Page MenuHomeFreeBSD

rang_acm.org (Anton Rang)
User

Projects

User does not belong to any projects.

User Details

User Since
Jun 18 2015, 9:08 PM (231 w, 12 h)

Recent Activity

Tue, Nov 5

rang_acm.org accepted D22239: Don't sanitize linker_set.
Tue, Nov 5, 5:55 AM

Mar 23 2019

rang_acm.org accepted D19665: x86: Use XSAVEOPT for fpusave(), when available.
Mar 23 2019, 1:44 AM

Mar 22 2019

rang_acm.org added inline comments to D19665: x86: Use XSAVEOPT for fpusave(), when available.
Mar 22 2019, 9:14 PM
rang_acm.org added inline comments to D19665: x86: Use XSAVEOPT for fpusave(), when available.
Mar 22 2019, 1:27 PM
rang_acm.org added inline comments to D19665: x86: Use XSAVEOPT for fpusave(), when available.
Mar 22 2019, 6:51 AM

Feb 28 2019

rang_acm.org accepted D19398: Fix a scribbler in the PMS driver..
Feb 28 2019, 5:29 PM

Jan 26 2018

rang_acm.org added a comment to D14062: Make memory mapped via pmap_qenter() non-executable for amd64/i386..

I think bpf jit would need a change to work with this — a kernel equivalent to this mprotect() call for the userspace case.

Jan 26 2018, 11:15 PM

Dec 13 2017

rang_acm.org accepted D13457: Implement ACPI CPU support when Processor object is not present.

Looks correct to me per the spec.

Dec 13 2017, 5:00 PM

Apr 13 2017

rang_acm.org added inline comments to D10379: savecore: fix space calculation with respect to `minfree` in check_space(..).
Apr 13 2017, 6:15 AM
rang_acm.org added inline comments to D10379: savecore: fix space calculation with respect to `minfree` in check_space(..).
Apr 13 2017, 6:11 AM
rang_acm.org added inline comments to D10379: savecore: fix space calculation with respect to `minfree` in check_space(..).
Apr 13 2017, 5:58 AM

Jan 11 2017

rang_acm.org retitled D9138: Enable ^C (and incidentally ^S/^Q) in ddb. from to Enable ^C (and incidentally ^S/^Q) in ddb..
Jan 11 2017, 5:52 PM

Oct 4 2016

rang_acm.org updated the diff for D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..

Add comments.

Oct 4 2016, 4:39 PM

Oct 3 2016

rang_acm.org updated the diff for D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..

Change generation number to 32 bit.

Oct 3 2016, 11:09 PM
rang_acm.org added a comment to D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..
In D8041#167776, @kib wrote:

The bit which I do not like about the patch is the use of uint64_t for generation. Why cannot it be uint32_t ?

Oct 3 2016, 5:05 PM

Sep 30 2016

rang_acm.org added a comment to D8098: Avoid unfruitful wakeups in bdone.
In D8098#167666, @kib wrote:

The main point is that bwait() is performed by threads which are not allowed to modify buffer until B_DONE is observed. Then, as was noted, mtxpool provides the exclusion between g_up thread (executing bdone()) and whatever threads happen to do bwait().

Sep 30 2016, 9:32 PM
rang_acm.org updated the diff for D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..

Add PCPU field for i386.

Sep 30 2016, 8:20 PM

Sep 29 2016

rang_acm.org added inline comments to D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..
Sep 29 2016, 3:56 AM

Sep 28 2016

rang_acm.org updated the diff for D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..

A very tiny microoptimization for invlrng_handler, while I'm there.

Sep 28 2016, 8:43 PM
rang_acm.org updated the diff for D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..

Move the read of smp_tlb_generation before serializing instructions.

Sep 28 2016, 8:33 PM
rang_acm.org added a comment to D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..
In D8041#166681, @kib wrote:

Are you sure about 2 cores ? Might be 4 ?

Sep 28 2016, 7:26 PM
rang_acm.org updated the diff for D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..

Remove unneeded wmb().

Sep 28 2016, 7:25 PM
rang_acm.org added inline comments to D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..
Sep 28 2016, 5:48 AM
rang_acm.org added a comment to D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..
In D8041#166525, @kib wrote:

Do you have any measurements that demostrate the improvement due to the patch ? I do not object against this change going in if e.g. even micromeasurements like hwpmc or intel pcm tool show reduction of inter-core traffic.

Sep 28 2016, 5:26 AM

Sep 26 2016

rang_acm.org retitled D8041: Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags. from to Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags..
Sep 26 2016, 11:17 PM