Add comments to vDSO hwcap loading process.

Loading of the vDSO pseudo-hwcap from the type 2 GNU note is
a rather arcane and poorly documented process. Given that I had
a chance to review this code today I thought I would add all
of the things I had to lookup to verify the validity of the
process.

With a single .note.GNU the vDSO can register up to 64 flags,
though in practice you are limited to 64 - _DL_FIRST_EXTRA
bits which on x86 is 12 bits.

The only use of this that I know of is in the Xen support
in Linux where they use the 1st bit to indicate "nosegneg".
I see "We use bit 1 to avoid bugs in some versions of glibc
when bit 0 is used; the choice is otherwise arbitrary.", but
no reference to a glibc bug anywhere. The code as-is should
support bit zero, so we still have that free for future use.

The kernel, glibc, and ld.so.cache must coordinate to ensure
that bit values don't go too high and are used consistently.

---

2013-05-13  Carlos O'Donell  <carlos@redhat.com>

	* elf/dl-hwcaps.c (_dl_important_hwcaps): Comment vDSO hwcap loading.
	* elf/ldconfig.c (is_hwcap_platform): Comment each hwcap check.
	(main): Comment "tls" pseudo-hwcap.

1 files changed, 13 insertions, 1 deletions

diff --git a/elf/dl-hwcaps.c b/elf/dl-hwcaps.c
index 8d49383d76..1b7fe52a6a 100644
--- a/elf/dl-hwcaps.c
+++ b/elf/dl-hwcaps.c
@@ -66,6 +66,11 @@ _dl_important_hwcaps (const char *platform, size_t platform_len, size_t *sz,
 	  {
 	    const ElfW(Addr) start = (phdr[i].p_vaddr
 				      + GLRO(dl_sysinfo_map)->l_addr);
+	    /* The standard ELF note layout is exactly as the anonymous struct.
+	       The next element is a variable length vendor name of length
+	       VENDORLEN (with a real length rounded to ElfW(Addr)), followed
+	       by the data of length DATALEN (with a real length rounded to
+	       ElfW(Addr)).  */
 	    const struct
 	    {
 	      ElfW(Word) vendorlen;
@@ -75,6 +80,11 @@ _dl_important_hwcaps (const char *platform, size_t platform_len, size_t *sz,
 	    while ((ElfW(Addr)) (note + 1) - start < phdr[i].p_memsz)
 	      {
 #define ROUND(len) (((len) + sizeof (ElfW(Word)) - 1) & -sizeof (ElfW(Word)))
+		/* The layout of the type 2, vendor "GNU" note is as follows:
+		   .long <Number of capabilities enabled by this note>
+		   .long <Capabilities mask> (as mask >> _DL_FIRST_EXTRA).
+		   .byte <The bit number for the next capability>
+		   .asciz <The name of the capability>.  */
 		if (note->type == NT_GNU_HWCAP
 		    && note->vendorlen == sizeof "GNU"
 		    && !memcmp ((note + 1), "GNU", sizeof "GNU")
@@ -84,7 +94,7 @@ _dl_important_hwcaps (const char *platform, size_t platform_len, size_t *sz,
 					   + ROUND (sizeof "GNU"));
 		    cnt += *p++;
 		    ++p;	/* Skip mask word.  */
-		    dsocaps = (const char *) p;
+		    dsocaps = (const char *) p; /* Pseudo-string "<b>name"  */
 		    dsocapslen = note->datalen - sizeof *p * 2;
 		    break;
 		  }
@@ -107,6 +117,8 @@ _dl_important_hwcaps (const char *platform, size_t platform_len, size_t *sz,
 #ifdef NEED_DL_SYSINFO_DSO
   if (dsocaps != NULL)
     {
+      /* dsocaps points to the .asciz string, and -1 points to the mask
+         .long just before the string.  */
       const ElfW(Word) mask = ((const ElfW(Word) *) dsocaps)[-1];
       GLRO(dl_hwcap) |= (uint64_t) mask << _DL_FIRST_EXTRA;
       /* Note that we add the dsocaps to the set already chosen by the