STM32的FSMC真是一个万能的总线控制器,不仅可以控制SRAM,NOR FLASH,NAND FLASH,PC Card,还能控制LCD,TFT.

一般越是复杂的东西,理解起来就很困难,但是使用上却很方便,如USB.

不过FSMC也有很诡异的地方.如
*(volatile uint16_t *)0x60400000=0x0; //
实际地址A21=1,而非A22.[注:0x60400000=0x60000000|(1UL<<22) ] *(volatile uint16_t *)
0x60800000=0x0; // 实际地址A22=1,而非A23 [注:0x60800000=0x60000000|(1UL<<23) ]
* 1
* 2
* 3
* 4
为什么呢?那时我还以为软件或硬件还是芯片有BUG, 
我就是从上面的不解中开始研究FSMC的…..

<>1.FSMC信号引脚


STM32的管脚排列很没有规律,而且分布在多个不同端口上,初始化要十分小心.需要用到的引脚都要先初始化成”复用功能推挽输出”模式.(GPIO_InitStructure.GPIO_Mode=GPIO_Mode_AF_PP

并且开启时钟 (RCC_APB2PeriphClockCmd(RCC_APB2Periph_GPIOx, ENABLE); )
像STM32F103Z(144脚)芯片有独立的地址和数据总线,而STM32F103V(100脚)就没有,
地址和数据总线要像51单片机一样分时复用,而在STM32F103R系列(64脚)就没有FSMC模块.

复用总线时管脚: 
PD14,//FSMC_DA0 
PD15,//FSMC_DA1 
PD0 ,//FSMC_DA2 
PD1 ,//FSMC_DA3 
PE7 ,//FSMC_DA4 
PE8 ,//FSMC_DA5 
PE9 ,//FSMC_DA6 
PE10,//FSMC_DA7 
PE11,//FSMC_DA8 
PE12,//FSMC_DA9 
PE13,//FSMC_DA10 
PE14,//FSMC_DA11 
PE15,//FSMC_DA12 
PD8 ,//FSMC_DA13 
PD9 ,//FSMC_DA14 
PD10,//FSMC_DA15 
PD11,//FSMC_A16 
PD12,//FSMC_A17 
PD13,//FSMC_A18 
PE3 ,//FSMC_A19 
PE4 ,//FSMC_A20 
PE5 ,//FSMC_A21 
PE6 ,//FSMC_A22 
PE2 ,//FSMC_A23 
PG13,//FSMC_A24//STM32F103Z 
PG14,//FSMC_A25//STM32F103Z

独立的地址总线管脚: 
[注:总线是16Bit情况下,FSMC通过FSMC_NBL1,FSMC_NBL0,区分高低字节.下面W代表WORD,即16BIT字.] 
PF0 ,//FSMC_A0 //2^1=2W =4 Bytes //144PIN STM32F103Z 
PF1 ,//FSMC_A1 //2^2=4W =8 Bytes//144PIN STM32F103Z 
PF2 ,//FSMC_A2 //2^3=8W= 16 Bytes //144PIN STM32F103Z 
PF3 ,//FSMC_A3 //2^4=16W =32 Bytes//144PIN STM32F103Z 
PF4 ,//FSMC_A4 //2^5=32W =64 Bytes//144PIN STM32F103Z 
PF5 ,//FSMC_A5 //2^6=64W =128 Bytes//144PIN STM32F103Z 
PF12,//FSMC_A6 //2^7=128W =256 Bytes //144PIN STM32F103Z 
PF13,//FSMC_A7 //2^8=256W =512 Bytes //144PIN STM32F103Z 
PF14,//FSMC_A8 //2^9= 512W =1k Bytes//144PIN STM32F103Z 
PF15,//FSMC_A9 //2^10=1kW =2k Bytes//144PIN STM32F103Z 
PG0 ,//FSMC_A10 //2^11=2kW =4k Bytes//144PIN STM32F103Z 
PG1 ,//FSMC_A11 //2^12=4kW =8k Bytes//144PIN STM32F103Z 
PG2 ,//FSMC_A12 //2^13=8kW =16k Bytes//144PIN STM32F103Z 
PG3 ,//FSMC_A13 //2^14=16kW =32k Bytes//144PIN STM32F103Z 
PG4 ,//FSMC_A14 //2^15=32kW =64k Bytes//144PIN STM32F103Z 
PG5 ,//FSMC_A15 //2^16=64kW =128k Bytes//144PIN STM32F103Z 
PD11,//FSMC_A16 //2^17=128kW =256k Bytes 
PD12,//FSMC_A17 //2^18=256kW =512k Bytes 
PD13,//FSMC_A18 //2^19=512kW =1M Bytes 
PE3 ,//FSMC_A19 //2^20=1MW =2M Bytes 
PE4 ,//FSMC_A20 //2^21=2MW =4M Bytes 
PE5 ,//FSMC_A21 //2^22=4MW =8M Bytes 
PE6 ,//FSMC_A22 //2^23=8MW =16M Bytes 
PE2 ,//FSMC_A23 //2^24=16MW =32M Bytes //100PIN STM32F103V MAX 
PG13,//FSMC_A24 //2^25=32MW =64M Bytes //144PIN STM32F103Z 
PG14,//FSMC_A25 //2^26=64MW =128M Bytes //144PIN STM32F103Z

独立的数据总线管脚: 
PD14,//FSMC_D0 
PD15,//FSMC_D1 
PD0 ,//FSMC_D2 
PD1 ,//FSMC_D3 
PE7 ,//FSMC_D4 
PE8 ,//FSMC_D5 
PE9 ,//FSMC_D6 
PE10,//FSMC_D7 
PE11,//FSMC_D8 
PE12,//FSMC_D9 
PE13,//FSMC_D10 
PE14,//FSMC_D11 
PE15,//FSMC_D12 
PD8 ,//FSMC_D13 
PD9 ,//FSMC_D14 
PD10,//FSMC_D15 
控制信号 
PD4,//FSMC_NOE,/RD 
PD5,//FSMC_NWE,/WR 
PB7,//FSMC_NADV,/ALE 
PE1,//FSMC_NBL1,/UB 
PE0,//FSMC_NBL0,/LB 
PD7,//FSMC_NE1,/CS1 
PG9,//FSMC_NE2,/CS2 
PG10,//FSMC_NE3,/CS3 
PG12,//FSMC_NE4,/CS4 
//PD3,//FSMC_CLK 
//PD6,//FSMC_NWAIT

<>2.地址的分配

地址与片选是挂勾的,也就是说器件挂载在哪个片选引脚上,就固定了访问地址范围和FsmcInitStructure.FSMC_Bank
//地址范围:0x60000000~0x63FFFFFF,片选引脚PD7(FSMC_NE1),最大支持容量64MB,
//[在STM32F103V(100脚)上地址范围为A0~A23,最大容量16MB] FsmcInitStructure.FSMC_Bank
=FSMC_Bank1_NORSRAM1;//地址范围:0x64000000~0x67FFFFFF, 片选引脚PG9(FSMC_NE2),最大支持容量64MB
FsmcInitStructure.FSMC_Bank =FSMC_Bank1_NORSRAM2;
//地址范围:0x68000000~0x6BFFFFFF,片选引脚PG10(FSMC_NE3),最大支持容量64MB
FsmcInitStructure.FSMC_Bank =FSMC_Bank1_NORSRAM3;
//地址范围:0x6C000000~0x6FFFFFFF,片选引脚(PG12 FSMC_NE4),最大支持容量64MB
FsmcInitStructure.FSMC_Bank =FSMC_Bank1_NORSRAM4;
* 1
* 2
* 3
* 4
* 5
* 6
* 7
* 8
* 9
* 10
* 11
* 12
* 13
<>3.时序测量

简单原理草图 
 
写数据的时序 

 
读数据的时序 
 
1.数据总线设定为16位宽情况下测量FSMC时序,即
FsmcInitStructure.FSMC_MemoryDataWidth = FSMC_MemoryDataWidth_16b;
* 1
使用逻辑分析仪测量(循环执行下面这条语句,下同)的波形
*(volatile uint16_t *)(0x60002468UL)=0xABCD;
* 1
* 2

可以看出NADV下降沿瞬间DATABUS上的数据被锁存器锁存,接着NWE低电平,总线输出0xABCD,数据0xABCD被写入0x1234这个地址.
*(volatile uint16_t*)(0x60002469UL )=0xABCD;
* 1
* 2

what?向这个地址写出现了两次总线操作.

为了一探究竟,我引出了控制线.
*(volatile uint16_t*)(0x60000468UL )=0xABCD;
* 1
* 2
向0x60000468UL写入0xABCD到底会发什么? 
 
从时序图中我们可以看到,
向0x60000468UL在地址(在范围:0x60000000~0x63FFFFFF内)写入数据,片选引脚PD7(FSMC_NE1)被拉低.而在这之前,数据总线上先产生0x234,于是在NADV下降沿瞬间,数据被锁存在地址锁存器上(A0~A15),与A16~A25(如果有配置的话,会在NE1下降沿同时送出)组合成完整的地址信号.然而有人会问这个0x234是哪来的,你是否注意到它正好等于0x468/2,难道是巧合吗?不是的,在16位数据总线情况下(NORSRAMInitStrc.FSMC_MemoryDataWidth=FSMC_MemoryDataWidth_16b;), 
像这样
*(volatile uint16_t*)(0x60000000|addr)=0xABCD;
* 1
* 2
写入一个值,实际在地址线上产生的值是addr/2(即addr>>2), 
所以如果我们一定要向addrx写入0xABCD则我们要这样写
*(volatile uint16_t*)(0x60000000|addrx<<1)=0xABCD;
* 1
NADV为高电平时, NEW被拉低,NOE为高,且NBL1,NBL0为低,随后数据总线线上产生0xABCD于是0xABCD被写进SRAM的地址0x234中

那如果我们向一个奇数地址像这样
*(volatile uint16_t*)(0x60000469UL )=0xABCD;写入值会发生什么呢?
* 1
* 2
 
从图中我们可以看到,STM32其实分成了两次字节写的过程,第一次向0x469/2写入0xCD,第二次向0x469/2+1写入0xAB, 

有人会问你为什么这样说,NWE为低时总线上不是0xCDAB吗?没错,但是注意NBL1,NBL0的电平组合,NBL1连接到SRAM的nUB,NBL0连接到SRAM的nLB.第一次NEW为低时NBL1为低,NBL0为高,0xCDAB的高位被写入SRAM的0x234,第二次NWE为低时NBL1为高,NBL0为低,0xCDAB的低位被写入SRAM的0x235. 
当我们查看反汇编时发现,指令是相同的
0x080036C4 0468 DCW 0x0468 0x080036C6 6000 DCW 0x6000 MOVW r0,#0xABCD LDR r1
,[pc,#420] ; @0x080036C4//r1=0x60000468 STRH r0,[r1,#0x00] 0x080036C4 0469 DCW
0x0469 0x080036C6 6000 DCW 0x6000 MOVW r0,#0xABCD LDR r1,[pc,#420] ;
@0x080036C4//r1=0x60000469 STRH r0,[r1,#0x00]
* 1
* 2
* 3
* 4
* 5
* 6
* 7
* 8
* 9
* 10
* 11
* 12
以上是写入的时序,下面测量读取的时序

首先我们向SRAM的真实地址0x234,0x235分别写入0x8824,0x6507
*(volatile uint16_t*)(0x60000000UL |0x234 <<1 )=0x8824; *(volatile uint16_t*)(
0x60000000UL |0x235 <<1 )=0x6507; *(volatile uint16_t*)(0x60000000UL |0x236 <<1
)=0x6735; *(volatile uint16_t*)(0x60000000UL |0x237 <<1 )=0x2003; *(volatile
uint16_t*)(0x60000000UL |0x238 <<1 )=0x6219;
* 1
* 2
* 3
* 4
* 5
然后读取:
tmp=*(volatile uint16_t*)(0x60000468UL );
* 1
* 2
 
如图tmp结果为0x8824 
再试
tmp=*(volatile uint16_t*)(0x60000469UL );
* 1
* 2

nUB=nLB=0;按16bit读 
从0x234读得0X8824取高字节”88”作tmp低8位 
从0x235读得0X6507取低字节”07”作tmp高8位 
最终tmp=0x0788

接下来验证更特殊的
*(volatile uint8_t*)(0x60000469UL )=0xABCD;
* 1
* 2
 
由于NBL1=0,NBL0=1,0xCD被写入0x234的高地址, 
数据总线上出现的值是0xCDNN, NN是随机数据,不过一般是和高位一样的值
*(volatile uint8_t*)(0x60000468UL )=0xABCD;
* 1
* 2
 
由于NBL1=1,NBL0=0,0xCD被写入0x234的低地址, 
数据总线上出现的值是0xNNCD,NN是随机数据

验证字节读取的 
首先我们向SRAM的真实地址0x234,0x235分别写入0x8824,0x6507
*(volatile uint16_t*)(0x60000000UL |0x234 <<1 )=0x8824; *(volatile uint16_t*)(
0x60000000UL |0x235 <<1 )=0x6507;
* 1
* 2
然后这样读取
tmp=*(volatile uint8_t*)(0x60000469UL );//对奇地址的单字节读取,数据总线的高8位被返回 tmp=0x88
* 1
* 2

tmp=*(volatile uint8_t*)(0x60000468UL );//对偶地址的单字节读取,数据总线的低8位被返回 tmp=0x24
* 1
* 2


还有更特殊的没有,有!
*(volatile int64_t*)(0x60000468UL)=0XABCDEF1234567890;//0XABCD EF12 3456
7890,如图,分别进行了4次操作才写完:
* 1
* 2

*(volatile int64_t*)(0x60000469UL)=0XABCDEF1234567890;//0XABCD EF12 3456
7890,如图,对奇地址写比偶地址多一次操作:
* 1
* 2

*(volatile uint16_t*)(0x60000000UL |0x234 <<1 )=0x8824; *(volatile uint16_t*)(
0x60000000UL |0x235 <<1 )=0x6507; *(volatile uint16_t*)(0x60000000UL |0x236 <<1
)=0x6735; *(volatile uint16_t*)(0x60000000UL |0x237 <<1 )=0x2003; *(volatile
uint16_t*)(0x60000000UL |0x238 <<1 )=0x6219;
* 1
* 2
* 3
* 4
* 5 tmp=*(volatile int64_t*)(0x60000469UL);// tmp=0x1920036735650788
* 1
* 2

tmp=*(volatile int64_t*)(0x60000468UL); //tmp=0x2003673565078824
* 1
* 2


1.数据总线设定为8位宽情况下测量FSMC时序,即 
FsmcInitStructure.FSMC_MemoryDataWidth = FSMC_MemoryDataWidth_8b;
*(volatile uint16_t*)(0x60000468UL )=0xABCD;
* 1
* 2

*(volatile uint16_t*)(0x60000469UL )=0xABCD;
* 1
* 2

*(volatile uint16_t*)(0x60000468UL )=0x3344; *(volatile
uint16_t*)(0x60000469UL )=0xABCD;
* 1
* 2
* 3
tmp=(volatile uint16_t)(0x60000469UL ); //tmp=0xabcd 

tmp=*(volatile uint16_t*)(0x60000468UL );
* 1
* 2
tmp=0xcd44
tmp=*(volatile uint8_t*)(0x60000468UL );
* 1
* 2
tmp=0x44 

tmp=*(volatile uint8_t*)(0x60000469UL ); tmp=0xcd
* 1
* 2
* 3

*(volatile uint8_t*)(0x60000469UL )=0xABCD;
* 1
* 2

*(volatile uint8_t*)(0x60000468UL )=0xABCD;
* 1
* 2

tmp=*(volatile uint64_t*)(0x60000468UL ); tmp=0x2003673565ABCD44
* 1
* 2
* 3

tmp=*(volatile uint64_t*)(0x60000469UL );//tmp=0x192003673565ABCD
* 1
* 2

*(volatile uint64_t*)(0x60000469UL )=0XABCDEF1234567890;
* 1
* 2

*(volatile uint64_t*)(0x60000468UL )=0XABCDEF1234567890;
* 1
* 2