TweetFollow Us on Twitter

September 94 - BALANCE OF POWER

BALANCE OF POWER

Tuning PowerPCMemory Usage

DAVE EVANS

[IMAGE 017-019_Balance_of_Pwr_h1.GIF]

If you care about the performance of code you write for the Power Macintosh, memory usage should be your foremost concern. With the PowerPCTM601 processor today, and even more important with future processors, memory usage of your code will have the greatest effect on its performance. Poorly written code will execute at a fraction of its potential, and often very simple changes will greatly improve the execution speed of your critical code.

Processors are improving much faster than the memory subsystems that support them. As the PowerPC chips move from 80 MHz to 100 MHz and beyond, their thirst for data to process and instructions to execute will increasingly tax memory. Memory caches attempt to mitigate that thirst, and all PowerPC processors come equipped with built-in caches. But your code can work well with a cache or it can work very poorly with a cache. I'll show you why and discuss what you can do to optimize your memory usage.

CACHE BASICS
As you know, a cache is simply very fast RAM that the processor can access quickly and that it uses to store recently referenced data and code. On the PowerPC processors, any data stored in the cache can be accessed without stalling the processor's pipeline. Accesses to data not in the cache will take about 20 times as long reading from main memory, or even 1 million times as long if the access causes a page fault with virtual memory. Getting and keeping your performance-critical code and data in the cache are therefore key to your execution speed.

A cache is divided into small blocks calledcache lines . On the PowerPC 601, for example, the cache has 1024 cache line blocks, each holding 32 bytes. In addition, the 601 will fetch two blocks when it can, making the cache line size effectively 64 bytes.

The first PowerPC processors have set associative caches of different sizes. The 601 has an eight-way set associative, unified cache that's 32K in size, and the 603 has a two-way set associative, split cache with 8K for data and 8K for instruction code. The termset associative refers to the way the cache relates to main memory, which is important to your performance. In some simple caching schemes, each cache line maps directly to specific areas of main memory; any access to one of these areas loads bytes into that cache line. But on the PowerPC processor, sets of cache lines are combined and then mapped to memory. There are eight cache lines in each set on the 601, and two in each set on the 603. An access to one of the areas mapped to a set will load bytes into the last-used cache line of the set, keeping the most frequently used cache lines from being purged. This more complicated scheme typically yields much better performance than the directly mapped cache.

CACHE THRASHING
The cache will most affect your performance when you're accessing large amounts of data. A typical example of this is walking through arrays to perform some operation. The best strategy is to minimize cache collisions during your accesses, and the best tactic for this is to access your data as sequentially as possible. If you walk through memory sequentially, you'll load the cache every 64 bytes, but all 64 bytes will be available for fast processing. Here's an example:

unsigned longdata[64][1024];
for (row = 2; row < 64; row++)
    for (column = 0; column < 1024; column++)
        data[row][column] =
            data[row-1][column] + data[row-2][column];

This example performs additions on each element of a large matrix and accesses that matrix sequentially in memory. It walks across each row, adding elements and storing the result. But just inverting the loops can significantly change the way memory is accessed:

unsigned longdata[64][1024];
for (column = 0; column < 1024; column++)
    for (row = 2; row < 64; row++)
        data[row][column] =
            data[row-1][column] + data[row-2][column];

Reversing the loops leads to less than optimal performance since we perform each addition for all the columns before moving to the next element of a row. Instead of sequential access, this access pattern jumps across memory in even steps of 4K. Unfortunately, on the PowerPC processor these accesses map to the same set of cache lines, and every operation causes the cache to reload from main memory. This second example takes twice as long to execute the same calculation on a Power Macintosh 6100/60.

By paying attention to how your code accesses memory, you can avoid serious cache thrashing like that done by the second example. Things to look out for are loops that iterate for a power of 2 steps (128, 256, and so on)and code whose memory accesses are not close together.

An approach called blocking may help your loops. Often your code isn't as simple as above, and your memory accesses aren't regular during the loop. If you're walking two different arrays with different increments through memory, it may be impossible to serialize your accesses. Blocking performs the calculations in blocks of rows and columns. Instead of iterating across all the columns and then proceeding to the next row, you divide the dimensional space into blocks and calculate one whole block at a time. In this next example, we calculate the multiplication of two matrices.

long result[64][64], foo[64][128], bar[128][64];
for (row = 0; row < 64; row++)
    for (column = 0; column < 64; column++) {
        long    sum = 0;
        for (i = 0; i < 128; i++)
            sum += foo[row][i] * bar[i][column];
        result[row][column] = sum;
    }

As this algorithm walks through memory, it accesses result and foo sequentially, but bar is accessed in 256-byte steps. Accessing bar by jumping through memory causes cache misses, and sequential elements of bar are flushed from the cache before they're needed.

By performing this operation in small blocks, we can better use the cache. The key is to use all the elements of foo and bar that are in a cache line before moving on. One way to do this is to expand the loop and perform four operations in a single iteration:

long result[64][64], foo[64][128], bar[128][64];
for (row = 0; row < 64; row++)
    for (column = 0; column < 64; column += 4) {
        long    sum1 = 0, sum2 = 0;
        long    sum3 = 0, sum4 = 0;
        for (i = 0; i < 128; i++) {
            sum1 += foo[row][i] * bar[i][column];
            sum2 += foo[row][i] * bar[i][column+1];
            sum3 += foo[row][i] * bar[i][column+2];
            sum4 += foo[row][i] * bar[i][column+3];
    }
    result[row][column] = sum1;
    result[row][column+1] = sum2;
    result[row][column+2] = sum3;
    result[row][column+3] = sum4;
    }

This expanded loop calculates a block of four cells in each iteration. This executes faster because elements of bar are read from the cache and don't always cause cache misses as in the earlier example. Notice that in the expanded inner loop, a cache line of the bar matrix will be loaded the first time that it's referenced; then the following three references to bar will occur without stalling. Using the bar elements while they're still in the cache gives us a significant improvement.

CODE STYLE
Good compilers can pay attention to your memory accesses and will optimize how you access memory. For example, load and store operations can be reordered by the compiler to occur when the data is most likely to be available. The first time data is accessed it tends to cause a cache line to load, and subsequent accesses to nearby data must also wait for the cache load to complete. The compiler may be able to help by inserting a few instructions between the loads. This way the cache line will be fully loaded when the subsequent accesses are needed.

For more information on loop expansion and instruction reordering, see the Balance of Power column in develop Issue 18.*

You can help your compiler by using local variables when you can. These tell the compiler exactly how the data will be used, enabling it to easily reorder the loads and stores for this data.

You should also carefully note memory dereferences, especially double dereferences. Although it may be obvious to you, the compiler often can't tell whether two pointers address the same object in memory. The compiler may be prevented from reordering instructions because it can't tell whether two operations are really dependent on each other, just because they contain dereferences. Here's an example:

paramBlock->size = myStructure->size;
paramBlock->offset = myStructure->offset;

Although it appears obvious, the compiler usually can't tell if paramBlock references the same memory as myStructure. In the resulting binary, the compiler will be conservative and not reorder these operations for best execution. Replacing the dereference of myStructure with local variables for size and offset will allow the compiler to fully optimize this example.

INSTRUCTION THRASHING
Your code binary itself can cause the cache to thrash as it loads to be executed. This is very hard to detect and optimize. The basic problem is that your subroutines may map to the same areas of the cache, and frequent calls among them will stall to reload the cache. Some code profilers for RISC workstations have attempted to detect this problem, but for the Macintosh I can't suggest much help. Just changing the link order of your code and then executing profiles may have an effect; some link orders will thrash more than others.

DATA STRUCTURES
The layout of your data structures can greatly affect your cache usage and your memory usage in general. For example, memory accesses that cross 64-bit memory boundaries take twice as long to process, as this forces two bus transactions. On the PowerPC 601 processor, any misaligned data access within a memory boundary takes the standard amount of time, which (because of typical Macintosh data structures) is a valuable feature of the chip; future PowerPC processors, however, may take longer to access misaligned data. If you can align your data structures, do so now. A good tactic is to keep 64-bit data at the top of your structure, followed by your 32-bit data, and so on to prevent accidental misalignment of elements. Pad the end of the structure to an even 64-bit increment if you will have arrays of structures or will allocate them on the stack. And if certain parts of your structure are accessed much more often than other parts, keep these together so that they stay in the cache, and make sure they're aligned.

DON'T FORGET
The memory usage of your speed-critical code will greatly affect its performance today, and current problems will just get worse when PowerPC processors go above 100 MHz. Profile your code to find the most critical bottlenecks; then pay close attention to how that code addresses memory. You'll be rewarded with an excellent return on your investment.

DAVE EVANS occasionally uses the combinatorics skills he learned at the Massachusetts Institute of Technology, but more often he's been practicing his combination punches at a Thai kickboxing gym. Designing fast algorithms for Apple's OS Platforms Group is definitely rewarding, but developing a fast left hook really gets him pumped up. *

Thanks to Tom Adams, Mike Cappella, Rob Johnston, and Mike Neil for reviewing this column. *

 

Community Search:
MacTech Search:

Software Updates via MacUpdate

Google Earth 7.1.8.3036 - View and contr...
Google Earth gives you a wealth of imagery and geographic information. Explore destinations like Maui and Paris, or browse content from Wikipedia, National Geographic, and more. Google Earth combines... Read more
QuickBooks 16.1.11.1556 R12 - Financial...
QuickBooks helps you manage your business easily and efficiently. Organize your finances all in one place, track money going in and out of your business, and spot areas where you can save. Built for... Read more
FileZilla 3.24.0 - Fast and reliable FTP...
FileZilla (ported from Windows) is a fast and reliable FTP client and server with lots of useful features and an intuitive interface. Version 3.24.0: New The context menu for remote file search... Read more
Bookends 12.7.8 - Reference management a...
Bookends is a full-featured bibliography/reference and information-management system for students and professionals. Bookends uses the cloud to sync reference libraries on all the Macs you use.... Read more
Duplicate Annihilator 5.8.3 - Find and d...
Duplicate Annihilator takes on the time-consuming task of comparing the images in your iPhoto library using effective algorithms to make sure that no duplicate escapes. Duplicate Annihilator detects... Read more
BusyContacts 1.1.6 - Fast, efficient con...
BusyContacts is a contact manager for OS X that makes creating, finding, and managing contacts faster and more efficient. It brings to contact management the same power, flexibility, and sharing... Read more
MarsEdit 3.7.10 - Quick and convenient b...
MarsEdit is a blog editor for OS X that makes editing your blog like writing email, with spell-checking, drafts, multiple windows, and even AppleScript support. It works with with most blog services... Read more
BusyCal 3.1.4 - Powerful calendar app wi...
BusyCal is an award-winning desktop calendar that combines personal productivity features for individuals with powerful calendar sharing capabilities for families and workgroups. Its unique features... Read more
VirtualBox 5.1.14 - x86 virtualization s...
VirtualBox is a family of powerful x86 virtualization products for enterprise as well as home use. Not only is VirtualBox an extremely feature rich, high performance product for enterprise customers... Read more
Bookends 12.7.8 - Reference management a...
Bookends is a full-featured bibliography/reference and information-management system for students and professionals. Bookends uses the cloud to sync reference libraries on all the Macs you use.... Read more

Super Mario Run dashes onto Android in M...
Super Mario Run was one of the biggest mobile launches in 2016 before it was met with a lukewarm response by many. While the game itself plays a treat, it's pretty hard to swallow the steep price for the full game. With that said, Android users... | Read more »
WarFriends Beginner's Guide: How to...
Chillingo's new game, WarFriends, is finally available world wide, and so far it's a refreshing change from common mobile game trends. The game's a mix of tower defense, third person shooter, and collectible card game. There's a lot to unpack here... | Read more »
Super Gridland (Entertainment)
Super Gridland 1.0 Device: iOS Universal Category: Entertainment Price: $1.99, Version: 1.0 (iTunes) Description: Match. Build. Survive. "exquisitely tuned" - Rock Paper Shotgun No in-app purches, and no ads! | Read more »
Red's Kingdom (Games)
Red's Kingdom 1.0 Device: iOS Universal Category: Games Price: $4.99, Version: 1.0 (iTunes) Description: Mad King Mac has kidnapped your father and stolen your golden nut! Solve puzzles and battle goons as you explore and battle your... | Read more »
Turbo League Guide: How to tame the cont...
| Read more »
Fire Emblem: Heroes coming to Google Pla...
Nintendo gave us our first look at Fire Emblem: Heroes, the upcoming mobile Fire Emblem game the company hinted at last year. Revealed at the Fire Emblem Direct event held today, the game will condense the series' tactical RPG combat into bite-... | Read more »
ReSlice (Music)
ReSlice 1.0 Device: iOS Universal Category: Music Price: $9.99, Version: 1.0 (iTunes) Description: Audio Slice Machine Slice your audio samples with ReSlice and create flexible musical atoms which can be triggered by MIDI notes or... | Read more »
Stickman Surfer rides in with the tide t...
Stickson is back and this time he's taken up yet another extreme sport - surfing. Stickman Surfer is out this Thursday on both iOS and Android, so if you've been following the other Stickman adventures, you might be interested in picking this one... | Read more »
Z-Exemplar (Games)
Z-Exemplar 1.4 Device: iOS Universal Category: Games Price: $3.99, Version: 1.4 (iTunes) Description: | Read more »
5 dastardly difficult roguelikes like th...
Edmund McMillen's popular roguelike creation The Binding of Isaac: Rebirth has finally crawled onto mobile devices. It's a grotesque dual-stick shooter that tosses you into an endless, procedurally generated basement as you, the pitiable Isaac,... | Read more »

Price Scanner via MacPrices.net

Twelve South Releases RelaxedLeather Cases fo...
Inspired by the laid-back luxury of burnished leather boots and crafted in rich tones of taupe, herb and marsala, RelaxedLeather cases deliver smart, easy protection for the iPhone 7. Each genuine... Read more
Week’s Best Deal: New 2016 13-inch 2.0GHz Mac...
Amazon has the new 2016 13″ 2.0GHz non-Touch Bar MacBook Pros on sale for a limited time for $225 off MSRP including free shipping: - 13″ 2.0GHz MacBook Pro, Space Gray (MLL42LL/A): $1274.99 $225 off... Read more
Back in stock: Apple refurbished Mac minis fr...
Apple has Certified Refurbished Mac minis available starting at $419. Apple’s one-year warranty is included with each mini, and shipping is free: - 1.4GHz Mac mini: $419 $80 off MSRP - 2.6GHz Mac... Read more
Apple Ranked ‘Most Intimate Brand’
The top ranked ‘”intimate” brands continued to outperform the S&P and Fortune 500 indices in revenue and profit over the past 10 years, according to MBLM’s Brand Intimacy 2017 Report, the largest... Read more
B-Eng introduces SSD Health Check for Mac OS
Fehraltorf, Switzerland based independant Swiss company- B-Eng has announced the release and immediate availability of SSD Health Check 1.0, the company’s new hard drive utility for Mac OS X. As the... Read more
Apple’s Education discount saves up to $300 o...
Purchase a new Mac or iPad using Apple’s Education Store and take up to $300 off MSRP. All teachers, students, and staff of any educational institution qualify for the discount. Shipping is free: -... Read more
4-core 3.7GHz Mac Pro on sale for $2290, save...
Guitar Center has the 3.7GHz 4-core Mac Pro (MD253LL/A) on sale for $2289.97 including free shipping or free local store pickup (if available). Their price is a $710 savings over standard MSRP for... Read more
128GB Apple iPad Air 2, refurbished, availabl...
Apple has Certified Refurbished 128GB iPad Air 2s WiFis available for $419 including free shipping. That’s an $80 savings over standard MSRP for this model. A standard Apple one-year warranty is... Read more
13-inch 2.7GHz Retina MacBook Pro on sale for...
B&H Photo has the 2015 13″ 2.7GHz/128GB Retina Apple MacBook Pro on sale for $100 off MSRP. Shipping is free, and B&H charges NY tax only: - 13″ 2.7GHz/128GB Retina MacBook Pro (MF839LL/A): $... Read more
Laptop Market – Flight To Quality? – The ‘Boo...
Preliminary quarterly PC shipments data released by Gartner Inc. last week reveal an interesting disparity between sales performance of major name PC vendors as opposed to that of less well-known... Read more

Jobs Board

*Apple* macOS Systems Integration Administra...
…most exceptional support available in the industry. SCI is seeking an Junior Apple macOS systems integration administrator that will be responsible for providing Read more
*Apple* Retail - Multiple Positions- Deer Pa...
Job Description: Sales Specialist - Retail Customer Service and Sales Transform Apple Store visitors into loyal Apple customers. When customers enter the store, Read more
*Apple* Retail - Multiple Positions - Apple,...
Job Description: Sales Specialist - Retail Customer Service and Sales Transform Apple Store visitors into loyal Apple customers. When customers enter the store, Read more
*Apple* Technician - nfrastructure (United S...
Let’s Work Together Apple Technician This position is based in Portland, ME Life at nfrastructure At nfrastructure, we understand that our success results from our Read more
*Apple* Mobile Master - Best Buy (United Sta...
**467692BR** **Job Title:** Apple Mobile Master **Location Number:** 000602-Columbia MO-Store **Job Description:** **What does a Best Buy Apple Mobile Master Read more
All contents are Copyright 1984-2011 by Xplain Corporation. All rights reserved. Theme designed by Icreon.