metal的带宽之前的blog有讲

这篇主要是vulkan 和ogles的解决方案

https://www.khronos.org/registry/vulkan/specs/1.1-extensions/man/html/VkAttachmentDescription.html

typedef struct VkAttachmentDescription {
VkAttachmentDescriptionFlags flags;
VkFormat format;
VkSampleCountFlagBits samples;
VkAttachmentLoadOp loadOp;
VkAttachmentStoreOp storeOp;
VkAttachmentLoadOp stencilLoadOp;
VkAttachmentStoreOp stencilStoreOp;
VkImageLayout initialLayout;
VkImageLayout finalLayout;
} VkAttachmentDescription;

这样 load store action和programmable blending

vulkan上都有对应实现了

(vulkan还能单独操作stencil的load store。。。)

这样 这两步在unity上应该 都是跨平台支持的 不需要分平台支持了

下面就是metal2那些 feature 在vulkan上的实现。。。

这里八成是要分平台写 因为metal里面的那些关键字别的平台又没有对应。。

=======================================================================================

https://developer.arm.com/solutions/graphics/developer-guides/understanding-render-passes/how-render-passes-work

opengl下 上述方案对应的解决方案

ogles是隐式控制pass切换的

  • The application calls glBindFramebuffer() to change the GL_FRAMEBUFFER or GL_DRAW_FRAMEBUFFER target.
  • The application calls glFramebufferTexture*() or glFramebufferRenderbuffer() to change the attachments of the currently bound draw framebuffer object when the drawing is queued.
  • The application calls eglSwapBuffers() to signal the end of a frame.
  • The application calls glFlush() or glFinish() to explicitly flush any queued rendering.
  • The application creates a glFenceSync() for some rendering in the current render pass and then calls glClientWaitSync() to wait on the completion of that work, or an equivalent behavior with a query object

vulkan和metal是显式控制切换的

this means avoiding:

  • Reading in older framebuffer values at the start of a render pass if they are going to be overdrawn.
  • Writing out values at the end of each render pass which are transient and are only needed for the duration of that render pass.

这两步就是load store

mali gpu

避免tile memory和ddr之间的traffic

这部分的traffic都是带宽 带宽除此之外还包括texture sample

https://developer.arm.com/solutions/graphics/developer-guides/understanding-render-passes/efficient-render-passes

1.避免切出去再切回来

2.尽量合并 共用renderpass

OGLES:

Load

  • glClear()
  • glClearBuffer*()
  • glInvalidateFramebuffer()

Caution: Only the start of tile clear is free. Calling glClear() or glClearBuffer*() after the first draw call in a render pass is not free, and this results in a per-fragment clear shader.

For Vulkan, set the loadOp for each attachment to either of:

  • VK_ATTACHMENT_LOAD_OP_CLEAR
  • VK_ATTACHMENT_LOAD_OP_DONT_CARE

Caution: If you call VkCmdClear*() commands to clear an attachment, or manually use a shader to write a constant color, it results in a per-fragment clear shader. To benefit from the fast fixed-function tile initialization, it is much more efficient to use the render pass loadOp operations.

不影响最终结果的情况下 mali里面invalidate operation 优于 a clear operation.

Store

用这个 glInvalidateFramebuffer

OpenGL ES, you can notify the driver that an attachment is transient by marking the content as invalid using a call to glInvalidateFramebuffer() as the last draw call in the render pass.

Note: If you write applications using OpenGL ES 2.0, you must use glDiscardFramebufferExt() from the [EXT_discard_framebuffer][EXT_dfb] extension.

depth能用这个么。。。GL_COLOR_ATTACHMENTiGL_DEPTH_ATTACHMENTGL_STENCIL_ATTACHMENT, and/or GL_DEPTH_STENCIL_ATTACHMENT....可以 处理depth

这样解决方案就ok了

For Vulkan, set the storeOp for each transient attachment to VK_ATTACHMENT_STORE_OP_DONT_CARE. For more efficiency, the application can even avoid allocating physical backing memory for transient attachments by allocating the backing memory using VK_MEMORY_PROPERTY_LAZILY_ALLOCATED_BIT and constructing the VkImagewith VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT.

这块是metal里 memoryless在vulkan上的解决方案

depthstencil在vulkan metal上的建议是 别包一起 我们确实遇到这个问题了 就一个mask却把depth一起开了 这种状况是要避免的

-----------------

    EXT_discard_framebuffer
这个unity有支持
===================
看了下unity代码 load store action它在ogles里也是支持的
那之后测下msaa下 是否生效就可以了
metal下msaa是不生效的 有bug要修
https://www.cnblogs.com/minggoddess/p/11447389.html

Load store action in vulkan & ogles 的解决方案的更多相关文章

  1. msaa mrt load store action unity

    unity buildin renderpipeline 和lightweight rp 对于开了msaa的rt 的load store action设置失效 buildin的时候set render ...

  2. unity里blit的load store action设置

    做blit的 load store action时 用 setrendertarget做 之后blit 参数用 BuiltinRenderTextureType.CurrentActive https ...

  3. TBDR下msaa 在metal vulkan和ogles的解决方案

    https://developer.arm.com/solutions/graphics/developer-guides/understanding-render-passes/multi-samp ...

  4. vulkan load store and memoryless

    https://www.jendrikillner.com/article_database/ https://community.arm.com/developer/tools-software/g ...

  5. Unordered load/store queue

    A method and processor for providing full load/store queue functionality to an unordered load/store  ...

  6. Load store and memoryless

    metal https://developer.apple.com/library/archive/documentation/3DDrawing/Conceptual/MTLBestPractice ...

  7. 报错:Unable to load configuration. - action - file:/E:/apache-tomcat-8.0.37/webapps/20161102-struts2-3/WEB-INF/classes/struts.xml:11:73

    第一种报错: 严重: Exception starting filter struts2Unable to load configuration. - action - file:/E:/apache ...

  8. 批量Load/Store指令的寻址方式

    批量Load/Store指令用于实现在一组寄存器和一块连续的内存单元之间传输数据.也称为多寄存器寻址方式,即一条指令可以完成多个寄存器值的传送.这种寻址方式可以用一条指令最多完成传送16个通用寄存器的 ...

  9. 出现错误:Unable to load configuration. - action - file:/E:/Java/Tomcat7.0/apache-tomcat-7.0.68-windows-x64/apache-tomcat-7.0.68/webapps/SSH2Integrate/WEB-INF/classes/struts.xml:8:43

    严重: Exception starting filter struts2 Unable to load configuration. - action - file:/E:/Java/Tomcat7 ...

随机推荐

  1. 在HTML5 中使用 kindeditor 的方法

    1.打开:http://kindeditor.net/ke4/examples/default.html 2.查看源代码,另存为 3.打开http://kindeditor.net/demo.php, ...

  2. Reactor系列(十一)take获取

    #java#reactor#take#获取# 获取Flux订阅数量 视频讲解: https://www.bilibili.com/video/av80322616/ FluxMonoTestCase. ...

  3. Reactor系列(九)collect集合

    #java#reactor#collect#collectMap#collectList# 集合 视频讲解: https://www.bilibili.com/video/av79860614/ Fl ...

  4. [转帖]IIS7配置Gzip压缩

    IIS7配置Gzip压缩 https://www.cnblogs.com/coce/p/6804373.html   II7中自带了gzip功能,理论上应该比ii6配置起来应该简单一点,但是容易出的问 ...

  5. MySQL数据库CPU飙升紧急处理方法

    MySQL数据库CPU飙升紧急处理方法 运行平稳的数据库,如果遇到CPU狂飙,到80%左右,那一定是开发写的烂SQL导致的,DBA首先要保证的是,数据库别跑挂了,所以我们要把那些运行慢的SQL杀死并记 ...

  6. oracle建表字段包含关键字注意事项

    SQL建表时最好不要把表定义成关键字字段 1 若不小心把表定义成关键字了,比如option, desc等等,oracle版本查询会有问题,短期内无法修改字段名称的,暂时用字段名+双引号解决: SELE ...

  7. Mysql解析json字符串/数组

    1 Mysql解析json字符串  解决方法:JSON_EXTRACT(原字段,'$.json字段名') 执行SQL: SELECT JSON_EXTRACT( t.result,'$.row'), ...

  8. Netty的学习

    看了几天高并发和NIO 今晚终于要开始学习Netty http://ifeve.com/netty5-user-guide/ Netty实现通信的步骤 1.创建两个NIO线程组,一个专门用于网络事件的 ...

  9. 第一次安装Dev-c++编译器如何设置?Dev-c++神奇功能

    安装完Dev-c++后,我们通常会根据自己的喜好调整设置,以下是个人的做法,欢迎参考. 1.字体 字体在安装界面就可以更改.默认字体是consolas,个人喜欢下面的courier new 有些字体很 ...

  10. 使用 js 简单的实现 bind、call 、aplly

    Function.prototype._call = function(obj,...arg){ var me = this; var k = Symbol("test"); // ...