May be the following notes will be useful to someone with an appropriate background . . .

I've tested head -r363932 under uefi/ACPI v1.20 with the 3072 limit disabled and it failed the huge-file duplicate and diff/cmp test:

# cp -aRx /usr/obj/clang-armv7-on-aarch64.tar /usr/obj/clang-armv7-on-aarch64.alt_tar
# diff /usr/obj/clang-armv7-on-aarch64.tar /usr/obj/clang-armv7-on-aarch64.alt_tar
Binary files /usr/obj/clang-armv7-on-aarch64.tar and /usr/obj/clang-armv7-on-aarch64.alt_tar differ
# cmp -l /usr/obj/clang-armv7-on-aarch64.tar /usr/obj/clang-armv7-on-aarch64.alt_tar | head -30
2633269249   3   0
2633269251   3   0
2633269252  55   0
2633269253   6   0
2633269254  21   0
2633269255 227   0
2633269256   1   0
2633269257 135   0
2633269258 336   0
2633269259  22 140
2633269260   0 100
2633269261 346   0
2633269262 353   0
2633269265 227   0
2633269266   1 160
2633269267 170 140
2633269268 336 100
2633269269  22   0
2633269271 362   0
2633269272 353   0
2633269275 227   0
2633269276   1   0
2633269277 225   0
2633269278 336   0
2633269279  22   0
2633269281 376   0
2633269282 353   1
2633269285   0   1
2633269289   0 223
2633269290   0 321

For reference:

# ls -ldT /usr/obj/clang-armv7-on-aarch64*
-rw-r--r--  1 root  wheel  11570948096 Jul 18 18:32:37 2020 /usr/obj/clang-armv7-on-aarch64.alt_tar
-rw-r--r--  1 root  wheel  11570948096 Jul 18 18:32:37 2020 /usr/obj/clang-armv7-on-aarch64.tar

(So the over 10 GiByte original file is significantly larger than the 8 GiByte RAM, although I've had large but much smaller files fail as well.)

Thanks for this review, it's now effectively enabled via rS364732 for the change that went into upstream, and rS364733 to change src.opts.mk.

I've submitted an upstream review here: https://reviews.llvm.org/D86397

I'll have a look at this now. I was not aware of this review at all.

ping

@andrew ping

It would also be good to test with 'cryptocheck -a all -d armv8crypto0 -z' though the NIST KAT are probably sufficient coverage already for XTS.

Looks fine. You should look at unrolling the loop to 3 or 4 rounds. Looking at the A72 optimization guide, it shows that there is a 3 cycle latency, but throughput of 1. Section 4.10 gives example showing three pairs to achieve max perf.

Already accepted, but just wanted to confirm this solved interrupt mapping problems we were having on LX2160a.

In D25373#565381, @greg_unrelenting.technology wrote:

In D25373#565380, @markm wrote:

ECAM hacks? When are those likely to appear? :-)

https://unrelentingtech.s3.dualstack.eu-west-1.amazonaws.com/flash-image-2020-07-01.bin try this image, it has the ECAM offset removed

@mw promised to make a toggle for ECAM offset in the setup menu.. :)

OK to close again.

That works, thanks! I'm at latest CURRENT, no backouts, my console works and so do my PCI ethernet port.

In D25373#565380, @markm wrote:

ECAM hacks? When are those likely to appear? :-)

ECAM hacks? When are those likely to appear? :-)

In D25373#565376, @markm wrote:

With this patch in, AND the flashe given above, I get my console back!

But I lose my PCI ethernet ports.

With this patch in, AND the flashe given above, I get my console back!

In D25373#565370, @markm wrote:

Worse. Previously working console now broken. Current with this change backed out now broken.

Is the console supposed to work without this fix applied?

Worse. Previously working console now broken. Current with this change backed out now broken.

Trying now. In the meanwhile, is there some place where the Collected Wisdom™ of all current mcbin firmware statusses, hints, binaries etc are kept?

Add conv=sync for dd
If it's the old version that GregV posted a while ago it's normal that this commit "breaks" for you, this image have a patched SPCR table to cope with the old bug that this code fixes.

Yup - my flash is a couple of months old, and a hand-rolled debug-version (not by me). The image you pointed to fails with checksum errors at boot, but it may have something to do with the way I made it. With the same MBR as the old one, I did:

It works fine on mine, are you using an old edk2 build maybe ?
Try https://people.freebsd.org/~manu/flash-image-mcbin.bin
This is mainline edk2 (latest stable version) and Marvell TF-A, ports will land soon.

This change breaks my MacchiattoBin DoubleShot. I've bisected to get to commit r362574, which is this. The symptoms are a hang right at the beginning of the boot, at this point:

I'm sorry that this is taking a while to get reviewed. I'd like to help get it in. Would be willing to rebase the patch and re-upload? It doesn't apply for me.