Messages in this thread |  | | From | Linus Torvalds <> | Date | Mon, 18 Jul 2022 12:06:52 -0700 | Subject | Re: mainline build failure of powerpc allmodconfig for prom_init_check |
| |
On Sun, Jul 17, 2022 at 9:41 PM Michael Ellerman <mpe@ellerman.id.au> wrote: > > > li 4,254 #, > > Here we load 254 into r4, which is the 2nd parameter to memset (c).
I love how even powerpc people know that "4" is bogus, and have to make it clear that it means "r4".
I don't understand why the powerpc assembler is so messed up, and uses random integer constants for register "names".
And it gets even worse, when you start mixing FP, vector and integer "names".
I've seen many bad assemblers (in fact, I have *written* a couple of bad assemblers myself), but I have never seen anything quite that broken on any other architecture.
Oddities, yes ("$" as a prefix for register? Alpha asm is also very odd), but nothing *quite* as broken as "simple constants have entirely different meanings depending on the exact instruction and argument position".
It's not even an IBM thing. S390 uses perfectly sane register syntax, and calls things '%r4" etc.
The human-written asm files have those #define's in headers just to make things slightly more legible, because apparently the assembler doesn't even *accept* the sane names. So it's not even a "the compiler generates this abbreviated illegible mess". It's literally that the assembler is so horrid.
Why do people put up with that?
Linus
|  |