-/*
- * void MapUserAddressSpaceInit(void)
- *
- * Initialized anything we need to in order to map user address space slices into
- * the kernel. Primarily used for copy in/out.
- *
- * Currently we only support one 512MB slot for this purpose. There are two special
- * mappings defined for the purpose: the special pmap nest, and linkage mapping.
- *
- * The special pmap nest (which is allocated in this function) is used as a place holder
- * in the kernel's pmap search list. It is 512MB long and covers the address range
- * starting at copyIOaddr. It points to no actual memory and when the fault handler
- * hits in it, it knows to look in the per_proc and start using the linkage
- * mapping contained therin.
- *
- * The linkage mapping is used to glue the user address space slice into the
- * kernel. It contains the relocation information used to transform the faulting
- * kernel address into the user address space. It also provides the link to the
- * user's pmap. This is pointed to by the per_proc and is switched in and out
- * whenever there is a context switch.
- *
- */
-
-void MapUserAddressSpaceInit(void) {
-
- addr64_t colladdr;
- int nlists, asize;
- mapping *mp;
-
- nlists = mapSetLists(kernel_pmap); /* Set number of lists this will be on */
-
- mp = mapping_alloc(nlists); /* Get a spare mapping block */
-
- mp->mpFlags = 0x01000000 |mpNest | mpSpecial | nlists; /* Set the flags. Make sure busy count is 1 */
- mp->mpSpace = kernel_pmap->space; /* Set the address space/pmap lookup ID */
- mp->mpBSize = 1; /* Set the size to 2 segments */
- mp->mpPte = 0; /* Means nothing */
- mp->mpPAddr = 0; /* Means nothing */
- mp->mpVAddr = copyIOaddr; /* Set the address range we cover */
- mp->mpNestReloc = 0; /* Means nothing */
-
- colladdr = hw_add_map(kernel_pmap, mp); /* Go add the mapping to the pmap */
-
- if(colladdr) { /* Did it collide? */
- panic("MapUserAddressSpaceInit: MapUserAddressSpace range already mapped\n");
- }
-
- return;
-}
-
-/*
- * addr64_t MapUserAddressSpace(vm_map_t map, vm_offset_t va, size)
- *
- * map = the vm_map that we are mapping into the kernel
- * va = start of the address range we are mapping
- * size = size of the range. No greater than 256MB and not 0.
- * Note that we do not test validty, we chose to trust our fellows...
- *
- * Maps a slice of a user address space into a predefined kernel range
- * on a per-thread basis. In the future, the restriction of a predefined
- * range will be loosened.
- *
- * Builds the proper linkage map to map the user range
- * We will round this down to the previous segment boundary and calculate
- * the relocation to the kernel slot
- *
- * We always make a segment table entry here if we need to. This is mainly because of
- * copyin/out and if we don't, there will be multiple segment faults for
- * each system call. I have seen upwards of 30000 per second.
- *
- * We do check, however, to see if the slice is already mapped and if so,
- * we just exit. This is done for performance reasons. It was found that
- * there was a considerable boost in copyin/out performance if we did not
- * invalidate the segment at ReleaseUserAddressSpace time, so we dumped the
- * restriction that you had to bracket MapUserAddressSpace. Further, there
- * is a yet further boost if you didn't need to map it each time. The theory
- * behind this is that many times copies are to or from the same segment and
- * done multiple times within the same system call. To take advantage of that,
- * we check cioSpace and cioRelo to see if we've already got it.
- *
- * We also need to half-invalidate the slice when we context switch or go
- * back to user state. A half-invalidate does not clear the actual mapping,
- * but it does force the MapUserAddressSpace function to reload the segment
- * register/SLBE. If this is not done, we can end up some pretty severe
- * performance penalties. If we map a slice, and the cached space/relocation is
- * the same, we won't reload the segment registers. Howver, since we ran someone else,
- * our SR is cleared and we will take a fault. This is reasonable if we block
- * while copying (e.g., we took a page fault), but it is not reasonable when we
- * just start. For this reason, we half-invalidate to make sure that the SR is
- * explicitly reloaded.
- *
- * Note that we do not go to the trouble of making a pmap segment cache
- * entry for these guys because they are very short term -- 99.99% of the time
- * they will be unmapped before the next context switch.
- *
- */
projects / apple / xnu.git / blobdiff