SohyeonKim(369)
-
[FreeBSD] Timing Services
3.6 Timing ServicesThe kernel provides several different timing services to processes. These services include timers that run in real time and timers that run only while a process is executing. Real TimeThe systemโs time offset since January 1, 1970, Universal Coordinated Time (UTC), also known as the Epoch, is returned by the system call gettimeofday. (ํ์ ์ธ๊ณ์) Most modern processors (includin..
2024.07.15 -
[FreeBSD] Memory Management Services
3.5 Memory-Management ServicesThe memory organization and layout associated with a FreeBSD process is shown in Figure 3.3. Each process begins execution with three memory segments: text, data, and stack. The data segment is divided into initialized data and uninitialized data (also known as bss). The text is read-only and is normally shared by all processes executing the file, whereas the data a..
2024.07.14 -
[wrtn] ๋คผํผ GenAI ์์ด๋์ดํค ๊ฒฝํฌ๋ํ๊ต ์์
2024. 05. 24 Friday๊ฒฝํฌ๋ ๋คผํผ ์์ด๋์ดํค ์์ ๐๋คผํผ ์บ ํผ์ค๋ฆฌ๋ ํ๋ ์ค๊ฐ์ฅ ์ค์ํ โจ ๊ต๋ด ์์ด๋์ดํค ์์ ์ ๊ฐ์ตํ์์ต๋๋ค :5์์ ์งํํ ์์ ๋ํ์ง๋ง, ์ํ ๊ธฐ๊ฐ ๋์์ ๋ฆฌ๋ฅผ ๋ชปํด๋๋ค๊ฐ, ์ด์ ์์ผ wrap-up์ ํด๋ด ๋๋ค.. ๐ฅณ๊ฒฐ๋ก : ์ฑ๊ณต์ ์ผ๋ก ๋ง๋ฌด๋ฆฌ๋์ด์, ๋ ๋คํ์ด๊ณ ๋ฟ๋ฏํ๋ค์ - ๐ฅณ์ค๋นํ๋ ๋์ ์ด์์ง ํ์๋ ์์ฃผ ํ๊ณ ,,ํ์ ์ค์์ ๋งํ๋ ๋ฐ๋์ ํ์ํ๋ถ๋ค๊ป๋์๋ ๋ง์ด ๋ฐ๊ณ , ๋คผํผ์์๋ ์ ๊ทน์ ์ผ๋ก ๋์์ฃผ์ ์ "์ ๋ง" ์ด๋ ต๊ฒ ์ด๋ฆฐ ๋ํ์๊ฑฐ๋ฑ์ .. . ์ด๋ฒ ๊ฒฝํ์ผ๋ก, ๊ต๋ด์์ ํ์ฌ๋ฅผ ์งํํ ๋์๋๋ฐ.๋.์ ๋จผ์ ํ์ํ๋ฅผ ์ปจํํด์ผ ํ๋๊ตฌ๋ .๋ผ์ ๋ฆฌ๊ฒ ๊นจ๋ฌ์๊ตฌ์ .. . โ ๏ธ ๐ ๊ทธ๋๋ ์ค์ค๋ก ๊ต๋ด ๋ํ๋ฅผ ๋ง๋ค์ด๋ณด๊ณ , ๊ตฌ์ฒด์ ์ธ ๊ธฐํ ..
2024.07.14 -
[FreeBSD] Clock Interrupts
3.4 Clock InterruptsThe system is driven by a clock that interrupts at regular intervals. Each interrupt is referred to as a tick. On the PC, the clock ticks 1000 times per second. At each tick, the system updates the current time of day as well as user-process and system timers. Handling 1000 interrupts per second can be time consuming. To reduce the interrupt load, the kernel computes the num..
2024.07.10 -
[FreeBSD] Traps and Interrupts
3.3 Traps and InterruptsTraps, like system calls, occur synchronously for a process. Traps normally occur because of unintentional errors, such as division by zero or indirection through an invalid pointer. The process becomes aware of the problem either by catching a signal or by being terminated. Traps can also occur because of a page fault, in which case the system makes the page available an..
2024.07.10 -
[FreeBSD] System Calls
3.2 System CallsThe most frequent trap into the kernel (after clock processing) is a request to do a system call. System performance requires that the kernel minimize the overhead in fielding and dispatching a system call. The system-call handler must do the following work:Verify that the parameters to the system call are located at a valid user address, and copy them from the userโs address spa..
2024.07.10 -
[FreeBSD] Run-time structure of the kernel
Run-Time OrganizationThe kernel can be logically divided into a top half and a bottom halfThe top half of the kernel provides services to processes in response to system calls or traps. This software can be thought of as a library of routines shared by all processes. The top half of the kernel executes in a privileged execution mode, in which it has access both to kernel data structures and to t..
2024.07.10 -
[FreeBSD] Kernel Organization
3.1 Kernel OrganizationThe FreeBSD kernel can be viewed as a service provider to user processes. Processes usually access these services through system calls. Some services, such as process scheduling and memory management, are implemented as processes that execute in kernel mode or as routines that execute periodically within the kernel.In this chapter, we describe how kernel services are provi..
2024.07.09 -
Byte-Pair Encoding tokenization and Tiktoken
Byte-Pair Encoding tokenizationhttps://youtu.be/HEikzVL-lZU์ด๋ป๊ฒ ํ ํฐํ์ ๋จ์๊ฐ ๊ฒฐ์ ๋๋์ง ์ ์ ์๋ค. + Byte-Pair Encoding (BPE) was initially developed as an algorithm to compress texts, and then used by OpenAI for tokenization when pretraining the GPT model. Itโs used by a lot of Transformer models, including GPT, GPT-2, RoBERTa, BART, and DeBERTa. 1) ์บ๋ฆญํฐ ๋ณ๋ก ๋ชจ๋ ๋ถ๋ฆฌํ๊ธฐ 2) Pair ๋จ์๋ก ๋น๋ ์ count 3) ๊ฐ์ฅ ๋ง์ ๋น๋๋ฅผ ๋ณด์ฌ์ฃผ๋ ..
2024.07.08 -
Transformer Tokenizer, Embedding and LLaMA
Tokenization and Embedding: Science Behind Large Language Model Every input that we are providing to GPT is nothing but a token (numerical id) or a sequence of tokens. GPT doesnโt understand the language the way humans do but it just processes sequence of numerical ids, that we call tokens. But how does it find the association among words(tokens) and provide human like response, here comes the c..
2024.07.06