Kernel: Use write(2) to /dev/(u)random as entropy source #24992

logkos · 2024-09-03T12:22:37Z

This feature allows userspace to easily supply kernel random event pool with use of write(2) on random device nodes. This function is widely adopted by other UNIX-like systems and allows use of software events as an entropy source. Before we make sure on effects of repetitive or malicious input, writing to random device nodes is exclusive to root.

Kernel/Devices/Generic/RandomDevice.cpp

supercomputer7 · 2024-09-03T14:05:39Z

Is there an actual use case for this? I don't see any userspace program using this functionality.
Are you expecting users to run echo 'something random' > /dev/urandom?
What happens if malicious program tries to make /dev/urandom unusable by entering the same input over and over again? Or is this scenario not possible? we should always be cautious with user input to things like this device.

logkos · 2024-09-03T15:49:31Z

Linux and BSDs allow writes to /dev/(u)random to fill the entropy pool (via rng-tools or something fancy like radioactive decay and CCTV footage of lava lambs). Entropy should not be pulled down in case of repetitive input since events get hashed before being sent to RNG and other sources still can get into pool. One mitigation would be changing file perms for random devices

Kernel/Devices/Generic/RandomDevice.cpp

supercomputer7 · 2024-09-06T06:29:52Z

Linux and BSDs allow writes to /dev/(u)random to fill the entropy pool (via rng-tools or something fancy like radioactive decay and CCTV footage of lava lambs). Entropy should not be pulled down in case of repetitive input since events get hashed before being sent to RNG and other sources still can get into pool. One mitigation would be changing file perms for random devices

The fact that you add support for this in the kernel is not an issue.
I just don’t understand what program will use this - I ask this because without an example of this, this would be essentially dead code.

logkos · 2024-09-06T10:50:18Z

I just don’t understand what program will use this - I ask this because without an example of this, this would be essentially dead code.

If anyone wants to supply software generated/collected entropy or aims to port rng-tools, this feature allows that. Might be used for an userspace driver

supercomputer7 · 2024-09-09T17:02:16Z

I will probably give this another review in Thursday or Friday. Anyway, we should aim at adding some sort of userspace code to use this functionality. Maybe some port can be added for this? Or just our own code…

Userland/Utilities/init.cpp

supercomputer7 · 2024-09-13T01:25:21Z

I still have mixed feelings on the topic. On one hand, I really want to see some userspace code doing something useful with this feature. On the other hand, it's really not a big change, so maybe we should just merge it as-is and think about this later on, if you say you will work on it in the future.

I will let the maintainers decide on this :)

Userland/Utilities/init.cpp

Kernel/Devices/Generic/RandomDevice.cpp

BuggieBot · 2024-09-13T16:10:49Z

Hello!

One or more of the commit messages in this PR do not match the SerenityOS code submission policy, please check the lint_commits CI job for more details on which commits were flagged and why.
Please do not close this PR and open another, instead modify your commit message(s) with git commit --amend and force push those changes to update this PR.

a write syscall from userspace can now supply entropy pool. This change stands in line with other UNIX-like systems.

We restrict write access to /dev/(u)random. So only root can feed the kernel entropy pool. This is a common practice for hardening.

timschumi

+1 on the "I would like to see something use this" front.

timschumi · 2024-09-14T11:01:59Z

Userland/Utilities/init.cpp

@@ -124,7 +124,7 @@ static ErrorOr<void> prepare_bare_minimum_devtmpfs_directory_structure()
    TRY(populate_device_node_with_symlink(DeviceNodeType::Character, "/dev/mem"sv, 0600, 1, 1));
    TRY(populate_device_node_with_symlink(DeviceNodeType::Character, "/dev/null"sv, 0666, 1, 3));
    TRY(populate_device_node_with_symlink(DeviceNodeType::Character, "/dev/full"sv, 0666, 1, 7));
-    TRY(populate_device_node_with_symlink(DeviceNodeType::Character, "/dev/random"sv, 0666, 1, 8));
+    TRY(populate_device_node_with_symlink(DeviceNodeType::Character, "/dev/random"sv, 0644, 1, 8));


Is there even such a thing as bad entropy? As far as I know/heard, you can't actually hurt the randomness of the RNG by mixing in bad data as long as it is mixed in properly.

RHEL and FreeBSD restrict writes to /dev/random for the sake of hardening so we might adopt this as well

(Or maybe no, there is much more to harvest in user apps)

github-actions bot added the 👀 pr-needs-review PR needs review from a maintainer or community member label Sep 3, 2024

supercomputer7 reviewed Sep 3, 2024

View reviewed changes

Kernel/Devices/Generic/RandomDevice.cpp Outdated Show resolved Hide resolved

logkos force-pushed the devs branch 2 times, most recently from 620103c to aada144 Compare September 3, 2024 14:10

logkos force-pushed the devs branch from aada144 to bbb979a Compare September 3, 2024 18:36

supercomputer7 reviewed Sep 6, 2024

View reviewed changes

Kernel/Devices/Generic/RandomDevice.cpp Outdated Show resolved Hide resolved

logkos force-pushed the devs branch from bbb979a to 7949469 Compare September 7, 2024 13:13

logkos requested a review from supercomputer7 September 9, 2024 07:36

supercomputer7 reviewed Sep 9, 2024

View reviewed changes

Userland/Utilities/init.cpp Show resolved Hide resolved

logkos force-pushed the devs branch from 7949469 to 6d6715a Compare September 9, 2024 17:09

supercomputer7 reviewed Sep 13, 2024

View reviewed changes

Userland/Utilities/init.cpp Show resolved Hide resolved

supercomputer7 reviewed Sep 13, 2024

View reviewed changes

Kernel/Devices/Generic/RandomDevice.cpp Show resolved Hide resolved

logkos changed the title ~~Kernel: use input to /dev/random as entropy source~~ Kernel: use write(2) to /dev/(u)random as entropy source Sep 13, 2024

logkos changed the title ~~Kernel: use write(2) to /dev/(u)random as entropy source~~ Kernel: Use write(2) to /dev/(u)random as entropy source Sep 13, 2024

logkos force-pushed the devs branch 2 times, most recently from b538524 to b49c330 Compare September 13, 2024 16:10

logkos force-pushed the devs branch from b49c330 to b028564 Compare September 13, 2024 17:13

logkos added 2 commits September 14, 2024 12:06

Kernel: Implement write(2) to /dev/random as an entropy source

247053a

a write syscall from userspace can now supply entropy pool. This change stands in line with other UNIX-like systems.

Userland/init: Change file permissions for /dev/(u)random device node

6eb7c25

We restrict write access to /dev/(u)random. So only root can feed the kernel entropy pool. This is a common practice for hardening.

logkos force-pushed the devs branch from b028564 to 6eb7c25 Compare September 14, 2024 09:07

timschumi reviewed Sep 14, 2024

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Kernel: Use write(2) to /dev/(u)random as entropy source #24992

Kernel: Use write(2) to /dev/(u)random as entropy source #24992

logkos commented Sep 3, 2024 •

edited

Loading

supercomputer7 commented Sep 3, 2024

logkos commented Sep 3, 2024 •

edited

Loading

supercomputer7 commented Sep 6, 2024

logkos commented Sep 6, 2024 •

edited

Loading

supercomputer7 commented Sep 9, 2024

supercomputer7 commented Sep 13, 2024

BuggieBot commented Sep 13, 2024

timschumi left a comment •

edited

Loading

timschumi Sep 14, 2024

logkos Sep 14, 2024 •

edited

Loading

Kernel: Use write(2) to /dev/(u)random as entropy source #24992

Are you sure you want to change the base?

Kernel: Use write(2) to /dev/(u)random as entropy source #24992

Conversation

logkos commented Sep 3, 2024 • edited Loading

supercomputer7 commented Sep 3, 2024

logkos commented Sep 3, 2024 • edited Loading

supercomputer7 commented Sep 6, 2024

logkos commented Sep 6, 2024 • edited Loading

supercomputer7 commented Sep 9, 2024

supercomputer7 commented Sep 13, 2024

BuggieBot commented Sep 13, 2024

timschumi left a comment • edited Loading

Choose a reason for hiding this comment

timschumi Sep 14, 2024

Choose a reason for hiding this comment

logkos Sep 14, 2024 • edited Loading

Choose a reason for hiding this comment

logkos commented Sep 3, 2024 •

edited

Loading

logkos commented Sep 3, 2024 •

edited

Loading

logkos commented Sep 6, 2024 •

edited

Loading

timschumi left a comment •

edited

Loading

logkos Sep 14, 2024 •

edited

Loading