Hearing the difference now isn't the reason to choose an optimal memcpy.
Some implementations use lossless copying, while other methods are 'lossy'. The differences between GNU libc and Boost are subtle, but important. What this means is that for each year your audio sits on your hard drive, it will lose roughly 12kbps, assuming you have NVMe - it's about 15kbps on M.2, but only 7kbps on eSATA, due to linear velocidensity. You don't want to know how much worse it is on CD-ROM or other optical media.
I started collecting MP3s in about 2001 (on Windows 98 using an ancient version of malloc and memcpy), and if I try to play any of the tracks I downloaded back then, even the stuff I grabbed at 320kbps, they just sound like crap. The bass is terrible, the midrange...well don't get me started. Some of those albums have degraded down to 32 or even 16kbps. If you ever used uClibc, I hope you kept your original vinyls! Proper memcpy rips (or better yet, ripping directly to factory-fresh RAM) from the same period still sound great, even if they weren't stored correctly, in a cool, dry place.
Seriously, stick to audiophile-grade library functions: you may not be able to hear the difference now, but in a year or two, you'll be glad you did.
How do SSDs compare for storing lossy media? And how does temperature factor in?
Out of a gut feeling I've been taking out my music SSD and storing it in the freezer when I'm gone for the weekend, or on holidays etc, as to retard the lossy decay process. But I've never actually measured, maybe you can throw in some expertise here?
Be careful. If your freezer is really cold, you can turn your CDs into superconductors, which may not lose bits, but they do introduce quantum shuffling of bits, where the same bits physically exchange position with other bits.
If you're serious about this, install Gentoo so you can compile everything yourself! Are you really going to entrust your music collection to a distro full of precompiled packages of unknown provenance, likely riddled with dirty bits from cheap second-hand hardware and possibly coal-fired electricity?
Some implementations use lossless copying, while other methods are 'lossy'. The differences between GNU libc and Boost are subtle, but important. What this means is that for each year your audio sits on your hard drive, it will lose roughly 12kbps, assuming you have NVMe - it's about 15kbps on M.2, but only 7kbps on eSATA, due to linear velocidensity. You don't want to know how much worse it is on CD-ROM or other optical media.
I started collecting MP3s in about 2001 (on Windows 98 using an ancient version of malloc and memcpy), and if I try to play any of the tracks I downloaded back then, even the stuff I grabbed at 320kbps, they just sound like crap. The bass is terrible, the midrange...well don't get me started. Some of those albums have degraded down to 32 or even 16kbps. If you ever used uClibc, I hope you kept your original vinyls! Proper memcpy rips (or better yet, ripping directly to factory-fresh RAM) from the same period still sound great, even if they weren't stored correctly, in a cool, dry place.
Seriously, stick to audiophile-grade library functions: you may not be able to hear the difference now, but in a year or two, you'll be glad you did.