From 23602d0b8098930a0c61556070d59706114580e7 Mon Sep 17 00:00:00 2001 From: mcsim Date: Wed, 16 Nov 2011 09:42:49 +0100 Subject: --- user/Maksym_Planeta.mdwn | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 6f4bda77..fc1b3784 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -55,12 +55,16 @@ since this parameter is unused TODO: Find out why and correct this. -#Translators vs FUSE +#Chalanges + +Translators vs FUSE: [[What can a translator do that FUSE can't?|http://lists.gnu.org/archive/html/bug-hurd/2010-07/msg00061.html]] [[Re: Hurd translators on FUSE|http://lists.gnu.org/archive/html/l4-hurd/2009-09/msg00146.html]] +[[Example of sane utilization of tmpfs|http://habrahabr.ru/blogs/gdev/131043/]] (Russian) + #Debugging To debug tmpfs, using libraries from "$PWD"/lib and trace rpc: -- cgit v1.2.3 From 15ec61e43d74d52d55944e54027b0f967f63faa4 Mon Sep 17 00:00:00 2001 From: mcsim Date: Wed, 16 Nov 2011 09:56:05 +0100 Subject: --- user/Maksym_Planeta.mdwn | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index fc1b3784..5bea897c 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -63,7 +63,7 @@ Translators vs FUSE: [[Re: Hurd translators on FUSE|http://lists.gnu.org/archive/html/l4-hurd/2009-09/msg00146.html]] -[[Example of sane utilization of tmpfs|http://habrahabr.ru/blogs/gdev/131043/]] (Russian) +[[Example of sane utilization of filesystem stored in RAM|http://habrahabr.ru/blogs/gdev/131043/]] (Russian). Author of this article copied some resources of game "World of Tanks" to RAM-drive and game started load much faster. Although he used Windows in this article, this could be good example of benefits, which filesystem, stored in RAM, could give. #Debugging -- cgit v1.2.3 From 02ae187badbf2978609e73002b05462bafa04245 Mon Sep 17 00:00:00 2001 From: mcsim Date: Sat, 19 Nov 2011 09:04:47 +0100 Subject: --- user/Maksym_Planeta.mdwn | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 5bea897c..f79b1864 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -42,6 +42,8 @@ since this parameter is unused 8.11.11 Symlinks work. + [[Patch by Ben Asselstine|http://lists.gnu.org/archive/html/bug-hurd/2005-07/msg00217.html]] + 4. Control of used space by tmpfs. TODO: Make tmpfs use not more space than it was allowed. @@ -97,11 +99,8 @@ For debugging ext2fs: and its target (relative to its directory) is foo/bar (which would mean /tmp/foo/foo/bar in canonical form) (10:29:42) braunr: youpi: tschwinge: what did ludovic achieve ? (10:30:06) tschwinge: mcsim: As Richard says, symlink targets are always relative to the directory they're contained in. - (10:30:51) tschwinge: braunr: This Hydra/Nix (I wtill mix it all up) thing is kind of a package managing system. - (10:31:17) tschwinge: He has written scripts for bootstrapping a Hurd toolchain. (10:31:26) braunr: oh ok (10:31:27) mcsim: so, if I want to create link in cd, first I need to cd there? - (10:31:28) tschwinge: And then uses that to build a whole bootable image. (10:31:36) mcsim: in foo* (10:31:36) braunr: mcsim: just provide the right paths (10:32:11) braunr: $ touch foo/bar -- cgit v1.2.3 From 62656d38beb34d1f12a6d89f190e8111032f2f6c Mon Sep 17 00:00:00 2001 From: mcsim Date: Sat, 19 Nov 2011 18:47:18 +0100 Subject: --- user/Maksym_Planeta.mdwn | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index f79b1864..e8bea72a 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -42,7 +42,7 @@ since this parameter is unused 8.11.11 Symlinks work. - [[Patch by Ben Asselstine|http://lists.gnu.org/archive/html/bug-hurd/2005-07/msg00217.html]] + [[Patch by Ben Asselstine.|http://lists.gnu.org/archive/html/bug-hurd/2005-07/msg00217.html]] 4. Control of used space by tmpfs. @@ -57,6 +57,8 @@ since this parameter is unused TODO: Find out why and correct this. +7. Passive translator doesn't work + #Chalanges Translators vs FUSE: -- cgit v1.2.3 From 916e352dd38d1a4b106f3eb4a47dcb4a1faaa77d Mon Sep 17 00:00:00 2001 From: mcsim Date: Sat, 19 Nov 2011 22:15:30 +0100 Subject: --- user/Maksym_Planeta.mdwn | 1 + 1 file changed, 1 insertion(+) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index e8bea72a..89e51dc8 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -87,6 +87,7 @@ For debugging ext2fs: 1. What are sequence numbers? What are they used for? 2. Is there any way to debug mach-defpager? When I set breakpoint to any function in it, pager never breaks. +3. Is it normal that defpager panics and breaks on any wrong data? #Links -- cgit v1.2.3 From d1ffe67c957a61d0e75fb290ea5f628d7c825ffe Mon Sep 17 00:00:00 2001 From: mcsim Date: Sun, 20 Nov 2011 11:33:13 +0100 Subject: --- user/Maksym_Planeta.mdwn | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 89e51dc8..02316dc5 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -42,7 +42,7 @@ since this parameter is unused 8.11.11 Symlinks work. - [[Patch by Ben Asselstine.|http://lists.gnu.org/archive/html/bug-hurd/2005-07/msg00217.html]] + [[Patch by Ben Asselstine.|http://thread.gmane.org/gmane.os.hurd.bugs/11829/focus=12098]] 4. Control of used space by tmpfs. -- cgit v1.2.3 From b659a0d4f515363676d43a36625a9774a5590a97 Mon Sep 17 00:00:00 2001 From: mcsim Date: Mon, 21 Nov 2011 17:06:24 +0100 Subject: --- user/Maksym_Planeta.mdwn | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 02316dc5..79e77ee1 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -23,20 +23,29 @@ My vision of the problem in short: when user tries to access to the memory backe ## Steps -1. Find out what causes crashes in tmpfs with defpager +### Find out what causes crashes in tmpfs with defpager [[http://www.gnu.org/s/hurd/hurd/translator/tmpfs/notes_various.html]] TODO: Consider deleting of parameter "port" in function mach-defpager/default_pager.c:pager_port_list_insert since this parameter is unused -2. Write own pager +Probably pager_request shouldn't be stored because request may arrive from different kernels (or from kernel and translator), so this parameter doesn't have any sense. + +In d_p_set_size memory_object_lock_request is used for manipulating of object size. There is a function memory_object_data_unavailable, which can increase space, provided for memory object. TODO: Consider using it. + +### Write own pager 6.11.11 Reading/writing for files that fit in vm_page_size works 7.11.11 Works for any size. -3. Make links work + TODO: During execution tmpfs hangs in random places. The most possible is variant is deadlocks, + because nothing was undertaken for thread safety. + + TODO: Make tmpfs use not more space than it was allowed. + +### Make links work Symlinks behavior: [[links|http://www.bddebian.com:8888/~hurd-web/user/Maksym_Planeta/#links81111]] @@ -44,20 +53,11 @@ since this parameter is unused [[Patch by Ben Asselstine.|http://thread.gmane.org/gmane.os.hurd.bugs/11829/focus=12098]] -4. Control of used space by tmpfs. - - TODO: Make tmpfs use not more space than it was allowed. - -5. Thread safety. - - TODO: During execution tmpfs hangs in random places. The most possible is variant is deadlocks, - because nothing was undertaken for thread safety. - -6. After sometime of inactivity tmpfs exits. +### After sometime of inactivity tmpfs exits. TODO: Find out why and correct this. -7. Passive translator doesn't work +### Passive translator doesn't work #Chalanges -- cgit v1.2.3 From 24b93f3cb0f3843e0687ad656ad0b97ffb65ec5c Mon Sep 17 00:00:00 2001 From: mcsim Date: Wed, 23 Nov 2011 07:37:26 +0100 Subject: --- user/Maksym_Planeta.mdwn | 33 ++++++++++++++++++++++++++++++++- 1 file changed, 32 insertions(+), 1 deletion(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 79e77ee1..5d6025f6 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -32,7 +32,7 @@ since this parameter is unused Probably pager_request shouldn't be stored because request may arrive from different kernels (or from kernel and translator), so this parameter doesn't have any sense. -In d_p_set_size memory_object_lock_request is used for manipulating of object size. There is a function memory_object_data_unavailable, which can increase space, provided for memory object. TODO: Consider using it. +22.11.11 Reading/writing for any size works, [[this|http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00127.html]] works, but fsx test fails ([[see|http://www.bddebian.com:8888/~hurd-web/user/Maksym_Planeta/#fsx_fail2211]]). ### Write own pager @@ -153,3 +153,34 @@ For debugging ext2fs: on system performance (15:56:54) antrik: both can be avoided by just passing a real anonymous memory object, i.e. one provided by the defpager (15:57:07) antrik: only problem is that the current defpager implementation can't really handle that... + +#Pastes + +## fsx test on 22.11.11 + $ ~/src/fsx/fsx -W -R -t 4096 -w 4096 -r 4096 bar + mapped writes DISABLED + truncating to largest ever: 0x32000 + truncating to largest ever: 0x39000 + READ BAD DATA: offset = 0x16000, size = 0xd9a0 + OFFSET GOOD BAD RANGE + 0x1f000 0x0000 0x01a0 0x 2d14 + operation# (mod 256) for the bad data may be 1 + LOG DUMP (16 total operations): + 1(1 mod 256): WRITE 0x1f000 thru 0x21d28 (0x2d29 bytes) HOLE ***WWWW + 2(2 mod 256): WRITE 0x10000 thru 0x15bfe (0x5bff bytes) + 3(3 mod 256): READ 0x1d000 thru 0x21668 (0x4669 bytes) + 4(4 mod 256): READ 0xa000 thru 0x16a59 (0xca5a bytes) + 5(5 mod 256): READ 0x8000 thru 0x8f2c (0xf2d bytes) + 6(6 mod 256): READ 0xa000 thru 0x17fe8 (0xdfe9 bytes) + 7(7 mod 256): READ 0x1b000 thru 0x20f33 (0x5f34 bytes) + 8(8 mod 256): READ 0x15000 thru 0x1c05b (0x705c bytes) + 9(9 mod 256): TRUNCATE UP from 0x21d29 to 0x32000 + 10(10 mod 256): READ 0x3000 thru 0x5431 (0x2432 bytes) + 11(11 mod 256): WRITE 0x29000 thru 0x34745 (0xb746 bytes) EXTEND + 12(12 mod 256): TRUNCATE DOWN from 0x34746 to 0x19000 ******WWWW + 13(13 mod 256): READ 0x14000 thru 0x186d8 (0x46d9 bytes) + 14(14 mod 256): TRUNCATE UP from 0x19000 to 0x39000 ******WWWW + 15(15 mod 256): WRITE 0x28000 thru 0x3548c (0xd48d bytes) + 16(16 mod 256): READ 0x16000 thru 0x2399f (0xd9a0 bytes) ***RRRR*** + Correct content saved for comparison + (maybe hexdump "bar" vs "bar.fsxgood") -- cgit v1.2.3 From 1f1604898a0d54a348628fa44ed71674fc4ae78a Mon Sep 17 00:00:00 2001 From: Thomas Schwinge Date: Wed, 23 Nov 2011 20:25:58 +0100 Subject: user/Maksym_Planeta: Some comments. --- user/Maksym_Planeta.mdwn | 31 +++++++++++++++++++++++++++++++ 1 file changed, 31 insertions(+) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 5d6025f6..d6a847b5 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -43,8 +43,15 @@ Probably pager_request shouldn't be stored because request may arrive from diffe TODO: During execution tmpfs hangs in random places. The most possible is variant is deadlocks, because nothing was undertaken for thread safety. +> Is this only a problem in your new *own pager* implementation, or also in the +> existing tmpfs implementation? --[[tschwinge]] + TODO: Make tmpfs use not more space than it was allowed. +> Is this only a problem in your new *own pager* implementation, or also in the +> existing tmpfs implementation? It is supposed to already obey this limit. +> --[[tschwinge]] + ### Make links work Symlinks behavior: [[links|http://www.bddebian.com:8888/~hurd-web/user/Maksym_Planeta/#links81111]] @@ -57,8 +64,21 @@ Probably pager_request shouldn't be stored because request may arrive from diffe TODO: Find out why and correct this. +> This may perhaps be the standard translator +> shutdown-after-a-period-of-inactivity functionality? This is of course not +> valid in the tmpfs case, as all its state (file system content) is not stored +> on any non-volatile backend. Can this auto-shutdown functionality be +> disabled (in [[hurd/libdiskfs]], perhaps? See `libdiskfs/init-first.c`, the +> call to `ports_manage_port_operations_multithread`, the `global_timeout` +> paramenter (see the [[hurd/reference_manual]]) (here: `server_timeout`). +> Probably this should be set to `0` in tmpfs to disable timing out? +> --[[tschwinge]] + ### Passive translator doesn't work +> Must be some bug: `diskfs_set_translator` is implemented in `node.c`, so this +> is supposed to work. --[[tschwinge]] + #Chalanges Translators vs FUSE: @@ -86,9 +106,20 @@ For debugging ext2fs: #Questions 1. What are sequence numbers? What are they used for? + +> See [[microkernel/mach/ipc/sequence_numbering]]. --[[tschwinge]] + 2. Is there any way to debug mach-defpager? When I set breakpoint to any function in it, pager never breaks. + +> Is that still an unresolved problem? If not, what was the problem? --[[tschwinge]] + 3. Is it normal that defpager panics and breaks on any wrong data? +> No server must ever panic (or reach any other invalid state), whichever input +> data it receives from untrusted sources (which may include every possible +> source). All input data must be checked for validity before use; anything +> else is a bug. --[[tschwinge]] + #Links 1. [[Cthreads manuals|http://www.cc.gatech.edu/classes/cs6432_99_winter/threads_man/]] -- cgit v1.2.3 From 5a99ae9aa6599ae949ee3652ed42c4c4f50cffe7 Mon Sep 17 00:00:00 2001 From: mcsim Date: Wed, 23 Nov 2011 20:39:12 +0100 Subject: --- user/Maksym_Planeta.mdwn | 7 ------- 1 file changed, 7 deletions(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index d6a847b5..a5299f53 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -43,15 +43,8 @@ Probably pager_request shouldn't be stored because request may arrive from diffe TODO: During execution tmpfs hangs in random places. The most possible is variant is deadlocks, because nothing was undertaken for thread safety. -> Is this only a problem in your new *own pager* implementation, or also in the -> existing tmpfs implementation? --[[tschwinge]] - TODO: Make tmpfs use not more space than it was allowed. -> Is this only a problem in your new *own pager* implementation, or also in the -> existing tmpfs implementation? It is supposed to already obey this limit. -> --[[tschwinge]] - ### Make links work Symlinks behavior: [[links|http://www.bddebian.com:8888/~hurd-web/user/Maksym_Planeta/#links81111]] -- cgit v1.2.3 From b5e48137f69574cc6396d9c0d143a4ce83f06378 Mon Sep 17 00:00:00 2001 From: mcsim Date: Wed, 23 Nov 2011 22:15:01 +0100 Subject: --- user/Maksym_Planeta.mdwn | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index a5299f53..67ef2137 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -47,11 +47,11 @@ Probably pager_request shouldn't be stored because request may arrive from diffe ### Make links work - Symlinks behavior: [[links|http://www.bddebian.com:8888/~hurd-web/user/Maksym_Planeta/#links81111]] +Symlinks behavior: [[links|http://www.bddebian.com:8888/~hurd-web/user/Maksym_Planeta/#links81111]] - 8.11.11 Symlinks work. +8.11.11 Symlinks work. - [[Patch by Ben Asselstine.|http://thread.gmane.org/gmane.os.hurd.bugs/11829/focus=12098]] +[[Patch by Ben Asselstine.|http://thread.gmane.org/gmane.os.hurd.bugs/11829/focus=12098]] ### After sometime of inactivity tmpfs exits. -- cgit v1.2.3 From 7c71157c69d769c007c02fc991dad2c1f519b5c5 Mon Sep 17 00:00:00 2001 From: mcsim Date: Thu, 24 Nov 2011 18:05:29 +0100 Subject: --- user/Maksym_Planeta.mdwn | 2 ++ 1 file changed, 2 insertions(+) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 67ef2137..3c330388 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -21,6 +21,8 @@ License|/fdl]]."]]"""]] My vision of the problem in short: when user tries to access to the memory backed by tmpfs first time, kernel asks defpager to initialize memory object and sets it's request port to some value. But sometimes user directly accesses to defpager and supplies it own request port. So defpager should handle different request port, what causes errors. +TODO: pager_extend always frees old pagemap and allocates new. Consider if it is necessary. + ## Steps ### Find out what causes crashes in tmpfs with defpager -- cgit v1.2.3 From e6aabed5cb7a1d5f5d472a36c9f4c5802ae95fc8 Mon Sep 17 00:00:00 2001 From: mcsim Date: Thu, 24 Nov 2011 23:20:31 +0100 Subject: --- user/Maksym_Planeta.mdwn | 31 ++++++++++++++++++++++++++++++- 1 file changed, 30 insertions(+), 1 deletion(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 3c330388..1c589ed8 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -7,7 +7,7 @@ any later version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license is included in the section entitled [[GNU Free Documentation License|/fdl]]."]]"""]] - + [[!toc]] #Notes on tmpfs @@ -36,6 +36,17 @@ Probably pager_request shouldn't be stored because request may arrive from diffe 22.11.11 Reading/writing for any size works, [[this|http://lists.gnu.org/archive/html/bug-hurd/2011-11/msg00127.html]] works, but fsx test fails ([[see|http://www.bddebian.com:8888/~hurd-web/user/Maksym_Planeta/#fsx_fail2211]]). +24.11.11 The problem with fsx. + +Here are follow operations: + +1. Write some data at address 0x100 with size 0x20 +2. Truncate file to size 0x80. First written data should be lost. +3. Write some data at address 0x200 size of 0x20. By this operation file size is increased up to 0x220. +4. Read data at address 0x110. Fsx expects here zeros, but in fact here is data, that was written at step 1. + +When fsx tries to read data kernel calls pager with seqno_memory_object_data_request, and pager returns on step 4 zeros either with memory_object_data_provided or memory_object_data_unavailable. Before this, in default_pager_set_size memory_object_lock_request called to flush any kernel caches, that could hold data to be truncated. When I set offset to 0 and size to limit in memory_object_lock_request it appeared another error ([[see|http://www.bddebian.com:8888/~hurd-web/user/Maksym_Planeta/#fsx_fail2411]]). Both these behaviors appear to be quite strange for me. It is quite late now, so i put these notes to not forget this and went sleep. Continue tomorrow. + ### Write own pager 6.11.11 Reading/writing for files that fit in vm_page_size works @@ -210,3 +221,21 @@ For debugging ext2fs: 16(16 mod 256): READ 0x16000 thru 0x2399f (0xd9a0 bytes) ***RRRR*** Correct content saved for comparison (maybe hexdump "bar" vs "bar.fsxgood") + +## fsx test on 24.11.11 + + $ ~/src/fsx/fsx bar + truncating to largest ever: 0x13e76 + READ BAD DATA: offset = 0x1f62e, size = 0x2152 + OFFSET GOOD BAD RANGE + 0x1f62e 0x0206 0x0000 0x 213e + operation# (mod 256) for the bad data unknown, check HOLE and EXTEND ops + LOG DUMP (6 total operations): + 1(1 mod 256): TRUNCATE UP from 0x0 to 0x13e76 + 2(2 mod 256): WRITE 0x17098 thru 0x26857 (0xf7c0 bytes) HOLE ***WWWW + 3(3 mod 256): READ 0xc73e thru 0x1b801 (0xf0c4 bytes) + 4(4 mod 256): MAPWRITE 0x32e00 thru 0x331fc (0x3fd bytes) + 5(5 mod 256): MAPWRITE 0x7ac1 thru 0x11029 (0x9569 bytes) + 6(6 mod 256): READ 0x1f62e thru 0x2177f (0x2152 bytes) ***RRRR*** + Correct content saved for comparison + (maybe hexdump "bar" vs "bar.fsxgood") -- cgit v1.2.3 From 05e7070db686aa8cb196dec23204cfc66734fbca Mon Sep 17 00:00:00 2001 From: mcsim Date: Mon, 5 Dec 2011 23:31:25 +0100 Subject: --- user/Maksym_Planeta.mdwn | 2 ++ 1 file changed, 2 insertions(+) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 1c589ed8..59b562c9 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -47,6 +47,8 @@ Here are follow operations: When fsx tries to read data kernel calls pager with seqno_memory_object_data_request, and pager returns on step 4 zeros either with memory_object_data_provided or memory_object_data_unavailable. Before this, in default_pager_set_size memory_object_lock_request called to flush any kernel caches, that could hold data to be truncated. When I set offset to 0 and size to limit in memory_object_lock_request it appeared another error ([[see|http://www.bddebian.com:8888/~hurd-web/user/Maksym_Planeta/#fsx_fail2411]]). Both these behaviors appear to be quite strange for me. It is quite late now, so i put these notes to not forget this and went sleep. Continue tomorrow. +5.12.11 Here is a problem with writing by address, which was freed already. It happens in function dealloc_direct in macros invalidate_block. This function is called from pager_truncate in branch when condition "if (!INDIRECT_PAGEMAP(old_size))" is true. But I didn't find why reference to freed object is kept. As workaround we can reduce kalloc_max in hurd/mach-defpager/kalloc.c to 3 to make allocator use vm_allocate always. The drawback is that allocator will allocate only multiple of vm_page_size, but this is temporary tradeoff. Till now fsx reaches operation number 14277. + ### Write own pager 6.11.11 Reading/writing for files that fit in vm_page_size works -- cgit v1.2.3 From f07fa1683332bcb6219b2cbec94b6725aa4f4a3c Mon Sep 17 00:00:00 2001 From: mcsim Date: Tue, 6 Dec 2011 01:12:17 +0100 Subject: --- user/Maksym_Planeta.mdwn | 2 ++ 1 file changed, 2 insertions(+) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 59b562c9..0f04d058 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -49,6 +49,8 @@ When fsx tries to read data kernel calls pager with seqno_memory_object_data_req 5.12.11 Here is a problem with writing by address, which was freed already. It happens in function dealloc_direct in macros invalidate_block. This function is called from pager_truncate in branch when condition "if (!INDIRECT_PAGEMAP(old_size))" is true. But I didn't find why reference to freed object is kept. As workaround we can reduce kalloc_max in hurd/mach-defpager/kalloc.c to 3 to make allocator use vm_allocate always. The drawback is that allocator will allocate only multiple of vm_page_size, but this is temporary tradeoff. Till now fsx reaches operation number 14277. +6.12.11 fsx works quite long and doesn't interrupt. + ### Write own pager 6.11.11 Reading/writing for files that fit in vm_page_size works -- cgit v1.2.3 From 0dc055beb675bee44dbde65f0dcd41234e010782 Mon Sep 17 00:00:00 2001 From: mcsim Date: Tue, 6 Dec 2011 01:31:50 +0100 Subject: --- user/Maksym_Planeta.mdwn | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 0f04d058..dc1e9d0e 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -49,7 +49,7 @@ When fsx tries to read data kernel calls pager with seqno_memory_object_data_req 5.12.11 Here is a problem with writing by address, which was freed already. It happens in function dealloc_direct in macros invalidate_block. This function is called from pager_truncate in branch when condition "if (!INDIRECT_PAGEMAP(old_size))" is true. But I didn't find why reference to freed object is kept. As workaround we can reduce kalloc_max in hurd/mach-defpager/kalloc.c to 3 to make allocator use vm_allocate always. The drawback is that allocator will allocate only multiple of vm_page_size, but this is temporary tradeoff. Till now fsx reaches operation number 14277. -6.12.11 fsx works quite long and doesn't interrupt. +6.12.11 fsx works quite long and doesn't interrupt. I've stopped at 124784. ### Write own pager -- cgit v1.2.3 From a527960787dfb358662626dcfef64442e1b76841 Mon Sep 17 00:00:00 2001 From: mcsim Date: Tue, 6 Dec 2011 18:49:37 +0100 Subject: --- user/Maksym_Planeta.mdwn | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index dc1e9d0e..71546caf 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -49,7 +49,7 @@ When fsx tries to read data kernel calls pager with seqno_memory_object_data_req 5.12.11 Here is a problem with writing by address, which was freed already. It happens in function dealloc_direct in macros invalidate_block. This function is called from pager_truncate in branch when condition "if (!INDIRECT_PAGEMAP(old_size))" is true. But I didn't find why reference to freed object is kept. As workaround we can reduce kalloc_max in hurd/mach-defpager/kalloc.c to 3 to make allocator use vm_allocate always. The drawback is that allocator will allocate only multiple of vm_page_size, but this is temporary tradeoff. Till now fsx reaches operation number 14277. -6.12.11 fsx works quite long and doesn't interrupt. I've stopped at 124784. +6.12.11 fsx works quite long and doesn't interrupt. I've stopped at 124784. Continued. It broke at 181091. ### Write own pager -- cgit v1.2.3 From 2030f5b1a7a5cc01def00ec4b78a4f0e22678884 Mon Sep 17 00:00:00 2001 From: mcsim Date: Wed, 7 Dec 2011 20:55:23 +0100 Subject: --- user/Maksym_Planeta.mdwn | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 71546caf..f7ef04c1 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -11,6 +11,19 @@ License|/fdl]]."]]"""]] [[!toc]] #Notes on tmpfs +## Current state + +7.12.11 + +There left 4 bugs, I found: +1. Passive translator doesn't work. (Didn't investigate yet) + +2. After sometime of inactivity tmpfs exits. (Didn't investigate yet) + +3. Writing by freed address somewhere. As workaround I set kalloc_max in mach-defpager/kalloc.c to 3, so vm_allocate is always used. + +4. Fsx still breaks with 11 from 12 seeds, I tested the problem was in first 4 Kb. In 12th case problem was in range between 4 Kb and 8 Kb. I find this quite odd. + ## mach-defpager [[defpager|http://www.bddebian.com:8888/~hurd-web/user/Maksym_Planeta/#defpager81111]] -- cgit v1.2.3 From f2637f96e5c944ec6d8dd964e0a764301f264ddf Mon Sep 17 00:00:00 2001 From: mcsim Date: Wed, 7 Dec 2011 21:06:45 +0100 Subject: --- user/Maksym_Planeta.mdwn | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index f7ef04c1..6458e114 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -16,9 +16,9 @@ License|/fdl]]."]]"""]] 7.12.11 There left 4 bugs, I found: -1. Passive translator doesn't work. (Didn't investigate yet) +1. Passive translator doesn't work. (Haven't investigate yet) -2. After sometime of inactivity tmpfs exits. (Didn't investigate yet) +2. After sometime of inactivity tmpfs exits. (Haven't investigate yet) 3. Writing by freed address somewhere. As workaround I set kalloc_max in mach-defpager/kalloc.c to 3, so vm_allocate is always used. -- cgit v1.2.3 From 8332e4d0dfc2f5a4c99e8e291ce571d6f4c86497 Mon Sep 17 00:00:00 2001 From: mcsim Date: Wed, 7 Dec 2011 21:15:22 +0100 Subject: --- user/Maksym_Planeta.mdwn | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 6458e114..77545bed 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -22,7 +22,7 @@ There left 4 bugs, I found: 3. Writing by freed address somewhere. As workaround I set kalloc_max in mach-defpager/kalloc.c to 3, so vm_allocate is always used. -4. Fsx still breaks with 11 from 12 seeds, I tested the problem was in first 4 Kb. In 12th case problem was in range between 4 Kb and 8 Kb. I find this quite odd. +4. Fsx still breaks with 11 from 12 seeds, I tested the problem was in first 4 Kb. In 12th case problem was in range between 4 Kb and 8 Kb. I find this quite odd. Usual amount of operations before fsx breaks passes 100 000. ## mach-defpager -- cgit v1.2.3 From 9e573d1215611b3faa3be6082dc7753f6b524491 Mon Sep 17 00:00:00 2001 From: mcsim Date: Wed, 7 Dec 2011 22:10:23 +0100 Subject: --- user/Maksym_Planeta.mdwn | 1 + 1 file changed, 1 insertion(+) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 77545bed..979c707a 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -16,6 +16,7 @@ License|/fdl]]."]]"""]] 7.12.11 There left 4 bugs, I found: + 1. Passive translator doesn't work. (Haven't investigate yet) 2. After sometime of inactivity tmpfs exits. (Haven't investigate yet) -- cgit v1.2.3 From a9a61a2b87240c558c1c18b699b89f02d7396ced Mon Sep 17 00:00:00 2001 From: mcsim Date: Fri, 9 Dec 2011 07:27:31 +0100 Subject: --- user/Maksym_Planeta.mdwn | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'user') diff --git a/user/Maksym_Planeta.mdwn b/user/Maksym_Planeta.mdwn index 979c707a..502f3e20 100644 --- a/user/Maksym_Planeta.mdwn +++ b/user/Maksym_Planeta.mdwn @@ -19,7 +19,7 @@ There left 4 bugs, I found: 1. Passive translator doesn't work. (Haven't investigate yet) -2. After sometime of inactivity tmpfs exits. (Haven't investigate yet) +2. After sometime of inactivity tmpfs exits. (8.12.11, Fixed) 3. Writing by freed address somewhere. As workaround I set kalloc_max in mach-defpager/kalloc.c to 3, so vm_allocate is always used. -- cgit v1.2.3 From 6c057cff39ff782e9155c07eee44884cd9c48c9c Mon Sep 17 00:00:00 2001 From: musial Date: Mon, 12 Dec 2011 18:23:45 +0100 Subject: --- user/musial.mdwn | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) (limited to 'user') diff --git a/user/musial.mdwn b/user/musial.mdwn index 24a526be..d9aaa599 100644 --- a/user/musial.mdwn +++ b/user/musial.mdwn @@ -8,12 +8,8 @@ Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license is included in the section entitled [[GNU Free Documentation License|/fdl]]."]]"""]] -Robert Musial +Robert Musial - Cleveland, OH -Cleveland, OH +musial/at/gnu/dot/org -musial [at] gnu [dot] org - -http://musial.sdf.org - -http://github.com/musial +http://musial.dyndns.info:8080/ -- cgit v1.2.3