From 6ab39f141f595f7bd5ac76a9d5e6a5e36223e1c8 Mon Sep 17 00:00:00 2001 From: Max Tyson <98maxt98@gmail.com> Date: Tue, 2 Sep 2025 15:47:57 +1200 Subject: [PATCH 1/7] Add resource management documentation --- 06_Userspace/06_Resource_Management | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) create mode 100644 06_Userspace/06_Resource_Management diff --git a/06_Userspace/06_Resource_Management b/06_Userspace/06_Resource_Management new file mode 100644 index 0000000..1c01bac --- /dev/null +++ b/06_Userspace/06_Resource_Management @@ -0,0 +1,24 @@ +# Resource Management +Intro Paragrapgh + +## Per Process Resource Table +- Each processs has a table mapping int -> rec +- Handles 0,1,2 reserved for stdio +- kernel uses this to clean up on exit + +## Resource Lifecycle +- Aquire with open rec syscall +- Use: generic read wrte +- Release with close resource + +## Resource Abstractation +- here is resource_t +- dispatch on type + +## Generic API +- the 3 funcs +- blokcing +- extended: resource specific syscalls / flags passed + +## Data Copying +- Explain speed of direct buffer use vs copy into a kernel buffer From be639e6481b9b937c584b92657bc5b82a9013d31 Mon Sep 17 00:00:00 2001 From: Max Tyson <98maxt98@gmail.com> Date: Tue, 2 Sep 2025 18:04:24 +1200 Subject: [PATCH 2/7] Enhance Resource Management documentation --- 06_Userspace/06_Resource_Management | 24 ----- 06_Userspace/06_Resource_Management.md | 129 +++++++++++++++++++++++++ 2 files changed, 129 insertions(+), 24 deletions(-) delete mode 100644 06_Userspace/06_Resource_Management create mode 100644 06_Userspace/06_Resource_Management.md diff --git a/06_Userspace/06_Resource_Management b/06_Userspace/06_Resource_Management deleted file mode 100644 index 1c01bac..0000000 --- a/06_Userspace/06_Resource_Management +++ /dev/null @@ -1,24 +0,0 @@ -# Resource Management -Intro Paragrapgh - -## Per Process Resource Table -- Each processs has a table mapping int -> rec -- Handles 0,1,2 reserved for stdio -- kernel uses this to clean up on exit - -## Resource Lifecycle -- Aquire with open rec syscall -- Use: generic read wrte -- Release with close resource - -## Resource Abstractation -- here is resource_t -- dispatch on type - -## Generic API -- the 3 funcs -- blokcing -- extended: resource specific syscalls / flags passed - -## Data Copying -- Explain speed of direct buffer use vs copy into a kernel buffer diff --git a/06_Userspace/06_Resource_Management.md b/06_Userspace/06_Resource_Management.md new file mode 100644 index 0000000..ed3cd96 --- /dev/null +++ b/06_Userspace/06_Resource_Management.md @@ -0,0 +1,129 @@ +# Resource Management +The kernel manages various things on behalf of the userspace process, including files, sockets, IPC, devices, and more, which we call 'resources'. It is a good idea to have a unified way to handle such a range of resources to reduce complexity. Rather than exposing a separate set of syscalls for each resource, a generic abstractation can be introduced to simplify everything whilst also keeping it all centralised in one place. + +To do this, we implement an API where every resource can be opened, read from, written to, and closed using the same syscalls. Through this design, the kernel is kept small, whilst also letting new resources be added in the future with minimal change to both kernel and user code. + +## Resource Abstractation +First, we can begin by defining a list of resource types: +``` +typedef enum { + FILE, + MESSAGE_ENDPOINT, + SHARED_MEM, + // can extend later +} resource_type_t; +``` + +And then internally each resource is represented as a `resource_t` struct: +``` +typedef struct { + resource_type_t type; + void* impl; +} resource_t; +``` +_NOTE: The `impl` pointer is where resource-specific structs can be stored, such as file descriptor states, IPC queues, shared memory regions, etc._ + +## Per Process Resource Table +With an abstract resource now defined, we can extend our previous definition of a process to include a **resource table**: +``` +typedef struct { +size_t pid; +status_t process_status; + +// Other fields + +resource_t* resource_table[MAX_RESOURCES]; +} process_t; +``` +Now each process has a resource table that is a map of integers, called handles, to the kernel resource objects. A handle is simply an identifier returned by the kernel when opening a resource that is later used by the user to inform what resource the operation should be performed upon. This way, the resource structure is not exposed to userspace. Because of this, the same handle number in different processes can refer to different resources. For example, in Unix, handles `0`, `1`, and `2` refer to stdio for each process. + +With this, we can also define a supporting function allowing the kernel to fetch a resource by handle: +``` +resource_t* get_resource(process_t* proc, int handle) { + + // Invalid handle + if (handle < 0 || handle >= MAX_HANDLES) + return NULL; + + return proc->table[handle]; +} +``` + + +## Resource Lifecycle +A resource follows a rather straightforward lifecycle, regardless of its type: +1. Firstly, a process acquires a handle by calling the `open_resource` system call. +2. While the handle is valid, the process can perform operations such as `read_resource` or `write_resource`. +3. Finally, when the process has finished using the resource, it calls `close_resource`, allowing the kernel to free any associated state. + +Typically, a process should `close()` a resource once it is done using it. However, that is not always the case, as processes may exit without cleaning up properly, and thus it is up to the kernel to ensure resources aren't leaked. +``` +for (int handle = 0; handle < MAX_RESOURCES; ++handle) { + + // Already closed or not used + if(process->resource_table[handle] == 0) + continue; + + close_resource(process, handle); +} +``` + + +## Generic API +The generic interface for a resource consists of four primary functions: `open`, `read`, `write`, and `close`. These functions form the minimum required API that every resource type must support. To begin the implementation of this, our `resource_t` needs extending to support these operations: +``` +typedef struct resource { + // ... + struct resource_ops* ops; +} resource_t; + +typedef struct resource_ops { + size_t (*read)(resource_t* res, void* buf, size_t len); + size_t (*write)(resource_t* res, const void* buf, size_t len); + void (*open)(resource_t* res); + void (*close)(resource_t* res); +} resource_ops_t; +``` +Operations are defined to be blocking by default, meaning that if a resource is not ready (for example, no data to read), the process is suspended until the operation can complete. Each resource type can override these generic operations to provide behavior specific to that resource. For example, a file resource can replace the write function with one that writes data to disk, while an IPC message resource could implement write to enqueue a message, allowing the same API call to behave differently depending on the resource. + +It has been left as an exercise to the user to decide on how they want to handle extending this design for extra resource-specific functionality (ie, renaming a file). There are two (of many) ways to do this, each with its own trade-off. Firstly, a simpler design would be to just add more syscalls to handle this; however, this means the ABI grows as your kernel manages more resources. Another approach would be to pass an additional `size_t flags` parameter and let the resource-specific operation handle it, which would keep the original four operations but with added complexity. + +The dispatch code would be as follows: +``` +size_t read_resource(process_t* proc, int handle, void* buf, size_t len) { + resource_t* res = get_resource(proc, handle); + if (!res || !res->ops->read) + return -1; + return res->ops->read(res, buf, len); +} + +size_t write_resource(process_t* proc, int handle, const void* buf, size_t len) { + resource_t* res = get_resource(proc, handle); + if (!res || !res->ops->write) + return -1; + return res->ops->write(res, buf, len); +} + +int open_resource(process_t* proc, resource_t* res) { + for (int i = 0; i < MAX_HANDLES; i++) { + if (proc->table[i] == NULL) { + proc->table[i] = res; + return i; // return handle + } + } + return -1; // no free slot +} + +void close_resource(process_t* proc, int handle) { + resource_t* res = get_resource(proc, handle); + if (!res) + return; + + if (res->ops->close) res->ops->close(res); // call resource-specific close + proc->table[handle] = NULL; +} +``` + +## Data Copying +Another thing left as an exercise to the user is to decide their method of copying data between userspace and the kernel. +One option is to use the userspace provided buffersm, which is efficient due to a single copy but does require sanitization of pointers and lengths to ensure safety. Some things to consider are other threads in the same address space modifying memory at the same address Another option is to copy into a kernel buffer first, which simplifies the sanitization but has the added overhead and loss of performance. From 80fbbb8c82ab0a7ffacad40b10772946fd1520cc Mon Sep 17 00:00:00 2001 From: Max Tyson <98maxt98@gmail.com> Date: Fri, 5 Sep 2025 18:24:38 +1200 Subject: [PATCH 3/7] Fix based on feedback --- 06_Userspace/06_Resource_Management.md | 70 ++++++++------------------ 1 file changed, 21 insertions(+), 49 deletions(-) diff --git a/06_Userspace/06_Resource_Management.md b/06_Userspace/06_Resource_Management.md index ed3cd96..eb45e51 100644 --- a/06_Userspace/06_Resource_Management.md +++ b/06_Userspace/06_Resource_Management.md @@ -4,7 +4,7 @@ The kernel manages various things on behalf of the userspace process, including To do this, we implement an API where every resource can be opened, read from, written to, and closed using the same syscalls. Through this design, the kernel is kept small, whilst also letting new resources be added in the future with minimal change to both kernel and user code. ## Resource Abstractation -First, we can begin by defining a list of resource types: +When talking about _resources_, we need a way to distinguish between the different types that the kernel may expect to provide to userspace. Each resource behaves differently internally, but from the view of the userspace process everyting should be acessable from the same set of syscalls. In order to achive this, we define an enum of resource types to allow the kernel to tag each resource with it's category. This way when a system call is made, the kernel knows how to dispatch the request. ``` typedef enum { FILE, @@ -13,15 +13,16 @@ typedef enum { // can extend later } resource_type_t; ``` +In this example `FILE` represents a file on the disk, `MESSAGE_ENDPOINT` is used for an IPC message queue and `SHARED_MEM` for a shared memory region between prcesses. As the kernel grows this struct can be extened to support more resource types. -And then internally each resource is represented as a `resource_t` struct: +Next, we need a generic representation of a resource inside the kernel. This can be defined by the `resource_t` struct: ``` typedef struct { resource_type_t type; void* impl; } resource_t; ``` -_NOTE: The `impl` pointer is where resource-specific structs can be stored, such as file descriptor states, IPC queues, shared memory regions, etc._ +The `type` field tells the kernel what kind of resource it is and the `impl` pointer allows the kernel to attach the resource specific implmentation of that resource. For example, a file's `impl` could point to a struct holding the file's offset and indoe or for shared memory it could point to the physical address of that region. ## Per Process Resource Table With an abstract resource now defined, we can extend our previous definition of a process to include a **resource table**: @@ -35,7 +36,7 @@ status_t process_status; resource_t* resource_table[MAX_RESOURCES]; } process_t; ``` -Now each process has a resource table that is a map of integers, called handles, to the kernel resource objects. A handle is simply an identifier returned by the kernel when opening a resource that is later used by the user to inform what resource the operation should be performed upon. This way, the resource structure is not exposed to userspace. Because of this, the same handle number in different processes can refer to different resources. For example, in Unix, handles `0`, `1`, and `2` refer to stdio for each process. +Now each process has a resource table that is a map of integers, called _handles_, to the kernel resource objects. A handle is simply an identifier returned by the kernel when opening a resource that is later used by the user to inform what resource the operation should be performed upon. This way, the resource structure is not exposed to userspace. Because of this, the same handle number in different processes can refer to different resources. For example, in Unix, handles `0`, `1`, and `2` refer to stdio for each process. With this, we can also define a supporting function allowing the kernel to fetch a resource by handle: ``` @@ -56,73 +57,44 @@ A resource follows a rather straightforward lifecycle, regardless of its type: 2. While the handle is valid, the process can perform operations such as `read_resource` or `write_resource`. 3. Finally, when the process has finished using the resource, it calls `close_resource`, allowing the kernel to free any associated state. -Typically, a process should `close()` a resource once it is done using it. However, that is not always the case, as processes may exit without cleaning up properly, and thus it is up to the kernel to ensure resources aren't leaked. -``` -for (int handle = 0; handle < MAX_RESOURCES; ++handle) { - - // Already closed or not used - if(process->resource_table[handle] == 0) - continue; - - close_resource(process, handle); -} -``` +Typically, a process should `close()` a resource once it is done using it. However, that is not always the case, as processes may exit without cleaning up properly, and thus it is up to the kernel to ensure resources aren't leaked. This could look like a loop through the process's resource table, calling `close_resource(process, handle);` for each open resource and letting the resource-specific `close()` function handle the work. ## Generic API -The generic interface for a resource consists of four primary functions: `open`, `read`, `write`, and `close`. These functions form the minimum required API that every resource type must support. To begin the implementation of this, our `resource_t` needs extending to support these operations: +Now that we have a way of representing resource, we need define how a process can interact with them. Generally, having a different syscall for each resource type can ___. Instead the kernel can expose a minmal and uniform API that every resource supports. The generic interface for a resource consists of four primary functions: `open`, `read`, `write`, and `close` and by restricting all resources to this same interface we can reduce the complexity of both the kernel and userspace. To begin the implementation of this, our `resource_t` needs extending with a table of function pointers to support these operations. Each resource can then provide it's own implementation of thse four functions whilst the generic interface remains the same. ``` typedef struct resource { - // ... - struct resource_ops* ops; + resource_type_t type; + void* impl; + struct resource_functions_t* funcs; } resource_t; -typedef struct resource_ops { +typedef struct resource_functions { size_t (*read)(resource_t* res, void* buf, size_t len); size_t (*write)(resource_t* res, const void* buf, size_t len); void (*open)(resource_t* res); void (*close)(resource_t* res); -} resource_ops_t; +} resource_functions_t; ``` +Here, `funcs` is the dispatch table that tells the kernel how to perform each operation for each resource. With this, each function pointer can be set differently dependin on wether the resource is a file, IPC endpoint or something else. + Operations are defined to be blocking by default, meaning that if a resource is not ready (for example, no data to read), the process is suspended until the operation can complete. Each resource type can override these generic operations to provide behavior specific to that resource. For example, a file resource can replace the write function with one that writes data to disk, while an IPC message resource could implement write to enqueue a message, allowing the same API call to behave differently depending on the resource. -It has been left as an exercise to the user to decide on how they want to handle extending this design for extra resource-specific functionality (ie, renaming a file). There are two (of many) ways to do this, each with its own trade-off. Firstly, a simpler design would be to just add more syscalls to handle this; however, this means the ABI grows as your kernel manages more resources. Another approach would be to pass an additional `size_t flags` parameter and let the resource-specific operation handle it, which would keep the original four operations but with added complexity. +It has been left as an exercise to the reader to decide on how they want to handle extending this design for extra resource-specific functionality (ie, renaming a file). There are two (of many) ways to do this, each with its own trade-off. Firstly, a simpler design would be to just add more syscalls to handle this; however, this means the ABI grows as your kernel manages more resources. Another approach would be to pass an additional `size_t flags` parameter and let the resource-specific operation handle it, which would keep the original four operations but with added complexity. -The dispatch code would be as follows: +On the kernel side of things, these syscalls can just act as dispatchers. For example, a `read_resource(...)` syscall would look up the process's resource table using the handle, retrieve the `resource_t`, and then forward the call to the correct, resource-specific, function: ``` size_t read_resource(process_t* proc, int handle, void* buf, size_t len) { resource_t* res = get_resource(proc, handle); - if (!res || !res->ops->read) - return -1; - return res->ops->read(res, buf, len); -} - -size_t write_resource(process_t* proc, int handle, const void* buf, size_t len) { - resource_t* res = get_resource(proc, handle); - if (!res || !res->ops->write) - return -1; - return res->ops->write(res, buf, len); -} -int open_resource(process_t* proc, resource_t* res) { - for (int i = 0; i < MAX_HANDLES; i++) { - if (proc->table[i] == NULL) { - proc->table[i] = res; - return i; // return handle - } - } - return -1; // no free slot -} - -void close_resource(process_t* proc, int handle) { - resource_t* res = get_resource(proc, handle); - if (!res) - return; + // Invalid handle or unsupported operation + if (!res || !res->funcs->read) + return -1; - if (res->ops->close) res->ops->close(res); // call resource-specific close - proc->table[handle] = NULL; + return res->funcs->read(res, buf, len); } ``` +The other operations (`write`, `open`, `close`) would follow the same pattern above: get the resource from the handle and then call the appropriate function from the `funcs` table if supported. With this indirect approach, the kernel's syscall layer is kept minimal whilst allowing for each resource type to have its own specialised behavior. ## Data Copying Another thing left as an exercise to the user is to decide their method of copying data between userspace and the kernel. From 8cfe2087b7feda45c2f90055aaa26ee877a0cf9d Mon Sep 17 00:00:00 2001 From: Max Tyson <98maxt98@gmail.com> Date: Mon, 8 Sep 2025 10:29:41 +1200 Subject: [PATCH 4/7] Small fixes --- 06_Userspace/06_Resource_Management.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/06_Userspace/06_Resource_Management.md b/06_Userspace/06_Resource_Management.md index eb45e51..4a7bac3 100644 --- a/06_Userspace/06_Resource_Management.md +++ b/06_Userspace/06_Resource_Management.md @@ -61,7 +61,7 @@ Typically, a process should `close()` a resource once it is done using it. Howev ## Generic API -Now that we have a way of representing resource, we need define how a process can interact with them. Generally, having a different syscall for each resource type can ___. Instead the kernel can expose a minmal and uniform API that every resource supports. The generic interface for a resource consists of four primary functions: `open`, `read`, `write`, and `close` and by restricting all resources to this same interface we can reduce the complexity of both the kernel and userspace. To begin the implementation of this, our `resource_t` needs extending with a table of function pointers to support these operations. Each resource can then provide it's own implementation of thse four functions whilst the generic interface remains the same. +Now that we have a way of representing resource, we to need define how a process can interact with them. Generally, having a different syscall for each resource type can lead to lots of repeated code segments and make the kernel interface harder to maintain and extend. Instead the kernel can expose a minmal and uniform API that every resource supports. The generic interface for a resource consists of four primary functions: `open`, `read`, `write`, and `close` and by restricting all resources to this same interface we can reduce the complexity of both the kernel and userspace. To begin the implementation of this, our `resource_t` needs extending with a table of function pointers to support these operations. Each resource can then provide it's own implementation of thse four functions whilst the generic interface remains the same. ``` typedef struct resource { resource_type_t type; From fe038807bc56e32c120fad893ad74e32e292bcaa Mon Sep 17 00:00:00 2001 From: Max Tyson <98maxt98@gmail.com> Date: Sun, 21 Sep 2025 16:31:23 +1200 Subject: [PATCH 5/7] Spelling --- 06_Userspace/06_Resource_Management.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/06_Userspace/06_Resource_Management.md b/06_Userspace/06_Resource_Management.md index 4a7bac3..e4df8c2 100644 --- a/06_Userspace/06_Resource_Management.md +++ b/06_Userspace/06_Resource_Management.md @@ -1,10 +1,10 @@ # Resource Management -The kernel manages various things on behalf of the userspace process, including files, sockets, IPC, devices, and more, which we call 'resources'. It is a good idea to have a unified way to handle such a range of resources to reduce complexity. Rather than exposing a separate set of syscalls for each resource, a generic abstractation can be introduced to simplify everything whilst also keeping it all centralised in one place. +The kernel manages various things on behalf of the userspace process, including files, sockets, IPC, devices, and more, which we call _resources_. It is a good idea to have a unified way to handle such a range of resources to reduce complexity. Rather than exposing a separate set of syscalls for each resource, a generic abstractation can be introduced to simplify everything while also keeping it all centralized in one place. To do this, we implement an API where every resource can be opened, read from, written to, and closed using the same syscalls. Through this design, the kernel is kept small, whilst also letting new resources be added in the future with minimal change to both kernel and user code. ## Resource Abstractation -When talking about _resources_, we need a way to distinguish between the different types that the kernel may expect to provide to userspace. Each resource behaves differently internally, but from the view of the userspace process everyting should be acessable from the same set of syscalls. In order to achive this, we define an enum of resource types to allow the kernel to tag each resource with it's category. This way when a system call is made, the kernel knows how to dispatch the request. +When talking about _resources_, we need a way to distinguish between the different types that the kernel may expect to provide to userspace. Each resource behaves differently internally, but from the view of the userspace process, everything should be accessible from the same set of syscalls. In order to achieve this, we define an enum of resource types to allow the kernel to tag each resource with its category. This way, when a system call is made, the kernel knows how to dispatch the request. ``` typedef enum { FILE, @@ -13,7 +13,7 @@ typedef enum { // can extend later } resource_type_t; ``` -In this example `FILE` represents a file on the disk, `MESSAGE_ENDPOINT` is used for an IPC message queue and `SHARED_MEM` for a shared memory region between prcesses. As the kernel grows this struct can be extened to support more resource types. +In this example, `FILE` represents a file on the disk, `MESSAGE_ENDPOINT` is used for an IPC message queue, and `SHARED_MEM` for a shared memory region between processes. As the kernel grows this struct can be extended to support more resource types. Next, we need a generic representation of a resource inside the kernel. This can be defined by the `resource_t` struct: ``` @@ -22,7 +22,7 @@ typedef struct { void* impl; } resource_t; ``` -The `type` field tells the kernel what kind of resource it is and the `impl` pointer allows the kernel to attach the resource specific implmentation of that resource. For example, a file's `impl` could point to a struct holding the file's offset and indoe or for shared memory it could point to the physical address of that region. +The `type` field tells the kernel what kind of resource it is, and the `impl` pointer allows the kernel to attach the resource-specific implementation of that resource. For example, a file's `impl` could point to a struct holding the file's offset and inode, or for shared memory, it could point to the physical address of that region. ## Per Process Resource Table With an abstract resource now defined, we can extend our previous definition of a process to include a **resource table**: @@ -61,7 +61,7 @@ Typically, a process should `close()` a resource once it is done using it. Howev ## Generic API -Now that we have a way of representing resource, we to need define how a process can interact with them. Generally, having a different syscall for each resource type can lead to lots of repeated code segments and make the kernel interface harder to maintain and extend. Instead the kernel can expose a minmal and uniform API that every resource supports. The generic interface for a resource consists of four primary functions: `open`, `read`, `write`, and `close` and by restricting all resources to this same interface we can reduce the complexity of both the kernel and userspace. To begin the implementation of this, our `resource_t` needs extending with a table of function pointers to support these operations. Each resource can then provide it's own implementation of thse four functions whilst the generic interface remains the same. +Now that we have a way of representing resources, we need to define how a process can interact with them. Generally, having a different syscall for each resource type can lead to lots of repeated code segments and make the kernel interface harder to maintain and extend. Instead, the kernel can expose a minimal and uniform API that every resource supports. The generic interface for a resource consists of four primary functions: `open`, `read`, `write`, and `close`, and by restricting all resources to this same interface, we can reduce the complexity of both the kernel and userspace. To begin the implementation of this, our `resource_t` needs extending with a table of function pointers to support these operations. Each resource can then provide its own implementation of these four functions whilst the generic interface remains the same. ``` typedef struct resource { resource_type_t type; @@ -98,4 +98,4 @@ The other operations (`write`, `open`, `close`) would follow the same pattern ab ## Data Copying Another thing left as an exercise to the user is to decide their method of copying data between userspace and the kernel. -One option is to use the userspace provided buffersm, which is efficient due to a single copy but does require sanitization of pointers and lengths to ensure safety. Some things to consider are other threads in the same address space modifying memory at the same address Another option is to copy into a kernel buffer first, which simplifies the sanitization but has the added overhead and loss of performance. +One option is to use the userspace provided buffersm, which is efficient due to a single copy but does require sanitization of pointers and lengths to ensure safety. Some things to consider are other threads in the same address space modifying memory at the same address. Another option is to copy into a kernel buffer first, which simplifies the sanitization but has the added overhead and loss of performance. From f05bffe3b0d42248ad5762ffab945e7589a86fe5 Mon Sep 17 00:00:00 2001 From: Max Tyson <98maxt98@gmail.com> Date: Sun, 21 Sep 2025 16:56:29 +1200 Subject: [PATCH 6/7] Requested Content Changes --- 06_Userspace/06_Resource_Management.md | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/06_Userspace/06_Resource_Management.md b/06_Userspace/06_Resource_Management.md index e4df8c2..ea04780 100644 --- a/06_Userspace/06_Resource_Management.md +++ b/06_Userspace/06_Resource_Management.md @@ -36,7 +36,7 @@ status_t process_status; resource_t* resource_table[MAX_RESOURCES]; } process_t; ``` -Now each process has a resource table that is a map of integers, called _handles_, to the kernel resource objects. A handle is simply an identifier returned by the kernel when opening a resource that is later used by the user to inform what resource the operation should be performed upon. This way, the resource structure is not exposed to userspace. Because of this, the same handle number in different processes can refer to different resources. For example, in Unix, handles `0`, `1`, and `2` refer to stdio for each process. +Now each process has a resource table that is a map of integers, called _handles_, to the kernel resource objects. A handle is simply an identifier returned by the kernel when opening a resource that is later used by the user to inform what resource the operation should be performed upon. This indirection is important because we do not want to expose any kernel pointers directly to a userspace process. Even if they cannot be used there, passing them could still create security or stability risks. This way, the resource structure is not exposed to userspace. Because of this, the same handle number in different processes can refer to different resources. For example, in Unix, handles `0`, `1`, and `2` refer to stdio for each process and are called "file descriptors". With this, we can also define a supporting function allowing the kernel to fetch a resource by handle: ``` @@ -61,7 +61,7 @@ Typically, a process should `close()` a resource once it is done using it. Howev ## Generic API -Now that we have a way of representing resources, we need to define how a process can interact with them. Generally, having a different syscall for each resource type can lead to lots of repeated code segments and make the kernel interface harder to maintain and extend. Instead, the kernel can expose a minimal and uniform API that every resource supports. The generic interface for a resource consists of four primary functions: `open`, `read`, `write`, and `close`, and by restricting all resources to this same interface, we can reduce the complexity of both the kernel and userspace. To begin the implementation of this, our `resource_t` needs extending with a table of function pointers to support these operations. Each resource can then provide its own implementation of these four functions whilst the generic interface remains the same. +Now that we have a way of representing resources, we need to define how a process can interact with them. Generally, having a different syscall for each resource type can lead to lots of repeated code segments and make the kernel interface harder to maintain and extend. Instead, the kernel can expose a minimal and uniform API that every resource supports. The generic interface for a resource consists of four primary functions: `open`, `read`, `write`, and `close`, and by restricting all resources to this same interface, we can reduce the complexity of both the kernel and userspace. To begin the implementation of this, our `resource_t` needs extending with a table of function pointers to support these operations. Each resource can then provide its own implementation of these four functions, whilst the generic interface remains the same. ``` typedef struct resource { resource_type_t type; @@ -76,11 +76,9 @@ typedef struct resource_functions { void (*close)(resource_t* res); } resource_functions_t; ``` -Here, `funcs` is the dispatch table that tells the kernel how to perform each operation for each resource. With this, each function pointer can be set differently dependin on wether the resource is a file, IPC endpoint or something else. +Here, `funcs` is the dispatch table that tells the kernel how to perform each operation for each resource. With this, each function pointer can be set differently depending on whether the resource is a file, IPC endpoint, or something else. Operations are defined to be blocking by default, meaning that if a resource is not ready (for example, no data to read), the process is suspended until the operation can complete. Each resource type can override these generic operations to provide behavior specific to that resource. -Operations are defined to be blocking by default, meaning that if a resource is not ready (for example, no data to read), the process is suspended until the operation can complete. Each resource type can override these generic operations to provide behavior specific to that resource. For example, a file resource can replace the write function with one that writes data to disk, while an IPC message resource could implement write to enqueue a message, allowing the same API call to behave differently depending on the resource. - -It has been left as an exercise to the reader to decide on how they want to handle extending this design for extra resource-specific functionality (ie, renaming a file). There are two (of many) ways to do this, each with its own trade-off. Firstly, a simpler design would be to just add more syscalls to handle this; however, this means the ABI grows as your kernel manages more resources. Another approach would be to pass an additional `size_t flags` parameter and let the resource-specific operation handle it, which would keep the original four operations but with added complexity. +It has been left as an exercise to the reader to decide on how they want to handle extending this design for extra resource-specific functionality (ie, renaming a file). A simpler design may be to just add more syscalls to handle this; however, this means the ABI grows as your kernel manages more resources. On the kernel side of things, these syscalls can just act as dispatchers. For example, a `read_resource(...)` syscall would look up the process's resource table using the handle, retrieve the `resource_t`, and then forward the call to the correct, resource-specific, function: ``` @@ -98,4 +96,6 @@ The other operations (`write`, `open`, `close`) would follow the same pattern ab ## Data Copying Another thing left as an exercise to the user is to decide their method of copying data between userspace and the kernel. -One option is to use the userspace provided buffersm, which is efficient due to a single copy but does require sanitization of pointers and lengths to ensure safety. Some things to consider are other threads in the same address space modifying memory at the same address. Another option is to copy into a kernel buffer first, which simplifies the sanitization but has the added overhead and loss of performance. +One option is to use the userspace provided buffers, which is efficient due to a single copy but does require sanitization of pointers and lengths to ensure safety. Some things to consider are other threads in the same address space modifying memory at the same address. Another option is to copy into a kernel buffer first, which simplifies the sanitization but has the added overhead and loss of performance. + +With using the user buffers, it's not necessarily a single copy, and you may be able to operate directly on the buffer (in which case it's zero-copy). Although, this can be dangerous as another user thread can write to, unmap, or remap the buffer while the kernel is operating on it. Holding a lock over the address space for that process throughout the entire duration of the resource operation is impractical, so the kernel must instead rely on fault handling. By faulting when the process tries to access the memory that the kernel is working with, it allows this behaviour to e caught and the kernel can try to abort or retry the operation safely. From 24aa63e25a495e189f88b0a0635e5611edd708e2 Mon Sep 17 00:00:00 2001 From: Max Tyson <98maxt98@gmail.com> Date: Sun, 21 Sep 2025 16:58:25 +1200 Subject: [PATCH 7/7] Proper Spacing --- 06_Userspace/06_Resource_Management.md | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/06_Userspace/06_Resource_Management.md b/06_Userspace/06_Resource_Management.md index ea04780..6d90f9b 100644 --- a/06_Userspace/06_Resource_Management.md +++ b/06_Userspace/06_Resource_Management.md @@ -1,10 +1,13 @@ # Resource Management + The kernel manages various things on behalf of the userspace process, including files, sockets, IPC, devices, and more, which we call _resources_. It is a good idea to have a unified way to handle such a range of resources to reduce complexity. Rather than exposing a separate set of syscalls for each resource, a generic abstractation can be introduced to simplify everything while also keeping it all centralized in one place. To do this, we implement an API where every resource can be opened, read from, written to, and closed using the same syscalls. Through this design, the kernel is kept small, whilst also letting new resources be added in the future with minimal change to both kernel and user code. ## Resource Abstractation + When talking about _resources_, we need a way to distinguish between the different types that the kernel may expect to provide to userspace. Each resource behaves differently internally, but from the view of the userspace process, everything should be accessible from the same set of syscalls. In order to achieve this, we define an enum of resource types to allow the kernel to tag each resource with its category. This way, when a system call is made, the kernel knows how to dispatch the request. + ``` typedef enum { FILE, @@ -13,19 +16,23 @@ typedef enum { // can extend later } resource_type_t; ``` + In this example, `FILE` represents a file on the disk, `MESSAGE_ENDPOINT` is used for an IPC message queue, and `SHARED_MEM` for a shared memory region between processes. As the kernel grows this struct can be extended to support more resource types. Next, we need a generic representation of a resource inside the kernel. This can be defined by the `resource_t` struct: + ``` typedef struct { resource_type_t type; void* impl; } resource_t; ``` + The `type` field tells the kernel what kind of resource it is, and the `impl` pointer allows the kernel to attach the resource-specific implementation of that resource. For example, a file's `impl` could point to a struct holding the file's offset and inode, or for shared memory, it could point to the physical address of that region. ## Per Process Resource Table With an abstract resource now defined, we can extend our previous definition of a process to include a **resource table**: + ``` typedef struct { size_t pid; @@ -36,9 +43,11 @@ status_t process_status; resource_t* resource_table[MAX_RESOURCES]; } process_t; ``` + Now each process has a resource table that is a map of integers, called _handles_, to the kernel resource objects. A handle is simply an identifier returned by the kernel when opening a resource that is later used by the user to inform what resource the operation should be performed upon. This indirection is important because we do not want to expose any kernel pointers directly to a userspace process. Even if they cannot be used there, passing them could still create security or stability risks. This way, the resource structure is not exposed to userspace. Because of this, the same handle number in different processes can refer to different resources. For example, in Unix, handles `0`, `1`, and `2` refer to stdio for each process and are called "file descriptors". With this, we can also define a supporting function allowing the kernel to fetch a resource by handle: + ``` resource_t* get_resource(process_t* proc, int handle) { @@ -50,8 +59,8 @@ resource_t* get_resource(process_t* proc, int handle) { } ``` - ## Resource Lifecycle + A resource follows a rather straightforward lifecycle, regardless of its type: 1. Firstly, a process acquires a handle by calling the `open_resource` system call. 2. While the handle is valid, the process can perform operations such as `read_resource` or `write_resource`. @@ -59,9 +68,10 @@ A resource follows a rather straightforward lifecycle, regardless of its type: Typically, a process should `close()` a resource once it is done using it. However, that is not always the case, as processes may exit without cleaning up properly, and thus it is up to the kernel to ensure resources aren't leaked. This could look like a loop through the process's resource table, calling `close_resource(process, handle);` for each open resource and letting the resource-specific `close()` function handle the work. - ## Generic API + Now that we have a way of representing resources, we need to define how a process can interact with them. Generally, having a different syscall for each resource type can lead to lots of repeated code segments and make the kernel interface harder to maintain and extend. Instead, the kernel can expose a minimal and uniform API that every resource supports. The generic interface for a resource consists of four primary functions: `open`, `read`, `write`, and `close`, and by restricting all resources to this same interface, we can reduce the complexity of both the kernel and userspace. To begin the implementation of this, our `resource_t` needs extending with a table of function pointers to support these operations. Each resource can then provide its own implementation of these four functions, whilst the generic interface remains the same. + ``` typedef struct resource { resource_type_t type; @@ -76,11 +86,13 @@ typedef struct resource_functions { void (*close)(resource_t* res); } resource_functions_t; ``` + Here, `funcs` is the dispatch table that tells the kernel how to perform each operation for each resource. With this, each function pointer can be set differently depending on whether the resource is a file, IPC endpoint, or something else. Operations are defined to be blocking by default, meaning that if a resource is not ready (for example, no data to read), the process is suspended until the operation can complete. Each resource type can override these generic operations to provide behavior specific to that resource. It has been left as an exercise to the reader to decide on how they want to handle extending this design for extra resource-specific functionality (ie, renaming a file). A simpler design may be to just add more syscalls to handle this; however, this means the ABI grows as your kernel manages more resources. On the kernel side of things, these syscalls can just act as dispatchers. For example, a `read_resource(...)` syscall would look up the process's resource table using the handle, retrieve the `resource_t`, and then forward the call to the correct, resource-specific, function: + ``` size_t read_resource(process_t* proc, int handle, void* buf, size_t len) { resource_t* res = get_resource(proc, handle); @@ -92,9 +104,11 @@ size_t read_resource(process_t* proc, int handle, void* buf, size_t len) { return res->funcs->read(res, buf, len); } ``` + The other operations (`write`, `open`, `close`) would follow the same pattern above: get the resource from the handle and then call the appropriate function from the `funcs` table if supported. With this indirect approach, the kernel's syscall layer is kept minimal whilst allowing for each resource type to have its own specialised behavior. ## Data Copying + Another thing left as an exercise to the user is to decide their method of copying data between userspace and the kernel. One option is to use the userspace provided buffers, which is efficient due to a single copy but does require sanitization of pointers and lengths to ensure safety. Some things to consider are other threads in the same address space modifying memory at the same address. Another option is to copy into a kernel buffer first, which simplifies the sanitization but has the added overhead and loss of performance.