From 0421f24ff1ffd85a1289045bf5dc91af54aa7238 Mon Sep 17 00:00:00 2001 From: Thomas Schwinge Date: Tue, 4 Aug 2009 22:57:50 +0200 Subject: open_issues/xen_crash_copy-size_le_page_size: IRC discussion. --- open_issues/xen_crash_copy-size_le_page_size.mdwn | 25 +++++++++++++++++++++++ 1 file changed, 25 insertions(+) (limited to 'open_issues') diff --git a/open_issues/xen_crash_copy-size_le_page_size.mdwn b/open_issues/xen_crash_copy-size_le_page_size.mdwn index 58784e96..0aa030fd 100644 --- a/open_issues/xen_crash_copy-size_le_page_size.mdwn +++ b/open_issues/xen_crash_copy-size_le_page_size.mdwn @@ -77,3 +77,28 @@ GDB on `mkfs.ext2`: #3 0x0805e87d in ext2fs_zero_blocks (fs=0x8082940, blk=524272, num=16, ret_blk=0x15ffb1c, ret_count=0x0) at ../../../git/lib/ext2fs/mkjournal.c:182 #4 0x0804ec56 in main (argc=131072, argv=0x80000) at ../../git/misc/mke2fs.c:2032 + +Discussion: + + I had a look at the code, but unfortunately don't really know + how this data transfers between Xen and the domU work. + Well, I know how it roughly works, but not the implementation + deatils. + well here it's not about the xen/domU transfers + it's about copying data to align it + i.e. when offset is not aligned, I need to copy it + Yes- + I was lazy, just implemented it for things smaller than a page + it just needs to be extended into copying several pages + youpi: Hmm, do we need to copy all the data to shift away the + offset or is there a better way? + the blkbackend needs data to be sector-aligned + just aligning on a page makes offset computation simpler + as it's rare that's not a problem + And a sector is the usual 512 bytes there, I assume? + But then we do need to copy all of it? + let me check + the sector is the granularity you can't go below + sector is the sector_size reported by the backend + but for sector_number and first/last_sect it's 512 + yes, that's weird -- cgit v1.2.3