the                80 arch/arc/include/asm/entry-arcv2.h 	; (B) Manually save the complete reg file below
the               142 arch/arc/include/asm/entry-arcv2.h 	; Saving pt_regs->sp correctly requires some extra work due to the way
the               145 arch/arc/include/asm/entry-arcv2.h 	;  - K mode: add the offset from current SP where H/w starts auto push
the               147 arch/arc/include/asm/entry-arcv2.h 	; 1. Utilize the fact that Z bit is set if Intr taken in U mode
the               224 arch/arc/include/asm/entry-arcv2.h 	; _SOFT clobbers r10 restored by _HARD hence the order
the               390 arch/arm/include/asm/assembler.h 	@ explicit IT instruction needed because of the label
the               391 arch/arm/include/asm/assembler.h 	@ introduced by the USER macro
the               402 arch/arm/include/asm/assembler.h 	@ Slightly optimised to avoid incrementing the pointer twice
the                14 arch/arm/include/asm/tls.h 	mrc	p15, 0, \tmp2, c13, c0, 2	@ get the user r/w register
the                16 arch/arm/include/asm/tls.h 	mcr	p15, 0, \tpuser, c13, c0, 2	@ and the user r/w register
the                26 arch/arm/include/asm/tls.h 	mrcne	p15, 0, \tmp2, c13, c0, 2	@ get the user r/w register
the                20 arch/arm/include/asm/vfpmacros.h 	@ read all the working registers back into the VFP
the                44 arch/arm/include/asm/vfpmacros.h 	@ write all the working registers out of the VFP
the                 8 arch/m68k/fpsp040/fpsp.h |       For details on the license for this file, please see the
the                13 arch/m68k/fpsp040/fpsp.h |	These equates are used to access the exception frame, the fsave
the                14 arch/m68k/fpsp040/fpsp.h |	frame and any local variables needed by the FPSP package.
the                24 arch/m68k/fpsp040/fpsp.h |	After initialization, the stack looks like this:
the                44 arch/m68k/fpsp040/fpsp.h |	Positive offsets from A6 refer to the exception frame.  Negative
the                45 arch/m68k/fpsp040/fpsp.h |	offsets refer to the Local Variable area and the fsave area.
the                46 arch/m68k/fpsp040/fpsp.h |	The fsave frame is also accessible from the top via A7.
the                48 arch/m68k/fpsp040/fpsp.h |	On exit, the handlers execute:
the                56 arch/m68k/fpsp040/fpsp.h |	and then either "bra fpsp_done" if the exception was completely
the                57 arch/m68k/fpsp040/fpsp.h |	handled	by the package, or "bra real_xxxx" which is an external
the                58 arch/m68k/fpsp040/fpsp.h |	label to a routine that will process a real exception of the
the                59 arch/m68k/fpsp040/fpsp.h |	type that was generated.  Some handlers may omit the "frestore"
the                60 arch/m68k/fpsp040/fpsp.h |	if the FPU state after the exception is idle.
the                62 arch/m68k/fpsp040/fpsp.h |	Sometimes the exception handler will transform the fsave area
the                63 arch/m68k/fpsp040/fpsp.h |	because it needs to report an exception back to the user.  This
the                64 arch/m68k/fpsp040/fpsp.h |	can happen if the package is entered for an unimplemented float
the                66 arch/m68k/fpsp040/fpsp.h |	a second fsave frame can be pushed onto the stack and the
the                67 arch/m68k/fpsp040/fpsp.h |	handler	exit code will reload the new frame and discard the old.
the                70 arch/m68k/fpsp040/fpsp.h |	restored from the "local variable" area and can be used as
the                72 arch/m68k/fpsp040/fpsp.h |	of these registers, it should modify the saved copy and let
the                73 arch/m68k/fpsp040/fpsp.h |	the handler exit code restore the value.
the                77 arch/m68k/fpsp040/fpsp.h |	Local Variables on the stack
the               124 arch/m68k/fpsp040/fpsp.h |	Offsets are defined from the end of an fsave because the last 10
the               125 arch/m68k/fpsp040/fpsp.h |	words of a busy frame are the same as the unimplemented frame.
the               266 arch/m68k/fpsp040/fpsp.h |	FPSR combinations used in the FPSP
the               325 arch/m68k/fpsp040/fpsp.h |					are in the $40s {$40-$4f}
the              8321 arch/m68k/ifpsp060/src/fplsp.S # if the input is exactly equal to one, then exit through ld_pzero.
the              10213 arch/m68k/ifpsp060/src/fplsp.S # if enabled so the operating system can log the event.			#
the               965 arch/m68k/ifpsp060/src/fpsp.S # if our emulation, after re-doing the operation, decided that
the               998 arch/m68k/ifpsp060/src/fpsp.S # if our emulation, after re-doing the operation, decided that
the              1220 arch/m68k/ifpsp060/src/fpsp.S # if the exception is an opclass zero or two unimplemented data type
the              1621 arch/m68k/ifpsp060/src/fpsp.S # if a disabled overflow occurred and inexact was enabled but the result
the              1725 arch/m68k/ifpsp060/src/fpsp.S # if the exception occurred from supervisor mode, check if
the              3294 arch/m68k/ifpsp060/src/fpsp.S # if the effective addressing mode was -() or ()+, then the address	#
the              3857 arch/m68k/ifpsp060/src/fpsp.S # if the F-Line instruction is an "fmovecr" w/ a non-zero <ea>. if
the              4143 arch/m68k/ifpsp060/src/fpsp.S # if the instruction was executed from supervisor mode and the addressing
the              8427 arch/m68k/ifpsp060/src/fpsp.S # if the input is exactly equal to one, then exit through ld_pzero.
the              11815 arch/m68k/ifpsp060/src/fpsp.S # if the rnd mode is anything but RZ, then we have to re-do the above
the              12044 arch/m68k/ifpsp060/src/fpsp.S # if the result would have overflowed/underflowed. If so, use unf_res()	#
the              12203 arch/m68k/ifpsp060/src/fpsp.S # if underflow or inexact is enabled, then go calculate the EXOP first.
the              14840 arch/m68k/ifpsp060/src/fpsp.S # if the precision is extended, this result could not have come from an
the              15293 arch/m68k/ifpsp060/src/fpsp.S # if the precision is extended, this result could not have come from an
the              16332 arch/m68k/ifpsp060/src/fpsp.S # if the NAN bit is set, in which case BSUN and AIOP will be set.	#
the              16960 arch/m68k/ifpsp060/src/fpsp.S # if the NAN bit is set, in which case BSUN and AIOP will be set.	#
the              17530 arch/m68k/ifpsp060/src/fpsp.S # if the NAN bit is set, in which case BSUN and AIOP will be set.	#
the              18080 arch/m68k/ifpsp060/src/fpsp.S # if the addressing mode is post-increment or pre-decrement,
the              18235 arch/m68k/ifpsp060/src/fpsp.S # if the bit string is a zero, then the operation is a no-op
the              19600 arch/m68k/ifpsp060/src/fpsp.S # if an immediate data access fails, the resulting fault status		#
the              19789 arch/m68k/ifpsp060/src/fpsp.S #		      if the ea is -() or ()+, need to know # of bytes.	#
the              21004 arch/m68k/ifpsp060/src/fpsp.S # if the mantissa is zero, I will zero the exponent, too.
the              23022 arch/m68k/ifpsp060/src/fpsp.S #	if negative, count the trailing zeros.  Set the adjusted	#
the              23193 arch/m68k/ifpsp060/src/fpsp.S #  if the exp was positive, and added if it was negative.  The purpose
the              23499 arch/m68k/ifpsp060/src/fpsp.S #		if it is a positive number, or the number of digits	#
the              24713 arch/m68k/ifpsp060/src/fpsp.S # if the effective addressing mode was predecrement or postincrement,
the               221 arch/m68k/ifpsp060/src/ilsp.S # if the register numbers are the same, only the quotient gets saved.
the               283 arch/m68k/ifpsp060/src/ilsp.S # if the divisor msw is 0, use simpler algorithm then the full blown
the               964 arch/m68k/ifpsp060/src/pfpsp.S # if our emulation, after re-doing the operation, decided that
the               997 arch/m68k/ifpsp060/src/pfpsp.S # if our emulation, after re-doing the operation, decided that
the              1219 arch/m68k/ifpsp060/src/pfpsp.S # if the exception is an opclass zero or two unimplemented data type
the              1620 arch/m68k/ifpsp060/src/pfpsp.S # if a disabled overflow occurred and inexact was enabled but the result
the              1724 arch/m68k/ifpsp060/src/pfpsp.S # if the exception occurred from supervisor mode, check if
the              3293 arch/m68k/ifpsp060/src/pfpsp.S # if the effective addressing mode was -() or ()+, then the address	#
the              4284 arch/m68k/ifpsp060/src/pfpsp.S # if the bit string is a zero, then the operation is a no-op
the              7905 arch/m68k/ifpsp060/src/pfpsp.S # if the mantissa is zero, I will zero the exponent, too.
the              8246 arch/m68k/ifpsp060/src/pfpsp.S # if the rnd mode is anything but RZ, then we have to re-do the above
the              8475 arch/m68k/ifpsp060/src/pfpsp.S # if the result would have overflowed/underflowed. If so, use unf_res()	#
the              8634 arch/m68k/ifpsp060/src/pfpsp.S # if underflow or inexact is enabled, then go calculate the EXOP first.
the              11271 arch/m68k/ifpsp060/src/pfpsp.S # if the precision is extended, this result could not have come from an
the              11724 arch/m68k/ifpsp060/src/pfpsp.S # if the precision is extended, this result could not have come from an
the              12982 arch/m68k/ifpsp060/src/pfpsp.S #	if negative, count the trailing zeros.  Set the adjusted	#
the              13153 arch/m68k/ifpsp060/src/pfpsp.S #  if the exp was positive, and added if it was negative.  The purpose
the              13459 arch/m68k/ifpsp060/src/pfpsp.S #		if it is a positive number, or the number of digits	#
the              14673 arch/m68k/ifpsp060/src/pfpsp.S # if the effective addressing mode was predecrement or postincrement,
the                70 arch/m68k/math-emu/fp_decode.h | first decoding of the instr type
the                71 arch/m68k/math-emu/fp_decode.h | this separates the conditional instr
the                82 arch/m68k/math-emu/fp_decode.h | second decoding of the instr type
the                96 arch/m68k/math-emu/fp_decode.h | extract the source specifier, specifies
the               112 arch/m68k/math-emu/fp_decode.h | extract the addressing mode
the               113 arch/m68k/math-emu/fp_decode.h | it depends on the instr which of the modes is valid
the               128 arch/m68k/math-emu/fp_decode.h | extract the register for the addressing mode
the               133 arch/m68k/math-emu/fp_decode.h | decode the 8bit displacement from the brief extension word
the               139 arch/m68k/math-emu/fp_decode.h | decode the index of the brief/full extension word
the               141 arch/m68k/math-emu/fp_decode.h 	bfextu	%d2{#17,#3},%d0		| get the register nr
the               167 arch/m68k/math-emu/fp_decode.h | decode the base displacement size
the               188 arch/m68k/math-emu/fp_decode.h | get the extension word and test for brief or full extension type
the               196 arch/m68k/math-emu/fp_decode.h | test if %pc is the base register for the indirect addr mode
the               202 arch/m68k/math-emu/fp_decode.h | test if %pc is the base register for one of the extended modes
the               405 arch/m68k/math-emu/fp_decode.h | get the absolute short address from user space
the               411 arch/m68k/math-emu/fp_decode.h | get the absolute long address from user space
the                21 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# addresses, and need to have the appropriate memory region set
the                22 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# by the kernel
the                29 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# Read the cavium mem control register
the                31 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# Clear the lower 6 bits, the CVMSEG size
the                34 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	dmtc0	v0, CP0_CVMMEMCTL_REG	# Write the cavium mem control register
the                35 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	dmfc0	v0, CP0_CVMCTL_REG	# Read the cavium control register
the                40 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# First clear off CvmCtl[IPPCI] bit and move the performance
the                67 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# Write the cavium control register
the                82 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# OCTEON II or better have bit 15 set.  Clear the error bits.
the                91 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# Jump the master to kernel_entry
the                98 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# All cores other than the master need to wait here for SMP bootstrap
the               114 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# This is the variable where the next core to boot is stored
the               116 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# Get the core id of the next to be booted
the               121 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# Get my GP from the global variable
the               124 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# Get my SP from the global variable
the               127 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# Set the SP global variable to zero so the master knows we've started
the               135 arch/mips/include/asm/mach-cavium-octeon/kernel-entry-init.h 	# Jump to the normal Linux SMP entry point
the                71 arch/nds32/include/asm/nds32.h 						 * We defined at the start of the physical memory */
the               112 arch/s390/include/asm/nospec-insn.h 	# the "br \reg" after the code has been patched.
the                 5 arch/x86/boot/code16gcc.h # This file is added to the assembler via -Wa when compiling 16-bit C code.
the               161 drivers/gpu/drm/nouveau/dispnv04/arb.c 						* the overlay. */
the               147 drivers/mtd/nftlmount.c The new DiskOnChip driver scans the MediaHeader itself, and presents a virtual
the               148 drivers/mtd/nftlmount.c erasesize based on UnitSizeFactor.  So the erasesize we read from the mtd
the               219 drivers/mtd/nftlmount.c The new DiskOnChip driver already scanned the bad block table.  Just query it.
the              1105 drivers/parisc/sba_iommu.c 	Tells where the dvi bits are located in the address.
the               269 include/scsi/scsi_transport_fc.h 							   is loop or the
the                14 include/sound/wavefront.h      it is necessary to pack the "wavefront_alias" structure to a size
the                16 include/sound/wavefront.h      the case on the original platform, MS-DOS). If this is not done,
the                22 include/sound/wavefront.h      still have the same (correct) size.