S390: Save and restore fprs/vrs while resolving symbols.

On s390, no fpr/vrs were saved while resolving a symbol
via _dl_runtime_resolve/_dl_runtime_profile.

According to the abi, the fpr-arguments are defined as call clobbered.
In leaf-functions, gcc 4.9 and newer can use fprs for saving/restoring gprs
instead of saving them to the stack.
If gcc do this in one of the resolver-functions, then the floating point
arguments of a library-function are invalid for the first library-function-call.
Thus, this patch saves/restores the fprs around the resolving code.

The same could occur for vector registers. Furthermore an ifunc-resolver
could also clobber the vector/floating point argument registers.
Thus this patch provides the further variants _dl_runtime_resolve_vx/
_dl_runtime_profile_vx, which are used if the kernel claims, that
we run on a machine with vector registers.

Furthermore, if _dl_runtime_profile calls _dl_call_pltexit,
the pointers to inregs-/outregs-structs were setup invalid.
Now they point to the correct location in the stack-frame.
Before branching back to the caller, the return values are now
restored instead of containing the return values of the
_dl_call_pltexit() call.
On s390-32, an endless loop occurs if _dl_call_pltexit() should be called.
Now, this code-path branches to this function instead of just after the
preceding basr-instruction.

ChangeLog:

	* sysdeps/s390/s390-32/dl-trampoline.S: Include dl-trampoline.h twice
	to create a non-vector/vector version for _dl_runtime_resolve and
	_dl_runtime_profile. Move implementation to ...
	* sysdeps/s390/s390-32/dl-trampoline.h: ... here.
	(_dl_runtime_resolve) Save and restore fpr/vrs.
	(_dl_runtime_profile) Save and restore vrs and fix some issues
	if _dl_call_pltexit is called.
	* sysdeps/s390/s390-32/dl-machine.h (elf_machine_runtime_setup):
	Choose the correct resolver function if running on a machine with vx.
	* sysdeps/s390/s390-64/dl-trampoline.S: Include dl-trampoline.h twice
	to create a non-vector/vector version for _dl_runtime_resolve and
	_dl_runtime_profile. Move implementation to ...
	* sysdeps/s390/s390-64/dl-trampoline.h: ... here.
	(_dl_runtime_resolve) Save and restore fpr/vrs.
	(_dl_runtime_profile) Save and restore vrs and fix some issues
	* sysdeps/s390/s390-64/dl-machine.h: (elf_machine_runtime_setup):
	Choose the correct resolver function if running on a machine with vx.

1 files changed, 11 insertions, 119 deletions

diff --git a/sysdeps/s390/s390-64/dl-trampoline.S b/sysdeps/s390/s390-64/dl-trampoline.S
index 6919ed0138..1b0c9e2680 100644
--- a/sysdeps/s390/s390-64/dl-trampoline.S
+++ b/sysdeps/s390/s390-64/dl-trampoline.S
@@ -16,126 +16,18 @@
    License along with the GNU C Library; if not, see
    <http://www.gnu.org/licenses/>.  */
 
-/* The PLT stubs will call _dl_runtime_resolve/_dl_runtime_profile
- * with the following linkage:
- *   r2 - r6 : parameter registers
- *   f0, f2, f4, f6 : floating point parameter registers
- *   48(r15), 56(r15) : PLT arguments PLT1, PLT2
- *   160(r15) : additional stack parameters
- * The normal clobber rules for function calls apply:
- *   r0 - r5 : call clobbered
- *   r6 - r13 :	 call saved
- *   r14 : return address (call clobbered)
- *   r15 : stack pointer (call saved)
- *   f1, f3, f5, f7 : call saved
- *   f0 - f3, f5, f7 - f15 : call clobbered
- */
-
 #include <sysdep.h>
 
 	.text
-	.globl _dl_runtime_resolve
-	.type _dl_runtime_resolve, @function
-	cfi_startproc
-	.align 16
-_dl_runtime_resolve:
-	stmg   %r2,%r5,64(15)	# save call-clobbered argument registers
-	stg    %r14,96(15)
-	cfi_offset (r14, -64)
-	lgr    %r0,%r15
-	aghi   %r15,-160	# create stack frame
-	cfi_adjust_cfa_offset (160)
-	stg    %r0,0(%r15)      # write backchain
-	lmg    %r2,%r3,208(%r15)# load args saved by PLT
-	brasl  %r14,_dl_fixup	# call fixup
-	lgr    %r1,%r2		# function addr returned in r2
-	aghi   %r15,160		# remove stack frame
-	cfi_adjust_cfa_offset (-160)
-	lg     %r14,96(15)	# restore registers
-	lmg    %r2,%r5,64(15)
-	br     %r1
-	cfi_endproc
-	.size _dl_runtime_resolve, .-_dl_runtime_resolve
-
-
-#ifndef PROF
-	.globl _dl_runtime_profile
-	.type _dl_runtime_profile, @function
-	cfi_startproc
-	.align 16
-_dl_runtime_profile:
-	stmg   %r2,%r6,64(%r15)		# save call-clobbered arg regs
-	std    %f0,104(%r15)		# + r6 needed as arg for
-	std    %f2,112(%r15)		#  _dl_profile_fixup
-	std    %f4,120(%r15)
-	std    %f6,128(%r15)
-	stg    %r12,24(%r15)		# r12 is used as backup of r15
-	stg    %r14,32(%r15)
-	cfi_offset (r6, -96)
-	cfi_offset (f0, -56)
-	cfi_offset (f2, -48)
-	cfi_offset (f4, -40)
-	cfi_offset (f6, -32)
-	cfi_offset (r12, -136)
-	cfi_offset (r14, -128)
-	lgr    %r12,%r15		# backup stack pointer
-	cfi_def_cfa_register (12)
-	aghi   %r15,-160		# create stack frame
-	stg    %r12,0(%r15)		# save backchain
-	lmg    %r2,%r3,48(%r12)		# load arguments saved by PLT
-	lgr    %r4,%r14			# return address as third parameter
-	la     %r5,64(%r12)		# pointer to struct La_s390_32_regs
-	la     %r6,40(%r12)		# long int * framesize
-	brasl  %r14,_dl_profile_fixup	# call resolver
-	lgr    %r1,%r2			# function addr returned in r2
-	lg     %r0,40(%r12)		# load framesize
-	ltgr   %r0,%r0
-	jnm    1f
-
-	lmg    %r2,%r6,64(%r12)		# framesize < 0 means no pltexit call
-	ld     %f0,104(%r12)		# so we can do a tail call without
-	ld     %f2,112(%r12)		# copying the arg overflow area
-	ld     %f4,120(%r12)
-	ld     %f6,128(%r12)
-
-	lgr    %r15,%r12		# remove stack frame
-	cfi_def_cfa_register (15)
-	lg     %r14,32(%r15)		# restore registers
-	lg     %r12,24(%r15)
-	br     %r1			# tail-call to resolved function
-
-	cfi_def_cfa_register (12)
-1:	jz     4f			# framesize == 0 ?
-	aghi   %r0,7			# align framesize to 8
-	nill   %r0,0xfff8
-	slgr   %r15,%r0			# make room for framesize bytes
-	stg    %r12,0(%r15)
-	la     %r2,160(%r15)
-	la     %r3,160(%r12)
-	srlg   %r0,%r0,3
-3:	mvc    0(8,%r2),0(%r3)		# copy additional parameters
-	la     %r2,8(%r2)
-	la     %r3,8(%r3)
-	brctg  %r0,3b
-4:	lmg    %r2,%r6,64(%r12)		# load register parameters
-	ld     %f0,104(%r12)            # restore call-clobbered arg regs
-	ld     %f2,112(%r12)
-	ld     %f4,120(%r12)
-	ld     %f6,128(%r12)
-	basr   %r14,%r1			# call resolved function
-	stg    %r2,136(%r12)
-	std    %f0,144(%r12)
-	lmg    %r2,%r3,48(%r12)		# load arguments saved by PLT
-	la     %r4,32(%r12)		# pointer to struct La_s390_32_regs
-	la     %r5,72(%r12)		# pointer to struct La_s390_32_retval
-	brasl  %r14,_dl_call_pltexit
-
-	lgr    %r15,%r12		# remove stack frame
-	cfi_def_cfa_register (15)
-	lg     %r14,32(%r15)		# restore registers
-	lg     %r12,24(%r15)
-	br     %r14
-
-	cfi_endproc
-	.size _dl_runtime_profile, .-_dl_runtime_profile
+/* Create variant of _dl_runtime_resolve/profile for machines before z13.
+   No vector registers are saved/restored.  */
+#include <dl-trampoline.h>
+
+#if defined HAVE_S390_VX_ASM_SUPPORT
+/* Create variant of _dl_runtime_resolve/profile for z13 and newer.
+   The vector registers are saved/restored, too.*/
+# define _dl_runtime_resolve _dl_runtime_resolve_vx
+# define _dl_runtime_profile _dl_runtime_profile_vx
+# define RESTORE_VRS
+# include <dl-trampoline.h>
 #endif