NetBSD/src j65TkKlsys/arch/x86/x86 identcpu.c

   Temporarily disable SMAP for CPUVENDOR_IDT vendor CPUs (if available).

   SMAP feature causes a supervisor page fault during boot on KX-7000 CPUs.

   port-amd64/60641
VersionDeltaFile
1.143+9-2sys/arch/x86/x86/identcpu.c
+9-21 files

NetBSD/src cY2kBxFusr.sbin/cpuctl/arch i386.c

   cpuctl(8): decode L2 cache associativity for VIA CPUs using the AMD cache lookup
   table for the 0x80000006 CPUID function associativity value.

   Use `cpuid leaf 4' to get cache parameters for Zhaoxin and late VIA CPU models.

   Mirrors the changes from identcpu.c rev. 1.142.

   Additionally, use via_cpu_probe() and via_cpu_cacheinfo() for CPUVENDOR_IDT
   family > 6 CPUs to identify Zhaoxin CPUs and their features.
   Also, print feature 5-7 bits for CPUVENDOR_IDT, which appear to be compatible
   with Intel.

   Use vendorname = "Zhaoxin" for family 7 models.

   Tested on various 32-bit and 64-bit VIA and Zhaoxin CPUs
   (Samuel 2 / Nehemiah, Esther (32-bit), multiple Nano models, and KX-U6580/7000).

   PR kern/60619

   pullups needed to netbsd-11 and netbsd-10.
VersionDeltaFile
1.149+34-15usr.sbin/cpuctl/arch/i386.c
+34-151 files

NetBSD/src hjvI2EFsys/arch/x86/x86 identcpu.c

   Decode L2 cache associativity for VIA CPUs using the AMD cache lookup table for
   the 0x80000006 CPUID function associativity value.

   Use `cpuid leaf 4' to get cache parameters for Zhaoxin and late VIA CPU models.

   VIA datasheets mention that associativity is stored as a decimal value. However,
   in practice, it is encoded the AMD way and does not store the actual value.
   Using the raw value leads to an incorrect CPU colors calculation, causing a
   panic on KX-7000 (since the result is not a power of two) and incorrect CPU
   colors values on other CPU models (e.g. as visible in `vmstat -s').
   Decoding value using the AMD cache lookup table produces correct values in all
   cases I have tested or verified using CPUID dumps available online.

   Additionally, Zhaoxin and late VIA CPUs (CNQ and newer) prefer the Intel
   `cpuid leaf 4' function to get cache parameters. Check whether cpuid_level
   is equal to or greater than 4 to identify these CPUs and override cache values
   (including L1) by calling cpu_dcp_cacheinfo(ci, 4). L3 cache information is
   available only through this method.


    [7 lines not shown]
VersionDeltaFile
1.142+25-7sys/arch/x86/x86/identcpu.c
+25-71 files

NetBSD/src 7F9f5YEsys/arch/riscv/conf GENERIC64, sys/arch/riscv/starfive files.starfive jh7110_wdt.c

   risc-v: Add a driver for the jh7110 watchdog timer

   ok skrll@
VersionDeltaFile
1.1+262-0sys/arch/riscv/starfive/jh7110_wdt.c
1.13+6-1sys/arch/riscv/starfive/files.starfive
1.28+2-1sys/arch/riscv/conf/GENERIC64
+270-23 files

NetBSD/src NXP9lyzdoc CHANGES-11.1

   Tickets #436 - #442
VersionDeltaFile
1.1.2.9+69-1doc/CHANGES-11.1
+69-11 files

NetBSD/src o4pimSZsys/arch/riscv/include asm.h

   Pull up following revision(s) (requested by skrll in ticket #442):

        sys/arch/riscv/include/asm.h: revision 1.13

   Ensure 4-byte (32bit alignment) in ENTRY_NP so that non-compressable
   instructions get the correct alignment.
VersionDeltaFile
1.11.2.2+2-2sys/arch/riscv/include/asm.h
+2-21 files

NetBSD/src S4OgvDGexternal/gpl3/gdb/dist/gdb Makefile.in

   Pull up following revision(s) (requested by tls in ticket #441):

        external/gpl3/gdb/dist/gdb/Makefile.in: revision 1.11

   Fix MKCROSSGDB build from release source tars that have had permissions
   whanged around
VersionDeltaFile
1.8.2.1+1-1external/gpl3/gdb/dist/gdb/Makefile.in
+1-11 files

NetBSD/src ehFkC1esys/arch/hppa/conf GENERIC, sys/arch/hppa/hppa process_machdep.c machdep.c

   Pull up following revision(s) (requested by tls in ticket #440):

        sys/arch/hppa/hppa/process_machdep.c: revision 1.19
        sys/arch/hppa/include/cpu.h: revision 1.14
        sys/arch/hppa/hppa/machdep.c: revision 1.28
        sys/arch/hppa/hppa/machdep.h: revision 1.20
        sys/arch/hppa/conf/GENERIC: revision 1.50
        sys/arch/hppa/conf/files.hppa: revision 1.30
        sys/arch/hppa/hppa/locore.S: revision 1.8
        sys/arch/hppa/hppa/hppa_machdep.c: revision 1.34
        sys/arch/hppa/hppa/fpu.c: revision 1.28

   Add eager FPU switching for hppa - works around QEMU bug, and safer -
   may eventually become the default.
VersionDeltaFile
1.27.28.1+105-4sys/arch/hppa/hppa/fpu.c
1.7.2.1+35-8sys/arch/hppa/hppa/locore.S
1.18.68.1+5-2sys/arch/hppa/hppa/process_machdep.c
1.22.2.2+5-2sys/arch/hppa/hppa/machdep.c
1.33.12.1+5-2sys/arch/hppa/hppa/hppa_machdep.c
1.49.2.1+4-2sys/arch/hppa/conf/GENERIC
+159-203 files not shown
+166-239 files

NetBSD/src JlAj6LCsys/dev/pci files.virtio vioscsi.c

   Pull up following revision(s) (requested by tls in ticket #439):

        sys/dev/pci/vioscsi.c: revision 1.37
        sys/dev/pci/files.virtio: revision 1.5

   Add VIOSCSI_MAX_TARGET and VIOSCSI_MAX_LUN options to reduce excessive probe time on QEMU
VersionDeltaFile
1.36.8.1+25-4sys/dev/pci/vioscsi.c
1.3.2.1+2-1sys/dev/pci/files.virtio
+27-52 files

NetBSD/src 2dlKRMksys/dev/goldfish gftty.c

   Pull up following revision(s) (requested by tls in ticket #438):

        sys/dev/goldfish/gftty.c: revision 1.6

   Make console input on virt68k 8-bit clean.

   On signed-char ports you can't feed a "char *" to the tty layer.  A cursory
   glance for more instances of this didn't find any but it's possible.
VersionDeltaFile
1.3.4.1+3-3sys/dev/goldfish/gftty.c
+3-31 files

NetBSD/src zfT6mP2sys/arch/alpha/alpha syscall.c, sys/arch/mips/mips syscall.c

   Pull up following revision(s) (requested by tls in ticket #437):

        sys/sys/systm.h: revision 1.307
        sys/sys/syscall.h: revision 1.330
        sys/kern/syscalls.c: revision 1.335
        sys/kern/syscalls_autoload.c: revision 1.51
        sys/kern/systrace_args.c: revision 1.57
        sys/arch/alpha/alpha/syscall.c: revision 1.46
        sys/arch/riscv/riscv/syscall.c: revision 1.7
        sys/arch/mips/mips/syscall.c: revision 1.52
        sys/kern/init_sysent.c: revision 1.347
        sys/kern/makesyscalls.sh: revision 1.189
        sys/sys/syscallargs.h: revision 1.313

   Fix a longstanding bug on 64 bit ports with ABIs requiring sign-extension when returning 32-bit values in registers.  Alpha, MIPS N64, RISCV.

   With these ABIs it is necessary for the syscall return code to manually
   sign-extend all 32-bit return values.  If we don't, everything may look
   just fine, until a syscall return value is spilled to the stack and then

    [13 lines not shown]
VersionDeltaFile
1.346.2.1+39-36sys/kern/init_sysent.c
1.187.8.1+29-1sys/kern/makesyscalls.sh
1.51.8.1+14-2sys/arch/mips/mips/syscall.c
1.6.2.1+12-2sys/arch/riscv/riscv/syscall.c
1.45.8.1+12-2sys/arch/alpha/alpha/syscall.c
1.50.2.1+3-3sys/kern/syscalls_autoload.c
+109-465 files not shown
+121-5611 files

NetBSD/src Ed4sa0Qsys/arch/sparc64/conf GENERIC

   sun4v: Add mfii(4) to sparc64 GENERIC config.

   Symbios Logic SAS3008 storage controller is present in SPARC S7-2 servers.

   Not tested yet.
VersionDeltaFile
1.252+3-2sys/arch/sparc64/conf/GENERIC
+3-21 files

NetBSD/src esgneVXsys/arch/evbmips/malta malta_bus_mem.c malta_bus_io.c, sys/arch/evbmips/malta/dev mainbus.c gt.c

   Pull up following revision(s) (requested by tls in ticket #436):

        sys/arch/evbmips/malta/pci/pcib.c: revision 1.29
        sys/arch/evbmips/malta/dev/mainbus.c: revision 1.21
        sys/arch/evbmips/malta/dev/gt.c: revision 1.20
        sys/arch/evbmips/malta/malta_intr.c: revision 1.30
        sys/arch/evbmips/malta/malta_bus_mem.c: revision 1.10
        sys/arch/evbmips/malta/machdep.c: revision 1.51
        sys/arch/evbmips/malta/malta_bus_io.c: revision 1.9
        sys/arch/evbmips/conf/MALTA: revision 1.107

   Address problems with MIPS Malta platform code found running under QEMU.

   QEMU's "malta" system emulates a MIPS Malta with the Gallileo host bridge
   and 32-bit or 64-bit CPUs of either endianness.  It is one of the only
   working QEMU system-level emulations that could run NetBSD with all
   combinations of endianness and address size.  After fixes to QEMU over the
   past several years, NetBSD has been unable to use the emulated PCI bus in
   big-endian and 64-bit configurations.

    [36 lines not shown]
VersionDeltaFile
1.27.26.1+16-14sys/arch/evbmips/malta/malta_intr.c
1.17.14.1+20-6sys/arch/evbmips/malta/dev/gt.c
1.18.4.1+18-4sys/arch/evbmips/malta/dev/mainbus.c
1.47.4.1+10-11sys/arch/evbmips/malta/machdep.c
1.7.126.1+11-2sys/arch/evbmips/malta/malta_bus_mem.c
1.6.126.1+11-2sys/arch/evbmips/malta/malta_bus_io.c
+86-392 files not shown
+99-458 files

NetBSD/src gUVMxmXdoc CHANGES-11.1

   Tickets #427 - #430, #432 - #434
VersionDeltaFile
1.1.2.8+57-1doc/CHANGES-11.1
+57-11 files

NetBSD/src CCSzEwudistrib/sets/lists/debug ad.mips ad.mipsel

   Pull up following revision(s) (requested by tls in ticket #434):

        distrib/sets/lists/debug/md.evbarm.earmv4: revision 1.2
        distrib/sets/sets.subr: revision 1.215
        sys/conf/Makefile.kern.inc: revision 1.306
        distrib/sets/lists/debug/md.evbarm.earmv5: revision 1.2
        distrib/sets/lists/debug/md.pmax: revision 1.5
        distrib/sets/lists/debug/md.emips: revision 1.4
        distrib/sets/lists/debug/ad.mipseb: revision 1.8
        distrib/sets/lists/debug/ad.mipsel: revision 1.8
        distrib/sets/lists/debug/mi: revision 1.513
        distrib/sets/lists/debug/md.evbarm.earmv5eb: revision 1.2
        distrib/sets/lists/debug/md.evbarm.earmv5eb: revision 1.3
        distrib/sets/lists/debug/ad.mips: revision 1.88
        share/mk/bsd.own.mk: revision 1.1482
        distrib/sets/lists/debug/md.evbppc: revision 1.6
        etc/Makefile: revision 1.476

   Fix MKDEBUGKERNEL vs MKDEBUG for kernel debug file sets.

    [21 lines not shown]
VersionDeltaFile
1.1.2.1+48-48distrib/sets/lists/debug/md.evbarm.earmv5
1.1.2.1+20-20distrib/sets/lists/debug/md.evbarm.earmv4
1.3.2.1+13-13distrib/sets/lists/debug/md.evbppc
1.1.2.1+11-11distrib/sets/lists/debug/md.evbarm.earmv5eb
1.6.2.1+10-10distrib/sets/lists/debug/ad.mipsel
1.84.2.2+7-7distrib/sets/lists/debug/ad.mips
+109-1098 files not shown
+138-13114 files

NetBSD/src BToUuHqdistrib/sets sets.subr

   Pull up following revision(s) (requested by tls in ticket #433):

        distrib/sets/sets.subr: revision 1.214

   Fix MKBINUTILS=no builds.  This message brought to you by the number
   33,554,432 and the word 'embedded'.
VersionDeltaFile
1.212.2.1+3-3distrib/sets/sets.subr
+3-31 files

NetBSD/src x12ytQksys/arch/sparc64/conf GENERIC

   sun4v: Add ixl(4) to sparc64 GENERIC config.

   Intel X710-T4 network interface is present in SPARC S7-2 servers.

   Not tested yet.

   From OpenBSD.
VersionDeltaFile
1.251+3-2sys/arch/sparc64/conf/GENERIC
+3-21 files

NetBSD/src yLHJXWvsys/dev/ic wdc.c

   Pull up following revision(s) (requested by tls in ticket #432):

        sys/dev/ic/wdc.c: revision 1.314

   Avoid crash due to `atactl atabusX reset` when no drives are attached
VersionDeltaFile
1.310.12.2+6-5sys/dev/ic/wdc.c
+6-51 files

NetBSD/src t7hZ2I8sys/dev/ic wdcvar.h wdc.c, sys/dev/pci svwsata.c artsata.c

   Pull up following revision(s) (requested by tls in ticket #430):

        sys/dev/pci/artsata.c: revision 1.31
        sys/dev/pci/pciide_common.c: revision 1.74
        sys/dev/pci/svwsata.c: revision 1.25
        sys/dev/ic/wdc.c: revision 1.313
        sys/dev/ic/wdcvar.h: revision 1.101

   Trying again to avoid that 3-second delay on PCI IDE controllers.

   This time, the change is conditionalized so it won't hit old-style wdc.
VersionDeltaFile
1.310.12.1+21-2sys/dev/ic/wdc.c
1.24.38.1+3-3sys/dev/pci/svwsata.c
1.30.8.1+3-3sys/dev/pci/artsata.c
1.72.2.1+4-2sys/dev/pci/pciide_common.c
1.100.28.1+2-1sys/dev/ic/wdcvar.h
+33-115 files

NetBSD/src GAwc65Esys/dev/pci if_pcn.c

   Pull up following revision(s) (requested by tls in ticket #429):

        sys/dev/pci/if_pcn.c: revision 1.82
        sys/dev/pci/if_pcn.c: revision 1.83

   Fix pcn MAC address read on QEMU and possibly some other emulated hardware.

   Correct typo in previous (s/1/i/) and remove gratuitous wsp (build fix)
VersionDeltaFile
1.80.2.1+21-4sys/dev/pci/if_pcn.c
+21-41 files

NetBSD/src AB8peNjsys/arch/alpha/alpha mcclock.c

   Pull up following revision(s) (requested by tls in ticket #428):

        sys/arch/alpha/alpha/mcclock.c: revision 1.24

   Fix alpha boot-time hang on unpatched QEMU by accommodating its broken mcclock.
VersionDeltaFile
1.23.4.1+15-2sys/arch/alpha/alpha/mcclock.c
+15-21 files

NetBSD/src yeMyy8Vsys/arch/riscv/riscv riscv_machdep.c

   Pull up following revision(s) (requested by tls in ticket #427):

        sys/arch/riscv/riscv/riscv_machdep.c: revision 1.49

   Support poweroff on RISCV
VersionDeltaFile
1.43.2.4+7-2sys/arch/riscv/riscv/riscv_machdep.c
+7-21 files

NetBSD/src mJXj40udoc CHANGES-11.1

   Tickets #421 - #424
VersionDeltaFile
1.1.2.7+21-1doc/CHANGES-11.1
+21-11 files

NetBSD/src Bb12ZHosys/conf std

   Pull up following revision(s) (requested by tls in ticket #424):

        sys/conf/std: revision 1.25

   Our security(7) manual page has long documented the functionality
   enabled by secmodel_extensions.  It's thus not really optional, and
   declaring it so causes an odd mismatch in behavior between MODULAR and
   non-modular kernels.  It's tiny, just add it to std.
VersionDeltaFile
1.24.4.1+2-1sys/conf/std
+2-11 files

NetBSD/src iUhcKJTetc/etc.hppa MAKEDEV.conf

   Pull up following revision(s) (requested by tls in ticket #423):

        etc/etc.hppa/MAKEDEV.conf: revision 1.5

   add ld to hppa MAKEDEV; without it, virtio disks are unusable
VersionDeltaFile
1.4.2.1+2-1etc/etc.hppa/MAKEDEV.conf
+2-11 files

NetBSD/src GnZTu5Usys/arch/riscv/riscv bus_space_generic.S

   risc-v: some fixes for the bus_space helper function

   - don't sign extend in generic_bs_r_4 for _LP64
   - use sll (not srl) for the delta = n << stride calculations
   - fix the delta calculation in generic_bs_rr_2
VersionDeltaFile
1.9+9-8sys/arch/riscv/riscv/bus_space_generic.S
+9-81 files

NetBSD/src 4ciI4jGdistrib/sets/lists/debug mi

   Pull up following revision(s) (requested by tls in ticket #422):

        distrib/sets/lists/debug/mi: revision 1.509

   dab is not built if MKCXX=no, so its debug cannot be built
VersionDeltaFile
1.485.2.12+2-2distrib/sets/lists/debug/mi
+2-21 files

NetBSD/src PzxVlTHexternal/mpl/mozilla-certdata Makefile

   Pull up following revision(s) (requested by tls in ticket #421):

        external/mpl/mozilla-certdata/Makefile: revision 1.2

   Avoid make descending into mozilla-certdata directory and finding no targets,
   which broke MKSHARE=no builds
VersionDeltaFile
1.1.6.1+2-2external/mpl/mozilla-certdata/Makefile
+2-21 files

NetBSD/src H1a62LLsbin/mount_lfs mount_lfs.8 mount_lfs.c

   Add options -k, to invoke a kernel cleaner thread, and -u, to invoke
   lfs_cleanerd(8).  -k is now the default, matching the test suite.
VersionDeltaFile
1.41+57-25sbin/mount_lfs/mount_lfs.c
1.16+17-9sbin/mount_lfs/mount_lfs.8
+74-342 files

NetBSD/src twJxXCysys/ufs/lfs lfs_vnops.c

   Hold the segment lock across the bytes reporting in LFCNRECLAIM.
   This is a trivial modification of riastradh's patch from PR 60504.
VersionDeltaFile
1.354+12-8sys/ufs/lfs/lfs_vnops.c
+12-81 files