完善资料让更多小伙伴认识你，还能领取20积分哦， 立即完善>
3天内不再提示
电子发烧友论坛|
嗨！我已经完成了我的申请，但在最后阶段我遇到了问题。我使用XC1.30和DSIC33 EP512MU810。我发现我的程序在我的应用程序的特定方面开始表现得非常怪异。在与我的代码斗争之后，我发现这是因为编译器。在某个时刻，如果我添加了更多的数据存储器（变量、数组或结构），程序在某个点上失败，但是XC1.30没有错误地编译，内存报告指示：数据存储器为62%，程序存储器为75%。程序存储器[Oracle＝0x3000，长度＝0x505FC]段地址长度（PC U）--------------------------Text 0x3000 0x5Da40x8C76（35958）.const 0x8Da40xA0C0xF12（3858）LbMxconst 0x97 B00x1Ac0x228（642）0533 C300 58AB0615 0x135C 0x13EEC 0x1DE62（122466）。nITS）长度（字节）（DEC）------------3281）文本0x34 dB80x9876 0xe4b1（58545）.Delay32 0x3e62e 0x1c 0x2a（42）.Sydi.0x3E64 A0x2 0x3（3）使用的总程序存储器（字节）：0x59172（364914）73% IVT存储器[原点＝0x4，长度＝0x1FC]段地址长度（PC单元）长度（字节）（-DEC）-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------IVT0x40x1300x1C8（456）Audio Flash存储器[Oracle＝0x7FC000，长度＝0x3FFA]段地址长度（PC单元）长度（字节）（DEC）-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------NSBS0x1000 0 0x2（2）。BS0x1002 0 0x35C2（13762）。数据0x41818 0 0x88（648）。数据0x77 A0 0 0x194（404）。BSS 0x7934 0 0xCC（204）053903C058AB06360X7A00，0x40 A40。地址对齐间隙总长度X48（72）。BS0x7C80 0 0x82/（44）。数据0x7d2e 0 0xa2（0）。数据0x7DD0*0x1a（α）。XAE（174）.0x7Aee 0 0x8e（142）. BSS 0x7B7C 0 0xBC（188）.数据0x7C38 0 (6).data 0x7e5a 0 0x4 (4).bss 0x7e5e 0 0x4 (4).data 0x7e62 0 0x8 (8).bss 0x7e6a 0 0xe (14).data 0x7e78 0 0x2 (2).bss 0x7e7a 0 0x2 (2).data 0x7e7c 0 0x2 (2).pbss 0x7e7e 0 0x2 (2).data 0x7e80 0 0x2 (2).pbss 0x7e82 0 0x2 (2).bss 0x7e84 0 0x2 (2).data 0x7e86 0 0x2 (2)_05B3DB4058ab0a06 0xce00 0 0x200 (512）02FIF3058AB0A07 0xD000 0 0x1000（4096）总数据存储器使用（字节）：0x8088（32904）61%动态存储器UAGERIEGION地址最大长度（DEC）--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------XC1.25，它报告内存错误，因此它是一个内存问题。Buff/Deult/Studio/Ext/ 488041537 /Fiel1.O:Link错误：无法分配Chans.PbSS，大小＝2字节，属性=持久构建/默认/生产/第二/ 488041537 /文件2. O：链接错误：无法分配节。大小＝2字节，属性=数据构建/默认/生产/ 488041537 /文件/文件2. O：链接错误：无法分配节。PBSS，大小＝2字节，属性=持久NBProject /MaFaCH默认值。MK：1263：目标“DIST/Deule/Prototh/Project”的配方。ON/OXEX/FIL3.0:FLIE3.0:链接错误：无法分配段。BSS，大小＝2字节，属性＝BSS使[2 ]：离开目录'c:/文件夹/Project。X'C30IUARTEP.EO:链接错误：无法分配节。数据，大小＝2字节，属性=数据链接错误：无法分配数据内存项Ct/MaCuff.DEFA.MK: 84：TAKE.CONFF’FuffEdTe[ 2 ]的配方：**[Dist/Deult/Projuts/Proj.x.Proto.HEX ]错误255：**[.Bug CONF]错误2:使**[构建IMPL]错误2IF，例如，我只减少一个数组，长度只有32个字节，问题消失了，所以我在内存问题的边缘。然而，数据内存使用率仅为62%！！！！为什么XC1.30编译，但程序工作不好。XC1.25给出了这些错误，但是当从数据存储器中删除一些字节时，它说它是在62%。XC1.26也表现为XC1.30i不能依赖XC1.30，因为它编译，但我的程序可能在某个地方出错，而不是真的信任。我遗漏了什么吗？谢谢你，我希望有人能澄清一下。
以上来自于百度翻译 以下为原文 Hi everobody!. I have got my application almost done but I am having problems at this final stage. I am using XC1.30 and DSPIC33EP512MU810. I have found that my program has started behaving very weird in specific aspects of my application. After struggling with my code I have found out that it is due to the compiler. At some point if I add more data memory (variables, arrays, or structures) the program fails at some point, but XC1.30 compiles with no error, and the memory report indicates: data memory at 62% and program memory at 75%. Program Memory [Origin = 0x3000, Length = 0x505fc] section address length (PC units) length (bytes) (dec) ------- ------- ----------------- -------------------- .text 0x3000 0x5da4 0x8c76 (35958) .const 0x8da4 0xa0c 0xf12 (3858) libmxconst 0x97b0 0x1ac 0x282 (642) _0533c30058ab0615 0x995c 0x13eec 0x1de62 (122466) .text 0x1d848 0x152da 0x1fc47 (130119) .dinit 0x32b22 0x2296 0x33e1 (13281) .text 0x34db8 0x9876 0xe4b1 (58545) .init.delay32 0x3e62e 0x1c 0x2a (42) .shared.dinit 0x3e64a 0x2 0x3 (3) Total program memory used (bytes): 0x59172 (364914) 73% Ivt Memory [Origin = 0x4, Length = 0x1fc] section address length (PC units) length (bytes) (dec) ------- ------- ----------------- -------------------- .ivt 0x4 0x130 0x1c8 (456) Auxflash Memory [Origin = 0x7fc000, Length = 0x3ffa] section address length (PC units) length (bytes) (dec) ------- ------- ----------------- -------------------- Total auxflash memory used (bytes): 0 (0) <1% Data Memory [Origin = 0x1000, Length = 0xd000] section address alignment gaps total length (dec) ------- ------- -------------- ------------------- .nbss 0x1000 0 0x2 (2) .bss 0x1002 0 0x35c2 (13762) .data 0x45c4 0 0x2f54 (12116) .bss 0x7518 0 0x288 (648) .data 0x77a0 0 0x194 (404) .bss 0x7934 0 0xcc (204) _053903C058ab0636 0x7a00 0 0x40 (64) .bss 0x7a40 0 0xae (174) .data 0x7aee 0 0x8e (142) .bss 0x7b7c 0 0xbc (188) .data 0x7c38 0 0x48 (72) .bss 0x7c80 0 0x82 (130) .data 0x7d02 0 0x2c (44) .bss 0x7d2e 0 0xa2 (162) .data 0x7dd0 0 0x1a (26) .bss 0x7dea 0 0x50 (80) program 0x7e3a 0 0xa (10) .bss 0x7e44 0 0xa (10) _0510918058ab062e 0x7e4e 0 0x6 (6) .bss 0x7e54 0 0x6 (6) .data 0x7e5a 0 0x4 (4) .bss 0x7e5e 0 0x4 (4) .data 0x7e62 0 0x8 (8) .bss 0x7e6a 0 0xe (14) .data 0x7e78 0 0x2 (2) .bss 0x7e7a 0 0x2 (2) .data 0x7e7c 0 0x2 (2) .pbss 0x7e7e 0 0x2 (2) .data 0x7e80 0 0x2 (2) .pbss 0x7e82 0 0x2 (2) .bss 0x7e84 0 0x2 (2) .data 0x7e86 0 0x2 (2) _05B3DB4058ab0a06 0xce00 0 0x200 (512) _02FEE30058ab0a07 0xd000 0 0x1000 (4096) Total data memory used (bytes): 0x8088 (32904) 61% Dynamic Memory Usage region address maximum length (dec) ------ ------- --------------------- heap 0 0 (0) stack 0x7e88 0x4f78 (20344) Maximum dynamic memory (bytes): 0x4f78 (20344) However when I compile using XC1.25, it reports a memory error, so it seems that is a memory problem. build/default/production/_ext/488041537/File1.o: Link Error: Could not allocate section .pbss, size = 2 bytes, attributes = persist build/default/production/_ext/488041537/File2.o: Link Error: Could not allocate section .data, size = 2 bytes, attributes = data build/default/production/_ext/488041537/File2.o: Link Error: Could not allocate section .pbss, size = 2 bytes, attributes = persist nbproject/Makefile-default.mk:1263: recipe for target 'dist/default/production/Project.X.production.hex' failed build/default/production/_ext/2016417865/File3.o: Link Error: Could not allocate section .bss, size = 2 bytes, attributes = bss make[2]: Leaving directory 'C:/Folder/Project.X' c30_uartep.eo: Link Error: Could not allocate section .data, size = 2 bytes, attributes = data Link Error: Could not allocate data memory nbproject/Makefile-default.mk:84: recipe for target '.build-conf' failed make[2]: *** [dist/default/production/Project.X.production.hex] Error 255 make[1]: *** [.build-conf] Error 2 make: *** [.build-impl] Error 2 If, for example, I decrease an array lenght only by 32 bytes, the problem disappears, so I am on the edge of the memory problem. However the data memory usage is only 62%!!! Why XC1.30 compiles but the program works bad. And XC1.25 gives those errors, but when removing a few bytes from the data memory it says that it is at 62%. XC1.26 behaves also as XC1.30 I can't rely on XC1.30 at all since it compiles but my program can run wrong at some point, not really trustwhorty. Am I missing anything?? Thanks in advance, i hope anyone may clarify somthing about it. CM 
0
|
|
相关推荐
11个回答
|
|
|
自1.30以来，XC8编译器发生了重大变化。我有一个版本多达1.33的许可证，发现一些问题用1.35（自由模式）消失了。您还可以尝试更新XC8 1.30支持的设备文件。我建议使用最新的XC8 1.41来查看它是否修复了您的问题。我不熟悉您的DSPIC设备，但这个问题可能与动态内存有关吗？动态内存用户地址最大长度（DEC）--------------------------HEP 0 0（0）堆栈0x7E88 0x4F78（20344）最大动态内存（字节）：0x4F78（20344）
以上来自于百度翻译 以下为原文 There are significant changes in the XC8 compiler since 1.30. I have a license for versions up to 1.33 and found some problems that went away with 1.35 (in free mode). You might also try updating the supported devices file for your XC8 1.30. I would suggest using the latest XC8 1.41 to see if it fixes your issue. I'm not familiar with your dsPIC device, but perhaps the problem has something to do with dynamic memory? Dynamic Memory Usage region address maximum length (dec) ------ ------- --------------------- heap 0 0 (0) stack 0x7e88 0x4f78 (20344) Maximum dynamic memory (bytes): 0x4f78 (20344) |
|
|
|
|
|
我猜你的意思是XC16。它是一个DSPIC，所以它必须是OP引用的16位编译器。只是我的2美分。
以上来自于百度翻译 以下为原文 Hi I guess you mean XC16. Its for a dsPIC so it must be the 16bit compiler the OP is refering. Just my 2 cents... Best regards Jorge |
|
|
|
|
|
是的，除了XC8或C18以外，我没有使用任何东西。我知道最新的XC16是1.30。
以上来自于百度翻译 以下为原文 Yes, I haven't used anything other than XC8 or C18. I see that the latest XC16 is 1.30. |
|
|
|
|
|
我的钱是在程序内存中的一个bug，它不应该。当内存随着每个源的改变而移动时，问题就会来来往往。
以上来自于百度翻译 以下为原文 My money is on a bug in the program accessing memory it shouldn't. Problems will come and go as the memory moves around with each change to the source. |
|
|
|
|
|
我想我已经发现数据变量必须从0x7FFF数据存储器地址中声明为“EDS”。无论如何，它将是一个错误，因为XC16 1.30没有报告错误并且显示了错误的APP功能性，但是XC16 1.25显示了具有这种“EDS”细节的内存分配错误。今天早上刚刚看到一个新的1.31版本已经发布了，我要去看看它是否被修复了。谢谢你的回答！溴
以上来自于百度翻译 以下为原文 Well, I think I have found out data variables have to be declared as "eds" from 0x7FFF data memory address. Anyway, it will be a bug, since XC16 1.30 doesn't report an error and shows bad app funcionality and however XC16 1.25 show a memory allocation error having this "eds" detail into account. I have seen just this morning a new 1.31 version has been released, I am gonna give it a go to see if it is fixed. Thanks for your answers! BR |
|
|
|
|
|
XC1.31显示了同样的问题。它不会报告内存溢出。
以上来自于百度翻译 以下为原文 XC1.31 shows the same problem. It doesn't report a memory overflow. |
|
|
|
|
|
你好，我刚开始在MPLABX得到一个更新消息，但是还没有下载它。所以我在使用之前最新的版本，并且有类似的问题。不幸的是，我不能说旧的MPLAB版本是否正确地存储了内存。我的DSPIC项目的量规显示了8。6%个RAM和90%个程序存储器。我可以在程序存储器中加一点，程序会编译得很好，但是在运行时会有问题。如果我给程序存储器增加一点，编译器就会给出错误，但是在这两个例子中，内存表显示的程序内存不超过91%个。RAM是S。IMILAR，因为如果我引入更多的全局变量，它将无法编译（至多），但是读数不超过86%。
以上来自于百度翻译 以下为原文 Hello, I just started getting an update message in MPLABX, but have not downloaded it yet. So I’m using the most current version prior to that, and have a similar problem. Unfortunately I can’t say if the older MPLAB versions got the memory gauges right or not. My gauges for a dsPic project show 86% RAM and 90% program memory. I can add to program memory a little, and the program will compile fine, but will have problems at run time. If I add a little more to program memory, the compiler with give errors, but on both examples, the memory gauges show no more than 91% program memory. The RAM is similar, in that it will fail to compile (at best) if I introduce any more global variables, but the gauge reads no more than the 86%. |
|
|
|
|
|
难道这不是堆栈溢出问题吗？添加变量、堆栈溢出和程序被破坏。
以上来自于百度翻译 以下为原文 Couldn't that be just stack overflow problem? You add variables, the stack overflows and program gets busted. |
|
|
|
|
|
嗨，NorthGuy，如果你回答我的问题，我就不会让人去窃听。这个问题看起来非常相似。要回答的是，我有一个情况，在函数的末尾在我的源代码的末尾添加一些工作代码，使函数挂起、崩溃或在代码被添加的地方起作用。写进空函数的完全相同的代码只调用FR。在同一点上，代码最初被添加，使事情像往常一样工作。这意味着我必须为空函数声明额外的私有变量，并调用另一个函数深，这不应该是任何解决方案。
以上来自于百度翻译 以下为原文 Hi NorthGuy, If you were answering me, I don’t men to thread hijack. The issue just seemed very similar. To answer, I’ve had a situation where adding some working code at the end of a function about halfway through my source made the function hang, crash, or otherwise act up where the code was added. The exact same code written into a void function and called only from the same point the code was originally added made things work as usual. That means I had to declare extra private variables for the void function, and call another function deep, which shouldn’t be a solution for anything. |
|
|
|
|
|
大家好，谢谢回答。我不认为这是一个堆栈溢出，因为我把堆栈内存增加到了32 K，而如果全局变量在运行时没有被使用，那么全局变量是一个堆栈负载吗？？不管怎样，如何正确看待MPLAB中的这个问题？现在我已经声明了很多变量，但是它有点…“Tyy”，因为我需要在函数域中重新声明函数来接收一个yxEdssl。在这个时候编程是有点危险的，我担心我的应用程序的任何部分都会腐烂，我无法承受测试。在任何时候我都会改变整个应用程序的行为，而且我对不同编译器的行为不太自信！溴
以上来自于百度翻译 以下为原文 Hi everyone, thanks for answering. I don't think it is a stack overflow since I increased the stack memory to 32K, and...are global variables a stack load if they are not used at that run time point??. Anyway, how is the proper way of looking at that problem in MPLAB? I have declared now lots of variables in the __eds__ domain, but it is a bit..."tiry" since I need to re-declarate the functions to receive a __eds__ in the argument field. At this point programing is a bit dangerous, I am afraid of any part of my app becoming corrupted and I can't afford to test the entire app behaviour at anytime I make changes. And I am not confident of different behaviours in different compilers! BR |
|
|
|
|
|
我仍然发现你原来的帖子惊人的相似，除了我只使用了一个版本的MPLABX。最近我发现一个限制我的全局变量被允许超过它应该是什么，程序仍然编译，但这是当它将重置在某个点在运行时。一个单独的无符号字符可以删除，程序正常运行。如果额外的无符号字符存在（顺便说一下，它是远程RAM属性），程序会崩溃，但是代码中有一些有趣的东西可以编译或编译。一个版本需要一些无用的东西。编译代码：只有当我删除额外的全局无符号字符时，我也可以删除上面无用的代码。
以上来自于百度翻译 以下为原文 I still find your original post strikingly similar except I’m only using one version of MPLabX. Recently I found out a limit to my global variables is allowed to be over what they should be, and the program still compiles, but that is when it will reset at some point at run time. I have found there’s one single unsigned char I can remove, and the program will function normally. If the extra unsigned char is present (it's far ram attribute by the way), the program will crash, but there are some funny things about the code that can make it compile or not compile. One version needed some useless code to compile: integer = 0; if (integer == 5) { // which it will never be DoSomethingQuickButUseless(); } // integer Only if I remove the extra global unsigned char, can I also remove the above useless code. |
|
|
|
|
只有小组成员才能发言，加入小组>>
5183 浏览 9 评论
2005 浏览 8 评论
1932 浏览 10 评论
请问是否能把一个ADC值转换成两个字节用来设置PWM占空比？
3178 浏览 3 评论
请问电源和晶体值之间有什么关系吗？PIC在正常条件下运行4MHz需要多少电压？
2229 浏览 5 评论
739浏览 1评论
626浏览 1评论
有偿咨询，关于MPLAB X IPE烧录PIC32MX所遇到的问题
510浏览 1评论
PIC Kit3出现目标设备ID(00000000)与预期的设备ID(02c20000)不匹配。是什么原因
637浏览 0评论
535浏览 0评论
/6 
关注我们的微信
下载发烧友APP
电子发烧友观察
版权所有 © 湖南华秋数字科技有限公司
电子发烧友 （电路图） 湘公网安备 43011202000918 号 电信与信息服务业务经营许可证：合字B2-20210191
工商网监
湘ICP备2023018690号
2740
评论
淘帖
721
