diff for duplicates of <1465329394.22178.223.camel@linux.intel.com> diff --git a/a/1.txt b/N1/1.txt index 6f5b67c..a127291 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -14,10 +14,10 @@ On Tue, 2016-06-07 at 12:23 -0400, Johannes Weiner wrote: > > Johannes, > > > > It seems like you are saying that the shadow entry is also present -> > for anonymous pages that are swapped out. A But once a page is swapped +> > for anonymous pages that are swapped out. But once a page is swapped > > out, its entry is removed from the radix tree and we won't be able -> > to store the shadow page entry as for file mapped pageA -> > in __remove_mapping. A Or are you thinking of modifying +> > to store the shadow page entry as for file mapped page +> > in __remove_mapping. Or are you thinking of modifying > > the current code to keep the radix tree entry? I may be missing something > > so will appreciate if you can clarify. > Sorry if this was ambiguously phrased. @@ -27,9 +27,9 @@ On Tue, 2016-06-07 at 12:23 -0400, Johannes Weiner wrote: > refaults and push back against cache, whereas cache only pushes > against anon if the cache workingset is determined to fit into memory. -Thanks. That makes sense. A I wasn't sure before whether you intend +Thanks. That makes sense. I wasn't sure before whether you intend to have a re-fault distance to determine if a -faulted in anonymous page is in working set. A I see now that +faulted in anonymous page is in working set. I see now that you always consider it to be in working set. > @@ -39,9 +39,9 @@ you always consider it to be in working set. > counting IO cost from anonymous refaults and pressuring page cache. > > [ Although you can set the effective cost of these refaults to 0 -> A (swappiness = 200) and reduce effective cache to a minimum - -> A possibly to a level where LRU rotations consume most of it. -> A But yeah. ] +> (swappiness = 200) and reduce effective cache to a minimum - +> possibly to a level where LRU rotations consume most of it. +> But yeah. ] > > So the current code works well when we assume that cache workingsets > might exceed memory, but anonymous workingsets don't. @@ -56,7 +56,7 @@ you always consider it to be in working set. The direct mapped swap case is trickier as we need a method to gauge how often a page was accessed in place in swap, to decide if we need to -bring it back to RAM. A The accessed bit in pte only tells +bring it back to RAM. The accessed bit in pte only tells us if it has been accessed, but not the frequency. If we simply try to mitigate IO cost, we may just have pages migrated and @@ -72,9 +72,3 @@ Tim > But it would have to build on top of > this series here. These patches are about eliminating or mitigating IO > by swapping idle or colder anon pages when the cache is thrashing. - --- -To unsubscribe, send a message with 'unsubscribe linux-mm' in -the body to majordomo@kvack.org. For more info on Linux MM, -see: http://www.linux-mm.org/ . -Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a> diff --git a/a/content_digest b/N1/content_digest index cd2f9e1..b924367 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -33,10 +33,10 @@ "> > Johannes,\n" "> > \n" "> > It seems like you are saying that the shadow entry is also present\n" - "> > for anonymous pages that are swapped out. A But once a page is swapped\n" + "> > for anonymous pages that are swapped out. \302\240But once a page is swapped\n" "> > out, its entry is removed from the radix tree and we won't be able\n" - "> > to store the shadow page entry as for file mapped pageA \n" - "> > in __remove_mapping. A Or are you thinking of modifying\n" + "> > to store the shadow page entry as for file mapped page\302\240\n" + "> > in __remove_mapping. \302\240Or are you thinking of modifying\n" "> > the current code to keep the radix tree entry? I may be missing something\n" "> > so will appreciate if you can clarify.\n" "> Sorry if this was ambiguously phrased.\n" @@ -46,9 +46,9 @@ "> refaults and push back against cache, whereas cache only pushes\n" "> against anon if the cache workingset is determined to fit into memory.\n" "\n" - "Thanks. That makes sense. A I wasn't sure before whether you intend\n" + "Thanks. That makes sense. \302\240I wasn't sure before whether you intend\n" "to have a re-fault distance to determine if a\n" - "faulted in anonymous page is in working set. A I see now that\n" + "faulted in anonymous page is in working set. \302\240I see now that\n" "you always consider it to be in working set.\n" "\n" "> \n" @@ -58,9 +58,9 @@ "> counting IO cost from anonymous refaults and pressuring page cache.\n" "> \n" "> [ Although you can set the effective cost of these refaults to 0\n" - "> A (swappiness = 200) and reduce effective cache to a minimum -\n" - "> A possibly to a level where LRU rotations consume most of it.\n" - "> A But yeah. ]\n" + "> \302\240 (swappiness = 200) and reduce effective cache to a minimum -\n" + "> \302\240 possibly to a level where LRU rotations consume most of it.\n" + "> \302\240 But yeah. ]\n" "> \n" "> So the current code works well when we assume that cache workingsets\n" "> might exceed memory, but anonymous workingsets don't.\n" @@ -75,7 +75,7 @@ "\n" "The direct mapped swap case is trickier as we need a method to gauge how often\n" "a page was accessed in place in swap, to decide if we need to\n" - "bring it back to RAM. A The accessed bit in pte only tells\n" + "bring it back to RAM. \302\240The accessed bit in pte only tells\n" "us if it has been accessed, but not the frequency.\n" "\n" "If we simply try to mitigate IO cost, we may just have pages migrated and\n" @@ -90,12 +90,6 @@ "\n" "> But it would have to build on top of\n" "> this series here. These patches are about eliminating or mitigating IO\n" - "> by swapping idle or colder anon pages when the cache is thrashing.\n" - "\n" - "--\n" - "To unsubscribe, send a message with 'unsubscribe linux-mm' in\n" - "the body to majordomo@kvack.org. For more info on Linux MM,\n" - "see: http://www.linux-mm.org/ .\n" - "Don't email: <a href=mailto:\"dont@kvack.org\"> email@kvack.org </a>" + > by swapping idle or colder anon pages when the cache is thrashing. -685fdf60c5bf3e9115b9376e1510e2e27a3c4b6e2c52436fefc1898e21fdb136 +5f9c6e678653f55d537fffebbf1dc7d3ab289ef309dde4ca708bc920ca86c374
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.