deps: bump uart_16550 to 0.6.0 by phip1611 · Pull Request #565 · rust-osdev/bootloader · GitHub

Let's use the latest and greatest version

Thank you for your contribution!

Should we replace SerialPort with Uart16550Tty? Both implement core::fmt::Write, so I don't see much of a reason to keep SerialPort around. Uart16550Tty also does the \n to \r\n conversion SerialPort does.

I tried but for some reason on of these checks fails:

This surprises me becaue in the integration test of uart_16550 it works in QEMU. I don't have time to investigate for now. Can we move forward?

I've added a TODO comment in the code

Looks to me like it's spinning somewhere. I think it's spinning in receive_bytes_exact.

could you help me investigating please? I'd love to understand why it fails

could you help me investigating please? I'd love to understand why it fails

I tried looking into this yesterday. I think it's some kind of race condition in the self-test code. I can't trigger the issue reliably though. Adding artificial system load seems to make it easier to reproduce, but it's still not completely reliable.
For some reason, I had a harder time reproducing the error with the latest HEAD for uart_16550. I'm not sure why.
I might be able to take another look at this later today.
I think my next step is going to be looking at the serial port implementation in QEMU to see if there's anything that could explain the race condition (maybe asynchronous processing?).

Interesting - the UEFI console takes ownership of the serial device by default. One must therefore disconnect the UEFI console from using the serial device (as long as boot services are active but one wants to use the serial device). I think this is the problem.

The forced disconnct is also fairly easy

The forced disconnct is also fairly easy

How would one do that?

Here's something interesting: Wrapping the call to Uart16550Tty::new_port in x86_64::instructions::interrupts::without_interrupts fixes the issue. Disabling interrupts in the tty config (config.interrupts = IER::empty();) does not work however.

The forced disconnct is also fairly easy

How would one do that?

https://github.com/phip1611/uefi-serial-chat/blob/47cc3961ab047d698790dc48131799aa10c18c65/src/main.rs#L104

Regarding interrupts:thanks! That's something I can use for my testing and debugging

BTW I'm preparing an update of the uefi dependency

Here's something interesting: Wrapping the call to Uart16550Tty::new_port in x86_64::instructions::interrupts::without_interrupts fixes the issue. Disabling interrupts in the tty config (config.interrupts = IER::empty();) does not work however.

Found the issue and fixed/documented everything. v0.8.0 will contain the fix. More info: rust-osdev/uart_16550#69

Uart16550 has a layout of (32 /* size*/, 4 /* align*/) whereas the old SerialPort had (2, 2). The test shows that this small change is sufficient to fail the test with the old stack size. We slightly increase the stack therefore.

LGTM, thanks!

I think this caused a regression on real hardware (see #573) because of rust-osdev/uart_16550#65.

Should we release a workaround until a fix is available?

I would be in favor. I see two options: either revert this PR for now or gate the serial printing on check_connected. The latter should fix the hang, but would skip serial output on typical TX/RX/GND serial adapters...

I would be in favor. I see two options: either revert this PR for now or gate the serial printing on check_connected. The latter should fix the hang, but would skip serial output on typical TX/RX/GND serial adapters...

Let's go with the former.

Experience has shown that enabling interrupts by default can lead to surprising behavior during initialization. In particular, if THR_EMPTY is enabled, init() may immediately trigger a THR_EMPTY interrupt before the function returns, unless initialization is performed with CPU interrupts disabled [0]. This is valid UART behavior, since the transmitter is empty when the interrupt is enabled, but it means that callers generally need to run the initialization sequence with CPU interrupts disabled to avoid handling an interrupt before init() has returned. Avoid imposing this requirement through the default configuration. Interrupts should instead be configured explicitly by API users according to their use case, without applying a smart or opinionated default selection. [0] rust-osdev/bootloader#565 (comment)

Experience has shown that enabling interrupts by default can lead to surprising behavior during initialization, unless initialization is performed with CPU interrupts disabled [0]. Let's use a safer and simpler default. [0] rust-osdev/bootloader#565 (comment)

phip1611 self-assigned this Jun 14, 2026

phip1611 force-pushed the uart branch from 3911077 to 2b1e66e Compare June 14, 2026 19:47

phip1611 changed the title deps: bump uart_16550 to 0.6.0 + streamline as workspace dep deps: bump uart_16550 to 0.6.0 Jun 14, 2026

phip1611 requested a review from phil-opp June 14, 2026 19:51

phip1611 force-pushed the uart branch from 2ebe1db to 99a0164 Compare June 14, 2026 19:54

phip1611 requested a review from Freax13 June 14, 2026 19:54

phip1611 force-pushed the uart branch 2 times, most recently from 69442f7 to 21dd2d3 Compare June 14, 2026 20:28

phip1611 commented Jun 14, 2026

View reviewed changes

Comment thread tests/test_kernels/min_stack/src/bin/basic_boot.rs Outdated

Freax13 reviewed Jun 15, 2026

View reviewed changes

phip1611 added 2 commits June 16, 2026 08:07

deps: bump uart_16550 to 0.6.0

6e92528

tests: increase stack size for min_stack test

d5935c0

Uart16550 has a layout of (32 /* size*/, 4 /* align*/) whereas the old SerialPort had (2, 2). The test shows that this small change is sufficient to fail the test with the old stack size. We slightly increase the stack therefore.

phip1611 force-pushed the uart branch from 21dd2d3 to d5935c0 Compare June 16, 2026 06:07

phip1611 requested a review from Freax13 June 16, 2026 06:08

Freax13 approved these changes Jun 16, 2026

View reviewed changes

Freax13 merged commit 617e92c into rust-osdev:main Jun 16, 2026
9 checks passed

phip1611 deleted the uart branch June 17, 2026 05:24

phil-opp mentioned this pull request Jul 7, 2026

Release 0.11.16 #570

Merged

phil-opp mentioned this pull request Jul 23, 2026

No boot on real hardware #573

Open

phil-opp mentioned this pull request Jul 26, 2026

Revert uart_16550 version bump to 0.6.0 #575

Merged

phip1611 mentioned this pull request Aug 16, 2026

deps: bump uart_16550 to 0.8.0 + fix UEFI weirdness #580

Merged

               use core::fmt;
+              use uart_16550::backend::PioBackend;
               pub struct SerialPort {

deps: bump uart_16550 to 0.6.0 by phip1611 · Pull Request #565 · rust-osdev/bootloader · GitHub

Uh oh!

deps: bump uart_16550 to 0.6.0 - #565

deps: bump uart_16550 to 0.6.0#565Freax13 merged 2 commits intorust-osdev:mainrust-osdev/bootloader:mainfrom phip1611:uartphip1611/bootloader:uartCopy head branch name to clipboard

Conversation

phip1611 commented Jun 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Freax13 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

phip1611 Jun 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

phip1611 Jun 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Freax13 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

phil-opp commented Jul 23, 2026

Uh oh!

Freax13 commented Jul 23, 2026

Uh oh!

phil-opp commented Jul 26, 2026

Uh oh!

Freax13 commented Jul 26, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

deps: bump uart_16550 to 0.6.0#565
Freax13 merged 2 commits into
rust-osdev:mainfrom
phip1611:uart

phip1611 commented Jun 14, 2026 •
edited
Loading

phip1611 Jun 17, 2026 •
edited
Loading

phip1611 Jun 17, 2026 •
edited
Loading