Comments for Hardwarebug

Comment on ARM inline asm secrets by Gregory Eckersley

Gregory Eckersley — Tue, 15 May 2012 02:38:01 +0000

Thanks for your web page. The GCC constraints are unfathomably complex. I wanted 2 lines of assembly code as follows " svc sym1", ".word symbol2". No constraint would allow symbol2 defined either as a symbolic address or 32 bit constant. Some crashed gcc. e.g:

#define svcgen(sym1,symbol2) __asm__("SVC %0\n\t .word %c1\n\t" : : "i" (sym1) , "i" (symbol2)  : )

will accept 32 bit constants, not symbols when invoked.

"svcgen(0x40,32_bit_constant)" works
"svcgen(0x40,symbol)" fails on constraint

My solution was:

#define svcgen(sym1,symbol2) __asm__("SVC %0\n\t .word " symbol2 "\n\t" : : "i" (sym1)   : )

Invoked by, e.g. :

svcgen(0x40,"c_variable_or_constant")

The main thing to realise is that gcc inline assembler constraints,however silly, cannot be circumvented. Quotes at at least allowed the constraints to be dodged! Your web page helped me realise this earlier - you are definitely not trolling - thanks again. GPE

Comment on ARM compiler shoot-out by A few things iOS developers ought to know about the ARM architecture « AVATAR.Dev - iOS Developer Tips, Tricks and Tutorials.

Mon, 23 Jan 2012 07:41:36 +0000

[...] While I’ve not tested it much, I have at least seen nice improvements on 64-bit integer code (a particularly weak point of GCC on ARM) when using LLVM. Hopefully, LLVM will prove to be an improvement over GCC in all domains. [...]

Comment on Ogg objections by Why didn't Google use OGG or MPEG-4 Part 12 as the container format for WebM? - Quora

Sun, 06 Nov 2011 22:48:42 +0000

[...] add comment if you are logged out. 2 Answers Kieran Kunhya, Homo Sapiens http://hardwarebug.org/2010/03/0… sums up the technical reasons behind why one would not want to use ogg. There's no technical [...]

Comment on Pointer peril by Mans

Mans — Thu, 20 Oct 2011 15:17:46 +0000

Your program prints the difference between two addresses, computation of which is, as you say, undefined. In this particular case, with your compiler, nothing unexpected happens. Undefined behaviour is not required to do anything strange.

Comment on Pointer peril by Alexandre

Alexandre — Thu, 20 Oct 2011 14:41:12 +0000

On my machine, the following function yields the same result with and without -O0 (gcc 4.5.2). According to my understanding of the specs, it relies on the same undefined behaviour. Or do I misunderstand "array objects"?

#include 
#include 

void foo(int* addr_a) {
  int b;
  ptrdiff_t diff = addr_a - &b;
  printf("&a = %p   --- &b = %p\n", addr_a, &b);
  printf("diff = 0x%x\n", diff);
}

int main(void) {
  int a;
  foo(&a);
  return 0;
}

Comment on Pointer peril by Rémi Denis-Courmont

Rémi Denis-Courmont — Tue, 18 Oct 2011 14:14:55 +0000

As far as I know, nothing in the C standard promises that a and b are on a sizeof(int) address boundary. So even without optimizations, you could still have rounding errors ;-) depending on the ABI obviously.

Comment on Pointer peril by Kostya

Kostya — Tue, 18 Oct 2011 07:23:15 +0000

I wonder how many bugs were reported for GCC producing “wrong” code when pointer aliasing is used. Or other things that ignore undefined behaviour.

All world runs on x86 CPUs after all…

Comment on ARM inline asm secrets by Julien

Julien — Tue, 10 May 2011 01:09:07 +0000

This seems fixed in gcc 4.5.

Comment on Shades of black by av500

av500 — Mon, 14 Feb 2011 22:38:07 +0000

46 DC EA …

Comment on Shades of black by Mans

Mans — Mon, 14 Feb 2011 15:48:46 +0000

I ended up getting a Sony VAIO Z1. Go on, ridicule me for paying the Sony tax. You know you’ll enjoy it.