PRX

scetool can decrypt SPRX's, producing an ELF... or is it? Not really. It has an ELF header but...

First LOAD segment, paddr points to the descriptor for the "dependency table" or "deptable" for short:

NOTE: All addresses inside the file assume the ELF header isn't there (basically add 0xE0 to all addresses.)

Offset	Type	Description
+0	long	Always(?) 0x101
+4	char[28]	Name of this PRX, padded out with zeroes. Doesn't necessarily match the file name(why?)
+32	long	A unique module ID? Gets used a lot later...
+36	long	Points to the deptable header
+40	long	Points to the END of the deptable header
+44	long	Points to the start of first deptable entry
+48	long	points to the end of the LAST deptable entry

Deptable header:

Offset	Type	Description
+0	long[5]	Always(?) 0x1C000000, 0x80000002, 0x10000, 0, 0
+20	long	Points to three longs which are always(?) 0xBC9A0086, 0xAB779874, 0xD7F43016. The start of the mystery table.
+24	long	Points to three pointers in the mystery table. Two go to pointers to subroutines in the funcpointer table, the third goes back to the deptable descriptor.

Entry in the deptable:

Offset	Type	Description
+0	short[2]	Always(?) 0x2C00, 1
+4	short	Flags? 0x1 is always set, 0x8 often, 0x2000 seems to indicate a non-PRX library (like "stdc" or "allocator") that comes from somewhere else (LV2?)
+6	short	The number of functions the depending PRX needs from the depended PRX. There's this many entries in the Mystery Table for that PRX, and there is this many function pointers in the stubtable (see below).
+8	short	Usually 0, but "paf" gets sometimes 5, sometimes 6. More flags maybe?
+10	short[3]	Always(?) 0. Probably for alignment.
+16	long	Pointer to the ASCII string of the depended PRX's name. These don't seem to consistently map with the PRX's file name in dev_flash.
+20	long	Pointer to this library's entries in the mystery table.
+24	long	Pointer to the the pointers to function stubs for this library in the stubtable.
+28	long[2]	0, 0 for every PRX except paf, which gets two more sets in the mystery table, pointed to by these. I suspect that +8 is somehow connected to this.

Second LOAD segment is all the relocation and symbol data, save for the names (as ASCII strings) of the exposed functions (which are in the first LOAD segment along with all the other strings the PRX uses).