From: BlaisorBlade <bla...@ya...> - 2004-04-01 17:44:24
Attachments:
Abstract_hostfs.patch
|
Alle 04:17, marted=EC 30 marzo 2004, Jeff Dike ha scritto: > ti...@un... said: > > Is there any clever way to reset the root password for users, aside > > from deleting the cow? "linux single" is fine, but aside from that? > > I'm finishing up a new hostfs-like filesystem which I want to become the > preferred mechanism to boot UMLs because of its ability to mmap data from > the host, eliminating duplicate copies of file data in memory. Al Viro said you that ubd=3Dmmap won't work, right? Would you give me the m= agic=20 words to search on the LKML or forward the messages to the -devel list? Also, I've started writing a little patch for 2.6 to hostfs to generalize i= t=20 like you said (which I attached). It is just 2-3 hours of work, since I had= =20 some doubt: do you like adding a hostfs_user_operations stored into=20 super_block->s_fs_info and calling all the functions of hostfs_user.c=20 indirectly (the trivial and long part I did not do yet)? Would it be useful for your "humfs", as it seems to me? =2D-=20 Paolo Giarrusso, aka Blaisorblade Linux registered user n. 292729 |
From: BlaisorBlade <bla...@ya...> - 2004-04-11 16:25:52
|
Alle 04:17, marted=EC 30 marzo 2004, Jeff Dike ha scritto: > ti...@un... said: > > Is there any clever way to reset the root password for users, aside > > from deleting the cow? "linux single" is fine, but aside from that? > > I'm finishing up a new hostfs-like filesystem which I want to become the > preferred mechanism to boot UMLs because of its ability to mmap data from > the host, eliminating duplicate copies of file data in memory. > > A side-effect of this is that UML filesystems will become visible on the > host, so password changing would be a matter of chrooting to the filesyst= em > and running passwd. > > Currently, there is no COW functionality, but I plan to add file-level > COWing with block-level COWing possibly following later. Have you heard of the discussion about COWlinks? I.e. creating a Copy - On = =2D=20 Write functionality for hard links? Or do you want to rewrite files one pag= e=20 at a time (i.e. if I change one byte inside a file, only the "page"=20 containing that byte is copied and changed)? http://lwn.net/Articles/77215/ =2D-=20 Paolo Giarrusso, aka Blaisorblade Linux registered user n. 292729 |
From: Jeff D. <jd...@ad...> - 2004-04-13 04:19:34
|
bla...@ya... said: > Have you heard of the discussion about COWlinks? I.e. creating a Copy > - On - Write functionality for hard links? Or do you want to rewrite > files one page at a time (i.e. if I change one byte inside a file, > only the "page" containing that byte is copied and changed)? COWlinks sound like they would be useful for doing file-level COWing for UML. I'm considering that, plus block-level COWing within a file, using the standard COW format we have today. This would be necessary if we do block devices by loop-mounting a rootfs from humfs and using that as the root device. Jeff |
From: roland <for...@gm...> - 2004-05-26 23:50:27
|
http://lwn.net/Articles/83790/ http://wohnheim.fh-wedel.de/~joern/cowlink/ ----- Original Message ----- From: "BlaisorBlade" <bla...@ya...> To: "Jeff Dike" <jd...@ad...>; "tim doyle" <ti...@un...> Cc: "UML" <use...@li...>; <use...@li...> Sent: Sunday, April 11, 2004 4:51 PM Subject: [uml-devel] Re: [uml-user] one drawback to using cows commercially... Alle 04:17, martedì 30 marzo 2004, Jeff Dike ha scritto: > ti...@un... said: > > Is there any clever way to reset the root password for users, aside > > from deleting the cow? "linux single" is fine, but aside from that? > > I'm finishing up a new hostfs-like filesystem which I want to become the > preferred mechanism to boot UMLs because of its ability to mmap data from > the host, eliminating duplicate copies of file data in memory. > > A side-effect of this is that UML filesystems will become visible on the > host, so password changing would be a matter of chrooting to the filesystem > and running passwd. > > Currently, there is no COW functionality, but I plan to add file-level > COWing with block-level COWing possibly following later. Have you heard of the discussion about COWlinks? I.e. creating a Copy - On - Write functionality for hard links? Or do you want to rewrite files one page at a time (i.e. if I change one byte inside a file, only the "page" containing that byte is copied and changed)? http://lwn.net/Articles/77215/ -- Paolo Giarrusso, aka Blaisorblade Linux registered user n. 292729 ------------------------------------------------------- This SF.Net email is sponsored by: IBM Linux Tutorials Free Linux tutorial presented by Daniel Robbins, President and CEO of GenToo technologies. Learn everything from fundamentals to system administration.http://ads.osdn.com/?ad_id70&alloc_id638&op=ick _______________________________________________ User-mode-linux-devel mailing list Use...@li... https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel |
From: Jeff D. <jd...@ad...> - 2004-04-01 19:18:19
|
On Thu, Apr 01, 2004 at 06:53:43PM +0100, BlaisorBlade wrote: > Al Viro said you that ubd=mmap won't work, right? Would you give me the magic > words to search on the LKML or forward the messages to the -devel list? It was on IRC. > Also, I've started writing a little patch for 2.6 to hostfs to generalize it > like you said (which I attached). It is just 2-3 hours of work, since I had > some doubt: do you like adding a hostfs_user_operations stored into > super_block->s_fs_info and calling all the functions of hostfs_user.c > indirectly (the trivial and long part I did not do yet)? > > Would it be useful for your "humfs", as it seems to me? Yup, and I've already done this, with allowing any number of userspace plugins to the hostfs kernel interface, calling the ops through the operations structure, and changing hostfs and and humfs to use it. I have humfs booting UML now, so it's close to being releasable. Jeff |
From: BlaisorBlade <bla...@ya...> - 2004-04-09 17:44:45
|
Alle 20:40, luned=EC 5 aprile 2004, hai scritto: > BTW, if you do IRC, you will have a better idea of what exactly I'm doing > if you hang on the UML IRC channel - #uml on oftc.net. I'm there quite > often, and I tend to mumble about what I'm working on at the moment. > > Also, would you be interested in being able to put downloads on the UML > site rather than having them on user-linux-linux.org, which isn't as > visible? Sorry for answering so late, but I've been away for a week, for theoretical= =20 studies in computer science. For now I must decrease my UML activity since= =20 I'll have to solve a very hard test soon; and I cannot do that myself. However, especially for the second question, the answer is "yes". At least for the 2.4.25 and 2.6.4 Skas patches, which are VERY important fo= r=20 users. Especially because there are 3 bugs in most SKAS patches floating=20 around: =2D the SMP bug we discussed about some time ago, fixed in my last release; =2D this change (the one in the 2nd one) from Ingo Molnar that everyone els= e=20 missed: =2Dstatic int write_ldt(void __user * ptr, unsigned long bytecount, int old= mode) +static int write_ldt(struct mm_struct * mm, void __user * ptr, unsigned lo= ng=20 bytecount, int oldmode) { =2D struct mm_struct * mm =3D current->mm; __u32 entry_1, entry_2, *lp; int error; struct user_desc ldt_info; @@ -195,7 +197,7 @@ down(&mm->context.sem); if (ldt_info.entry_number >=3D mm->context.size) { ///// ---------------> NO ONE NOTICED this! =2D error =3D alloc_ldt(¤t->mm->context, =20 ldt_info.entry_number+1, 1); + error =3D alloc_ldt(&mm->context, ldt_info.entry_number+1, = 1); if (error < 0) goto out_unlock; } =2D and finally this bug he fixed in ldt.c =2D if (reload) { + if (reload && (¤t->active_mm->context =3D=3D pc)) { They are perfectly, with only one issue: since there is CONFIG_PROC_MM in t= he=20 headers, which is created by make oldconfig, but you don't have a=20 user-visible option, user often miscompile the kernel (which then creates a= =20 "Checking for Skas3 patch ... yes Checking for /proc/mm ... no"). Also, I see no reason for which the bug with switch_mm() that involved the= =20 Skas patch in 2.6 should not happen also in 2.4; in 2.4.25 we have this=20 check, which has never been triggered (include/asm-i386/mmu_context.h): if(cpu_tlbstate[cpu].active_mm !=3D next) out_of_line_bug(); So, maybe the 2.4 Skas patch could need the same change the 2.6 has in=20 include/asm-i386/mmu_context.h. However, please review and upload them by yourself. I'm actually busy at th= e=20 moment - VERY busy. Bye =2D-=20 Paolo Giarrusso, aka Blaisorblade Linux registered user n. 292729 |