Live data from Hacker News

Building GCC 1.27 (first GCC with x86 support) (2019)

kristerw.blogspot.com

31–40 of 68 posts

Re: Building GCC 1.27 (first GCC with x86 support) (2019)

#31
post #15

Earlier quoted context omitted.

My high school was a “free school” in the same sense of the term. Founded 1640 by royal charter.

According to this seemingly exhaustive list ( https://privycouncil.independent.gov.uk/wp-content/uploads/2... ), no royal charters were granted in 1640.

My guess: the royal Governor of an (at-the-time) British colony issued a charter in the name of the crown, and people in that colony called it a “royal charter” since it was issued in the King’s name, but while that kind of thing may pass for a “royal charter” in the colonies, in London it was not considered to be one

Re: Building GCC 1.27 (first GCC with x86 support) (2019)

#32
post #15

Earlier quoted context omitted.

My high school was a “free school” in the same sense of the term. Founded 1640 by royal charter.

According to this seemingly exhaustive list ( https://privycouncil.independent.gov.uk/wp-content/uploads/2... ), no royal charters were granted in 1640.

That's a pretty interesting list! A lot of interesting entries there, including a couple (e.g. Needlemakers) signed by Cromwell.

Given that it was issued by the privy counsel it is interesting that it has some notable omissions (though how comprehensive can the records be over such a long period, anyway?). I thought at first it might have been limited to charters in the current UK, but it does have the Massachusetts Company (1629), though not Virginia's charter (1606). Has the East and West India Companies, but not the College of William and Mary (1693) -- there are images of that one on the web, though not the signature.

My school was also in the Massachusetts colony, the somewhat pompous Roxbury Latin, 1643, charter 1645 (the unfortunate Charles I). It's been more than 40 years since I clapped eyes on its charter - astonishing they haven't put it online. My memory that it was 1640 was corrected by a bit of web searching.

Really these things are archaic silly relics, like monarchies themselves.

Re: Building GCC 1.27 (first GCC with x86 support) (2019)

#33

Earlier quoted context omitted.

According to this seemingly exhaustive list ( https://privycouncil.independent.gov.uk/wp-content/uploads/2... ), no royal charters were granted in 1640.

My guess: the royal Governor of an (at-the-time) British colony issued a charter in the name of the crown, and people in that colony called it a “royal charter” since it was issued in the King’s name, but while that kind of thing may pass for a “royal charter” in the colonies, in London it was not considered to be one

I don't know. While looking around to see if I could find my school's charter online I discovered that Harvard never had a royal charter, though at various times they sought one. On the other hand College of William and Mary got one (from the eponymous co-rulers) in 1693.

As my parallel comment noted, there are some notable omissions in that list, quite unsurprising given its thousand year scope.

Re: Building GCC 1.27 (first GCC with x86 support) (2019)

#34
post #7

Nitpick: calling this an "x86" compiler is a modernism. There were, indeed, multiple successor chips to the 8086 in the market. But gcc supported only one of them. This is a 386 compiler, not an x86 compiler.

Unless it is not the first version that targets any x86, then there is no nit to pick.

The point to a nitpick is to point out something is correct, but misleading. In 1988, no one alive was calling this an "x86" compiler. It was specifically targeting the (then quite new -- about 1.5 years old) 80386 chip, which at the time was the only implementation of the architecture that we understand today as "x86".

Re: Building GCC 1.27 (first GCC with x86 support) (2019)

#35
post #34

Earlier quoted context omitted.

Unless it is not the first version that targets any x86, then there is no nit to pick.

The point to a nitpick is to point out something is correct, but misleading. In 1988, no one alive was calling this an "x86" compiler. It was specifically targeting the (then quite new -- about 1.5 years old) 80386 chip, which at the time was the only implementation of the architecture that we understand today as "x86".

x86 was originally used to refer to any intel cpu of x86 name, even in the late 80's

Re: Building GCC 1.27 (first GCC with x86 support) (2019)

#36
post #34

Earlier quoted context omitted.

The point to a nitpick is to point out something is correct, but misleading. In 1988, no one alive was calling this an "x86" compiler. It was specifically targeting the (then quite new -- about 1.5 years old) 80386 chip, which at the time was the only implementation of the architecture that we understand today as "x86".

x86 was originally used to refer to any intel cpu of x86 name, even in the late 80's

By which definition, this is absolutely not an "x86" compiler, since it doesn't generate code for any of those other chips. Thus my nitpick.

Re: Building GCC 1.27 (first GCC with x86 support) (2019)

#37
post #20
post #18

Earlier quoted context omitted.

A 750 wasn't very fast back in 1987 if I remember it correctly? But a 784 was still decent?

Not very fast, but it was a bit of a standard in the BSD world as most development there happened on 750s at Berkeley. https://gunkies.org/wiki/4.1_BSD

Yup. My point was that I can understand if someone rejects a 750 in 1987 because it is fairly slow. I think home computers like the Amiga and Atari ST had at least that much raw CPU power?

But VAXen in general in 1987 weren't yet hopelessly slow, right?

I can understand if people didn't like VAXes for personal reasons. I had a pick of both a fast VAX and fast Sun workstations in 1987, the Suns were clear winners for me.

Re: Building GCC 1.27 (first GCC with x86 support) (2019)

#38
post #26
post #18

Earlier quoted context omitted.

A 750 wasn't very fast back in 1987 if I remember it correctly? But a 784 was still decent?

No, but it had been lying around unused in a machine room, powered up, for a few years. RMS published the GNU manifesto in 83 or 84.

Wasn't VAX assembly very orthogonal and nice? I can understand if it was used as a compiler target, whereas x86 strikes me as harder to deal with?

So if you have the space and free electricity in 1987, sure, give me a fast VAX.

Not so much if you live in a student dorm, but I'm sure some have done that too.

Re: Building GCC 1.27 (first GCC with x86 support) (2019)

#39
post #34

Earlier quoted context omitted.

The point to a nitpick is to point out something is correct, but misleading. In 1988, no one alive was calling this an "x86" compiler. It was specifically targeting the (then quite new -- about 1.5 years old) 80386 chip, which at the time was the only implementation of the architecture that we understand today as "x86".

x86 was originally used to refer to any intel cpu of x86 name, even in the late 80's

In my circles, I think we just used the CPU name since there weren't that many of them yet, and unlike 386/486/586/etc, the 286 and lower CPUs weren't real 32-bitters. So, "386", "286" etc. Or just "Intel" to refer to the whole CPU family.

IIRC, I started hearing "x86" once the 486 came out in 1989.

Re: Building GCC 1.27 (first GCC with x86 support) (2019)

#40
post #33

Earlier quoted context omitted.

My guess: the royal Governor of an (at-the-time) British colony issued a charter in the name of the crown, and people in that colony called it a “royal charter” since it was issued in the King’s name, but while that kind of thing may pass for a “royal charter” in the colonies, in London it was not considered to be one

I don't know. While looking around to see if I could find my school's charter online I discovered that Harvard never had a royal charter, though at various times they sought one. On the other hand College of William and Mary got one (from the eponymous co-rulers) in 1693. As my parallel comment noted, there are some notable omissions in that list, quite unsurprising given its thousand year scope.

Wikipedia’s article on Thomas Dudley, governor of Massachusetts Bay Colony, says he issued one of Harvard’s charters, and provided land and funds for the founding of Roxbury Latin. Since as Governor he was technically the monarch’s representative, I could see why both Harvard and Roxbury might have elided the technical distinction between “chartered by the King’s official representative” (vice regal) and “chartered by the King personally” (regal)

> As my parallel comment noted, there are some notable omissions in that list, quite unsurprising given its thousand year scope.

The only "Royal Australian" body it has is the Royal Australian Chemical Institute (RACI, to which my father belongs), which is the national professional body for chemistry. But, it is missing many other "Royal Australian” organisations, including all the “Royal Australian” medical professional colleges (e.g. the Royal Australian College of General Practice, to which belongs my mother). I know under Australian law, it is illegal to call yourself “Royal” without approval from Buckingham Palace. But, possibly, that approval isn’t technically considered a “royal charter”-since 1946, the Australian federal and state governments effectively control the use of “Royal”, the Australian government sends the paperwork to Buckingham Palace for the monarch’s signature, but the monarch will sign anything if the right person gives it to them. However, that’s out of the Privy Council’s jurisdiction (which is the UK only), so the Privy Council might not know about it. The RACI is on this list, because it was 1934, and that was before control over the use of the word “Royal” was transferred from the British government to the Australian ones, and hence the Privy Council would have processed the application. EDIT: One page on their website notes that getting permission to call yourself "Royal X" is technically a separate process from getting a royal charter, you can have one without the other

Another guess: the College of William and Mary got a letter from the eponymous monarchs approving its use of their names, but the Privy Council doesn’t technically consider that letter to be a “royal charter”, because it didn’t use the right magic code words to count as one

Post reply on HN