| /third_party/openGLES/extensions/NV/ |
| H A D | NV_texgen_emboss.txt | 70 NO. Just odd numbered units. This meets a constraint of the
74 Can you just enable one coordinate of a texture unit for embossing?
87 Could the emboss constant just be the reciprocal of the width and
|
| /third_party/openssl/crypto/asn1/ |
| H A D | a_int.c | 58 * complement and 0 by the add one (due to carry) so just copy as many trailing
151 * NULL just return length.
249 * Absolute value of INT64_MIN: we can't just use -INT64_MIN as gcc produces
|
| /third_party/openssl/crypto/bn/ |
| H A D | bn_div.c | 96 * just replace 0 with 1.
221 * just rely on bn_check_top() here (bn_check_top() works only for in BN_div()
426 * ignore top values of the bignums just sub the two BN_ULONG arrays in bn_div_fixed_top()
|
| /third_party/openssl/crypto/engine/ |
| H A D | eng_dyn.c | 133 * (this is a bug in the design of CRYPTO_EX_DATA). As such, we just declare
268 * release our just-created reference. in engine_load_dynamic_int()
527 * created leaks. We just have to fail where we are, after in dynamic_load()
|
| /third_party/openGLES/extensions/SGIX/ |
| H A D | SGIX_vertex_array_object.txt | 53 * Why can't you just use display lists for this?
84 * What about just accelerating vertex arrays in display lists? The idea
183 only after they are first bound (just as if the name were unused).
|
| /third_party/openssl/crypto/ |
| H A D | init.c | 477 * have created optsdone_lock yet so we can't use it). This is just an in OPENSSL_init_crypto()
483 * This doesn't matter - it just means we will run the full function in OPENSSL_init_crypto()
581 /* If called recursively from OBJ_ calls, just skip it. */ in OPENSSL_init_crypto()
|
| /third_party/openGLES/extensions/SGIS/ |
| H A D | SGIS_shared_multisample.txt | 101 available for copying outside the multisample subrectangle, just
129 and expensive just like multisample buffers. LBE apps still may
175 non-multisampled windowed apps just fine concurrently with a
|
| /third_party/skia/third_party/externals/angle2/src/common/third_party/smhasher/src/ |
| H A D | PMurHash.cpp | 14 off reading the original source. Just point your browser at:
32 Just copy PMurHash.h and PMurHash.c and you're ready to go.
121 /* Without a known fast bswap32 we're just as well off doing this */
|
| /third_party/skia/third_party/externals/angle2/src/compiler/translator/ |
| H A D | SymbolTable.cpp | 152 // it would have just been put in the symbol table. Otherwise, we're looking up an earlier in setFunctionParameterNamesFromDefinition()
362 ASSERT(level >= 0); // Just to be safe. Should not happen. in getDefaultPrecision()
403 // We need just one precision stack level for predefined precisions.
|
| /third_party/skia/third_party/externals/swiftshader/third_party/llvm-10.0/llvm/lib/CodeGen/GlobalISel/ |
| H A D | Utils.cpp | 95 // FIXME: Just bailing out like this here could be not enough, unless we in constrainOperandRegClass()
102 // rule producing such extra copies, not just an input GMIR with them as in constrainOperandRegClass()
399 // FP operations quiet. For now, just handle the ones inserted during in isKnownNeverNaN()
|
| /third_party/vk-gl-cts/external/vulkan-docs/src/appendices/ |
| H A D | VK_NV_device_generated_commands.adoc | 39 since an application is free to define its own data structures, and just
195 * one could just reference the inputs, and have the processing done in
291 The current focus is just to allow VertexInput changes as well as shaders,
|
| H A D | VK_NV_device_generated_commands.txt | 39 since an application is free to define its own data structures, and just
195 * one could just reference the inputs, and have the processing done in
291 The current focus is just to allow VertexInput changes as well as shaders,
|
| /third_party/skia/third_party/externals/angle2/src/tests/gl_tests/ |
| H A D | DrawElementsTest.cpp | 425 // Draw with just the second set of 6 items, then first 6, and then the entire index buffer in TEST_P()
439 // Draw with just the first 6 indices, and then with the entire index buffer in TEST_P()
451 // Draw with just the second set of 6 items, then first 6, and then the entire index buffer in TEST_P()
|
| /third_party/skia/site/docs/dev/design/raster_tragedy/ |
| H A D | _index.md | 51 of the glyphs for best effect. Otherwise the glyphs should just keep looking like they used to
68 this generally applies to the whole framebuffer, not just individual draws. For various reasons,
197 rather complicated way and amounts to an optical sizing tweak. Just as hinting can be used to
|
| /third_party/skia/src/gpu/geometry/ |
| H A D | GrShape.cpp | 46 // Use the path's full fill type instead of just whether or not it's inverted. in stateKey()
113 // Either just the starting point, or a line from the center to the start in simplifyArc()
451 // CreateDrawArcPath resets the output path and configures its fill type, so we just in asPath()
|
| /third_party/skia/src/image/ |
| H A D | SkImage_Lazy.cpp | 304 // life time of the proxy and not just upload. For non-DDL draws we should look into in textureProxyViewFromPlanes()
465 // will fall back to just using the non mipped proxy; See skbug.com/7094. in lockTextureProxyView()
508 // with *our* key, not a key derived from bitmap. We're just making the proxy here. in lockTextureProxyView()
|
| /third_party/skia/src/gpu/d3d/ |
| H A D | GrD3DCommandList.cpp | 97 // D3D will apply barriers in order so we can just add onto the end in resourceBarrier()
113 // D3D will apply barriers in order so we can just add onto the end in uavBarrier()
130 // D3D will apply barriers in order so we can just add onto the end in aliasingBarrier()
|
| /third_party/skia/src/core/ |
| H A D | SkEdge.cpp | 30 // away data in value, so just perform a modify up-shift in SkFDot6ToFixedDiv2()
254 * A and B at 1/2 of their actual value, and just apply a 2x scale during in setQuadraticWithoutUpdate()
399 // so we try just looking at the two off-curve points in setCubicWithoutUpdate()
|
| /third_party/skia/src/gpu/ops/ |
| H A D | TriangulatingPathRenderer.cpp | 337 // expensive so we just roll with it. in createNonAAMesh()
475 // just go ahead and use it.
481 // expensive so we just roll with it.
|
| /third_party/openssl/crypto/x509/ |
| H A D | x_name.c | 298 * comparison of Name structures can be rapidly performed by just using
393 /* If type not in bitmask just copy string across */ in asn1_string_canon()
435 /* If not ASCII set just copy across */ in asn1_string_canon()
|
| /third_party/openssl/doc/man3/ |
| H A D | EVP_PKEY_ASN1_METHOD.pod | 296 item_sign() did everything, OpenSSL internals just needs to pass the
307 OpenSSL internals should just sign using those algorithms.
321 item_sign() did everything, OpenSSL internals just needs to pass the
|
| H A D | d2i_X509.pod | 404 at I<*ppout>, and increments it to point after the data just written.
410 to the start of the data just written.
555 it was incremented after the call to point after the data just written.
|
| /third_party/openssl/doc/man7/ |
| H A D | provider-keymgmt.pod | 295 selector bits must consider all those subsets, not just one. An
299 RSA keys the function should just return 1 as the selected subset
315 RSA keys the function should just return 1 as there is nothing to validate for
|
| /third_party/openssl/ssl/ |
| H A D | s3_cbc.c | 323 * mac_end_offset is the index just past the end of the data to be MACed. in ssl3_cbc_digest_record()
397 * greater than |md_block_size|. However we add a sanity check just in ssl3_cbc_digest_record()
452 * and we're past the 0x80 value then just write zero. in ssl3_cbc_digest_record()
|
| /third_party/skia/third_party/externals/libjpeg-turbo/ |
| H A D | jidctred.c | 21 * 1x1 is trivial: just take the DC coefficient divided by 8.
57 * (With a reasonable C compiler, you can just rely on the FIX() macro...)
399 /* We hardly need an inverse DCT routine for this: just take the in jpeg_idct_1x1()
|