Windows NT4 for PowerPC guest on OPAL on POWER8 in qemu

Sometimes, programming is just for fun. This is what PREPHV is for Andrei Warkentin. To quote the README:

“This is mostly a huge ugly hack, derived from my
ppc64le_hello code. The running philosophy here is
to throw things together late at night with my family
asleep and see how far I get without a real design
or without a real desire to implement boring things
like IDE (*sigh*) emulation”

Since my day job is maintaining the firmware that it runs on, I decided to have a go (it also ties in with the retro stuff I’ve been blogging about). So…

screenshot-from-2016-10-30-17-22-20and I’m off! (yes, this is the very latest qemu and skiboot):

screenshot-from-2016-10-30-17-23-32screenshot-from-2016-10-30-17-23-48Yes, prephv does clear all thirty two megabytes of guest memory

screenshot-from-2016-10-30-17-24-15A quick diversion, if you try Windows NT 3.51 for PowerPC, you get this:

screenshot-from-2016-10-30-18-17-35

But on NT4, you continue unharmed:

screenshot-from-2016-10-30-17-22-32A sign I needed to hack my filesystem of bits of NT installer bits a bit more:

screenshot-from-2016-10-30-17-22-45But, on my next try:

screenshot-from-2016-10-30-17-25-26Well… looks like there’s an instruction that needs to be emulated (and there’s no code to currently do that). Mind you… this is decently far into booting before we hit anything fatal, which is a pretty impressive effort – and it is tempting to continue and see if it’ll run on real hardware and if it could be made to work well enough to not find any disks :)

TianoCore (UEFI) ported to OpenPower

Recently, there’s been (actually two) ports of TianoCore (the reference implementation of UEFI firmware) to run on POWER on top of OPAL (provided by skiboot) – and it can be run in the Qemu PowerNV model.

More details:

1 Million SQL Queries per second: GA MariaDB 10.1 on POWER8

A couple of days ago, MariaDB announced that MariaDB 10.1 is stable GA – around 19 months since the GA of MariaDB 10.0. With MariaDB 10.1 comes some important scalabiity improvements, especially for POWER8 systems. On POWER, we’re a bit unique in that we’re on the higher end of CPUs, have many cores, and up to 8 threads per core (selectable at runtime: 1, 2, 4 or 8/core) – so a dual socket system can easily be a 160 thread machine.

Recently, we (being IBM) announced availability of a couple of new POWER8 machines – machines designed for Linux and cloud environments. They are very much OpenPower machines, and more info is available here: http://www.ibm.com/marketplace/cloud/commercial-computing/us/en-us

Combine these two together, with Axel Schwenke running some benchmarks and you get 1 Million SQL Queries per second with MariaDB 10.1 on POWER8.

Having worked a lot on both MySQL for POWER and the firmware that ships in the S882LC, I’m rather happy that 1 Million queries per second is beyond what it was in June 2014, which was a neat hack on MySQL 5.7 that showed the potential of MySQL on POWER8 but wasn’t yet a product. Now, you can run a GA release of MariaDB on GA POWER8 hardware designed for scale-out cloud environments and get 1 Million SQL queries/second (with fewer cores than my initial benchmark last year!)

What’s even more impressive is that this million queries per second is in a KVM guest!

PAPR spec publicly available to download

PAPR is the Power Architecture Platform Reference document. It’s a short read at only 890 pages and defines the virtualised environment that guests run in on PowerKVM and PowerVM (i.e. what is referred to as ‘pseries’ platform in the Linux kernel).

https://members.openpowerfoundation.org/document/dl/469

As part of the OpenPower Foundation, we’re looking at ensuring this is up to date, documents KVM specific things as well as splitting out the bits that are common to OPAL and PAPR into their own documents.

doing nothing on modern CPUs

Sometimes you don’t want to do anything. This is understandably human, and probably a sign you should either relax or get up and do something.

For processors, you sometimes do actually want to do absolutely nothing. Often this will be while waiting for a lock. You want to do nothing until the lock is free, but you want to be quick about it, you want to start work once that lock is free as soon as possible.

On CPU cores with more than one thread (e.g. hyperthreading on Intel, SMT on POWER) you likely want to let the other threads have all of the resources of the core if you’re sitting there waiting for something.

So, what do you do? On x86 there’s been the PAUSE instruction for a while and on POWER there’s been the SMT priority instructions.

The x86 PAUSE instruction delays execution of the next instruction for some amount of time while on POWER each executing thread in a core has a priority and this is how chip resources are handed out (you can set different priorities using special no-op instructions as well as setting the Relative Priority Register to map how these coarse grained priorities are interpreted by the chip).

So, when you’re writing spinlock code (or similar, such as the implementation of mutexes in InnoDB) you want to check if the lock is free, and if not, spin for a bit, but at a lower priority than the code running in the other thread that’s doing actual work. The idea being that when you do finally acquire the lock, you bump your priority back up and go do actual work.

Usually, you don’t continually check the lock, you do a bit of nothing in between checking. This is so that when the lock is contended, you don’t just jam every thread in the system up with trying to read a single bit of memory.

So you need a trick to do nothing that the complier isn’t going to optimize away.

Current (well, MySQL 5.7.5, but it’s current in MariaDB 10.0.17+ too, and other MySQL versions) code in InnoDB to “do nothing” looks something like this:

ulint ut_delay(ulint   delay)
{
        ulint   i, j;
        UT_LOW_PRIORITY_CPU();
        j = 0;
        for (i = 0; i < delay * 50; i++) {
                j += i;
                UT_RELAX_CPU();
        }
        if (ut_always_false) {
                ut_always_false = (ibool) j;
        }
        UT_RESUME_PRIORITY_CPU();
        return(j);
}

On x86, UT_RELAX_CPU() ends up being the PAUSE instruction.

On POWER, the UT_LOW_PRIORITY_CPU() and UT_RESUME_PRIORITY_CPU() tunes the SMT thread priority (and on x86 they’re defined as nothing).

If you want an idea of when this was all written, this comment may be a hint:

/*!< in: delay in microseconds on 100 MHz Pentium */

But, if you’re not on x86 you don’t have the PAUSE instruction, instead, you end up getting this code:

# elif defined(HAVE_ATOMIC_BUILTINS)
#  define UT_RELAX_CPU() do { \
     volatile lint      volatile_var; \
     os_compare_and_swap_lint(&volatile_var, 0, 1); \
   } while (0)

Which you may think “yep, that does nothing and is not optimized away by the compiler”. Except you’d be wrong! What it actually does is generates a lot of memory traffic. You’re now sitting in a tight loop doing atomic operations, which have to be synchronized between cores (and sockets) since there’s no real way that the hardware is going to be able to work out that this is only a local variable that is never accessed from anywhere.

Additionally, the ut_always_false and j variable there is also attempts to trick the complier into not optimizing the loop away, and since ut_always_false is a global, you’re generating traffic to a single global variable too.

Instead, what’s needed is a compiler barrier. This simple bit of nothing tells the compiler “pretend memory has changed, so you can’t optimize around this point”.

__asm__ __volatile__ ("":::"memory")

So we can eliminate all sorts of useless non-work and instead do what we want: do nothing (a for loop for X iterations that isn’t optimized away by the compiler) and don’t have side effects.

In MySQL bug 74832 I detailed this with the appropriately produced POWER assembler. Unfortunately, this patch (submitted under the OCA) has sat since November 2014 (so, over 9 months) with no action. I’m a bit disappointed by that to be honest.

Anyway, the real moral of this story is: don’t implement your own locking primitives. You’re either going to get it wrong or you’ll be wrong in a few years when everything changes under you.

See also:

The sad state of MySQL and NUMA

Way back in 2010, MySQL Bug 57241 was filed, pointing out that the “swap insanity” problem was getting serious on x86 systems – with NUMA being more and more common back then.

The swapping problem is due to running out of memory on a NUMA node and having to swap things to other nodes (see Jeremy Cole‘s blog entry also from 2010 on the topic of swap insanity). This was back when 64GB and dual quad core CPUs was big – in the past five years big systems have gotten bigger.

Back then there were two things you could do to have your system be usable: 1) numa=off as kernel boot parameter (this likely has other implications though) and 2) “numactl –interleave all” in mysqld_safe script (I think MariaDB currently has this built in if you set an option but I don’t think MySQL does, otherwise perhaps the bug would have been closed).

Anyway, it’s now about 5 years since this bug was opened and even when there’s been a patch in the Twitter MySQL branch for a while (years?) and my Oracle Contributor Agreement signed patch attached to bug 72811 since May 2014 (over a year) we still haven’t seen any action.

My patch takes the approach of you want things allocated at server startup to be interleaved across nodes (e.g. buffer pool) while runtime allocations are probably per connection and are thus fine (in fact, better) to do node local allocations.

Without a patch like this, or without running mysqld with the right numactl incantation, you end up either having all your memory on one NUMA node (potentially not utilising full memory bandwidth of the hardware), or you end up with swap insanity, or you end up with some other not exactly what you’d expect situation.

While we could have MySQL be more NUMA aware and perhaps do a buffer pool instance per NUMA node or some such thing, it’s kind of disappointing that for dedicated database servers bought in the past 7+ years (according to one comment on one of the bugs) this crippling issue hasn’t been addressed upstream.

Just to make it even more annoying, on certain workloads you end up with a lot of mutex contention, which can end up meaning that binding MySQL to fewer NUMA nodes (memory and CPU) ends up increasing performance (cachelines don’t have as far to travel) – this is a different problem than swap insanity though, and one that is being addressed.

Update: My patch as part of https://bugs.mysql.com/bug.php?id=72811 has been merged! MySQL on NUMA machines just got a whole lot better. I just hope it’s enabled by default…

OPAL firmware specification, conformance and documentation

Now that we have an increasing amount of things that run on top of OPAL:

  1. Linux
  2. hello_world (in skiboot tree)
  3. ppc64le_hello (as I wrote about yesterday)
  4. FreeBSD

and that the OpenPower ecosystem is rapidly growing (especially around people building OpenPower machines), the need for more formal specification, conformance testing and documentation for OPAL is increasing rapidly.

If you look at the documentation in the skiboot tree late last year, you’d notice a grand total of seven text files. Now, we’re a lot better (although far from complete).

I’m proud to say that I won’t merge new code that adds/modifies an OPAL API call or anything in the device tree that doesn’t come with accompanying documentation, and this has meant that although it may not be perfect, we have something that is a decent starting point.

We’re in the interesting situation of starting with a working system, with mainline Linux kernels now for over a year (maybe even 18 months) being able to be booted by skiboot and run on powernv hardware (the more modern the kernel the better though).

So…. if anyone loves going through deeply technical documentation… do I have a project you can contribute to!

FreeBSD on OpenPower

There’s been some work on porting FreeBSD over to run natively on top of OPAL, that is, on bare metal OpenPower machines (not just under KVM).

This is one of four possible things to run natively on an OPAL system:

  1. Linux
  2. hello_world (in skiboot tree)
  3. ppc64le_hello (as I wrote about yesterday)
  4. FreeBSD

It’s great to see that another fully featured OS is getting ported to POWER8 and OPAL. It’s not yet at a stage where you could say it was finished or anything (PCI support is pretty preliminary for example, and fancy things like disks and networking live on PCI).

skiboot-4.1

I just posted this to the mailing list, but I’ve tagged skiboot-4.1, so we have another release! There’s a good amount of changes since 4.0 nearly a month ago and this is the second release since we hit github back in July.

For the full set of changes, “git log” is your friend, but a summary of them follows:

  • We now build with -fstack-protector and -Werror
  • Stack checking extensions when built with STACK_CHECK=1
  • Reduced stack usage in some areas, -Wstack-usage=1024 now.
    • Some functions could use 2kb stack, now all are <1kb
  • Unsafe libc functions such as sprintf() have been removed
  • Symbolic backtraces
  • expose skiboot symbol map to OS (via device-tree)
  • removed machine check interrupt patching in OPAL
  • occ/hbrt: Call stopOCC() for implementing reset OCC command from FSP
  • occ: Fix the low level ACK message sent to FSP on receiving {RESET/LOAD}_OCC
  • hardening to errors of various FSP code
    • fsp: Avoid NULL dereference in case of invalid class_resp bits
    • abort if device tree parsing fails
    • FSP: Validate fsp_msg in fsp_queue_msg
    • fsp-elog: Add various NULL checks
  • Finessing of when to use error log vs prerror()
  • More i2c work
  • Can now run under Mambo simulator (see external/mambo/skiboot.tcl) (commonly known as “POWER8 Functional Simulator”)
  • Document skiboot versioning scheme
  • opal: Handle more TFAC errors.
    • TB_RESIDUE_ERR, FW_CONTROL_ERR and CHIP_TOD_PARITY_ERR
  • ipmi: populate FRU data
  • rtc: Add a generic rtc cache
  • ipmi/rtc: use generic cache
  • Error Logging backend for bmc based machines
  • PSI: Drive link down on HIR
  • occ: Fix clearing of OCC interrupt on remote fix

So, who worked on this release? We had 84 csets from 17 developers. A total of 3271 lines were added, 1314 removed (delta 1957).

Developers with the most changesets
Stewart Smith 24 28.6%
Benjamin Herrenschmidt 17 20.2%
Alistair Popple 8 9.5%
Vasant Hegde 6 7.1%
Ananth N Mavinakayanahalli 5 6.0%
Neelesh Gupta 4 4.8%
Mahesh Salgaonkar 4 4.8%
Cédric Le Goater 3 3.6%
Wei Yang 3 3.6%
Anshuman Khandual 2 2.4%
Shilpasri G Bhat 2 2.4%
Ryan Grimm 1 1.2%
Anton Blanchard 1 1.2%
Shreyas B. Prabhu 1 1.2%
Joel Stanley 1 1.2%
Vaidyanathan Srinivasan 1 1.2%
Dan Streetman 1 1.2%
Developers with the most changed lines
Benjamin Herrenschmidt 1290 35.1%
Alistair Popple 963 26.2%
Stewart Smith 344 9.4%
Mahesh Salgaonkar 308 8.4%
Ananth N Mavinakayanahalli 198 5.4%
Neelesh Gupta 186 5.1%
Vasant Hegde 122 3.3%
Shilpasri G Bhat 39 1.1%
Vaidyanathan Srinivasan 24 0.7%
Joel Stanley 21 0.6%
Wei Yang 20 0.5%
Anshuman Khandual 15 0.4%
Cédric Le Goater 12 0.3%
Shreyas B. Prabhu 9 0.2%
Ryan Grimm 3 0.1%
Anton Blanchard 2 0.1%
Dan Streetman 2 0.1%
Developers with the most lines removed
Mahesh Salgaonkar 287 21.8%
Developers with the most signoffs (total 54)
Stewart Smith 44 81.5%
Vasant Hegde 4 7.4%
Benjamin Herrenschmidt 4 7.4%
Vaidyanathan Srinivasan 2 3.7%
Developers with the most reviews (total 2)
Vasant Hegde 2 100.0%

skiboot/OPAL versioning

skiboot is boot and runtime firmware for OpenPower systems. There are other components that make up all the firmware you need, but if you’re, say, a Linux kernel, you’re going to be interacting with skiboot.

I recently committed doc/versioning.txt to skiboot to try and explain our current thoughts on versioning releases.

It turns out that picking version numbers is a bit harder than you’d expect, especially when you want to construct a version string to display in places that has semantic meaning. In fact, the writing on Semantic Versioning influenced us heavily.

Since we’re firmware, making incompatible API changes is something we should basically never, ever do. Old kernels should must boot and work on new firmware and new kernels should boot and function on old firmware (and if they don’t, it plainly be a kernel bug). So, ignore the Major version parts of Semantic Versioning for us :)

For each new release, we plan to bump the minor version for mostly bug fix releases, while bump the major version for added functionality. Any additional information is to describe the version on that particular platform – as everybody shipping OPAL is likely to build it themselves with possibly some customizations (e.g. YOUR COMPANY NAME HERE, support for some on board RAID card or on-board automated coffee maker). See doc/versioning.txt for details.

You may wonder why we started at 4.0 for our first real version number. Well… this is purely a cunning plan to avoid confusion with other things, the details of which will only be extracted out of my when plied with a suitable amount of excellent craft beer (because if I’m going to tell a boring story, I may as well have awesome craft beer).

C bitfields considered harmful

In C (and C++) you can specify that a variable should take a specific number of bits of storage by doing “uint32_t foo:4;” rather than just “uint32_t foo”. In this example, the former uses 4 bits while the latter uses 32bits. This can be useful to pack many bit fields together.

Or, that’s what they’d like you to think.

In reality, the C spec allows the compiler to do just about anything it wants with these bitfields – which usually means it’s something you didn’t expect.

For a start, in a struct -e.g. “struct foo { uint32_t foo:4; uint32_t blah; uint32_t blergh:20; }” the compiler could go and combine foo and blergh into a single uint32_t and place it somewhere… or it could not. In this case, sizeof(struct foo) isn’t defined and may vary based on compiler, platform, compiler version, phases of the moon or if you’ve washed your hands recently.

Where this can get interesting is in network protocols (OMG DO NOT DO IT), APIs (OMG DO NOT DO IT), protecting different parts of a struct with different mutexes (EEP, don’t do it!) and performance.

I recently filed MySQL bug 74831 which relates to InnoDB performance on POWER8. InnoDB uses C bitfields which are themselves bitfields (urgh) for things like “flag to say if this table is compressed”. At various parts of the code, this flag is checked.

When you apply this simple patch:

--- mysql-5.7.5-m15.orig/storage/innobase/include/dict0mem.h
+++ mysql-5.7.5-m15/storage/innobase/include/dict0mem.h
@@ -1081,7 +1081,7 @@ struct dict_table_t {
        Use DICT_TF_GET_COMPACT(), DICT_TF_GET_ZIP_SSIZE(),
        DICT_TF_HAS_ATOMIC_BLOBS() and DICT_TF_HAS_DATA_DIR() to parse this
        flag. */
-       unsigned                                flags:DICT_TF_BITS;
+       unsigned                                flags;

I get 10,000 key lookups/sec more than without it!

Why is this? If you go and read the bug, you’ll see that the amount of CPU time spent on the instruction checking the bit flag is actually about the same… and this puzzled me for a while. That is, until Anton reminded me that the PMU can be approximate and perhaps I should look at the loads.

Sure enough, the major difference is that with the bitfield in place (i.e. MySQL 5.7.5 as it stands today), there is a ld instruction doing the load – which is a 64bit load. In my patched version, it’s a lwx instruction – which is a 32bit load.

So, basically, we were loading 8 bytes instead of 4 every time we were checking if it was a compressed table.

So, along with yesterday’s lesson of never, ever, ever use volatile, today’s lesson is never, ever, ever use bitfields.