2/24/2016 CS152, Spring 2016
CS 152 Computer Architecture and Engineering
Lecture 9 - Virtual Memory
Dr. George MichelogiannakisEECS, University of California at Berkeley
CRD, Lawrence Berkeley National Laboratory
http://inst.eecs.berkeley.edu/~cs152
2/24/2016 CS152, Spring 2016
Administrivia
PS2 and lab 2 due Wednesday next week (March 2nd)
Quiz 2 is on Monday March 7th
2
2/24/2016 CS152, Spring 2016
Last time in Lecture 9
Protection and translation required for multiprogramming– Base and bounds was early simple scheme
Page-based translation and protection avoids need for memory compaction, easy allocation by OS
– But need to indirect in large page table on every access
Can use multi-level page table to hold translation/protection information, but implies multiple memory accesses per reference
Can use “translation lookaside buffer” (TLB) to cache address translations (sometimes known as address translation cache)
– Still have to walk page tables on TLB miss, can be hardware or software talk
Virtual memory uses DRAM as a “cache” of disk memory, allows very cheap main memory
3
2/24/2016 CS152, Spring 2016
Question of the Day
How would you design a TLB prefetcher?
4
2/24/2016 CS152, Spring 2016
Memory Management
Can separate into orthogonal functions:– Translation (mapping of virtual address to physical address)
– Protection (permission to access word in memory)
– Virtual memory (transparent extension of memory space using slower disk or flash storage)
But most modern systems provide support for all the above functions with a single page-based system
5
2/24/2016 CS152, Spring 2016
Modern Virtual Memory SystemsIllusion of a large, private, uniform store
6
Protection & Privacyseveral users, each with their private address space and one or more shared address spaces
page table <-> name space
Demand PagingProvides the ability to run programs larger than the primary memory
Hides differences in machine configurations
The price is address translation on each memory reference
OS
useri
PrimaryMemory
SecondaryStorage
VA PAmapping
TLB
2/24/2016 CS152, Spring 2016
Hierarchical Page Table
7
Level 1 Page Table
Level 2
Page Tables
Data Pages
page in primary memory page in secondary memory
Root of CurrentPage Table
p1
offset
p2
Virtual Address
(Processor Register)
PTE of a nonexistent page
p1 p2 offset
01112212231
10-bitL1 index
10-bit L2 index
Ph
ysic
al M
emo
ry
2/24/2016 CS152, Spring 2016
Page-Based Virtual-Memory Machine(Hardware Page-Table Walk)
Assumes page tables held in untranslated physical memory
8
PCInst. TLB
Inst. Cache D Decode E M
Data Cache W+
Page Fault?
Protection violation?
Page Fault?
Protection violation?
Data TLB
Main Memory (DRAM)
Memory ControllerPhysical Address
Physical Address
Physical Address
Physical Address
Page-Table Base Register
Virtual Address Physical
Address
Virtual Address
Hardware Page Table Walker
Miss? Miss?
2/24/2016 CS152, Spring 2016
Address Translation:putting it all together
9
Virtual Address
TLBLookup
Page TableWalk
Update TLBPage Fault(OS loads page)
ProtectionCheck
PhysicalAddress(to cache)
miss hit
the page is
memory memory denied permitted
ProtectionFault
hardwarehardware or softwaresoftware
SEGFAULTWhere?
2/24/2016 CS152, Spring 2016
Page Fault Handler
When the referenced page is not in DRAM:– The missing page is located (or created)
– It is brought in from disk, and page table is updated
• Another job may be run on the CPU while the first job waits for the requested page to be read from disk
• Could be different thread of the same address space
– If no free pages are left, a page is swapped out
• Pseudo-LRU replacement policy, implemented in software
Since it takes a long time to transfer a page (msecs), page faults are handled completely in software by the OS
– Untranslated addressing mode is essential to allow kernel to access page tables
10
2/24/2016 CS152, Spring 2016
Handling VM-related exceptions
Handling a TLB miss needs a hardware or software mechanism to refill TLB
Handling a page fault (e.g., page is on disk) needs a restartableexception so software handler can resume after retrieving page
– Precise exceptions are easy to restart
– Can be imprecise but restartable, but this complicates OS software
Handling protection violation may abort process– But often handled the same as a page fault
11
PCInstTLB
Inst. Cache D Decode E M
Data TLB
Data Cache W+
TLB miss? Page Fault?Protection violation?
TLB miss? Page Fault?Protection violation?
2/24/2016 CS152, Spring 2016
Address Translation in CPU Pipeline
Need to cope with additional latency of TLB:– slow down the clock?
– pipeline the TLB and cache access?
– virtual address caches
– parallel TLB/cache access
12
PCInstTLB
Inst. Cache D Decode E M
Data TLB
Data Cache W+
TLB miss? Page Fault?Protection violation?
TLB miss? Page Fault?Protection violation?
2/24/2016 CS152, Spring 2016
Virtual-Address Caches
one-step process in case of a hit (+)
cache needs to be flushed on a context switch unless address space identifiers (ASIDs) included in tags (-)
aliasing problems due to the sharing of pages (-)
maintaining cache coherence (-) (see later in course)
13
CPUPhysicalCache
TLBPrimaryMemory
VA PA PA
Alternative: place the cache before the TLB
CPUVA (StrongARM)Virtual
Cache
PATLB
PrimaryMemory
VA
2/24/2016 CS152, Spring 2016
Virtually Addressed Cache(Virtual Index/Virtual Tag)
14
PC
Inst. TLB
Inst. Cache D Decode E M
Data Cache W+
Data TLB
Main Memory (DRAM)
Memory Controller
Physical Address
Instruction dataPhysical Address
Physical Address
Page-Table BaseRegister
Virtual Address
Virtual Address
Hardware Page Table Walker
Miss?Miss?
Translate on miss
2/24/2016 CS152, Spring 2016
Question
What is the benefit of separating instruction and data TLBs?
And the downside?
15
2/24/2016 CS152, Spring 2016
Aliasing in Virtual-Address Caches
16
VA1
VA2
Page Table
Data Pages
PA
VA1
VA2
1st Copy of Data at PA
2nd Copy of Data at PA
Tag Data
Two virtual pages share one physical page
Virtual cache can have two copies of same physical data. Writes to one copy not visible to reads of other!
General Solution: Prevent aliases coexisting in cache
Software (i.e., OS) solution for direct-mapped cache
VAs of shared pages must agree in cache index bits; this ensures all VAs accessing same PA will conflict in direct-mapped cache (early SPARCs)
2/24/2016 CS152, Spring 2016
Question
How can we parallelize TLB and cache access?
17
Cache
TLB
2/24/2016 CS152, Spring 2016
Concurrent Access to TLB & Cache(Virtual Index/Physical Tag)
18
Index L is available without consulting the TLB=>cache and TLB accesses can begin simultaneously!
Tag comparison is made after both accesses are completed
Cases: L + b = k, L + b < k, L + b > k
VPN L b
TLBDirect-map Cache
2L blocks2b-byte blockPPN Page Offset
=hit?
DataPhysical TagTag
VA
PA
VirtualIndex
k
2/24/2016 CS152, Spring 2016
Virtual-Index Physical-Tag Caches: Associative Organization
19
How does this scheme scale to larger caches?
VPN a L = k-b b
TLBDirect-map
2L blocks
PPN Page Offset
=hit?
Data
Phy.Tag
Tag
VA
PA
VirtualIndex
kDirect-map
2L blocks
2a
=2a
After the PPN is known, 2a physical tags are compared
What about fully-set associative?
2/24/2016 CS152, Spring 2016
Question
Does virtual tag, physical index make sense?
20
2/24/2016 CS152, Spring 2016
Concurrent Access to TLB & Large L1The problem with L1 > Page size
21
Can VA1 and VA2 both map to PA ?
VPN a Page Offset b
TLB
PPN Page Offset b
Tag
VA
PA
Virtual IndexL1 PA cacheDirect-map
= hit?
PPNa Data
PPNa Data
VA1
VA2
If they differ in the lower ‘a’ bits alone, and share a physical page.
2/24/2016 CS152, Spring 2016
A solution via Second Level Cache
22
Usually a common L2 cache backs up both Instruction and Data L1 caches
L2 is “inclusive” of both Instruction and Data caches• Inclusive means L2 has copy of any line in either L1
CPU
L1 Data Cache
L1 Instruction
Cache Unified L2 Cache
RF Memory
Memory
Memory
Memory
2/24/2016 CS152, Spring 2016
Anti-Aliasing Using L2 [MIPS R10000,1996]
Suppose VA1 and VA2 both map to PA and
VA1 is already in L1, L2 (VA1 VA2)
After VA2 is resolved to PA, a collision will be detected in L2 (L2 is in physical space).
VA1 will be purged from L1 and L2, and VA2 will be loaded no aliasing !
23
VPN a Page Offset b
TLB
PPN Page Offset b
Tag
VA
PA
Virtual IndexL1 PA cacheDirect-map
= hit?
PPNa Data
PPNa Data
VA1
VA2
Direct-Mapped L2
PA a1 Data
PPN
into L2 tag
2/24/2016 CS152, Spring 2016
Anti-Aliasing using L2 for a VirtuallyAddressed L1
24
VPN Page Offset b
TLB
PPN Page Offset b
Tag
VA
PA
VirtualIndex & Tag
PhysicalIndex & Tag
L1 VA Cache
L2 PA Cache L2 “contains” L1
PA VA1 Data
VA1 Data
VA2 Data
“Virtual Tag”
Physically-addressed L2 can also be used to avoid aliases in virtually-addressed L1
2/24/2016 CS152, Spring 2016
Atlas Revisited
One PAR for each physical page
PAR’s contain the VPN’s of the pages resident in primary memory
Advantage: The size is proportional to the size of the primary memory
What is the disadvantage ?
How does this work with caches?– Is a TLB necessary?
25
VPN
PAR’s
PPN
2/24/2016 CS152, Spring 2016
Hashed Page Table:Approximating Associative Addressing
Hashed Page Table is typically 2 to 3 times larger than the number of PPN’s to reduce collision probability. What does this mean?
It can also contain DPN’s for some non-resident pages (not common)
If a translation cannot be resolved in this table then the software consults a data structure that has an entry for every existing page (e.g., full page table)
26
hashOffset
Base of Table
+PA of PTE
PrimaryMemory
VPN PID PPN
Page Table
VPN d Virtual Address
VPN PID DPN
VPN PID
PID
2/24/2016 CS152, Spring 2016
Power PC: Hashed Page Table
27
Each hash table slot has 8 PTE's <VPN,PPN> that are searched sequentially
If the first hash slot fails, an alternate hash function is used to look in another slot
All these steps are done in hardware!
Hashed Table is typically 2 to 3 times larger than the number of physical pages
The full backup Page Table is managed in software
Base of Table
hashOffset +
PA of Slot
PrimaryMemory
VPN PPN
Page TableVPN d 80-bit VA
VPN
2/24/2016 CS152, Spring 2016
VM features track historical uses: Bare machine, only physical addresses
– One program owned entire machine
Batch-style multiprogramming– Several programs sharing CPU while waiting for I/O– Base & bound: translation and protection between programs (supports
swapping entire programs but not demand-paged virtual memory)– Problem with external fragmentation (holes in memory), needed occasional
memory defragmentation as new jobs arrived
Time sharing– More interactive programs, waiting for user. Also, more jobs/second.– Motivated move to fixed-size page translation and protection, no external
fragmentation (but now internal fragmentation, wasted bytes in page)– Motivated adoption of virtual memory to allow more jobs to share limited
physical memory resources while holding working set in memory
Virtual Machine Monitors– Run multiple operating systems on one machine– Idea from 1970s IBM mainframes, now common on laptops
• e.g., run Windows on top of Mac OS X– Hardware support for two levels of translation/protection
• Guest OS virtual -> Guest OS physical -> Host machine physical
28
2/24/2016 CS152, Spring 2016
Virtual Memory Use Today - 1
Servers/desktops/laptops/smartphones have full demand-paged virtual memory
– Portability between machines with different memory sizes
– Protection between multiple users or multiple tasks
– Share small physical memory among active tasks
– Simplifies implementation of some OS features
Vector supercomputers have translation and protection butrarely complete demand-paging
(Older Crays: base&bound, Japanese & Cray X1/X2: pages)– Don’t waste expensive CPU time thrashing to disk (make jobs fit in memory)
– Mostly run in batch mode (run set of jobs that fits in memory)
– Difficult to implement restartable vector instructions
29
2/24/2016 CS152, Spring 2016
Virtual Memory Use Today - 2
Most embedded processors and DSPs provide physical addressing only
– Can’t afford area/speed/power budget for virtual memory support
– Often there is no secondary storage to swap to!
– Programs custom written for particular memory configuration in product
– Difficult to implement restartable instructions for exposed architectures
30
2/24/2016 CS152, Spring 2016
Question of the Day
How would you design a TLB prefetcher?
31
2/24/2016 CS152, Spring 2016
Acknowledgements
These slides contain material developed and copyright by:– Arvind (MIT)
– Krste Asanovic (MIT/UCB)
– Joel Emer (Intel/MIT)
– James Hoe (CMU)
– John Kubiatowicz (UCB)
– David Patterson (UCB)
MIT material derived from course 6.823
UCB material derived from course CS252
32