alignedw * alignedh * 3;
            break;
        case HAL_PIXEL_FORMAT_RGB_565:
        case HAL_PIXEL_FORMAT_RGBA_5551:
        case HAL_PIXEL_FORMAT_RGBA_4444:
            size = alignedw * alignedh * 2;
            break;

        // adreno formats
        case HAL_PIXEL_FORMAT_YCrCb_420_SP_ADRENO:  // NV21
            size  = ALIGN(alignedw*alignedh, 4096);
            size += ALIGN(2 * ALIGN(w/2, 32) * ALIGN(h/2, 32), 4096);
            break;
        case HAL_PIXEL_FORMAT_YCbCr_420_SP_TILED:   // NV12
            // The chroma plane is subsampled,
            // but the pitch in bytes is unchanged
            // The GPU needs 4K alignment, but the video decoder needs 8K
            alignedw = ALIGN(w, 128);
            size  = ALIGN( alignedw * alignedh, 8192);
            size += ALIGN( alignedw * ALIGN(h/2, 32), 4096);
            break;

        case HAL_PIXEL_FORMAT_YV12:
            if ((w&1) || (h&1)) {
                LOGE("w or h is odd for HAL_PIXEL_FORMAT_YV12");
                return -EINVAL;
            }
            alignedw = ALIGN(w, 16);
            alignedh = h;
            size = alignedw*alignedh +
                    (ALIGN(alignedw/2, 16) * (alignedh/2))*2;
            break;

        default:
            LOGE("unrecognized pixel format: %d", format);
            return -EINVAL;
    }

    if ((ssize_t)size <= 0)
        return -EINVAL;

    int err;
    if (usage & GRALLOC_USAGE_HW_FB) {
        err = gralloc_alloc_framebuffer(size, usage, pHandle);
    } else {
        err = gralloc_alloc_buffer(size, usage, pHandle);
    }

    if (err < 0) {
        return err;
    }

    *pStride = alignedw;
    return 0;
}

int gpu_context_t::free_impl(private_handle_t const* hnd) {
    private_module_t* m = reinterpret_cast<private_module_t*>(common.module);
    if (hnd->flags & private_handle_t::PRIV_FLAGS_FRAMEBUFFER) {
        // free this buffer
        const size_t bufferSize = m->finfo.line_length * m->info.yres;
        int index = (hnd->base - m->framebuffer->base) / bufferSize;
        m->bufferMask &= ~(1<<index); 
    } else { 
        PmemAllocator* pmem_allocator = 0;
        if (hnd->flags & private_handle_t::PRIV_FLAGS_USES_PMEM) {
            pmem_allocator = &pmemAllocator;
        } else if (hnd->flags & private_handle_t::PRIV_FLAGS_USES_PMEM_ADSP) {
            pmem_allocator = &pmemAdspAllocator;
        }
        if (pmem_allocator) {
            pmem_allocator->free_pmem_buffer(hnd->size, (void*)hnd->base,
                    hnd->offset, hnd->fd);
        }
        deps.terminateBuffer(&m->base, const_cast<private_handle_t*>(hnd));
    }

    deps.close(hnd->fd);
    delete hnd; // XXX JMG: move this to the deps
    return 0;
}

/******************************************************************************
 * Static functions
 *****************************************************************************/

int gpu_context_t::gralloc_alloc(alloc_device_t* dev, int w, int h, int format,
        int usage, buffer_handle_t* pHandle, int* pStride)
{
    if (!dev) {
        return -EINVAL;
    }
    gpu_context_t* gpu = reinterpret_cast<gpu_context_t*>(dev);
    return gpu->alloc_impl(w, h, format, usage, pHandle, pStride);
}

int gpu_context_t::gralloc_free(alloc_device_t* dev,
                                    buffer_handle_t handle)
{
    if (private_handle_t::validate(handle) < 0)
        return -EINVAL;

    private_handle_t const* hnd = reinterpret_cast<private_handle_t const*>(handle);
    gpu_context_t* gpu = reinterpret_cast<gpu_context_t*>(dev);
    return gpu->free_impl(hnd);
}

/*****************************************************************************/

int gpu_context_t::gralloc_close(struct hw_device_t *dev)
{
    gpu_context_t* ctx = reinterpret_cast<gpu_context_t*>(dev);
    if (ctx) {
        /* TODO: keep a list of all buffer_handle_t created, and free them
         * all here.
         */
        delete ctx;
    }
    return 0;
}


gpu_context_t::Deps::~Deps() {}