arch/x86_64: Implement ukarch_random #695
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
4 tasks
There was a problem hiding this comment.
Hi, @michpappas!
I tested the functions and they seem to work just as expected! I will let here a couple of comments.
michpappas
force-pushed
the
x86_arch_random
branch
from
307909f to
8729b62
Compare
razvanvirtan
approved these changes
Jan 28, 2023
There was a problem hiding this comment.
Thank you for this!
Reviewed-by: Razvan Virtan virtanrazvan@gmail.com
mariasfiraiala
approved these changes
Jan 28, 2023
There was a problem hiding this comment.
Looks good to me. Thank you @michpappas !
Reviewed-by: Maria Sfiraiala maria.sfiraiala@gmail.com
Implement the ukarch_random API for x86_64. The implementation requires that both the RDRAND and RDSEED instructions are available in the system. Signed-off-by: Michalis Pappas <mpappas@fastmail.fm>
michpappas
force-pushed
the
x86_arch_random
branch
from
8729b62 to
58f0fb2
Compare
marcrittinghaus
approved these changes
Feb 2, 2023
There was a problem hiding this comment.
Thanks! 🚀
Approved-by: Marc Rittinghaus marc.rittinghaus@unikraft.io
|
✅ Checkpatch passed Beep boop! I ran Unikraft's
|
dragosp27
added a commit
to dragosp27/unikraft
that referenced
this pull request
Feb 7, 2023
commit a20323a40e14f98a1e7b793a47eea755dbfe8f34
Author: Simon Kuenzer <simon@unikraft.io>
Date: Wed Feb 1 23:13:05 2023 +0100
lib/syscall_shim: Pretty-print format: clone
This commit introduces initial print format and decoding definitions for
the following system calls: `SYS_clone`
Checkpatch-Ignore: MISSING_BREAK
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit 239b989d3bef7787b984cc52096b96baba458b10
Author: Simon Kuenzer <simon@unikraft.io>
Date: Wed Feb 1 23:12:13 2023 +0100
lib/syscall_shim: Pretty-print format: socket, bind, sendto, recvmsg
This commit introduces initial print format and decoding definitions for
the following system calls: `SYS_socket`, `SYS_bind`, `SYS_sendto`,
`SYS_recvmsg`
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit e93760fa0d750547e5bf9d052e983bf3a394756f
Author: Simon Kuenzer <simon@unikraft.io>
Date: Wed Feb 1 23:10:26 2023 +0100
lib/syscall_shim: Pretty-print format: clock_gettime
This commit introduces initial print format and decoding definitions for
the following system calls: `SYS_clock_gettime`
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit d7ca4f7bdb6355e7d894fb2f3e21710b61655a9a
Author: Simon Kuenzer <simon@unikraft.io>
Date: Wed Feb 1 23:08:59 2023 +0100
lib/syscall_shim: Pretty-print format: futex
This commit introduces initial print format and decoding definitions for
the following system calls: `SYS_futex`
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit 7883f875f73b2296fba7c6e23e5fdacf6eac764d
Author: Simon Kuenzer <simon@unikraft.io>
Date: Wed Feb 1 23:07:24 2023 +0100
lib/syscall_shim: Pretty-print format: mmap, munmap, mprotect
This commit introduces initial print format and decoding definitions for
the following system calls: `SYS_mmap`, `SYS_munmap`, `SYS_mprotect`
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit c036b1528118dcb07ba2cb906670c5d838506379
Author: Simon Kuenzer <simon@unikraft.io>
Date: Wed Feb 1 18:07:16 2023 +0100
lib/posix-futex: Remove circular config dependency to syscall_shim
This commit removes the unit test dependency of `lib/posix-futex` to
syscall_shim which caused a circular dependency over `lib/uktest`:
`lib/uktest` started to be dependent on `lib/syscall_shim` and because the
`lib/syscall_shim` depends on `lib/ukstreambuf`, the unit tests of
`lib/ukstreambuf` depend on `lib/uktest` too.
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit a7630b1e8c5697a1817086900379552156945e7a
Author: Simon Kuenzer <simon@unikraft.io>
Date: Wed Feb 1 16:22:22 2023 +0100
lib/syscall_shim: strace-like output for binary system calls
This commit introduces the option to enable a strace-like output for binary
system calls. The option can be found under the debugging options of
`lib/syscall_shim`. The implementation utilizes the pretty-print function
`uk_snprsyscall()` for system call requests.
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit 26dfabdfdc43a9d89dc00c5284e2af19ddb44265
Author: Simon Kuenzer <simon@unikraft.io>
Date: Wed Feb 1 22:36:24 2023 +0100
lib/syscall_shim: Functions to pretty-print system call requests
This commit introduces an initial implementation of `uk_snprsyscall()` and
`uk_vsnprsyscall()`, two functions that pretty print a system call request
to a given C-string buffer. The system call name, the given arguments
(including flags and definitions) and the response are decoded.
As a start, this commit implements the formatting of the following system
calls: brk, open, openat, write, read, stat, fstat, close, dup, dup2,
gettid, getpid
Yet unsupported system calls are printed with the raw values of arguments.
Checkpatch-Ignore: DO_WHILE_MACRO_WITH_TRAILING_SEMICOLON
Checkpatch-Ignore: MULTISTATEMENT_MACRO_USE_DO_WHILE
Checkpatch-Ignore: TRAILING_SEMICOLON
Checkpatch-Ignore: MISSING_BREAK
Checkpatch-Ignore: COMPLEX_MACRO
Checkpatch-Ignore: SPACING
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit 5f0510de86507b2e1745020acc862ad1b1a2bb97
Author: Simon Kuenzer <simon@unikraft.io>
Date: Mon Jan 30 15:09:24 2023 +0100
lib/syscall_shim: Use `UK_NARGS()` from `<uk/essentials.h>`
This commit removes the macros `__UK_SYSCALL_NARGS()` and
`__UK_SYSCALL_DEF_NARGS()` and replaces their usage with `UK_NARGS()`
provided by `<uk/essentials.h>`.
Checkpatch-Ignore: LONG_LINE
Checkpatch-Ignore: COMPLEX_MACRO
Checkpatch-Ignore: SPACING
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit c47bff7b4a7f19981524a2c60605ffc92aa42268
Author: Simon Kuenzer <simon@unikraft.io>
Date: Mon Jan 30 15:05:32 2023 +0100
include/essentials: UK_NARGS(): Support up to 25 arguments
Increase the number of supported arguments of `UK_NARGS()` from 7 to 25.
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit 85b938c4678c6f2e365510cf575b8131098b5b61
Author: Simon Kuenzer <simon@unikraft.io>
Date: Mon Jan 30 03:53:36 2023 +0100
lib/ukstreambuf: Unit tests: Append operations for binary data
Registers `libuktest` unit tests for append operations of binary data to a
streambuf object.
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit 3f39a18376498028ed8fe11fe678b2d2fc7743cb
Author: Simon Kuenzer <simon@unikraft.io>
Date: Mon Jan 30 03:51:50 2023 +0100
lib/ukstreambuf: Unit tests: Append operations for C-strings
Registers `libuktest` unit tests for append operations of C-strings to a
streambuf object.
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit b52f6ab7bda929c1342f3fc323bf35a3475839c1
Author: Simon Kuenzer <simon@unikraft.io>
Date: Mon Jan 30 03:46:37 2023 +0100
lib/ukstreambuf: Unit tests: Initialization and basic operations
Registers `libuktest` unit tests for initialization and basic operations of
a streambuf object.
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit c28496eae076b29ef2fa1dcce9eca8fd8e00ef82
Author: Simon Kuenzer <simon@unikraft.io>
Date: Mon Jan 30 03:34:07 2023 +0100
lib/ukstreambuf: Append operations for binary data
This commit introduces append operations for binary data:
- `uk_streambuf_memcpy()`: Append a copy of binary data
- `uk_streambuf_reserve()`: In-place appending of binary data (zero copy)
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit 2c9065022de71f63796af0d83df356590ea7e379
Author: Simon Kuenzer <simon@unikraft.io>
Date: Mon Jan 30 03:30:49 2023 +0100
lib/ukstreambuf: Append operations for C-strings
This commit introduces append operations for C-strings:
- `uk_streambuf_printf()`, `uk_streambuf_vprintf()`:
Append a printf-formatted string
- `uk_streambuf_strcpy`: Append a copy of a C-string
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit 44e2e7f7034003168130d472584c09cfa1c21821
Author: Simon Kuenzer <simon@unikraft.io>
Date: Mon Jan 30 03:25:09 2023 +0100
lib/ukstreambuf: Initialization and basic operations
This commit introduces functions to allocate and/or initialize a streambuf
object for a given memory buffer. It also adds basic operations, like
resetting a streambuf buffer and querying its current state.
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit f15a21681bb53727aa2537b606cac81662f081af
Author: Simon Kuenzer <simon@unikraft.io>
Date: Mon Jan 30 03:13:20 2023 +0100
lib/ukstreambuf: Add library skeleton
The library will provide helper functions and macros to simplify dealing
with appending of data and/or C-string to buffers.
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Reviewed-by: Andra Paraschiv <andra@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #737
commit 6092385bf2a5a3fdec4bf6d9f9551998ae6209ee
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Fri Feb 3 11:58:19 2023 +0100
lib/uksched: Silence warning about unused variable
The parent variable was not used outside the assert. This leads to a
warning when asserts are disabled. This commit fixes the warning
be removing the variable.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #739
commit 1fbd1045bffa17ff1b38cd7f2488b0843aea5960
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Fri Feb 3 11:54:58 2023 +0100
lib/ukschedcoop: Silence warning about unused args
When asserts are disabled we get warnings about unused args. This
commit silences the warnings.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #739
commit cdeccfb9343d352a7bf63c66506275df663919d4
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Fri Feb 3 11:53:20 2023 +0100
lib/ukdebug: Silence warnings about unused args
If printing of source line and file is disabled, we get warnings about
srcline and srcname not being used. This commit silences the warning.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #739
commit 9910feb8c3efea57f3de284b0b0a28f0facbbbd6
Author: Simon Kuenzer <simon@unikraft.io>
Date: Fri Feb 3 20:40:05 2023 +0100
lib/vfscore: Cast initrd vbase to (void *)
PR #722 (Boot code refactoring) introduced an update to
`struct ukplat_memregion_desc` where the virtual base of a memory region is
now specified as a __vaddr_t integer value. Since it is a virtual address
we can cast it to (void *) to avoid build warnings in
`lib/vfscore/rootfs.c`.
Signed-off-by: Simon Kuenzer <simon@unikraft.io>
Reviewed-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Approved-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #744
commit e742f9d1da1008be8f35b51247b5182ab2b0eae4
Author: Michalis Pappas <mpappas@fastmail.fm>
Date: Sun Dec 11 13:17:58 2022 +0100
plat/kvm: Enable Pointer Authentication in KVM
Signed-off-by: Michalis Pappas <mpappas@fastmail.fm>
Reviewed-by: Maria Sfiraiala <maria.sfiraiala@gmail.com>
Reviewed-by: Razvan Virtan <virtanrazvan@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #686
commit fb02d665a3a382fd875ec4f6588a792def31c261
Author: Michalis Pappas <mpappas@fastmail.fm>
Date: Sun Dec 11 13:13:37 2022 +0100
plat/common: Use ukarch_random to generate PAuth key
With the addition of the ukarch_random() API it is now possible to
generate PAuth keys without delegating the implementation of key
generation to the platform. Remove ukplat_pauth_gen_key() from the
platform API and generate PAuth keys using the ukarch_random() API.
Signed-off-by: Michalis Pappas <mpappas@fastmail.fm>
Reviewed-by: Maria Sfiraiala <maria.sfiraiala@gmail.com>
Reviewed-by: Razvan Virtan <virtanrazvan@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #686
commit e0c98d94be9c9b870de9ef146b94bb19a0fda540
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Fri Feb 3 19:35:25 2023 +0100
include/uk/plat: Fix initrd0 macro
This commit fixes a typo in the ukplat_memregion_find_initrd0 macro.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Simon Kuenzer <simon@unikraft.io>
Approved-by: Simon Kuenzer <simon@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #742
commit 1064d46d60d8e8fdbf0735d002acba8a58e92bfb
Author: Delia-Maria Pavel <delia_maria.pavel@stud.acs.upb.ro>
Date: Tue Dec 27 16:43:48 2022 +0200
lib/ramfs: Add README.md file
Signed-off-by: Delia-Maria Pavel <delia_maria.pavel@stud.acs.upb.ro>
Reviewed-by: Stefan Jumarea <stefanjumarea02@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #705
commit 9d907d4df807c705313c2ad30a411fa02d0f14b2
Author: Delia-Maria Pavel <delia_maria.pavel@stud.acs.upb.ro>
Date: Tue Dec 27 16:34:45 2022 +0200
lib/ramfs: Add comments to ramfs.h file
Signed-off-by: Delia-Maria Pavel <delia_maria.pavel@stud.acs.upb.ro>
Reviewed-by: Stefan Jumarea <stefanjumarea02@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #704
commit e627b2d3c3fbec090cdbcd886d3edc80be05ebf0
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 20:30:04 2023 +0100
lib/uknofault: Fix warning with ukvmem
When ukvmem is activated a warning may be observed during build
about ps being potentially uninitialized. This commit fixes this.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit 425eb8468f56a07ebe7ed3a673523fc12db2ae84
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 19:47:36 2023 +0100
lib/posix-mmap: Introduce POSIX memory functions
This commit adds the posix-mmap library that implements the POSIX
memory-mapping related functions such as mmap and munmap based
on ukvmem.
Checkpatch-Ignore: NON_RAW_SYSCALL
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit d3626a6ce1061b34408d999775eda4fe12df3f4b
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 20:03:37 2023 +0100
drivers/virtio: Use DMA mapping with ukvmem
The current implementation uses the heap to allocate memory for the
virtio queues. This is problematic when used in conjunction with
on-demand paging for the heap. It can cause pages of the virtio
queue to be non-contiguous in physical memory. Some pages may
not even mapped to physical memory, yet. Since the hypervisor
does not care about the virtual mapping but only gets the physical
start address of the virtio queue (i.e., the physical address of the
first page), it will write on any physical page in the range of the
virtio queue (i.e., start physical address + size of queue). However,
these physical pages may belong to arbitrary other mappings. And
even if they belong to the same virtqueue, they might not be mapped
in the same order as in the virtual space. This is especially hard to
debug, because the guest is not performing the corrupting memory
writes.
To fix this, we request contiguous memory from the frame allocator
and map this in the same order in the virtual address space using
uk_vma_map_dma().
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit f335edee3e87b422efc9cce0cb2984bc9c9bdce2
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 19:45:51 2023 +0100
lib/ukboot: Initialize VAS + Heap
When ukvmem is available, we can represent the heap as a virtual
memory area. This has the advantage of enabling on-demand paging
for the heap.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit 0313587052254819ea1d59d6bf0f1946bc354930
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 19:43:54 2023 +0100
lib/ukvmem: Introduce virtual memory library
ukvmem provides the means to perform virtual address space management
where each virtual address space is build from a set of virtual memory
areas. This allows the library to select virtual addresses for mappings.
This is in contrast to the primitive paging API which does not manage
virtual address allocation. The library implements different VMA types
via VMA operations that define the VMA's behavior and contents.
Just like on Linux contiguous VMAs are merged and split depending on
the compatibility of their flags and the behavior defined by the VMA.
The library provides functions for mapping, unmapping, and
changing of memory attributes. The initial version comes with
support for x86_64, only.
Checkpatch-Ignore: LONG_LINE_COMMENT
Checkpatch-Ignore: FUNCTION_WITHOUT_ARGS
Checkpatch-Ignore: SPLIT_STRING
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit eb5696faa938e8e85ed99099a9c8da72ab30525e
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 19:19:37 2023 +0100
include/plat: Update paging header to new CC
This commit updates the function documentation in the platform paging
header to the new coding convention so it is inline with the newly added
functions.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit 17d6ae38922124a56494e72c84d753cc4f9a8f47
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 19:37:19 2023 +0100
plat/common: Ignore non-existent PTEs
ukplat_page_unmap() ignores non-existent PTEs and just goes on to the
next PTE in the memory range. The same behavior can be achieved with
ukplat_page_mapx(). However, ukplat_page_set_attr() stops at the first
non-existent PTE. This is problematic when on-demand paging has
lead to some pages in a memory range being present and others not.
This commit changes the behavior of ukplat_page_set_attr() to just
ignore this like the other mapping functions.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit 2bc86d1f93407d44d4d35b3b4b032e2977233ee5
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 19:30:13 2023 +0100
plat/common: Ignore -ENOMEM errors in pg_ffree()
Currently, the pg_ffree() function already ignored any errors, but it
has an assert to check if the error could be expected. We add
-ENOMEM to this, which signals that the memory has not been
allocated. While this would indicate a true error condition in case
a physical page is mapped only once, this is not the case if
the same page is mapped multiple times in a certain range and this
range is unmap en-bloc. In this case, the first free would free the
frame in the allocator while the other mappings would cause an
expected error.
An example for such a mapping could be a memory range which
should be zero-initialized by mapping all pages to the same
zero frame and using a CoW-style mapping.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit 956e45ceb5de4c93d9147df6f8ccfc08afa20b5a
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 19:17:43 2023 +0100
plat/common: Introduce ukplat_page_k(un)map()
For temporary mappings of physical memory within the unikernel,
it is useful to have a fast alternative to ukplat_page_mapx() that also
automatically selects a virtual address for the mapping. This commit
introduces ukplat_page_kmap() and ukplat_page_kunmap() for this
purpose. However, they are not guaranteed to succeed if for some
reason no kernel mapping can be established.
However, for direct-mapped physical memory implementations the
functions just return the virtual address of the direct mapping. This
makes the kmap() very fast. In addition, this implementation cannot
fail and the unmap is a noop. The commit comes with an
implementation for x86_64 and arm64.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit a51d65c446c2dbe204c5466e80f66e5974e8c8c6
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 19:11:39 2023 +0100
plat/common: Introduce ukplat_page_mapx()
The current ukplat_page_map() functions allows to map a range of virtual
pages to physical ones with a defined set of protections. This function
has the following limitations that create problems in the context of
virtual memory management:
1) If a mapping the virtual range already exists, the function
aborts with an error. The caller has no knowledge about the virtual
address that caused the error.
2) If the page attributes are set to read only while physical memory
should be allocated at the same time with paddr=__PADDR_ANY, the
caller has no chance of initializing the frame contents.
This commit replaces ukplat_page_map() with ukplat_page_mapx()
with the x being execute/extended. The mapx version allows the caller
to specify a function that should be called during mapping before
writing a PTE. The function can modify the PTE, instruct the mapping to
skip the current page, or switch to a smaller page size. The original
function behavior is provided when the mapx argument is NULL.
The mapx function can be used in case 1) to either overwrite or skip
existing mappings. The mapx function can also be used together with
a temporary mapping to initialize the physical memory of the current
page before establishing the mapping. This way, we avoid
map/rw -> remap/ro windows, which can also pose a security problem
besides the runtime overhead. If the caller sets paddr=0, it is also
possible to completely leave physical memory allocation to the mapx
function, for example, to serve memory from a cache.
For compatibility, a wrapper for ukplat_page_map() is provided.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit 49d97bd83846166ab09007405fcbd712e5fad51b
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 18:52:40 2023 +0100
include/arch/x86_64: Add page fault error bits
This commit adds bit definitions for parsing x86_64 page fault
error codes.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit a93c1366b980edd84f34f39f0cc0743c889eae7e
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 18:48:40 2023 +0100
include/arch/x86_64: Add PT_Lx_PTE_SET_PADDR()
This commit introduces PT_Lx_PTE_SET_PADDR() which allows
to set the physical address in a page table entry. This commit
comes only with the x86_64 implementation.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit 075cd8d62c0940feaa742cb7826e8c6b9266b6a1
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 18:44:33 2023 +0100
include/arch/[x86_64,arm64]: Add PAGE_SHIFT_Lx()
This commit introduces a macro called PAGE_SHIFT_Lx() that is
the reverse operation of PAGE_Lx_SHIFT() and allows to compute
the page table level from the page size order.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit 69a268de0c87f1b4eeb67f4304d6ed85641c7abf
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 18:36:36 2023 +0100
include/uk/plat: Fix PAGE_FLAG_SIZE_MASK
Currently, the macro uses the shift to define the mask. However, the
mask is determined by the number of bits reserved for the size, not
by the number of bits shifted in the flag word.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit 7d8ffd5047d335a19398f9587a7e257366934517
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 18:34:18 2023 +0100
include/uk: Move PAGE_ATTR_PROT_* values
The PAGE_ATTR_PROT_RW and PAGE_ATTR_PROT_RWX definitions should
be available even when the <uk/plat/paging.h> header is not available,
just like the other PAGE_ATTR_* values. We thus move them to the
arch header.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit e2183e285a2fce7c10ff3fc27fdb643826b18139
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 18:31:05 2023 +0100
lib/ukfallocbuddy: Build as ISR-safe
To be able to invoke the frame allocator from within a page fault
handler, we need to compile it as ISR-safe.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit 6408fadae4e67ae642484fbf8af7bdef7ed0ee68
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Thu Feb 2 18:28:41 2023 +0100
plat/kvm: Build paging code as ISR-safe
To be able to invoke the paging code in the context of a page fault
handler, we need to build it as ISR-safe.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Michalis Pappas <michalis.pappas@opensynergy.com>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #603
commit 47ec124fff0551ba3c4eccd4264cacc2f5f980f3
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Fri Feb 3 17:08:24 2023 +0100
plat/kvm/x86: Fix cmdline parsing
In the boot refactoring, we replaced the statically allocated command
line buffer with a dynamically allocated one. However, the length
argument was not updated, which leads to the cmdline being not
properly parsed.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #741
commit 85ef21541f61e8ec3b633d8ccc9e1453911c609e
Author: Marco Schlumpp <marco.schlumpp@gmail.com>
Date: Thu Dec 16 10:48:56 2021 +0100
plat/common/pci: Implement subclass matching
The class is usually not precise enough to identify the correct driver
for a device. For example, an NVM driver can only talk to devices which
are in the subclass of NVM storage devices but not to other devices in
the larger storage class.
Signed-off-by: Marco Schlumpp <marco.schlumpp@gmail.com>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Approved-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #624
commit 6d686ab21fd2961b341413b0415db3d68253a9a7
Author: Maria Sfiraiala <maria.sfiraiala@gmail.com>
Date: Wed Dec 21 23:57:49 2022 +0200
plat/pci_ecam.c: Fix gcc-ism for clang compilation
This commit replaces a small gcc-ism from plat/common/pci_ecam.c that
makes impossible any clang compilation on AArch64.
The `dummy_imask[] = { [0 ... 16] = cpu_to_fdt32(~0) };` syntax is a GNU
extension, ergo, it is not recognized by LLVM's clang.
Signed-off-by: Maria Sfiraiala <maria.sfiraiala@gmail.com>
Reviewed-by: Michalis Pappas <mpappas@fastmail.fm>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #701
commit 554e64136c136011d2a835c90909fd264721c1c8
Author: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Date: Mon Nov 28 15:44:12 2022 +0100
plat/virtio_net: Notify host of RX bufs on start
Currently, we don't notify the virtio-net device of available RX buffers
on driver start, meaning it might never send us incoming packets. Make
sure that we send this notification once the driver has started.
Signed-off-by: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #669
commit add665661dec7b38e9db08a2e296d49e8167a44b
Author: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Date: Mon Nov 28 15:43:57 2022 +0100
plat/virtio_net: Fix device MTU detection
When looking for MTU information, instead of checking for the
VIRTIO_NET_F_MTU feature flag, the virtio-net driver ends up checking
VIRTIO_NET_F_STATUS instead. This commit fixes this check, and reports
the device's MTU to the netdev architecture.
Signed-off-by: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #669
commit e2696bf36c9a45a894a978afd84533c9101010f0
Author: Nour-eddine Taleb <contact@noureddine.xyz>
Date: Wed Dec 14 15:31:35 2022 +0100
lib/ukallocbbuddy: fix a memory leak
fix a leak in the buddy allocator init routine.
and an extra fix which subtract the bitmap size
from the available pages in the region.
GitHub-Fixes: #688
Signed-off-by: Nour-eddine Taleb <contact@noureddine.xyz>
Reviewed-by: Florin Postolache <florin.postolache80@gmail.com>
Approved-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #689
commit babad72979a85c021b41ef2b7d2605f2c556c6ed
Author: Razvan Deaconescu <razvan.deaconescu@cs.pub.ro>
Date: Thu Jul 21 09:19:01 2022 +0300
lib/nolibc: Use Linux-compatible stat structure
Use fields and field types for nolibc `stat` structure to make it
compatible with Linux. It is required for binary compatibility mode. The
structure is copied from Musl (arch/x86_64/bits/stat.h).
Signed-off-by: Razvan Deaconescu <razvan.deaconescu@cs.pub.ro>
Reviewed-by: Delia Pavel <delia_maria.pavel@stud.acs.upb.ro>
Approved-by: Simon Kuenzer <simon@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #675
commit 8f0ddca1bb8a0869a27f111b6f4687eca9886cce
Author: Marco Schlumpp <marco@unikraft.io>
Date: Fri Jan 20 13:46:39 2023 +0100
plat/*: Handle errors returned from IRQ event handlers
PR #691 adds the possibility to return error codes from the event
handler. In the context of interrupts, we interpret an error code as a
fatal error and emit a crash.
Signed-off-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #699
commit 419062ff01ee42a6eaffa9ef2a55372c055d5087
Author: Marco Schlumpp <marco@unikraft.io>
Date: Tue Dec 20 15:46:35 2022 +0100
plat/*: Add event for IRQs
Adds an event which is triggered when an IRQ occurs. The event handler
has access to the IRQ number and the registers of the interrupted
thread.
Signed-off-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #699
commit f65c7267794e347711cb652542dac68e0f10a2ae
Author: Marco Schlumpp <marco@unikraft.io>
Date: Fri Dec 16 15:55:42 2022 +0100
plat/*: Expose registers to main irq handler function
The trap handlers already pass on the registers to the handler
functions. This passes them to the KVM-internal dispatch function.
Signed-off-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #699
commit 3ef521c6b99e8685040be1d6bee59ca82c3165c8
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Fri Feb 3 11:02:21 2023 +0100
include/event: Expose HANDLED_CONT return value
The previous changes to the event hide if the last handler that
successfully handled the event allowed further event processing
(i.e., on success always UK_EVENT_HANDLED is returned). This
commit fixes this so that the caller of uk_raise_event() is informed
about if event processing can continue by other means or if it
should stop.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #738
commit 3520375f78ca30aa073b1449bf6b24be0fb60f0e
Author: Marco Schlumpp <marco@unikraft.io>
Date: Fri Dec 16 16:03:05 2022 +0100
lib/ukdebug: Use correct variable name for sect-strip.py tool
The variable used didn't match the one defined in the platform part and
therefore the specified sections were never stripped.
Signed-off-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Radu Nichita <radunichita99@gmail.com>
Approved-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #690
commit 19694171104fbe7931c8a041629bcfa59652f1e6
Author: Razvan Deaconescu <razvan.deaconescu@cs.pub.ro>
Date: Fri May 13 00:34:49 2022 +0300
lib/posix-info: Comment out unused sysinfo buffer
The sysinfo structure defines the `reserved` buffer. This is however not
present on Linux, meaning that binary-compatibile applications use a
shorter buffer. This results in a buffer overflow when transferring
information from Unikraft to the buffer in binary-compatible app.
This commit fixes this by commenting out the `reserved` buffer in the
`sysinfo` structure.
Signed-off-by: Razvan Deaconescu <razvan.deaconescu@cs.pub.ro>
Reviewed-by: Radu Nichita <radunichita99@gmail.com>
Approved-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #673
commit ba8d5cab8e750653d6f55eb292b322818cf9966b
Author: Marco Schlumpp <marco@unikraft.io>
Date: Fri Oct 28 12:37:00 2022 +0200
plat/kvm: Ensure time monotonicity for tscclock
The TSC is not synchronized perfectly on some systems. Examples of such
systems are multi-socket system or newer AMD processors. This patch
ensure that this does not cause an unwanted underflow in the delta
calculation.
GitHub-Fixes: #334
Signed-off-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #617
commit 3c194d7efe22c79fdda213de939db692e9f7f279
Author: Marco Schlumpp <marco@unikraft.io>
Date: Mon Nov 28 11:27:30 2022 +0100
build: Check doc comments when compiling with clang
The diagnostic can check for various issues related to doxygen
documentation comments. A list of detectable issues can be found at [1].
[1]:https://clang.llvm.org/docs/DiagnosticsReference.html#wdocumentation
Signed-off-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Maria Sfiraiala <maria.sfiraiala@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #664
commit 323cd8630be2db717576263a0273a55902861a0e
Author: Radu Nichita <radunichita99@gmail.com>
Date: Sun Jan 1 22:28:33 2023 +0200
lib/vfscore: Add comments to vfs.h file
Signed-off-by: Radu Nichita <radunichita99@gmail.com>
Reviewed-by: Stefan Jumarea <stefanjumarea02@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #707
commit 5d93938fcd07c0e24c70e18567acf07c5951649f
Author: Marco Schlumpp <marco.schlumpp@gmail.com>
Date: Tue Oct 19 10:19:21 2021 +0200
lib/vfscore: Introduce no-access-time flag
The ported ext* filesystem supports tracks the access time by
default. By specifying this flag, the user can disable the feature to
reduce the amount of inode updates.
Signed-off-by: Marco Schlumpp <marco.schlumpp@gmail.com>
Reviewed-by: Stefan Jumarea <stefanjumarea02@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #623
commit fb5d3a866d42f2b7c2aaf07b002d7806be625300
Author: Marco Schlumpp <marco.schlumpp@gmail.com>
Date: Tue Oct 19 10:14:54 2021 +0200
lib/vfscore: Add additional vnode modes
The ext* filesystem port supports the usage of these flags.
Signed-off-by: Marco Schlumpp <marco.schlumpp@gmail.com>
Reviewed-by: Stefan Jumarea <stefanjumarea02@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #623
commit 20d3d242e939886284190d62089d9867374ef44c
Author: Stefan Jumarea <stefanjumarea02@gmail.com>
Date: Sat Sep 17 13:55:17 2022 +0300
lib/uktest: Add README.md file
Signed-off-by: Stefan Jumarea <stefanjumarea02@gmail.com>
Reviewed-by: Razvan Deaconescu <razvand@unikraft.io>
Reviewed-by: Alexander Jung <alex@unikraft.io>
Approved-by: Alexander Jung <alex@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #555
commit 8a366930a3e6548d369db76382ef469b04affe12
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 18:47:08 2023 +0100
plat/kvm/x86: Remove old bootinfo structure
This commit removes the no longer needed uk_bootinfo structure header.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 20bdc0556e36f257e199a534d98741b01cbbc0ec
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 18:45:10 2023 +0100
lib/ukboot: Add heap initialization
Previously, the heap has been initialized in the platform code and
information about the heap areas has been communicated via
ALLOCATABLE memory regions. With the new boot information
heap initialization is fully moved to libukboot.
If paging is not enabled, we use the free physical memory areas to
initialize the heap. The platform takes care of only exposing mapped
physical memory. As before, non-contiguous areas are added to the
allocator one-by-one.
If paging is enabled, we create a contiguous mapping at a customizable
base address (default @16GiB) and initialize the allocator with this single
area. As before, using paging allows to access more memory than
mapped by the boot page table.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 08a9ebf3492d22229ebc7143a3ca87634e64361c
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 18:36:04 2023 +0100
plat/kvm: Add generation of boot info
The information about the kernel segments available as part of the
boot information at runtime is generated by the mkbootinfo script.
The script generates a binary blob based on the linked unikernel
image. This blob is then stored in the .uk_bootinfo section to replace
the dummy data.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit f0b9508e0884e807c02b07ea2f3501f7f55c136c
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 18:32:17 2023 +0100
plat/kvm: Remove platform symbol hiding
To use protocol-specific names for boot protocol entry functions
we need to remove the platform symbol hiding step. This step
does not serve any useful purpose and can take considerable
time for large images. Removing it thus has also other benefits.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 65e37e2a03aaee979fd342dd824c0a4534f5a482
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 18:28:52 2023 +0100
plat/kvm: Use multiboot entry function
With different boot protocols providing different entry functions
we can no longer statically define a single entry function or would
need to give it a generic name. This, however, would prevent us
from building images that come with support for multiple boot
protocols (i.e., entry functions). Also keeping a protocol specific
name make the function name more descriptive.
We thus remove the static ENTRY from the linker script and
provide the (default) entry symbol via the linker command line.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit fd30438ed96ead1e6855a9704a9ba182b5b7c7c2
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 18:16:26 2023 +0100
plat/kvm: Replace platform initialization code
This commit introduces a boot protocol-independent
platform initialization based on the unikraft boot information.
Major changes to the previous code are:
+ Simplified and no longer specific to multiboot
+ Early boot memory allocation from free memory regions to avoid
static memory reservation for the command line
+ Prepares paging initialization to be moved out of platform code
+ The paging-based implementation performs (un)mappings generically
based on the boot information, leading to more flexibility. For example,
a chainloader can specify its own memory mappings as regions to be
unmapped during boot by the chainloaded unikernel.
+ No heap initialization within the platform.
+ Detailed printing of boot memory map, kernel command line, and
others via the ukplat_bootinfo_print() method.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 82a1219db835f597ae15bda4ae3caa9ac41f55d2
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 18:03:37 2023 +0100
plat/kvm: Add multiboot entry
With the new boot information every boot protocol must come with
entry code that transforms the protocol-specific information into the
generic unikraft format. This commit inserts the corresponding
entry function for multiboot v1.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit b23cc9ced90fa1b24052f271674b71065e75883e
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 17:57:13 2023 +0100
plat/kvm: Back memregions by boot information
Previously, the entries of the memory region list were dynamically
created at runtime based on the section symbols. This commit
replaces the entire implementation and uses the boot information
as source for the list instead. However, this is only done on
x86 for now. We keep the old method for ARM until we have switched
to boot information.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit e46704df909aa3b9d1e8cff61f125b876bcfe3ee
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 17:50:23 2023 +0100
plat/common: Introduce boot information
Currently, every boot protocol (e.g., multiboot, Linux Boot Protocol)
must come with their own complete platform initialization as the
basic platform code directly works on the specific boot data
structures. The generalized boot information abstracts away from
the specifics of a certain boot protocol, thereby making it possible to
write generic platform initialization code. Every boot protocol is
thus expected to transform protocol-specific boot information to
the generic form before entering platform initialization code. In case
of chainloading via a custom unikraft loader, the loader may directly
supply boot information in the generalized format.
An important piece of boot information is the memory layout
at the time of boot. That includes the load address of the kernel,
the address of a command line, the load addresses of additional
data blobs (e.g., initrd, devicetree), and where free memory is
located. The boot information structure expresses this as a list
of memory region descriptors. For sanity checks and
compatibility a magic value and version information is provided
at the beginning of the structure.
The boot information is stored in a dedicated section (.uk_bootinfo)
that is initialized with information about the kernel image segments
and their desired access permissions at link-time. To this end, the
linker script needs to call the mkbootinfo.py script and replace the
dummy information in the allocated section with the generated
data.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit ef17c1daef1f82bd88518e2e940480f9d27e7245
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 17:27:43 2023 +0100
plat/kvm: Complement multiboot header
Multiboot uses bits in the flags field to specify what fields are
available. The commit adds definitions for the flag bits. It also
adds any other missing definitions and fixes some style issues.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 232567ec69726e491e6acaa670a5e6449b815fd8
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 17:18:04 2023 +0100
plat/kvm: Combine multiboot headers
Currently the multiboot definitions span across two headers. The one
that is used in C code only, and one that is safe to include from
assembly. Since we have the __ASSEMBLY__ macro, which is used
extensively in the project, we reduce the number of headers by
combining both multiboot headers into one.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 8cd52c8acb60977fd6921e04968e2bf7d942e704
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 17:10:57 2023 +0100
include/uk/plat/common: Add memregion list
This commit adds a platform internal memory region list type based
on the public memory region descriptor. The list is effectively a
pre-allocated array that can be filled up to its capacity.
Non-overlapping regions in the list are sorted in ascending order
according to the physical address of the memory regions.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 732df5ae694410634d978048a029b1719266e2cd
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 19:08:48 2023 +0100
plat/linuxu: Adapt to new memregion desc
Adapts the linuxu code to the new memory region descriptor
and listing functions.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 3ec186b558abb0c563d2ef2e10ec9cf30573fd89
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 18:51:31 2023 +0100
plat/xen: Adapt to new memregion desc
Adapts the XEN code to the new memory region descriptor
and listing functions.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit d732a7f6bcbb3e5d3bc9c0bc6c4df05e5d5ecf0d
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 18:50:18 2023 +0100
plat/xen/x86: Adapt to new memregion desc
Adapts the XEN/x86 code to the new memory region descriptor
and listing functions.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit cb00309fa70055b29c9f6b5957a086ebd581cb47
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 18:48:46 2023 +0100
plat/xen/arm: Adapt to new memregion desc
Adapts the XEN/arm code to the new memory region descriptor
and listing functions.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 442b1e5038364c4b079ce438d56a8e417214019c
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 16:59:10 2023 +0100
plat/common: Adapt W^X to memregion desc
Adapts the existing code to the changed declaration of the memory
region descriptor.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit fac37da39f4ba7626361d3cbd6cb4ee3b41f3b73
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 16:51:28 2023 +0100
lib/vfscore: Adapt to new memregion desc
Adapts the existing code to work with the new definition of the
memory region descriptor.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 99bac7bd2f967115e951cc3a99daadbae34bf5d6
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 16:49:54 2023 +0100
include/uk/plat: Adapt documentation style
This commit just adapts the documentation style of the unmodified
functions to be inline with the other functions and the coding
convention RFC.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 47dfe0b72d55141672010d6eb3c3eb00180e24fc
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 16:46:37 2023 +0100
include/uk/plat: Extend memory region iteration
The previous search function only accepted exact matches. The new
search function extends this by allowing to specify multiple region
types that are accepted and also select the region flags that should
be incorporated in the search. For example, this allows to list all
readable kernel segments irrespective if they are also writeable or
executable.
For improved performance, the new functions only return a pointer
to the memory region and do not create a copy. While this permits
changing the memory region descriptor, this can be useful in early
boot phases to adapt the region list.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit c2430d42a2860d1b59b86b320deb4df3cf5a3755
Author: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Date: Tue Jan 17 16:28:41 2023 +0100
include/uk/plat: Extend memory region descriptor
The commit extends the memory region descriptor in multiple ways to
make it ready to be used within the boot protocol to generically
express the memory layout of the system.
The new structure adds separate virtual and physical address fields
so that scenarios are supported where we don't have a 1:1 mapping
(e.g., when loading the kernel to a different virtual address). Types
are changed accordingly. We also embed the name so that the
descriptor becomes self-contained. The type and flags have been
separated for ease of use and because search semantics for these
are different.
Signed-off-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Reviewed-by: Marco Schlumpp <marco@unikraft.io>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Marco Schlumpp <marco@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #722
commit 693d8775dfebeff1cdb3da5dbb36f5a71d43ce4d
Author: Michalis Pappas <mpappas@fastmail.fm>
Date: Sun Nov 20 17:19:35 2022 +0100
arch/x86_64: Implement ukarch_random
Implement the ukarch_random API for x86_64. The implementation
requires that both the RDRAND and RDSEED instructions are available
in the system.
Signed-off-by: Michalis Pappas <mpappas@fastmail.fm>
Reviewed-by: Razvan Virtan <virtanrazvan@gmail.com>
Reviewed-by: Maria Sfiraiala <maria.sfiraiala@gmail.com>
Approved-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #695
commit d1d05128d0a432b805106c12fff7f43f089144d2
Author: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Date: Tue Nov 29 15:01:21 2022 +0000
lib/ukparam: Accept arguments containing `=`
Mount options of some filesystems (e.g. 9pfs) may contain a `=` sign.
Pass the whole argument as-is in that case.
Signed-off-by: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #671
commit 31f8173d0fc59442200aad0eaefc49ad905dfeb3
Author: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Date: Tue Nov 29 10:43:32 2022 +0000
lib/9pfs: Add mount option parsing
Allow specifying protocol version, uname and aname in mount options.
Signed-off-by: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #671
commit 50751c585315591728ba00c53b3ef2a598299918
Author: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Date: Thu Jun 30 18:22:28 2022 +0000
lib/9pfs: Add 9p2000.L support
Signed-off-by: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #671
commit 804cbb39c2b528019e1779af7528e8db6b8a50db
Author: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Date: Fri Oct 14 14:02:22 2022 +0000
lib/uk9p: Fix writes not accepting iounit of zero
The 9P protocol specifies that iounit may be zero. Make sure that writes
accept this value like reads already do.
Signed-off-by: Tu Dinh Ngoc <dinhngoc.tu@irit.fr>
Reviewed-by: Eduard Vintilă <eduard.vintila47@gmail.com>
Approved-by: Razvan Deaconescu <razvand@unikraft.io>
Tested-by: Unikraft CI <monkey@unikraft.io>
GitHub-Closes: #671
commit 0a70f192cf70f331e566a87299764ff6e3cd137f
Author: Marco Schlumpp <marco.schlumpp@gmail.com>
Date: Tue Oct 19 10:11:36 2021 +0200
lib/nolibc: Expose rmdir/rename from nolibc
These two functions were not exposed by the nolibc library but are
implemented by vfscore.
Signed-off-by: Marco Schlumpp <marco.schlumpp@gmail.com>
Reviewed-by: Sergiu Moga <sergiu.moga@protonmail.com>
Approved-by: Razvan Deacones…
noureddine-taleb
pushed a commit
to noureddine-taleb/unikraft
that referenced
this pull request
Mar 23, 2023
Implement the ukarch_random API for x86_64. The implementation requires that both the RDRAND and RDSEED instructions are available in the system. Signed-off-by: Michalis Pappas <mpappas@fastmail.fm> Reviewed-by: Razvan Virtan <virtanrazvan@gmail.com> Reviewed-by: Maria Sfiraiala <maria.sfiraiala@gmail.com> Approved-by: Marc Rittinghaus <marc.rittinghaus@unikraft.io> Tested-by: Unikraft CI <monkey@unikraft.io> GitHub-Closes: unikraft#695
4 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Prerequisite checklist
checkpatch.plon your commit series before opening this PR;Base target
x86_64]Additional configuration
Introduce a "Processor Features" menu under "Architecture Selection" when x86 is selected.
The
CONFIG_X86_64_HAVE_RANDOMoption enables the processor-generated randomness.Description of changes
This PR introduces an implement the ukarch_random API for x86_64. The implementation uses the
RDRAND/RDSEEDinstructions provided by Intel's DRNG technology [1]. Each of these instructions has been introduced at different stages of the architecture's timeline:This implementation requires that both instructions are supported by the processor.
[1] https://www.intel.com/content/www/us/en/developer/articles/guide/intel-digital-random-number-generator-drng-software-implementation-guide.html