Re: Perl release v5.40.1 on a 32-bit box showing failed tests for atime and ctime in Time::HiRes

You should be worried about more than Time::HiRes. Although the module shouldn't fail, you have potentially bigger problems:

You seem to be using Debian 6 "Squeeze". Long-term support for that ended in 2014 (ten years ago) if Wikipedia is to be trusted. That potentially means you are missing a decade of security bugfixes.
You are using a 32bit operating system. These have all kinds of limitations. Including that your OS (or major libraries installed in it) potentially still use 32bit timestamps. Or use a mix of 32/64 bit timestamps, which certainly could explain some of the problems if your filesystem timestamp bitsize mismatches Perl.
It seems even the Debian maintainers are thinking of sunsetting 32bit distros.

As for me, i can't even test Time::HiRes on a 32 bit system anymore, because i have upgraded the last one more than a decade ago. I might have some ancient Raspberry PI around in some spare parts box that i could potentially get running again. But this bug might be processor/file system specific. What filesystem are you running on?

PerlMonks XP is useless? Not anymore: XPD - Do more with your PerlMonks XP
Also check out my sisters artwork and my weekly webcomics

Comment on Re: Perl release v5.40.1 on a 32-bit box showing failed tests for atime and ctime in Time::HiRes

Replies are listed 'Best First'.
Re^2: Perl release v5.40.1 on a 32-bit box showing failed tests for atime and ctime in Time::HiRes by Intrepid (Curate) on Mar 05, 2025 at 19:49 UTC
Hi cavac, thanks for the reply. You raised good questions. I'd like to get a couple more years use out of this laptop but I see that I could run into problems. You wrote: >You are using a 32bit operating system. These have all kinds of limitations. Including that your OS (or major libraries installed in it) potentially still use 32bit timestamps. Or use a mix of 32/64 bit timestamps, which certainly could explain some of the problems if your filesystem timestamp bitsize mismatches Perl. It's a bit strange because I similarly "renovated" an older laptop a few months ago with a fresh OS install (the last Debian available for 32-bit machines, 4.19) and didn't see any test failures like these when I built perl and modules, that machine runs a 32-bit AMD processor and the machine in question here runs an Intel i5. >What filesystem are you running on? I'm using the ext4 filesystem. Although probably based on Squeeze as you say, this machine actually got Linux Mint (LMDE) installed on it. Mint seems to be supporting 32-bit boxes long than Debian themselves. Thanks again for your time and attention and suggestions. Mar 05, 2025 at 19:46 UTC	[reply]
Re^3: Perl release v5.40.1 on a 32-bit box showing failed tests for atime and ctime in Time::HiRes by cavac (Prior) on Mar 07, 2025 at 12:50 UTC
Is this laptop running a 32bit-only processor? How old is it? I have a 14 to 15 year old laptop around that i still occasionally use. I run Xubuntu 22.04 LTS (64 bit) on it and it performs reasonably well. For me, the thing that makes modern Linux on older hardware work is selecting a Desktop environment that uses few resources but still has all the modern convenience. In my case, that's XFCE. It's highly configurable, and it got a decent terminal emulator (and if you need a high performance terminal, xterm is a good choice, at least for me terminal graphics demos). I'd recommend trying to run a few 64 bit Linux distris as live system from an USB stick to see if they boot. (X)Ubuntu also has a settings program usually called "Additional drivers" that, so far, hasn't let me down when it comes to supporting graphics cards. PerlMonks XP is useless? Not anymore: XPD - Do more with your PerlMonks XP Also check out my sisters artwork and my weekly webcomics	[reply]