From 5b7c4cabbb65f5c469464da6c5f614cbd7f730f2 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 21 Feb 2023 18:24:12 -0800 Subject: Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next Pull networking updates from Jakub Kicinski: "Core: - Add dedicated kmem_cache for typical/small skb->head, avoid having to access struct page at kfree time, and improve memory use. - Introduce sysctl to set default RPS configuration for new netdevs. - Define Netlink protocol specification format which can be used to describe messages used by each family and auto-generate parsers. Add tools for generating kernel data structures and uAPI headers. - Expose all net/core sysctls inside netns. - Remove 4s sleep in netpoll if carrier is instantly detected on boot. - Add configurable limit of MDB entries per port, and port-vlan. - Continue populating drop reasons throughout the stack. - Retire a handful of legacy Qdiscs and classifiers. Protocols: - Support IPv4 big TCP (TSO frames larger than 64kB). - Add IP_LOCAL_PORT_RANGE socket option, to control local port range on socket by socket basis. - Track and report in procfs number of MPTCP sockets used. - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path manager. - IPv6: don't check net.ipv6.route.max_size and rely on garbage collection to free memory (similarly to IPv4). - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986). - ICMP: add per-rate limit counters. - Add support for user scanning requests in ieee802154. - Remove static WEP support. - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate reporting. - WiFi 7 EHT channel puncturing support (client & AP). BPF: - Add a rbtree data structure following the "next-gen data structure" precedent set by recently added linked list, that is, by using kfunc + kptr instead of adding a new BPF map type. - Expose XDP hints via kfuncs with initial support for RX hash and timestamp metadata. - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to better support decap on GRE tunnel devices not operating in collect metadata. - Improve x86 JIT's codegen for PROBE_MEM runtime error checks. - Remove the need for trace_printk_lock for bpf_trace_printk and bpf_trace_vprintk helpers. - Extend libbpf's bpf_tracing.h support for tracing arguments of kprobes/uprobes and syscall as a special case. - Significantly reduce the search time for module symbols by livepatch and BPF. - Enable cpumasks to be used as kptrs, which is useful for tracing programs tracking which tasks end up running on which CPUs in different time intervals. - Add support for BPF trampoline on s390x and riscv64. - Add capability to export the XDP features supported by the NIC. - Add __bpf_kfunc tag for marking kernel functions as kfuncs. - Add cgroup.memory=nobpf kernel parameter option to disable BPF memory accounting for container environments. Netfilter: - Remove the CLUSTERIP target. It has been marked as obsolete for years, and we still have WARN splats wrt races of the out-of-band /proc interface installed by this target. - Add 'destroy' commands to nf_tables. They are identical to the existing 'delete' commands, but do not return an error if the referenced object (set, chain, rule...) did not exist. Driver API: - Improve cpumask_local_spread() locality to help NICs set the right IRQ affinity on AMD platforms. - Separate C22 and C45 MDIO bus transactions more clearly. - Introduce new DCB table to control DSCP rewrite on egress. - Support configuration of Physical Layer Collision Avoidance (PLCA) Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of shared medium Ethernet. - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing preemption of low priority frames by high priority frames. - Add support for controlling MACSec offload using netlink SET. - Rework devlink instance refcounts to allow registration and de-registration under the instance lock. Split the code into multiple files, drop some of the unnecessarily granular locks and factor out common parts of netlink operation handling. - Add TX frame aggregation parameters (for USB drivers). - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning messages with notifications for debug. - Allow offloading of UDP NEW connections via act_ct. - Add support for per action HW stats in TC. - Support hardware miss to TC action (continue processing in SW from a specific point in the action chain). - Warn if old Wireless Extension user space interface is used with modern cfg80211/mac80211 drivers. Do not support Wireless Extensions for Wi-Fi 7 devices at all. Everyone should switch to using nl80211 interface instead. - Improve the CAN bit timing configuration. Use extack to return error messages directly to user space, update the SJW handling, including the definition of a new default value that will benefit CAN-FD controllers, by increasing their oscillator tolerance. New hardware / drivers: - Ethernet: - nVidia BlueField-3 support (control traffic driver) - Ethernet support for imx93 SoCs - Motorcomm yt8531 gigabit Ethernet PHY - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA) - Microchip LAN8841 PHY (incl. cable diagnostics and PTP) - Amlogic gxl MDIO mux - WiFi: - RealTek RTL8188EU (rtl8xxxu) - Qualcomm Wi-Fi 7 devices (ath12k) - CAN: - Renesas R-Car V4H Drivers: - Bluetooth: - Set Per Platform Antenna Gain (PPAG) for Intel controllers. - Ethernet NICs: - Intel (1G, igc): - support TSN / Qbv / packet scheduling features of i226 model - Intel (100G, ice): - use GNSS subsystem instead of TTY - multi-buffer XDP support - extend support for GPIO pins to E823 devices - nVidia/Mellanox: - update the shared buffer configuration on PFC commands - implement PTP adjphase function for HW offset control - TC support for Geneve and GRE with VF tunnel offload - more efficient crypto key management method - multi-port eswitch support - Netronome/Corigine: - add DCB IEEE support - support IPsec offloading for NFP3800 - Freescale/NXP (enetc): - support XDP_REDIRECT for XDP non-linear buffers - improve reconfig, avoid link flap and waiting for idle - support MAC Merge layer - Other NICs: - sfc/ef100: add basic devlink support for ef100 - ionic: rx_push mode operation (writing descriptors via MMIO) - bnxt: use the auxiliary bus abstraction for RDMA - r8169: disable ASPM and reset bus in case of tx timeout - cpsw: support QSGMII mode for J721e CPSW9G - cpts: support pulse-per-second output - ngbe: add an mdio bus driver - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing - r8152: handle devices with FW with NCM support - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation - virtio-net: support multi buffer XDP - virtio/vsock: replace virtio_vsock_pkt with sk_buff - tsnep: XDP support - Ethernet high-speed switches: - nVidia/Mellanox (mlxsw): - add support for latency TLV (in FW control messages) - Microchip (sparx5): - separate explicit and implicit traffic forwarding rules, make the implicit rules always active - add support for egress DSCP rewrite - IS0 VCAP support (Ingress Classification) - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS etc.) - ES2 VCAP support (Egress Access Control) - support for Per-Stream Filtering and Policing (802.1Q, 8.6.5.1) - Ethernet embedded switches: - Marvell (mv88e6xxx): - add MAB (port auth) offload support - enable PTP receive for mv88e6390 - NXP (ocelot): - support MAC Merge layer - support for the the vsc7512 internal copper phys - Microchip: - lan9303: convert to PHYLINK - lan966x: support TC flower filter statistics - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x - lan937x: support Credit Based Shaper configuration - ksz9477: support Energy Efficient Ethernet - other: - qca8k: convert to regmap read/write API, use bulk operations - rswitch: Improve TX timestamp accuracy - Intel WiFi (iwlwifi): - EHT (Wi-Fi 7) rate reporting - STEP equalizer support: transfer some STEP (connection to radio on platforms with integrated wifi) related parameters from the BIOS to the firmware. - Qualcomm 802.11ax WiFi (ath11k): - IPQ5018 support - Fine Timing Measurement (FTM) responder role support - channel 177 support - MediaTek WiFi (mt76): - per-PHY LED support - mt7996: EHT (Wi-Fi 7) support - Wireless Ethernet Dispatch (WED) reset support - switch to using page pool allocator - RealTek WiFi (rtw89): - support new version of Bluetooth co-existance - Mobile: - rmnet: support TX aggregation" * tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits) page_pool: add a comment explaining the fragment counter usage net: ethtool: fix __ethtool_dev_mm_supported() implementation ethtool: pse-pd: Fix double word in comments xsk: add linux/vmalloc.h to xsk.c sefltests: netdevsim: wait for devlink instance after netns removal selftest: fib_tests: Always cleanup before exit net/mlx5e: Align IPsec ASO result memory to be as required by hardware net/mlx5e: TC, Set CT miss to the specific ct action instance net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG net/mlx5: Refactor tc miss handling to a single function net/mlx5: Kconfig: Make tc offload depend on tc skb extension net/sched: flower: Support hardware miss to tc action net/sched: flower: Move filter handle initialization earlier net/sched: cls_api: Support hardware miss to tc action net/sched: Rename user cookie and act cookie sfc: fix builds without CONFIG_RTC_LIB sfc: clean up some inconsistent indentings net/mlx4_en: Introduce flexible array to silence overflow warning net: lan966x: Fix possible deadlock inside PTP net/ulp: Remove redundant ->clone() test in inet_clone_ulp(). ... --- arch/m68k/fpsp040/do_func.S | 558 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 558 insertions(+) create mode 100644 arch/m68k/fpsp040/do_func.S (limited to 'arch/m68k/fpsp040/do_func.S') diff --git a/arch/m68k/fpsp040/do_func.S b/arch/m68k/fpsp040/do_func.S new file mode 100644 index 000000000..3eff99a80 --- /dev/null +++ b/arch/m68k/fpsp040/do_func.S @@ -0,0 +1,558 @@ +| +| do_func.sa 3.4 2/18/91 +| +| Do_func performs the unimplemented operation. The operation +| to be performed is determined from the lower 7 bits of the +| extension word (except in the case of fmovecr and fsincos). +| The opcode and tag bits form an index into a jump table in +| tbldo.sa. Cases of zero, infinity and NaN are handled in +| do_func by forcing the default result. Normalized and +| denormalized (there are no unnormalized numbers at this +| point) are passed onto the emulation code. +| +| CMDREG1B and STAG are extracted from the fsave frame +| and combined to form the table index. The function called +| will start with a0 pointing to the ETEMP operand. Dyadic +| functions can find FPTEMP at -12(a0). +| +| Called functions return their result in fp0. Sincos returns +| sin(x) in fp0 and cos(x) in fp1. +| + +| Copyright (C) Motorola, Inc. 1990 +| All Rights Reserved +| +| For details on the license for this file, please see the +| file, README, in this same directory. + +DO_FUNC: |idnt 2,1 | Motorola 040 Floating Point Software Package + + |section 8 + +#include "fpsp.h" + + |xref t_dz2 + |xref t_operr + |xref t_inx2 + |xref t_resdnrm + |xref dst_nan + |xref src_nan + |xref nrm_set + |xref sto_cos + + |xref tblpre + |xref slognp1,slogn,slog10,slog2 + |xref slognd,slog10d,slog2d + |xref smod,srem + |xref sscale + |xref smovcr + +PONE: .long 0x3fff0000,0x80000000,0x00000000 |+1 +MONE: .long 0xbfff0000,0x80000000,0x00000000 |-1 +PZERO: .long 0x00000000,0x00000000,0x00000000 |+0 +MZERO: .long 0x80000000,0x00000000,0x00000000 |-0 +PINF: .long 0x7fff0000,0x00000000,0x00000000 |+inf +MINF: .long 0xffff0000,0x00000000,0x00000000 |-inf +QNAN: .long 0x7fff0000,0xffffffff,0xffffffff |non-signaling nan +PPIBY2: .long 0x3FFF0000,0xC90FDAA2,0x2168C235 |+PI/2 +MPIBY2: .long 0xbFFF0000,0xC90FDAA2,0x2168C235 |-PI/2 + + .global do_func +do_func: + clrb CU_ONLY(%a6) +| +| Check for fmovecr. It does not follow the format of fp gen +| unimplemented instructions. The test is on the upper 6 bits; +| if they are $17, the inst is fmovecr. Call entry smovcr +| directly. +| + bfextu CMDREG1B(%a6){#0:#6},%d0 |get opclass and src fields + cmpil #0x17,%d0 |if op class and size fields are $17, +| ;it is FMOVECR; if not, continue + bnes not_fmovecr + jmp smovcr |fmovecr; jmp directly to emulation + +not_fmovecr: + movew CMDREG1B(%a6),%d0 + andl #0x7F,%d0 + cmpil #0x38,%d0 |if the extension is >= $38, + bge serror |it is illegal + bfextu STAG(%a6){#0:#3},%d1 + lsll #3,%d0 |make room for STAG + addl %d1,%d0 |combine for final index into table + leal tblpre,%a1 |start of monster jump table + movel (%a1,%d0.w*4),%a1 |real target address + leal ETEMP(%a6),%a0 |a0 is pointer to src op + movel USER_FPCR(%a6),%d1 + andl #0xFF,%d1 | discard all but rounding mode/prec + fmovel #0,%fpcr + jmp (%a1) +| +| ERROR +| + .global serror +serror: + st STORE_FLG(%a6) + rts +| +| These routines load forced values into fp0. They are called +| by index into tbldo. +| +| Load a signed zero to fp0 and set inex2/ainex +| + .global snzrinx +snzrinx: + btstb #sign_bit,LOCAL_EX(%a0) |get sign of source operand + bnes ld_mzinx |if negative, branch + bsr ld_pzero |bsr so we can return and set inx + bra t_inx2 |now, set the inx for the next inst +ld_mzinx: + bsr ld_mzero |if neg, load neg zero, return here + bra t_inx2 |now, set the inx for the next inst +| +| Load a signed zero to fp0; do not set inex2/ainex +| + .global szero +szero: + btstb #sign_bit,LOCAL_EX(%a0) |get sign of source operand + bne ld_mzero |if neg, load neg zero + bra ld_pzero |load positive zero +| +| Load a signed infinity to fp0; do not set inex2/ainex +| + .global sinf +sinf: + btstb #sign_bit,LOCAL_EX(%a0) |get sign of source operand + bne ld_minf |if negative branch + bra ld_pinf +| +| Load a signed one to fp0; do not set inex2/ainex +| + .global sone +sone: + btstb #sign_bit,LOCAL_EX(%a0) |check sign of source + bne ld_mone + bra ld_pone +| +| Load a signed pi/2 to fp0; do not set inex2/ainex +| + .global spi_2 +spi_2: + btstb #sign_bit,LOCAL_EX(%a0) |check sign of source + bne ld_mpi2 + bra ld_ppi2 +| +| Load either a +0 or +inf for plus/minus operand +| + .global szr_inf +szr_inf: + btstb #sign_bit,LOCAL_EX(%a0) |check sign of source + bne ld_pzero + bra ld_pinf +| +| Result is either an operr or +inf for plus/minus operand +| [Used by slogn, slognp1, slog10, and slog2] +| + .global sopr_inf +sopr_inf: + btstb #sign_bit,LOCAL_EX(%a0) |check sign of source + bne t_operr + bra ld_pinf +| +| FLOGNP1 +| + .global sslognp1 +sslognp1: + fmovemx (%a0),%fp0-%fp0 + fcmpb #-1,%fp0 + fbgt slognp1 + fbeq t_dz2 |if = -1, divide by zero exception + fmovel #0,%FPSR |clr N flag + bra t_operr |take care of operands < -1 +| +| FETOXM1 +| + .global setoxm1i +setoxm1i: + btstb #sign_bit,LOCAL_EX(%a0) |check sign of source + bne ld_mone + bra ld_pinf +| +| FLOGN +| +| Test for 1.0 as an input argument, returning +zero. Also check +| the sign and return operr if negative. +| + .global sslogn +sslogn: + btstb #sign_bit,LOCAL_EX(%a0) + bne t_operr |take care of operands < 0 + cmpiw #0x3fff,LOCAL_EX(%a0) |test for 1.0 input + bne slogn + cmpil #0x80000000,LOCAL_HI(%a0) + bne slogn + tstl LOCAL_LO(%a0) + bne slogn + fmovex PZERO,%fp0 + rts + + .global sslognd +sslognd: + btstb #sign_bit,LOCAL_EX(%a0) + beq slognd + bra t_operr |take care of operands < 0 + +| +| FLOG10 +| + .global sslog10 +sslog10: + btstb #sign_bit,LOCAL_EX(%a0) + bne t_operr |take care of operands < 0 + cmpiw #0x3fff,LOCAL_EX(%a0) |test for 1.0 input + bne slog10 + cmpil #0x80000000,LOCAL_HI(%a0) + bne slog10 + tstl LOCAL_LO(%a0) + bne slog10 + fmovex PZERO,%fp0 + rts + + .global sslog10d +sslog10d: + btstb #sign_bit,LOCAL_EX(%a0) + beq slog10d + bra t_operr |take care of operands < 0 + +| +| FLOG2 +| + .global sslog2 +sslog2: + btstb #sign_bit,LOCAL_EX(%a0) + bne t_operr |take care of operands < 0 + cmpiw #0x3fff,LOCAL_EX(%a0) |test for 1.0 input + bne slog2 + cmpil #0x80000000,LOCAL_HI(%a0) + bne slog2 + tstl LOCAL_LO(%a0) + bne slog2 + fmovex PZERO,%fp0 + rts + + .global sslog2d +sslog2d: + btstb #sign_bit,LOCAL_EX(%a0) + beq slog2d + bra t_operr |take care of operands < 0 + +| +| FMOD +| +pmodt: +| ;$21 fmod +| ;dtag,stag + .long smod | 00,00 norm,norm = normal + .long smod_oper | 00,01 norm,zero = nan with operr + .long smod_fpn | 00,10 norm,inf = fpn + .long smod_snan | 00,11 norm,nan = nan + .long smod_zro | 01,00 zero,norm = +-zero + .long smod_oper | 01,01 zero,zero = nan with operr + .long smod_zro | 01,10 zero,inf = +-zero + .long smod_snan | 01,11 zero,nan = nan + .long smod_oper | 10,00 inf,norm = nan with operr + .long smod_oper | 10,01 inf,zero = nan with operr + .long smod_oper | 10,10 inf,inf = nan with operr + .long smod_snan | 10,11 inf,nan = nan + .long smod_dnan | 11,00 nan,norm = nan + .long smod_dnan | 11,01 nan,zero = nan + .long smod_dnan | 11,10 nan,inf = nan + .long smod_dnan | 11,11 nan,nan = nan + + .global pmod +pmod: + clrb FPSR_QBYTE(%a6) | clear quotient field + bfextu STAG(%a6){#0:#3},%d0 |stag = d0 + bfextu DTAG(%a6){#0:#3},%d1 |dtag = d1 + +| +| Alias extended denorms to norms for the jump table. +| + bclrl #2,%d0 + bclrl #2,%d1 + + lslb #2,%d1 + orb %d0,%d1 |d1{3:2} = dtag, d1{1:0} = stag +| ;Tag values: +| ;00 = norm or denorm +| ;01 = zero +| ;10 = inf +| ;11 = nan + lea pmodt,%a1 + movel (%a1,%d1.w*4),%a1 + jmp (%a1) + +smod_snan: + bra src_nan +smod_dnan: + bra dst_nan +smod_oper: + bra t_operr +smod_zro: + moveb ETEMP(%a6),%d1 |get sign of src op + moveb FPTEMP(%a6),%d0 |get sign of dst op + eorb %d0,%d1 |get exor of sign bits + btstl #7,%d1 |test for sign + beqs smod_zsn |if clr, do not set sign big + bsetb #q_sn_bit,FPSR_QBYTE(%a6) |set q-byte sign bit +smod_zsn: + btstl #7,%d0 |test if + or - + beq ld_pzero |if pos then load +0 + bra ld_mzero |else neg load -0 + +smod_fpn: + moveb ETEMP(%a6),%d1 |get sign of src op + moveb FPTEMP(%a6),%d0 |get sign of dst op + eorb %d0,%d1 |get exor of sign bits + btstl #7,%d1 |test for sign + beqs smod_fsn |if clr, do not set sign big + bsetb #q_sn_bit,FPSR_QBYTE(%a6) |set q-byte sign bit +smod_fsn: + tstb DTAG(%a6) |filter out denormal destination case + bpls smod_nrm | + leal FPTEMP(%a6),%a0 |a0<- addr(FPTEMP) + bra t_resdnrm |force UNFL(but exact) result +smod_nrm: + fmovel USER_FPCR(%a6),%fpcr |use user's rmode and precision + fmovex FPTEMP(%a6),%fp0 |return dest to fp0 + rts + +| +| FREM +| +premt: +| ;$25 frem +| ;dtag,stag + .long srem | 00,00 norm,norm = normal + .long srem_oper | 00,01 norm,zero = nan with operr + .long srem_fpn | 00,10 norm,inf = fpn + .long srem_snan | 00,11 norm,nan = nan + .long srem_zro | 01,00 zero,norm = +-zero + .long srem_oper | 01,01 zero,zero = nan with operr + .long srem_zro | 01,10 zero,inf = +-zero + .long srem_snan | 01,11 zero,nan = nan + .long srem_oper | 10,00 inf,norm = nan with operr + .long srem_oper | 10,01 inf,zero = nan with operr + .long srem_oper | 10,10 inf,inf = nan with operr + .long srem_snan | 10,11 inf,nan = nan + .long srem_dnan | 11,00 nan,norm = nan + .long srem_dnan | 11,01 nan,zero = nan + .long srem_dnan | 11,10 nan,inf = nan + .long srem_dnan | 11,11 nan,nan = nan + + .global prem +prem: + clrb FPSR_QBYTE(%a6) |clear quotient field + bfextu STAG(%a6){#0:#3},%d0 |stag = d0 + bfextu DTAG(%a6){#0:#3},%d1 |dtag = d1 +| +| Alias extended denorms to norms for the jump table. +| + bclr #2,%d0 + bclr #2,%d1 + + lslb #2,%d1 + orb %d0,%d1 |d1{3:2} = dtag, d1{1:0} = stag +| ;Tag values: +| ;00 = norm or denorm +| ;01 = zero +| ;10 = inf +| ;11 = nan + lea premt,%a1 + movel (%a1,%d1.w*4),%a1 + jmp (%a1) + +srem_snan: + bra src_nan +srem_dnan: + bra dst_nan +srem_oper: + bra t_operr +srem_zro: + moveb ETEMP(%a6),%d1 |get sign of src op + moveb FPTEMP(%a6),%d0 |get sign of dst op + eorb %d0,%d1 |get exor of sign bits + btstl #7,%d1 |test for sign + beqs srem_zsn |if clr, do not set sign big + bsetb #q_sn_bit,FPSR_QBYTE(%a6) |set q-byte sign bit +srem_zsn: + btstl #7,%d0 |test if + or - + beq ld_pzero |if pos then load +0 + bra ld_mzero |else neg load -0 + +srem_fpn: + moveb ETEMP(%a6),%d1 |get sign of src op + moveb FPTEMP(%a6),%d0 |get sign of dst op + eorb %d0,%d1 |get exor of sign bits + btstl #7,%d1 |test for sign + beqs srem_fsn |if clr, do not set sign big + bsetb #q_sn_bit,FPSR_QBYTE(%a6) |set q-byte sign bit +srem_fsn: + tstb DTAG(%a6) |filter out denormal destination case + bpls srem_nrm | + leal FPTEMP(%a6),%a0 |a0<- addr(FPTEMP) + bra t_resdnrm |force UNFL(but exact) result +srem_nrm: + fmovel USER_FPCR(%a6),%fpcr |use user's rmode and precision + fmovex FPTEMP(%a6),%fp0 |return dest to fp0 + rts +| +| FSCALE +| +pscalet: +| ;$26 fscale +| ;dtag,stag + .long sscale | 00,00 norm,norm = result + .long sscale | 00,01 norm,zero = fpn + .long scl_opr | 00,10 norm,inf = nan with operr + .long scl_snan | 00,11 norm,nan = nan + .long scl_zro | 01,00 zero,norm = +-zero + .long scl_zro | 01,01 zero,zero = +-zero + .long scl_opr | 01,10 zero,inf = nan with operr + .long scl_snan | 01,11 zero,nan = nan + .long scl_inf | 10,00 inf,norm = +-inf + .long scl_inf | 10,01 inf,zero = +-inf + .long scl_opr | 10,10 inf,inf = nan with operr + .long scl_snan | 10,11 inf,nan = nan + .long scl_dnan | 11,00 nan,norm = nan + .long scl_dnan | 11,01 nan,zero = nan + .long scl_dnan | 11,10 nan,inf = nan + .long scl_dnan | 11,11 nan,nan = nan + + .global pscale +pscale: + bfextu STAG(%a6){#0:#3},%d0 |stag in d0 + bfextu DTAG(%a6){#0:#3},%d1 |dtag in d1 + bclrl #2,%d0 |alias denorm into norm + bclrl #2,%d1 |alias denorm into norm + lslb #2,%d1 + orb %d0,%d1 |d1{4:2} = dtag, d1{1:0} = stag +| ;dtag values stag values: +| ;000 = norm 00 = norm +| ;001 = zero 01 = zero +| ;010 = inf 10 = inf +| ;011 = nan 11 = nan +| ;100 = dnrm +| +| + leal pscalet,%a1 |load start of jump table + movel (%a1,%d1.w*4),%a1 |load a1 with label depending on tag + jmp (%a1) |go to the routine + +scl_opr: + bra t_operr + +scl_dnan: + bra dst_nan + +scl_zro: + btstb #sign_bit,FPTEMP_EX(%a6) |test if + or - + beq ld_pzero |if pos then load +0 + bra ld_mzero |if neg then load -0 +scl_inf: + btstb #sign_bit,FPTEMP_EX(%a6) |test if + or - + beq ld_pinf |if pos then load +inf + bra ld_minf |else neg load -inf +scl_snan: + bra src_nan +| +| FSINCOS +| + .global ssincosz +ssincosz: + btstb #sign_bit,ETEMP(%a6) |get sign + beqs sincosp + fmovex MZERO,%fp0 + bras sincoscom +sincosp: + fmovex PZERO,%fp0 +sincoscom: + fmovemx PONE,%fp1-%fp1 |do not allow FPSR to be affected + bra sto_cos |store cosine result + + .global ssincosi +ssincosi: + fmovex QNAN,%fp1 |load NAN + bsr sto_cos |store cosine result + fmovex QNAN,%fp0 |load NAN + bra t_operr + + .global ssincosnan +ssincosnan: + movel ETEMP_EX(%a6),FP_SCR1(%a6) + movel ETEMP_HI(%a6),FP_SCR1+4(%a6) + movel ETEMP_LO(%a6),FP_SCR1+8(%a6) + bsetb #signan_bit,FP_SCR1+4(%a6) + fmovemx FP_SCR1(%a6),%fp1-%fp1 + bsr sto_cos + bra src_nan +| +| This code forces default values for the zero, inf, and nan cases +| in the transcendentals code. The CC bits must be set in the +| stacked FPSR to be correctly reported. +| +|**Returns +PI/2 + .global ld_ppi2 +ld_ppi2: + fmovex PPIBY2,%fp0 |load +pi/2 + bra t_inx2 |set inex2 exc + +|**Returns -PI/2 + .global ld_mpi2 +ld_mpi2: + fmovex MPIBY2,%fp0 |load -pi/2 + orl #neg_mask,USER_FPSR(%a6) |set N bit + bra t_inx2 |set inex2 exc + +|**Returns +inf + .global ld_pinf +ld_pinf: + fmovex PINF,%fp0 |load +inf + orl #inf_mask,USER_FPSR(%a6) |set I bit + rts + +|**Returns -inf + .global ld_minf +ld_minf: + fmovex MINF,%fp0 |load -inf + orl #neg_mask+inf_mask,USER_FPSR(%a6) |set N and I bits + rts + +|**Returns +1 + .global ld_pone +ld_pone: + fmovex PONE,%fp0 |load +1 + rts + +|**Returns -1 + .global ld_mone +ld_mone: + fmovex MONE,%fp0 |load -1 + orl #neg_mask,USER_FPSR(%a6) |set N bit + rts + +|**Returns +0 + .global ld_pzero +ld_pzero: + fmovex PZERO,%fp0 |load +0 + orl #z_mask,USER_FPSR(%a6) |set Z bit + rts + +|**Returns -0 + .global ld_mzero +ld_mzero: + fmovex MZERO,%fp0 |load -0 + orl #neg_mask+z_mask,USER_FPSR(%a6) |set N and Z bits + rts + + |end -- cgit v1.2.3