I tried a similar version of this change and my eMAG reports CPU 0: APM eMAG 8180 r3p2 affinity: 0 0
- Queries
- All Stories
- Search
- Advanced Search
- Transactions
- Transaction Logs
Feed Advanced Search
Advanced Search
Advanced Search
Aug 21 2019
Aug 21 2019
D21314: Identify emag 8180 CPU is now accepted and ready to land.
makefs: share fat.h between kernel msdosfs and makefs
makefs: use `char *` not `void *` for buf b_data, drop casts in msdos
emaste added a comment to D21348: makefs: use `char *` not `void *` for buf b_data and drop casts from msdos.
You've build-checked this, right?
Yes on amd64.
Aug 20 2019
Aug 20 2019
makefs: avoid "dereferencing 'void *' pointer" warnings
emaste added a comment to D11197: msdosfs: expose and hide definitions/functions for makefs FAT fs support.
fat.h deduped in D21346
shuffle includes to avoid struct declarations
makefs: add msdosfs (FAT) support
MFC r351254: mqueuefs: fix compat32 struct file leak
MFC r351254: mqueuefs: fix compat32 struct file leak
mqueuefs: fix compat32 struct file leak
mounting the ESP at /boot/efi LGTM, this will give paths like /boot/efi/EFI/... but this is consistent with what Linux distros do
rebase and remove extraneous changes
Aug 19 2019
Aug 19 2019
makefs.8: expand description of image size
makefs.8: style updates from igor
Rework advisory "Perform one of the following" text
Aug 16 2019
Aug 16 2019
D20562: The efifat files are no longer used: remove the code to build them. now requires changes to proceed.
This change is not valid, it removes the templates but leaves the Makefile rules in place to install them.
LGTM other than the minor comments inline
D7897: bspatch: add integer overflow checks now requires review to proceed.
rebase, add inline as suggested by @delphij
aarch64: make pmap_change_attr public like on other platforms
This is just another "upstream only considered Linux, we need to be included in the same way as Linux" case.
D18936: Remove the unused and unusable MDIOCLIST ioctl. is now accepted and ready to land.
Can you show a diff of the NetBSD version to this? i.e. what you had to change in the port?
(I assume that we don't want to treat this as contrib code and will just adopt it.)
Wrap in .if !defined(_SKIP_BUILD)
Add @generated tag to libc syscall asm wrappers
emaste committed rS351121: arch.7: indicate expectation that arm (i.e., v5) will be removed before 13.0.
arch.7: indicate expectation that arm (i.e., v5) will be removed before 13.0
Regen src.conf.5 after r351027 (enable BSD_CRTBEGIN on powerpc)
(I've looked over the change other than the just update dictionary section so far)
Aug 15 2019
Aug 15 2019
emaste added a comment to D21060: Stop installing clang, clang++, and clang-cpp hardlinks in /usr/bin..
No objection from me
emaste updated subscribers of D18758: Refactor skein_block_asm.s to assemble with LLVM integrated assembler.
emaste added inline comments to D18758: Refactor skein_block_asm.s to assemble with LLVM integrated assembler.
gptzfsboot: remove CLANG_NO_IAS
As it turns out gptzfsboot is identical, whether GAS or IAS is used to assemble gptldr.S:
all other boot components have been addressed; only gptzfsboot needs to be investigated now
stand: remove CLANG_NO_IAS from pxeldr
rebase after committing boot2 and cdboot changes
cdboot in rS351092
stand: remove CLANG_NO_IAS from cdboot
cdboot differences are a large number of addr32 prefixes and using nop nop for alignment rather than xchg %eax,%eax, so it is ok too
Do we know if the differences are similar there as well?
stand: remove CLANG_NO_IAS from boot2
rS351073 for boot2
side by side comparison with the .o generated
telnet: remove 3rd clause from Berkeley copyrights
do not enable userland retpoline if not supported by compiler/linker
arch.7: be explicit about the future for unmigrated GCC 4.2.1 archs
Aug 14 2019
Aug 14 2019
compiler-rt: enable __bswapsi2/__bswapdi2 for RISC-V
arch.7: clarify that the table shows the default tool chain
arch.7: add deprecation notice for GCC 4.2.1
arch.7: add tool chain table
MFC r350497: ppp: correct echo-req magic number on big endian archs
MFC r350497: ppp: correct echo-req magic number on big endian archs
Aug 13 2019
Aug 13 2019
sys/{x86,amd64}: remove one of doubled ;s
In D21110#461916, @asomers wrote:Submitted as r350665. I don't know why Phabricator didn't close the revision automatically.
advisory-template: add shutdown command to template
Remove some more leftover rlogin man page xrefs
crunch: remove rsh and rlogin from example config file
Regen src.conf.5 after r350980 (remove rsh/rlogin references)
Remove rlogin/rsh references from src.conf(5) WITHOUT_BLACKLIST_SUPPORT
Remove rsh/rlogin references from security man page
Aug 12 2019
Aug 12 2019
nandfs: avoid integer overflow in nandfs_get_dat_bdescs_ioctl
In D21232#461469, @imp wrote:These are fine but (a) you need to have mount privs to execute this ioctl; (b) nandfs is a panic trap due to bad locking and the system can't stay up once there's any vnode pressure at all; and (c) the set of nandfs users is the empty set due to (b). This is not exploitable in any meaningful way.
emaste added a comment to D21230: Force ports depending on a fetch target to actually run checksum..
Seems reasonable to me. I might expand the comment slightly to "prevents a MITM attack on the dependency."
MFC r350498: telnet: use asprintf for r350140 [MFC of r349890] change
MFC r350498: telnet: use asprintf for r350139 [MFC of r349890] change
Aug 8 2019
Aug 8 2019
@ngie will you commit this?
armv8crypto: add missing unlock for session (after r336439)
emaste added a comment to D21012: armv8crypto: add missing unlock for session (fix "sleeping thread owns a non-sleepable lock").
Looks like this was broken in rS336439
Aug 6 2019
Aug 6 2019
MFC r350645: Correct ICMPv6/MLDv2 out-of-bounds memory access