Remove unrelated bits.
- Queries
- All Stories
- Search
- Advanced Search
- Transactions
- Transaction Logs
Advanced Search
Apr 8 2017
Add missing devel/Makefile bit
Apr 7 2017
.. works for me. Get that in, chase perfection at own schedule later...
Apr 6 2017
Indeed, corrected.
Mar 23 2017
In D10086#208559, @jmallett wrote:This looks reasonable, but I share the confusion about correctness. It seems like this should be in libgcc generally. I know we spoke about this, but I've forgotten; can you remind me why libgcc isn't providing this symbol?
Mar 16 2017
Mar 7 2017
Mar 4 2017
Quick way to find out is to build mipselhf world using external GCC. It basically gets all things wrong re:default endianness and softfloat.
Does LDFLAGS contain -m32/64 -mabi and -EB/-EL as appropriate on MIPS? If not, this change will definitely break things.
Feb 27 2017
Feb 22 2017
Feb 20 2017
This kinda somewhat works for mbufs, but not much more.
Feb 15 2017
Feb 6 2017
Feb 1 2017
It looks like compiled code gets relaxed by the assembler to have correct nops in there. I am not convinced we should be dependent on this behavior in the long run, but pretty much every atomic inline depends on it now, so I think I am going to commit this version.
0xffffffff8020945c <__mtx_lock_flags+232>: ll v1,16(s0) 0xffffffff80209460 <__mtx_lock_flags+236>: bne v1,v0,0x80209480 <__mtx_lock_flags+268> 0xffffffff80209464 <__mtx_lock_flags+240>: nop 0xffffffff80209468 <__mtx_lock_flags+244>: move v1,a2 0xffffffff8020946c <__mtx_lock_flags+248>: sc v1,16(s0) 0xffffffff80209470 <__mtx_lock_flags+252>: beqz v1,0x8020945c <__mtx_lock_flags+232> 0xffffffff80209474 <__mtx_lock_flags+256>: nop 0xffffffff80209478 <__mtx_lock_flags+260>: j 0x80209488 <__mtx_lock_flags+276> 0xffffffff8020947c <__mtx_lock_flags+264>: nop 0xffffffff80209480 <__mtx_lock_flags+268>: sw v1,24(sp) 0xffffffff80209484 <__mtx_lock_flags+272>: li v1,0
Jan 31 2017
In D9391#194079, @brooks wrote:Subject to me being wrong to worry about branch delay slots, this looks fine.
Jan 30 2017
Jan 27 2017
Jan 14 2017
Yup
Jan 12 2017
Jan 6 2017
Jan 5 2017
Jan 4 2017
Dec 30 2016
Nope, it is better to just exclude the file again. It is utter waste on mips64 indeed and I somehow though it servded useful purpose
Dec 29 2016
Dec 28 2016
Missed one more place
Dec 24 2016
Do dot define VM_MEMATTR_WRITE_COMBINING as uncached as the fallback option, some userland tools expect VM_MEMATTR_WRITE_COMBINING and VM_MEMATTR_UNCACHEABLE be numerically different.
Dec 23 2016
Fix typo
Dec 18 2016
Dec 17 2016
Dec 15 2016
Dec 14 2016
Dec 12 2016
Dec 2 2016
Nov 20 2016
Nov 18 2016
Copy ARM way of dealing with differences through suffixes per suggestion by Ruslan
Nov 17 2016
Nov 16 2016
Nov 13 2016
In D8505#176873, @imp wrote:I've had issues in the past trying to do this. There are some files in the kernel that don't actually work when you set these features, especially the compressed trampoline. I also think there's one in either csu or libc that has issues.
Best way to test this is to spike gcc to be the wrong endian by default and do a buildworld and buildkernel for one of the platforms that has a trampiline. If that works in addition to the external toolchain you're using, I'm game.
May 4 2016
Apr 27 2016
Glad to see this stuff getting kicked out of the 'normal' KASSERT path.