1、The SPU actually loads two instructions at a time from a doubleword-aligned boundary.
SPU實際上會從一個雙字對齊的邊界開始一次加載兩條指令。

2、Here are some of the more common SPU intrinsics.
下面是比較常見的一些SPUintrinsic(類型沒有給出，這是因為它們大多為多態(tài))。

3、Debugging SPU programs creates a new set of problems.
調(diào)試s PU程序會出現(xiàn)一些新的問題。

4、The SPU processes buffer #1.
SPU處理一號緩沖區(qū)。

5、Also, exceptions resulting from executed code aren't reported to the SPU itself.
而且，執(zhí)行代碼產(chǎn)生的異常結(jié)果也不會向SPU進(jìn)行匯報。

6、The SPU waits for the buffer to fill up.
SPU等待緩沖區(qū)被填充。

7、Instead, the SPU can access any 128-bit word in the local store at L1 cache speed.
相反，SPU可以以L1緩存的速度來訪問本地存儲器中的任何128位的字。

8、The SPU has no special registers.
SPU沒有特殊的寄存器。

9、The SPU processes buffer #2.
SPU處理二號緩沖區(qū)。

10、The SPU waits for buffer #2 to finish filling.
SPU等待二號緩沖區(qū)被填充。

11、Often this is called by the SPU code itself.
通常這是由SPU代碼本身調(diào)用的。

12、The SPU has no real hardware support for branch hinting.
SPU對于分支提示并沒有沒有實際硬件支持。

13、The SPU queues a DMA PUT to transmit the buffer back to main memory.
SPU對DMA PUT進(jìn)行排隊，從而將緩沖區(qū)傳輸回主內(nèi)存。

14、It turns out that several SPU instructions deal with only one of the register's values.
幾個SPU指令只能處理這些寄存器值中的一個。

15、Remember that the SPU registers are 128 bits wide, but our value is only 32 bits long.
SPU寄存器均為128位寬，但我們的值只有32位長。

16、A signal might need to be delivered to a thread that is executing SPU code.
我們可能需要將信號發(fā)送給正在執(zhí)行SPU代碼的線程。

17、Every directory in spufs refers to a logical spu context.
spufs中的每個目錄都是指一個邏輯的spu上下文。

18、Branch hinting is necessary on the SPU because mispredicted branches come at a high cost.
分支提示之所以在SPU是必需的是因為預(yù)測錯誤的分支可能會帶來高耗費。

19、An SPU resembles a cross between a simple CPU design and a digital signal processor.
SPU就像是簡單的CPU設(shè)計與數(shù)字信號處理器之間的交叉。

20、The SPU queues a DMA GET to pull a portion of the problem data set from main memory to a buffer.
SPU對DMA GET進(jìn)行排隊，從而將涉及的數(shù)據(jù)集的一部分從主內(nèi)存?zhèn)鬏數(shù)揭粋€緩沖區(qū)。

21、The two transmission steps do not involve the SPU at all, but only the MFC.
兩個傳輸步驟根本沒有涉及SPU，只涉及MFC(這是比較大的SPE的一部分)。

22、The SPU waits for the buffer to finish being transmitted.
SPU等待緩沖區(qū)完成傳輸。

23、starts execution of SPU code.
開始執(zhí)行SPU代碼。

24、SPU code is compiled separately from the PPC code and gets loaded at run time.
SPU代碼與PPC代碼分開進(jìn)行編譯，并在運行時進(jìn)行加載。

SPU翻譯

【計】 標(biāo)量處理部件, 系統(tǒng)劃分部件, 系統(tǒng)處理部件 詳情