Rev |
Age |
Author |
Path |
Log message |
Diff |
2144 |
6350 d 22 h |
jermar |
/trunk/kernel/arch/sparc64/src/mm/ |
Fix TSB size. |
|
2141 |
6351 d 21 h |
jermar |
/trunk/ |
The Ultimate Solution To Illegal Virtual Aliases.
It is better to avoid them completely than to fight them.
Switch the sparc64 port to 16K pages. The TLBs and TSBs
continue to operate with 8K pages only. Page tables and
other generic parts operate with 16K pages.
Because the MMU doesn't support 16K directly, each 16K
page is emulated by a pair of 8K pages. With 16K pages,
illegal aliases cannot be created in 16K D-cache. |
|
2134 |
6353 d 1 h |
jermar |
/trunk/kernel/ |
Reworked handling of illegal virtual aliases caused by frame reuse.
We moved the incomplete handling from backend's frame method to
backend's page_fault method. The page_fault method is the one that
can create an illegal alias if it writes the userspace frame using
kernel address with a different page color than the page to which is
this frame mapped in userspace. When we detect this, we do D-cache
shootdown on all processors (!!!).
If we add code that accesses userspace memory from kernel address
space, we will have to check for illegal virtual aliases at all such
places.
I tested this on a 4-way simulated E6500 and a real-world Ultra 5,
which has unfortunatelly only one processor.
This solves ticket #26. |
|
2089 |
6410 d 7 h |
decky |
/trunk/ |
huge type system cleanup
remove cyclical type dependencies across multiple header files
many minor coding style fixes |
|
2078 |
6417 d 4 h |
jermar |
/trunk/kernel/arch/sparc64/src/ |
When invalidating entire TLBs on sparc64, make sure to also invalidate any (locked) global entries.
This fixes Ticket #21.
Fix a comment in start.S stating that the kernel installs a global entry for itself. All entries installed by the kernel
are local to some memory context. |
|
2076 |
6419 d 8 h |
jermar |
/trunk/kernel/ |
Beat the implicit illegal virtual alias created by reusing userspace frames.
In the anonymous and ELF backends, if the architecture has virtually indexed D-cache,
selectively flush cachelines belonging to the frame being freed.
This fixes Ticket #20. |
|
2071 |
6421 d 0 h |
jermar |
/trunk/ |
(c) versus (C) |
|
2048 |
6453 d 1 h |
jermar |
/trunk/ |
Formatting and indentation changes. |
|
2016 |
6458 d 7 h |
jermar |
/trunk/kernel/arch/sparc64/src/mm/ |
The D-cache line size is actually 32 bytes on UltraSPARC II, IIi (and also IIe). |
|
2015 |
6460 d 0 h |
jermar |
/trunk/ |
Rework support for virtually indexed cache.
Instead of repeatedly flushing the data cache, which was a huge overkill, refuse to create an illegal address alias
in the kernel (again) and allocate appropriate page color in userspace instead. Extend the detection also to
SYS_PHYSMEM_MAP syscall.
Add support for tracking physical memory areas mappable by SYS_PHYSMEM_MAP.
Lots of coding style changes. |
|
2010 |
6464 d 23 h |
jermar |
/trunk/kernel/arch/sparc64/src/mm/ |
Fix comment. |
|
2009 |
6464 d 23 h |
jermar |
/trunk/kernel/ |
Initial support for handling illegal virtual aliases on sparc64. |
|
2008 |
6467 d 0 h |
jermar |
/trunk/kernel/arch/sparc64/ |
Add dcache_flush() function that flushes D-Cache on sparc64. |
|
2001 |
6470 d 21 h |
jermar |
/trunk/kernel/ |
Indentation and comment fixes. |
|
1996 |
6477 d 2 h |
jermar |
/trunk/kernel/ |
Add option to compile the sparc64 kernel without the TTE_CV bit support.
The bit is not used by default now.
Enabling it may theoretically lead to physical memory inconsistencies until code that
mitigates the problem is written. |
|
1988 |
6478 d 22 h |
jermar |
/trunk/kernel/arch/sparc64/src/mm/ |
Update comment and move initialization code away from application processor codepath. |
|
1987 |
6479 d 1 h |
jermar |
/trunk/kernel/arch/sparc64/src/ |
Take the possible difference between kernel and physical address into account in TSB and userspace window buffer
allocation and deallocation code. |
|
1983 |
6481 d 4 h |
jermar |
/trunk/kernel/arch/sparc64/src/mm/ |
sparc64 work:
- fix computation of the frame allocator configuration frame
- mark the first physical frame unavailable, no matter what is its address
Because of that, we can now boot past the frame_arch_init() on the Ultra 60 now. |
|
1982 |
6481 d 8 h |
jermar |
/trunk/ |
Modify the sparc64 startup code to not cause MMU traps before it takes over the TLB and
the trap table. Fix several PA2KA and KA2PA omittions or errors. Fix configuration to pass
the DEFS variable along. |
|
1960 |
6503 d 0 h |
jermar |
/trunk/kernel/arch/sparc64/src/mm/ |
Fix the following bug:
Ticket #3 data_access_exception after killing task on sparc64 |
|