--- tags: Linux2020 --- # I09: kcalc contributed by < `YLowy` > [code explain](https://hackmd.io/@YLowy/S1cmeqBtw) 順便對學習 kernel 時做些額外筆記 ## 前期準備 確認 ubuntu 版本 ``` $ uname -r 5.4.0-52-generic ``` 載入 linux headers （版本要對） ``` $ dpkg -L linux-headers-5.4.0-52-generic | grep "/lib/modules" /lib/modules /lib/modules/5.4.0-52-generic /lib/modules/5.4.0-52-generic/build ``` ## 取得 kcalc-fixed 程式碼並測試 順利 make 之後載入該 .ko 核心檔案 ``` $ sudo insmod calc.ko ``` 在核心所吐出訊息可看到其註冊成功 ``` $ dmesg [ 224.902852] calc: Initializing the module [ 224.902854] calc: registered correctly with major number 237 [ 224.902860] calc: device class registered correctly [ 224.902955] calc: device class created correctly ``` 在 /dev 目錄下可以觀察到存在 cal 檔案，這也是程式碼中所註冊的名稱 ``` /dev$ ls -al | grep calc crw------- 1 root root 237, 0 十一 9 23:46 calc ``` 為了方便測試更改讀寫權限，以方便在 user space 做讀寫 ``` /dev$ sudo chmod 0666 calc ``` ``` dev$ ls -al | grep calc crw-rw-rw- 1 root root 237, 0 十一 9 23:46 calc ``` 對該 calc 做寫入，這也是呼叫 write API 方式之一 ``` echo -ne "3*5\0" > /dev/calc ``` 此時可以對 calc 檔案做讀取，可以發現內已經有表達輸出式 ``` $ cat calc 64424509440 ``` 對應到 kernel 輸出訊息 ``` $ dmesg ``` ``` [ 749.002247] calc: Received 4 -> 3*5 [ 749.002262] calc: Result: 64424509440 [ 749.002270] calc: Device successfully closed [ 796.015508] calc: size: 12 result: 64424509440 [ 796.015548] calc: Device successfully closed ``` 卸載檔案 ``` $ sudo rmmod calc ` [ 1042.867565] kauditd_printk_skb: 3153 callbacks suppressed [ 1042.867568] audit: type=1400 audit(1604937605.718:3247): apparmor="DENIED" operation="capable" profile="/usr/sbin/cups-browsed" pid=4191 comm="cups-browsed" capability=23 capname="sys_nice" [ 1340.545745] calc: Goodbye! ``` --- ## 對於 kcalc-fixed 所做之資測 (benchforKcalc) 為了檢驗數值正確性，故需要對kcalc-fixed 做資測 [benchforKcalc](https://github.com/YLowy/benchforKcalc) 是為了比較 ＭathEx 以及 kcalc-fixed 所做測驗 ### benchforKcalc 解釋 ::: danger 1. 此程式碼必須在 calc 載入 kernel 情況下執行 2. 必須在其父目錄底下同時存在 MathEx 以及 kcalc-fixed 3. 會對 kcalc-fixed 做極小更動，以下講解 ::: ### 執行資測程式碼 make 之後生成的檔案 bench 會比較原本寫在裡面的資測內容做比對，進而判斷其正確性 ``` ~/linux2020/benchforkalc-fixed$ ./bench ---Start Bench--- BENCH [PASS] 10+1.1 MathEx: 11.10000 kcalc-fixed: 11.10000 BENCH [PASS] 10*1.1 MathEx: 11.00000 kcalc-fixed: 11.00000 BENCH [X] 10/1.1 MathEx: 9.09091 kcalc-fixed: 9.00000 BENCH [PASS] 10-1.1 MathEx: 8.90000 kcalc-fixed: 8.90000 BENCH [X] 1.1*1.1 MathEx: 1.21000 kcalc-fixed: 42949672.00000 ---End Bench--- ``` ### 對於 bench 程式碼解釋 ```c=1 #include "../MathEx/expression.h" #include "../MathEx/expression.c" #include "../kcalc-fixed/fixed-point.h" ``` `test_benchmark` 為接受輸入字串並比較之程式 ```c=37 static void test_benchmark(const char *s){ .. ... //此處計算 MathEx 輸出 for (long i = 0; i < N; i++) { result = expr_eval(e); } ... //此處計算 kcalc-fixed 輸出 kresult.data = write(fp,s,strlen(s)); float kout = eval(kresult); ... } ``` 只需要對 main 裡寫入測試內容即可 ```c=65 int main() { fp = open("/dev/calc",02); if(fp < 0){ exit(1); } printf("---Start Bench---\n"); /*--- write bench test code ---*/ test_benchmark("10+1.1"); test_benchmark("10*1.1"); test_benchmark("10/1.1"); test_benchmark("10-1.1"); test_benchmark("1.1*1.1"); /*------------------------------*/ printf("---End Bench---\n"); close(fp); return 0; } ``` ### 對 calc-fixed 程式碼之修改 原先對該 system call API 的 write 中該函式會回傳 len 也就是該輸入表達式長度，而現在我們目標為對呼叫該 system call 時可以取得該輸出表達式，以方便讓我們在資測時期做 eval 函式 ```c=86 static ssize_t dev_write(struct file *filep, const char *buffer, size_t len, loff_t *offset) { memset(message, 0, sizeof(char) * BUFF_SIZE); if (len >= BUFF_SIZE) { pr_alert("Expression too long"); return 0; } copy_from_user(message, buffer, len); pr_info("Received %ld -> %s\n", len, message); calc(); return len; } ``` 將 len 更改成為 result，如下 ： ```c=86 static ssize_t dev_write(struct file *filep, const char *buffer, size_t len, loff_t *offset) { memset(message, 0, sizeof(char) * BUFF_SIZE); if (len >= BUFF_SIZE) { pr_alert("Expression too long"); return 0; } copy_from_user(message, buffer, len); pr_info("Received %ld -> %s\n", len, message); calc(); return result; } ``` --- ## 擴充數值運算器功能 新增 sqrt($\sqrt{}{}$), Sigma ($\Sigma$), Pi ($\pi$) ### Sqrt ### Sigma ### Pi --- ## 錯誤修正 ### mul 給定程式碼中對於乘法少了小數與小樹之間的相乘（也就是缺少 ＋bd ） ```c=248 static uint64_t mult(uint64_t a, uint64_t b) { fixedp fa = {.data = a}, fb = {.data = b}; /* (a + b) * (c + d) = ac + ad + bc + bd */ fixedp result = {.inte = fa.inte * fb.inte}; uint64_t ad = (GET_NUM(a) >> 32) * GET_FRAC(b); uint64_t bc = GET_FRAC(a) * (GET_NUM(b) >> 32); return result.data + ad + bc; } ``` 更改成為 ```c=248 static uint64_t mult(uint64_t a, uint64_t b) { fixedp fa = {.data = a}, fb = {.data = b}; /* (a + b) * (c + d) = ac + ad + bc + bd */ fixedp result = {.inte = fa.inte * fb.inte}; uint64_t ad = (GET_NUM(a) >> 32) * GET_FRAC(b); uint64_t bc = GET_FRAC(a) * (GET_NUM(b) >> 32); uint64_t bd = GET_FRAC(a) * GET_FRAC(b); return result.data + ad + bc + (bd>>32); } ``` 但是這裡還是有錯，錯誤部分為兩點 1. overflow 問題 2. 負數整數乘法問題
Last changed by  
 
YLowy
282

Read more

eBPF

Debug always make developer sad, especially in a big system. How to debug in Linux Kernel? Tracing In early kernel patch, we have some tracing tools, e.g. strace, gdb. They are all based on the old tracing system call: ptrace. GDB gdb -q test.o strace

7/8/2022
How io_uring and eBPF Will Revolutionize Programming in Linux

refer: https://thenewstack.io/how-io_uring-and-ebpf-will-revolutionize-programming-in-linux/ io_uring Classic IO system call For UDP's receive system call, it can be divided into two stages, wait for data and copy data to user space. Kernel offered the following blocking system calls to deal with file descriptors, be they storage files or sockets: size_t read(int fd, void *buf, size_t count); size_t write(int fd, const void *buf, size_t count);

6/15/2022
Optimed Strlen (cont.)

1. uclib vs glib uclib ->https://git.uclibc-ng.org/git/uclibc-ng.git uClibc-ng is a small C library for developing embedded Linux systems. It is much smaller than the GNU C Library, but nearly all applications supported by glibc also work perfectly with uClibc-ng. In strlen() function, the "finding zero in long size string" is different between glib and uclib. See Here //uclib

6/6/2022
Optimed Strlen

We often use the strlen(str) function to get the length of the string. This is broad-use, and simply implemented. In C, we cannot consider a string as an object. We need to find the '\0'(NULL) character to determine the end of the string. C11 6.4.5 6. a byte or code of value zero is appended to each multibyte character sequence that results from a string literal or literals. C11 7.1.1 A string is a contiguous sequence of characters terminated by and including the first null character. The term multibyte string is sometimes used instead to emphasize special processing given to multibyte characters contained in the string or to avoid confusion with a wide string. A pointer to a string is a pointer to its initial (lowest addressed)

5/22/2022
Read more from YLowy
Published on HackMD