diff for duplicates of <1427149549.4770.240.camel@kernel.crashing.org> diff --git a/a/1.txt b/N1/1.txt index 45f321a..5223642 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -10,7 +10,7 @@ On Mon, 2015-03-23 at 12:54 -0400, Sowmini Varadhan wrote: > - Caller of iommu_tbl_range_alloc() can do the flush_all if they > see start <= end for the one single pool > - lose the other ->flush_all invocation (i.e., the one that is -> done when iommu_area_alloc() fails for pass == 0, and we reset +> done when iommu_area_alloc() fails for pass = 0, and we reset > start to 0 to roll-back) > > that would avoid the need for any iommu_tbl_ops in my patch-set. @@ -21,7 +21,7 @@ kaboom. However if that's the only callback left, pass it as an argument. > But it would imply that you would still take the perf hit for the roll-back -> if we failed the pass == 0 iteration through iommu_area_alloc(). +> if we failed the pass = 0 iteration through iommu_area_alloc(). > Perhaps this is an acceptable compromise in favor of cleaner code > (again, assuming that current/future archs will all follow the HV > based design). diff --git a/a/content_digest b/N1/content_digest index 744a172..3247bb9 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -5,7 +5,7 @@ "ref\020150323165406.GG14061@oracle.com\0" "From\0Benjamin Herrenschmidt <benh@kernel.crashing.org>\0" "Subject\0Re: Generic IOMMU pooled allocator\0" - "Date\0Tue, 24 Mar 2015 09:25:49 +1100\0" + "Date\0Mon, 23 Mar 2015 22:25:49 +0000\0" "To\0Sowmini Varadhan <sowmini.varadhan@oracle.com>\0" "Cc\0aik@au1.ibm.com" aik@ozlabs.ru @@ -28,7 +28,7 @@ "> - Caller of iommu_tbl_range_alloc() can do the flush_all if they \n" "> see start <= end for the one single pool \n" "> - lose the other ->flush_all invocation (i.e., the one that is\n" - "> done when iommu_area_alloc() fails for pass == 0, and we reset\n" + "> done when iommu_area_alloc() fails for pass = 0, and we reset\n" "> start to 0 to roll-back)\n" "> \n" "> that would avoid the need for any iommu_tbl_ops in my patch-set.\n" @@ -39,7 +39,7 @@ "argument.\n" "\n" "> But it would imply that you would still take the perf hit for the roll-back\n" - "> if we failed the pass == 0 iteration through iommu_area_alloc().\n" + "> if we failed the pass = 0 iteration through iommu_area_alloc().\n" "> Perhaps this is an acceptable compromise in favor of cleaner code\n" "> (again, assuming that current/future archs will all follow the HV\n" "> based design).\n" @@ -47,4 +47,4 @@ "> --Sowmini\n" > -53e77cc4d70a3ed8f5c5b07e31928824b1c02697bdb843b1ed536e00a4d8730e +d57a6644c6a8f71ec330ef88b36d8830a7ac3cac996441b08c8afdd722fedf6b
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.