sanchayanmaity/minix
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
minix/servers/vfs/device.c

1228 lines
35 KiB
C
Raw Normal View History

Initial revision
2005-04-21 16:53:53 +02:00
/* When a needed block is not in the cache, it must be fetched from the disk.
* Special character files also require I/O. The routines for these are here.
*
* The entry points in this file are:
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
* dev_open: open a character device
* dev_reopen: reopen a character device after a driver crash
* dev_close: close a character device
* bdev_open: open a block device
* bdev_close: close a block device
Initial revision
2005-04-21 16:53:53 +02:00
* dev_io: FS does a read or write on a device
FS understanding revives and selects as alerts (new notifies)
2005-07-27 15:08:52 +02:00
* dev_status: FS processes callback request alert
Initial revision
2005-04-21 16:53:53 +02:00
* gen_opcl: generic call to a task to perform an open/close
* gen_io: generic call to a task to perform an I/O operation
* no_dev: open/close processing for devices that don't exist
FS: . unmap device drivers from dmap when PM signals they are dead . new null-io function (no_dev_io) to fill in for io functions of unmapped drivers . driver (process number) of unmapped drivers is NONE instead of 0 (a valid process number) IS: . print mutable flag of dmap table too FS changes require sync() to be done 'manually' (currently by reboot/shutdown) at shutdown time; could be caught by SIGTERM in the future.
2005-10-05 17:38:15 +02:00
* no_dev_io: i/o processing for devices that don't exist
Initial revision
2005-04-21 16:53:53 +02:00
* tty_opcl: perform tty-specific processing for open/close
* ctty_opcl: perform controlling-tty-specific processing for open/close
* ctty_io: perform controlling-tty-specific processing for I/O
* do_ioctl: perform the IOCTL system call
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
* do_setsid: perform the SETSID system call (FS side)
Initial revision
2005-04-21 16:53:53 +02:00
*/
#include "fs.h"
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
#include <string.h>
Initial revision
2005-04-21 16:53:53 +02:00
#include <fcntl.h>
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
#include <assert.h>
VFS: bugfixes for handling block-special files: - on driver restarts, reopen devices on a per-file basis, not per-mount - do not assume that there is just one vnode per block-special device - update block-special files in the uncommon mounting success paths, too - upon mount, sync but also invalidate affected buffers on the root FS - upon unmount, check whether a vnode is in use before updating it
2011-03-25 11:56:43 +01:00
#include <sys/stat.h>
Initial revision
2005-04-21 16:53:53 +02:00
#include <minix/callnr.h>
#include <minix/com.h>
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
#include <minix/endpoint.h>
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
#include <minix/ioctl.h>
First cut at 64-bit file offsets in block devices for mkfs/fsck.
2006-11-27 15:21:43 +01:00
#include <minix/u64.h>
Initial revision
2005-04-21 16:53:53 +02:00
#include "file.h"
#include "fproc.h"
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
#include <minix/vfsif.h>
#include "vnode.h"
#include "vmnt.h"
Initial revision
2005-04-21 16:53:53 +02:00
#include "param.h"
kernel and servers send diagnostic messages to IS; IS sends them to TTY and the new log driver if enabled. new usyslogd is started from /usr/etc/rc. New device created by MAKEDEV.sh. /var/log created by etc/mtree/minix.tree (on root for now). Made select() slightly more generic, with less code duplication.
2005-07-08 19:30:01 +02:00
#define ELEMENTS(a) (sizeof(a)/sizeof((a)[0]))
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
FORWARD _PROTOTYPE( int safe_io_conversion, (endpoint_t, cp_grant_id_t *,
int *, cp_grant_id_t *, int,
endpoint_t *, void **, int *,
vir_bytes, u32_t *) );
3rd fix for improper behaviour on crashing driver with safe_io_conversion
2006-06-29 16:24:07 +02:00
FORWARD _PROTOTYPE( void safe_io_cleanup, (cp_grant_id_t, cp_grant_id_t *,
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
int) );
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
FORWARD _PROTOTYPE( void restart_reopen, (int maj) );
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
kernel and servers send diagnostic messages to IS; IS sends them to TTY and the new log driver if enabled. new usyslogd is started from /usr/etc/rc. New device created by MAKEDEV.sh. /var/log created by etc/mtree/minix.tree (on root for now). Made select() slightly more generic, with less code duplication.
2005-07-08 19:30:01 +02:00
extern int dmap_size;
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
PRIVATE int dummyproc;
Initial revision
2005-04-21 16:53:53 +02:00
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
Initial revision
2005-04-21 16:53:53 +02:00
/*===========================================================================*
* dev_open *
*===========================================================================*/
Remove the types Dev_t, _mnx_Gui, _mnx_Uid, and similar. Use ANSI-style function declarations where necessary.
2010-04-13 12:58:41 +02:00
PUBLIC int dev_open(
dev_t dev, /* device to open */
int proc, /* process to open for */
int flags /* mode bits and flags */
)
Initial revision
2005-04-21 16:53:53 +02:00
{
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
/* Open a character device. */
Initial revision
2005-04-21 16:53:53 +02:00
int major, r;
struct dmap *dp;
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
/* Determine the major device number so as to call the device class specific
Initial revision
2005-04-21 16:53:53 +02:00
* open/close routine. (This is the only routine that must check the
* device number for being in range. All others can trust this check.)
*/
major = (dev >> MAJOR) & BYTE;
Removed debug dumps from the PM and FS servers. The dumps are now done by the IS servers, which obtains a copy of the data through the getsysinfo() system call. CTRL-F1 now is a special TTY key to shows function key mappings.
2005-06-07 16:43:35 +02:00
if (major >= NR_DEVICES) major = 0;
Initial revision
2005-04-21 16:53:53 +02:00
dp = &dmap[major];
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (dp->dmap_driver == NONE) return(ENXIO);
Initial revision
2005-04-21 16:53:53 +02:00
r = (*dp->dmap_opcl)(DEV_OPEN, dev, proc, flags);
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
return(r);
}
/*===========================================================================*
* dev_reopen *
*===========================================================================*/
Remove the types Dev_t, _mnx_Gui, _mnx_Uid, and similar. Use ANSI-style function declarations where necessary.
2010-04-13 12:58:41 +02:00
PUBLIC int dev_reopen(
dev_t dev, /* device to open */
int filp_no, /* filp to reopen for */
int flags /* mode bits and flags */
)
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
{
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
/* Reopen a character device after a failing device driver. */
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
int major, r;
struct dmap *dp;
/* Determine the major device number call the device class specific
* open/close routine. (This is the only routine that must check the
* device number for being in range. All others can trust this check.)
*/
major = (dev >> MAJOR) & BYTE;
if (major >= NR_DEVICES) major = 0;
dp = &dmap[major];
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (dp->dmap_driver == NONE) return(ENXIO);
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
r = (*dp->dmap_opcl)(DEV_REOPEN, dev, filp_no, flags);
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
if (r == OK) panic("OK on reopen from: %d", dp->dmap_driver);
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (r == SUSPEND) r = OK;
Initial revision
2005-04-21 16:53:53 +02:00
return(r);
}
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
Initial revision
2005-04-21 16:53:53 +02:00
/*===========================================================================*
* dev_close *
*===========================================================================*/
Remove the types Dev_t, _mnx_Gui, _mnx_Uid, and similar. Use ANSI-style function declarations where necessary.
2010-04-13 12:58:41 +02:00
PUBLIC int dev_close(
dev_t dev, /* device to close */
int filp_no
)
Initial revision
2005-04-21 16:53:53 +02:00
{
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
/* Close a character device. */
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
int r;
Don't panic on missing drivers.
2005-10-21 21:31:38 +02:00
/* See if driver is roughly valid. */
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (dmap[(dev >> MAJOR)].dmap_driver == NONE) return(ENXIO);
r = (*dmap[(dev >> MAJOR) & BYTE].dmap_opcl)(DEV_CLOSE, dev, filp_no, 0);
return(r);
Initial revision
2005-04-21 16:53:53 +02:00
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
/*===========================================================================*
* bdev_open *
*===========================================================================*/
PUBLIC int bdev_open(dev_t dev, int access)
{
/* Open a block device. */
int major;
major = major(dev);
if (major < 0 || major >= NR_DEVICES) return(ENXIO);
if (dmap[major].dmap_driver == NONE) return(ENXIO);
return (*dmap[major].dmap_opcl)(BDEV_OPEN, dev, 0, access);
}
/*===========================================================================*
* bdev_close *
*===========================================================================*/
PUBLIC int bdev_close(dev_t dev)
{
/* Close a block device. */
int major;
major = major(dev);
if (major < 0 || major >= NR_DEVICES) return(ENXIO);
if (dmap[major].dmap_driver == NONE) return(ENXIO);
return (*dmap[major].dmap_opcl)(BDEV_CLOSE, dev, 0, 0);
}
/*===========================================================================*
* bdev_ioctl *
*===========================================================================*/
PRIVATE int bdev_ioctl(dev_t dev, endpoint_t proc_e, int req, void *buf)
{
/* Perform an I/O control operation on a block device. */
struct dmap *dp;
u32_t dummy;
cp_grant_id_t gid;
message dev_mess;
int op, safe, major_dev, minor_dev, vec_grants;
major_dev = major(dev);
minor_dev = minor(dev);
/* Determine task dmap. */
dp = &dmap[major_dev];
if (dp->dmap_driver == NONE) {
printf("VFS: dev_io: no driver for major %d\n", major_dev);
return(ENXIO);
}
/* Set up a grant if necessary. */
op = VFS_DEV_IOCTL;
safe = safe_io_conversion(dp->dmap_driver, &gid, &op, NULL, 0, &proc_e, &buf,
&vec_grants, 0, &dummy);
/* Set up the message passed to the task. */
memset(&dev_mess, 0, sizeof(dev_mess));
dev_mess.m_type = BDEV_IOCTL;
dev_mess.BDEV_MINOR = minor_dev;
dev_mess.BDEV_REQUEST = req;
dev_mess.BDEV_GRANT = gid;
dev_mess.BDEV_ID = 0;
/* Call the task. */
(*dp->dmap_io)(dp->dmap_driver, &dev_mess);
/* Clean up. */
if (safe) safe_io_cleanup(gid, NULL, vec_grants);
if (dp->dmap_driver == NONE) {
printf("VFS: block driver gone!?\n");
return(EIO);
}
/* Return the result. */
return dev_mess.BDEV_STATUS;
}
FS understanding revives and selects as alerts (new notifies)
2005-07-27 15:08:52 +02:00
/*===========================================================================*
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
* suspended_ep *
*===========================================================================*/
endpoint_t suspended_ep(endpoint_t driver, cp_grant_id_t g)
{
/* A process is suspended on a driver for which FS issued
* a grant. Find out which process it was.
*/
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
struct fproc *rfp;
for (rfp = &fproc[0]; rfp < &fproc[NR_PROCS]; rfp++) {
if(rfp->fp_pid == PID_FREE) continue;
Removed dependency of vfs on NR_TASKS macro - all macros in consts.h that depend on NR_TASKS replaced by a FP_BLOCKED_ON_* - fp_suspended removed and replaced by fp_blocked_on. Testing whether a process is supended is qeual to testing whether fp_blocked_on is FP_BLOCKED_ON_NONE or not - fp_task is valid only if fp_blocked_on == FP_BLOCKED_ON_OTHER - no need of special values that do not colide with valid and special endpoints since they are not used as endpoints anymore - suspend only takes FP_BLOCKED_ON_* values not endpoints anymore - suspend(task) replaced by wait_for(task) which sets fp_task so we remember who are we waiting for and suspend sets fp_blocked_on to FP_BLOCKED_ON_OTHER to signal that we are waiting for some other process - some functions should take endpoint_t instead of int, fixed
2009-09-22 23:48:26 +02:00
if(rfp->fp_blocked_on == FP_BLOCKED_ON_OTHER &&
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
rfp->fp_task == driver && rfp->fp_grant == g)
return rfp->fp_endpoint;
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
return(NONE);
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
/*===========================================================================*
* dev_status *
FS understanding revives and selects as alerts (new notifies)
2005-07-27 15:08:52 +02:00
*===========================================================================*/
PUBLIC void dev_status(message *m)
{
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
message st;
int d, get_more = 1;
endpoint_t endpt;
FS understanding revives and selects as alerts (new notifies)
2005-07-27 15:08:52 +02:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
for(d = 0; d < NR_DEVICES; d++)
if (dmap[d].dmap_driver != NONE && dmap[d].dmap_driver == m->m_source)
break;
FS understanding revives and selects as alerts (new notifies)
2005-07-27 15:08:52 +02:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (d >= NR_DEVICES) return;
Driver mapping refactory. VFS CHANGES: - dmap table no longer statically initialized in VFS - Dropped FSSIGNON svrctl call no longer used by INET INET CHANGES: - INET announces its presence to VFS just like any other driver RS CHANGES: - The boot image dev table contains all the data to initialize VFS' dmap table - RS interface supports asynchronous up and update operations now - RS interface extended to support driver style and flags
2010-04-09 23:56:44 +02:00
if (dmap[d].dmap_style == STYLE_DEVA) {
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
printf("dev_status: not doing dev_status for async driver %d\n",
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
m->m_source);
return;
}
FS understanding revives and selects as alerts (new notifies)
2005-07-27 15:08:52 +02:00
do {
int r;
st.m_type = DEV_STATUS;
Driver refactory for live update and crash recovery. SYSLIB CHANGES: - DS calls to publish / retrieve labels consider endpoints instead of u32_t. VFS CHANGES: - mapdriver() only adds an entry in the dmap table in VFS. - dev_up() is only executed upon reception of a driver up event. INET CHANGES: - INET no longer searches for existing drivers instances at startup. - A newtwork driver is (re)initialized upon reception of a driver up event. - Networking startup is now race-free by design. No need to waste 5 seconds at startup any more. DRIVER CHANGES: - Every driver publishes driver up events when starting for the first time or in case of restart when recovery actions must be taken in the upper layers. - Driver up events are published by drivers through DS. - For regular drivers, VFS is normally the only subscriber, but not necessarily. For instance, when the filter driver is in use, it must subscribe to driver up events to initiate recovery. - For network drivers, inet is the only subscriber for now. - Every VFS driver is statically linked with libdriver, every network driver is statically linked with libnetdriver. DRIVER LIBRARIES CHANGES: - Libdriver is extended to provide generic receive() and ds_publish() interfaces for VFS drivers. - driver_receive() is a wrapper for sef_receive() also used in driver_task() to discard spurious messages that were meant to be delivered to a previous version of the driver. - driver_receive_mq() is the same as driver_receive() but integrates support for queued messages. - driver_announce() publishes a driver up event for VFS drivers and marks the driver as initialized and expecting a DEV_OPEN message. - Libnetdriver is introduced to provide similar receive() and ds_publish() interfaces for network drivers (netdriver_announce() and netdriver_receive()). - Network drivers all support live update with no state transfer now. KERNEL CHANGES: - Added kernel call statectl for state management. Used by driver_announce() to unblock eventual callers sendrecing to the driver.
2010-04-08 15:41:35 +02:00
r = sendrec(m->m_source, &st);
Don't use kernel responses in servers
2010-06-24 09:37:26 +02:00
if(r == OK && st.REP_STATUS == ERESTART) r = EDEADEPT;
Driver refactory for live update and crash recovery. SYSLIB CHANGES: - DS calls to publish / retrieve labels consider endpoints instead of u32_t. VFS CHANGES: - mapdriver() only adds an entry in the dmap table in VFS. - dev_up() is only executed upon reception of a driver up event. INET CHANGES: - INET no longer searches for existing drivers instances at startup. - A newtwork driver is (re)initialized upon reception of a driver up event. - Networking startup is now race-free by design. No need to waste 5 seconds at startup any more. DRIVER CHANGES: - Every driver publishes driver up events when starting for the first time or in case of restart when recovery actions must be taken in the upper layers. - Driver up events are published by drivers through DS. - For regular drivers, VFS is normally the only subscriber, but not necessarily. For instance, when the filter driver is in use, it must subscribe to driver up events to initiate recovery. - For network drivers, inet is the only subscriber for now. - Every VFS driver is statically linked with libdriver, every network driver is statically linked with libnetdriver. DRIVER LIBRARIES CHANGES: - Libdriver is extended to provide generic receive() and ds_publish() interfaces for VFS drivers. - driver_receive() is a wrapper for sef_receive() also used in driver_task() to discard spurious messages that were meant to be delivered to a previous version of the driver. - driver_receive_mq() is the same as driver_receive() but integrates support for queued messages. - driver_announce() publishes a driver up event for VFS drivers and marks the driver as initialized and expecting a DEV_OPEN message. - Libnetdriver is introduced to provide similar receive() and ds_publish() interfaces for network drivers (netdriver_announce() and netdriver_receive()). - Network drivers all support live update with no state transfer now. KERNEL CHANGES: - Added kernel call statectl for state management. Used by driver_announce() to unblock eventual callers sendrecing to the driver.
2010-04-08 15:41:35 +02:00
if (r != OK) {
Two 'dynamic driver' features in FS: . When drivers disappear that have pending select()s, wake up those user processes with EAGAIN so that they can retry their select() and won't hang forever on it. . When drivers re-appear and are mapped into the dmap, run through the list of mounted filesystems and re-dev_open() every one (for partition tables and such). This can't happen before the driver has exec()ced itself, so processes that have fork()ed but not exec()ced yet are marked as DMAP_BABY in the dmap table if they are dmapped before they are execced. If that happens, the above procedure happens after the exec(). If the exec() happens before the dmapping, it (the dev_open()ing) happens right away.
2005-10-20 21:39:32 +02:00
printf("DEV_STATUS failed to %d: %d\n", m->m_source, r);
Don't use kernel responses in servers
2010-06-24 09:37:26 +02:00
if (r == EDEADSRCDST || r == EDEADEPT) return;
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("couldn't sendrec for DEV_STATUS: %d", r);
Two 'dynamic driver' features in FS: . When drivers disappear that have pending select()s, wake up those user processes with EAGAIN so that they can retry their select() and won't hang forever on it. . When drivers re-appear and are mapped into the dmap, run through the list of mounted filesystems and re-dev_open() every one (for partition tables and such). This can't happen before the driver has exec()ced itself, so processes that have fork()ed but not exec()ced yet are marked as DMAP_BABY in the dmap table if they are dmapped before they are execced. If that happens, the above procedure happens after the exec(). If the exec() happens before the dmapping, it (the dev_open()ing) happens right away.
2005-10-20 21:39:32 +02:00
}
FS understanding revives and selects as alerts (new notifies)
2005-07-27 15:08:52 +02:00
switch(st.m_type) {
case DEV_REVIVE:
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
endpt = st.REP_ENDPT;
Remove legacy MM, FS, and FS_PROC_NR macros
2010-06-08 15:58:01 +02:00
if(endpt == VFS_PROC_NR) {
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
endpt = suspended_ep(m->m_source,
st.REP_IO_GRANT);
if(endpt == NONE) {
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
printf("FS: proc with grant %d"
" from %d not found (revive)\n",
st.REP_IO_GRANT, st.m_source);
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
continue;
}
}
revive(endpt, st.REP_STATUS);
FS understanding revives and selects as alerts (new notifies)
2005-07-27 15:08:52 +02:00
break;
case DEV_IO_READY:
Clean up and fix multiple bugs in select: - Remove redundant code. - Always wait for the initial reply from an asynchronous select request, even if the select has been satisfied on another file descriptor or was canceled due to a serious error. - Restart asynchronous selects if upon reply from the driver turns out that there are deferred operations (and do not forget we're still interested in the results of the deferred operations). - Do not hang a non-blocking select when another blocking select on the same filp is still blocking. - Split blocking operations in read, write, and exceptions (i.e., blocking on read does not imply the write will block as well). - Some loops would iterate over OPEN_MAX file descriptors instead of the "highest" file descriptor. - Use proper internal error return values. - A secondary reply from a synchronous driver is essentially the same as from an asynchronous driver (the only difference being how the answer is received). Merge. - Return proper error code after a driver failure. - Auto-detect whether a driver is synchronous or asynchronous. - Remove some code duplication. - Clean up code (coding style, add missing comments, put all select related code together).
2011-04-13 15:25:34 +02:00
select_reply2(st.m_source, st.DEV_MINOR,
st.DEV_SEL_OPS);
FS understanding revives and selects as alerts (new notifies)
2005-07-27 15:08:52 +02:00
break;
default:
3rd fix for improper behaviour on crashing driver with safe_io_conversion
2006-06-29 16:24:07 +02:00
printf("FS: unrecognized reply %d to "
"DEV_STATUS\n", st.m_type);
FS understanding revives and selects as alerts (new notifies)
2005-07-27 15:08:52 +02:00
/* Fall through. */
case DEV_NO_STATUS:
get_more = 0;
break;
}
} while(get_more);
return;
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
/*===========================================================================*
* safe_io_conversion *
*===========================================================================*/
3rd fix for improper behaviour on crashing driver with safe_io_conversion
2006-06-29 16:24:07 +02:00
PRIVATE int safe_io_conversion(driver, gid, op, gids, gids_size,
. readall: use lseek64() to read more than 4GB of a device . vfs: 64-bit offset support for character device i/o (also remove unused dev_bio function) . memory: /dev/null and /dev/zero are infinitely large, don't stop reading/writing at 4GB
2007-04-24 15:27:33 +02:00
io_ept, buf, vec_grants, bytes, pos_lo)
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
endpoint_t driver;
cp_grant_id_t *gid;
int *op;
cp_grant_id_t *gids;
int gids_size;
endpoint_t *io_ept;
3rd fix for improper behaviour on crashing driver with safe_io_conversion
2006-06-29 16:24:07 +02:00
void **buf;
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
int *vec_grants;
vir_bytes bytes;
. readall: use lseek64() to read more than 4GB of a device . vfs: 64-bit offset support for character device i/o (also remove unused dev_bio function) . memory: /dev/null and /dev/zero are infinitely large, don't stop reading/writing at 4GB
2007-04-24 15:27:33 +02:00
u32_t *pos_lo;
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
{
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
int access = 0, size, j;
iovec_t *v;
static iovec_t new_iovec[NR_IOREQS];
/* Number of grants allocated in vector I/O. */
*vec_grants = 0;
/* Driver can handle it - change request to a safe one. */
*gid = GRANT_INVALID;
switch(*op) {
case VFS_DEV_READ:
case VFS_DEV_WRITE:
/* Change to safe op. */
*op = *op == VFS_DEV_READ ? DEV_READ_S : DEV_WRITE_S;
*gid = cpf_grant_magic(driver, *io_ept, (vir_bytes) *buf, bytes,
*op == DEV_READ_S ? CPF_WRITE : CPF_READ);
if (*gid < 0)
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("cpf_grant_magic of buffer failed");
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
break;
case VFS_DEV_GATHER:
case VFS_DEV_SCATTER:
/* Change to safe op. */
*op = *op == VFS_DEV_GATHER ? DEV_GATHER_S : DEV_SCATTER_S;
/* Grant access to my new i/o vector. */
*gid = cpf_grant_direct(driver, (vir_bytes) new_iovec,
bytes * sizeof(iovec_t), CPF_READ|CPF_WRITE);
if (*gid < 0)
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("cpf_grant_direct of vector failed");
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
v = (iovec_t *) *buf;
/* Grant access to i/o buffers. */
for(j = 0; j < bytes; j++) {
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
if(j >= NR_IOREQS) panic("vec too big: %d", bytes);
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
new_iovec[j].iov_addr =
gids[j] =
cpf_grant_direct(driver, (vir_bytes) v[j].iov_addr, v[j].iov_size,
*op == DEV_GATHER_S ? CPF_WRITE : CPF_READ);
if(!GRANT_VALID(gids[j]))
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("grant to iovec buf failed");
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
new_iovec[j].iov_size = v[j].iov_size;
(*vec_grants)++;
}
/* Set user's vector to the new one. */
*buf = new_iovec;
break;
case VFS_DEV_IOCTL:
*pos_lo = *io_ept; /* Old endpoint in POSITION field. */
*op = DEV_IOCTL_S;
if(_MINIX_IOCTL_IOR(m_in.REQUEST)) access |= CPF_WRITE;
if(_MINIX_IOCTL_IOW(m_in.REQUEST)) access |= CPF_READ;
if(_MINIX_IOCTL_BIG(m_in.REQUEST))
size = _MINIX_IOCTL_SIZE_BIG(m_in.REQUEST);
else
size = _MINIX_IOCTL_SIZE(m_in.REQUEST);
/* Do this even if no I/O happens with the ioctl, in
* order to disambiguate requests with DEV_IOCTL_S.
*/
*gid = cpf_grant_magic(driver, *io_ept, (vir_bytes) *buf, size,
access);
if (*gid < 0)
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("cpf_grant_magic failed (ioctl)");
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
break;
case VFS_DEV_SELECT:
*op = DEV_SELECT;
break;
default:
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("safe_io_conversion: unknown operation: %d", *op);
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
}
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
/* If we have converted to a safe operation, I/O
* endpoint becomes FS if it wasn't already.
*/
if(GRANT_VALID(*gid)) {
Remove legacy MM, FS, and FS_PROC_NR macros
2010-06-08 15:58:01 +02:00
*io_ept = VFS_PROC_NR;
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
return 1;
}
/* Not converted to a safe operation (because there is no
* copying involved in this operation).
*/
return 0;
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
}
/*===========================================================================*
* safe_io_cleanup *
*===========================================================================*/
PRIVATE void safe_io_cleanup(gid, gids, gids_size)
cp_grant_id_t gid;
cp_grant_id_t *gids;
int gids_size;
{
/* Free resources (specifically, grants) allocated by safe_io_conversion(). */
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
int j;
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
cpf_revoke(gid);
for(j = 0; j < gids_size; j++)
cpf_revoke(gids[j]);
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
}
New interface between PM and FS.
2006-05-11 16:57:23 +02:00
Initial revision
2005-04-21 16:53:53 +02:00
/*===========================================================================*
* dev_io *
*===========================================================================*/
Remove the types Dev_t, _mnx_Gui, _mnx_Uid, and similar. Use ANSI-style function declarations where necessary.
2010-04-13 12:58:41 +02:00
PUBLIC int dev_io(
int op, /* DEV_READ, DEV_WRITE, DEV_IOCTL, etc. */
dev_t dev, /* major-minor device number */
int proc_e, /* in whose address space is buf? */
void *buf, /* virtual address of the buffer */
u64_t pos, /* byte position */
int bytes, /* how many bytes to transfer */
int flags, /* special flags, like O_NONBLOCK */
int suspend_reopen /* Just suspend the process */
)
Initial revision
2005-04-21 16:53:53 +02:00
{
/* Read or write from a device. The parameter 'dev' tells which one. */
struct dmap *dp;
. readall: use lseek64() to read more than 4GB of a device . vfs: 64-bit offset support for character device i/o (also remove unused dev_bio function) . memory: /dev/null and /dev/zero are infinitely large, don't stop reading/writing at 4GB
2007-04-24 15:27:33 +02:00
u32_t pos_lo, pos_high;
Initial revision
2005-04-21 16:53:53 +02:00
message dev_mess;
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
cp_grant_id_t gid = GRANT_INVALID;
static cp_grant_id_t gids[NR_IOREQS];
Remove useless variables and the computations on them.
2010-02-19 11:00:32 +01:00
int vec_grants = 0, safe;
3rd fix for improper behaviour on crashing driver with safe_io_conversion
2006-06-29 16:24:07 +02:00
void *buf_used;
. fix for wild store at startup time by calling dev_io without fp initialized . fix for writing exactly PIPE_SIZE to a pipe with O_NONBLOCK blocking anyway because of incomplete logic in pipe_check
2006-08-10 13:51:11 +02:00
endpoint_t ioproc;
Initial revision
2005-04-21 16:53:53 +02:00
. readall: use lseek64() to read more than 4GB of a device . vfs: 64-bit offset support for character device i/o (also remove unused dev_bio function) . memory: /dev/null and /dev/zero are infinitely large, don't stop reading/writing at 4GB
2007-04-24 15:27:33 +02:00
pos_lo= ex64lo(pos);
pos_high= ex64hi(pos);
First cut at 64-bit file offsets in block devices for mkfs/fsck.
2006-11-27 15:21:43 +01:00
Initial revision
2005-04-21 16:53:53 +02:00
/* Determine task dmap. */
dp = &dmap[(dev >> MAJOR) & BYTE];
FS: . unmap device drivers from dmap when PM signals they are dead . new null-io function (no_dev_io) to fill in for io functions of unmapped drivers . driver (process number) of unmapped drivers is NONE instead of 0 (a valid process number) IS: . print mutable flag of dmap table too FS changes require sync() to be done 'manually' (currently by reboot/shutdown) at shutdown time; could be caught by SIGTERM in the future.
2005-10-05 17:38:15 +02:00
/* See if driver is roughly valid. */
Don't panic on missing drivers.
2005-10-21 21:31:38 +02:00
if (dp->dmap_driver == NONE) {
printf("FS: dev_io: no driver for dev %x\n", dev);
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
return(ENXIO);
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (suspend_reopen) {
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
/* Suspend user. */
fp->fp_grant = GRANT_INVALID;
fp->fp_ioproc = NONE;
Removed dependency of vfs on NR_TASKS macro - all macros in consts.h that depend on NR_TASKS replaced by a FP_BLOCKED_ON_* - fp_suspended removed and replaced by fp_blocked_on. Testing whether a process is supended is qeual to testing whether fp_blocked_on is FP_BLOCKED_ON_NONE or not - fp_task is valid only if fp_blocked_on == FP_BLOCKED_ON_OTHER - no need of special values that do not colide with valid and special endpoints since they are not used as endpoints anymore - suspend only takes FP_BLOCKED_ON_* values not endpoints anymore - suspend(task) replaced by wait_for(task) which sets fp_task so we remember who are we waiting for and suspend sets fp_blocked_on to FP_BLOCKED_ON_OTHER to signal that we are waiting for some other process - some functions should take endpoint_t instead of int, fixed
2009-09-22 23:48:26 +02:00
wait_for(dp->dmap_driver);
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
fp->fp_flags |= SUSP_REOPEN;
return(SUSPEND);
}
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
if(isokendpt(dp->dmap_driver, &dummyproc) != OK) {
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
printf("FS: dev_io: old driver for dev %x (%d)\n",dev,dp->dmap_driver);
return(ENXIO);
Don't panic on missing drivers.
2005-10-21 21:31:38 +02:00
}
FS: . unmap device drivers from dmap when PM signals they are dead . new null-io function (no_dev_io) to fill in for io functions of unmapped drivers . driver (process number) of unmapped drivers is NONE instead of 0 (a valid process number) IS: . print mutable flag of dmap table too FS changes require sync() to be done 'manually' (currently by reboot/shutdown) at shutdown time; could be caught by SIGTERM in the future.
2005-10-05 17:38:15 +02:00
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
/* By default, these are right. */
Server/driver protocols: no longer allow third-party copies. Before safecopies, the IO_ENDPT and DL_ENDPT message fields were needed to know which actual process to copy data from/to, as that process may not always be the caller. Now that we have full safecopy support, these fields have become useless for that purpose: the owner of the grant is *always* the caller. Allowing the caller to supply another endpoint is in fact dangerous, because the callee may then end up using a grant from a third party. One could call this a variant of the confused deputy problem. From now on, safecopy calls should always use the caller's endpoint as grant owner. This fully obsoletes the DL_ENDPT field in the inet/ethernet protocol. IO_ENDPT has other uses besides identifying the grant owner though. This patch renames IO_ENDPT to USER_ENDPT, not only because that is a more fitting name (it should never be used for I/O after all), but also in order to intentionally break any old system source code outside the base system. If this patch breaks your code, fixing it is fairly simple: - DL_ENDPT should be replaced with m_source; - IO_ENDPT should be replaced with m_source when used for safecopies; - IO_ENDPT should be replaced with USER_ENDPT for any other use, e.g. when setting REP_ENDPT, matching requests in CANCEL calls, getting DEV_SELECT flags, and retrieving of the real user process's endpoint in DEV_OPEN. The changes in this patch are binary backward compatible.
2011-04-11 19:35:05 +02:00
dev_mess.USER_ENDPT = proc_e;
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
dev_mess.ADDRESS = buf;
/* Convert DEV_* to DEV_*_S variants. */
3rd fix for improper behaviour on crashing driver with safe_io_conversion
2006-06-29 16:24:07 +02:00
buf_used = buf;
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
safe = safe_io_conversion(dp->dmap_driver, &gid, &op, gids, NR_IOREQS,
Server/driver protocols: no longer allow third-party copies. Before safecopies, the IO_ENDPT and DL_ENDPT message fields were needed to know which actual process to copy data from/to, as that process may not always be the caller. Now that we have full safecopy support, these fields have become useless for that purpose: the owner of the grant is *always* the caller. Allowing the caller to supply another endpoint is in fact dangerous, because the callee may then end up using a grant from a third party. One could call this a variant of the confused deputy problem. From now on, safecopy calls should always use the caller's endpoint as grant owner. This fully obsoletes the DL_ENDPT field in the inet/ethernet protocol. IO_ENDPT has other uses besides identifying the grant owner though. This patch renames IO_ENDPT to USER_ENDPT, not only because that is a more fitting name (it should never be used for I/O after all), but also in order to intentionally break any old system source code outside the base system. If this patch breaks your code, fixing it is fairly simple: - DL_ENDPT should be replaced with m_source; - IO_ENDPT should be replaced with m_source when used for safecopies; - IO_ENDPT should be replaced with USER_ENDPT for any other use, e.g. when setting REP_ENDPT, matching requests in CANCEL calls, getting DEV_SELECT flags, and retrieving of the real user process's endpoint in DEV_OPEN. The changes in this patch are binary backward compatible.
2011-04-11 19:35:05 +02:00
(endpoint_t*) &dev_mess.USER_ENDPT, &buf_used,
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
&vec_grants, bytes, &pos_lo);
3rd fix for improper behaviour on crashing driver with safe_io_conversion
2006-06-29 16:24:07 +02:00
if(buf != buf_used)
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("dev_io: safe_io_conversion changed buffer");
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
/* If the safe conversion was done, set the ADDRESS to
* the grant id.
*/
if(safe) dev_mess.IO_GRANT = (char *) gid;
/* Set up the rest of the message passed to task. */
Initial revision
2005-04-21 16:53:53 +02:00
dev_mess.m_type = op;
dev_mess.DEVICE = (dev >> MINOR) & BYTE;
. readall: use lseek64() to read more than 4GB of a device . vfs: 64-bit offset support for character device i/o (also remove unused dev_bio function) . memory: /dev/null and /dev/zero are infinitely large, don't stop reading/writing at 4GB
2007-04-24 15:27:33 +02:00
dev_mess.POSITION = pos_lo;
Initial revision
2005-04-21 16:53:53 +02:00
dev_mess.COUNT = bytes;
. readall: use lseek64() to read more than 4GB of a device . vfs: 64-bit offset support for character device i/o (also remove unused dev_bio function) . memory: /dev/null and /dev/zero are infinitely large, don't stop reading/writing at 4GB
2007-04-24 15:27:33 +02:00
dev_mess.HIGHPOS = pos_high;
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
. fix for wild store at startup time by calling dev_io without fp initialized . fix for writing exactly PIPE_SIZE to a pipe with O_NONBLOCK blocking anyway because of incomplete logic in pipe_check
2006-08-10 13:51:11 +02:00
/* This will be used if the i/o is suspended. */
Server/driver protocols: no longer allow third-party copies. Before safecopies, the IO_ENDPT and DL_ENDPT message fields were needed to know which actual process to copy data from/to, as that process may not always be the caller. Now that we have full safecopy support, these fields have become useless for that purpose: the owner of the grant is *always* the caller. Allowing the caller to supply another endpoint is in fact dangerous, because the callee may then end up using a grant from a third party. One could call this a variant of the confused deputy problem. From now on, safecopy calls should always use the caller's endpoint as grant owner. This fully obsoletes the DL_ENDPT field in the inet/ethernet protocol. IO_ENDPT has other uses besides identifying the grant owner though. This patch renames IO_ENDPT to USER_ENDPT, not only because that is a more fitting name (it should never be used for I/O after all), but also in order to intentionally break any old system source code outside the base system. If this patch breaks your code, fixing it is fairly simple: - DL_ENDPT should be replaced with m_source; - IO_ENDPT should be replaced with m_source when used for safecopies; - IO_ENDPT should be replaced with USER_ENDPT for any other use, e.g. when setting REP_ENDPT, matching requests in CANCEL calls, getting DEV_SELECT flags, and retrieving of the real user process's endpoint in DEV_OPEN. The changes in this patch are binary backward compatible.
2011-04-11 19:35:05 +02:00
ioproc = dev_mess.USER_ENDPT;
Initial revision
2005-04-21 16:53:53 +02:00
/* Call the task. */
(*dp->dmap_io)(dp->dmap_driver, &dev_mess);
Driver disappearing during dev_io() is not a fatal error.
2006-03-15 22:55:32 +01:00
if(dp->dmap_driver == NONE) {
/* Driver has vanished. */
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
printf("Driver gone?\n");
if(safe) safe_io_cleanup(gid, gids, vec_grants);
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
return(EIO);
Driver disappearing during dev_io() is not a fatal error.
2006-03-15 22:55:32 +01:00
}
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
Initial revision
2005-04-21 16:53:53 +02:00
/* Task has completed. See if call completed. */
if (dev_mess.REP_STATUS == SUSPEND) {
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
if(vec_grants > 0) panic("SUSPEND on vectored i/o");
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
. fix for wild store at startup time by calling dev_io without fp initialized . fix for writing exactly PIPE_SIZE to a pipe with O_NONBLOCK blocking anyway because of incomplete logic in pipe_check
2006-08-10 13:51:11 +02:00
/* fp is uninitialized at init time. */
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
if(!fp) panic("SUSPEND on NULL fp");
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
Driver mapping refactory. VFS CHANGES: - dmap table no longer statically initialized in VFS - Dropped FSSIGNON svrctl call no longer used by INET INET CHANGES: - INET announces its presence to VFS just like any other driver RS CHANGES: - The boot image dev table contains all the data to initialize VFS' dmap table - RS interface supports asynchronous up and update operations now - RS interface extended to support driver style and flags
2010-04-09 23:56:44 +02:00
if ((flags & O_NONBLOCK) && !(dp->dmap_style == STYLE_DEVA)) {
Initial revision
2005-04-21 16:53:53 +02:00
/* Not supposed to block. */
dev_mess.m_type = CANCEL;
Server/driver protocols: no longer allow third-party copies. Before safecopies, the IO_ENDPT and DL_ENDPT message fields were needed to know which actual process to copy data from/to, as that process may not always be the caller. Now that we have full safecopy support, these fields have become useless for that purpose: the owner of the grant is *always* the caller. Allowing the caller to supply another endpoint is in fact dangerous, because the callee may then end up using a grant from a third party. One could call this a variant of the confused deputy problem. From now on, safecopy calls should always use the caller's endpoint as grant owner. This fully obsoletes the DL_ENDPT field in the inet/ethernet protocol. IO_ENDPT has other uses besides identifying the grant owner though. This patch renames IO_ENDPT to USER_ENDPT, not only because that is a more fitting name (it should never be used for I/O after all), but also in order to intentionally break any old system source code outside the base system. If this patch breaks your code, fixing it is fairly simple: - DL_ENDPT should be replaced with m_source; - IO_ENDPT should be replaced with m_source when used for safecopies; - IO_ENDPT should be replaced with USER_ENDPT for any other use, e.g. when setting REP_ENDPT, matching requests in CANCEL calls, getting DEV_SELECT flags, and retrieving of the real user process's endpoint in DEV_OPEN. The changes in this patch are binary backward compatible.
2011-04-11 19:35:05 +02:00
dev_mess.USER_ENDPT = ioproc;
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
dev_mess.IO_GRANT = (char *) gid;
/* This R_BIT/W_BIT check taken from suspend()/unpause()
* logic. Mode is expected in the COUNT field.
*/
dev_mess.COUNT = 0;
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (call_nr == READ) dev_mess.COUNT = R_BIT;
else if (call_nr == WRITE) dev_mess.COUNT = W_BIT;
Initial revision
2005-04-21 16:53:53 +02:00
dev_mess.DEVICE = (dev >> MINOR) & BYTE;
(*dp->dmap_io)(dp->dmap_driver, &dev_mess);
if (dev_mess.REP_STATUS == EINTR) dev_mess.REP_STATUS = EAGAIN;
} else {
don't suspend the process as a side-effect if device returns SUSPEND if it's select; select already does this.
2009-05-08 15:50:29 +02:00
/* select() will do suspending itself. */
if(op != DEV_SELECT) {
/* Suspend user. */
Removed dependency of vfs on NR_TASKS macro - all macros in consts.h that depend on NR_TASKS replaced by a FP_BLOCKED_ON_* - fp_suspended removed and replaced by fp_blocked_on. Testing whether a process is supended is qeual to testing whether fp_blocked_on is FP_BLOCKED_ON_NONE or not - fp_task is valid only if fp_blocked_on == FP_BLOCKED_ON_OTHER - no need of special values that do not colide with valid and special endpoints since they are not used as endpoints anymore - suspend only takes FP_BLOCKED_ON_* values not endpoints anymore - suspend(task) replaced by wait_for(task) which sets fp_task so we remember who are we waiting for and suspend sets fp_blocked_on to FP_BLOCKED_ON_OTHER to signal that we are waiting for some other process - some functions should take endpoint_t instead of int, fixed
2009-09-22 23:48:26 +02:00
wait_for(dp->dmap_driver);
don't suspend the process as a side-effect if device returns SUSPEND if it's select; select already does this.
2009-05-08 15:50:29 +02:00
}
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
assert(!GRANT_VALID(fp->fp_grant));
fp->fp_grant = gid; /* revoke this when unsuspended. */
. fix for wild store at startup time by calling dev_io without fp initialized . fix for writing exactly PIPE_SIZE to a pipe with O_NONBLOCK blocking anyway because of incomplete logic in pipe_check
2006-08-10 13:51:11 +02:00
fp->fp_ioproc = ioproc;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
if (flags & O_NONBLOCK) {
/* Not supposed to block, send cancel message */
dev_mess.m_type = CANCEL;
Server/driver protocols: no longer allow third-party copies. Before safecopies, the IO_ENDPT and DL_ENDPT message fields were needed to know which actual process to copy data from/to, as that process may not always be the caller. Now that we have full safecopy support, these fields have become useless for that purpose: the owner of the grant is *always* the caller. Allowing the caller to supply another endpoint is in fact dangerous, because the callee may then end up using a grant from a third party. One could call this a variant of the confused deputy problem. From now on, safecopy calls should always use the caller's endpoint as grant owner. This fully obsoletes the DL_ENDPT field in the inet/ethernet protocol. IO_ENDPT has other uses besides identifying the grant owner though. This patch renames IO_ENDPT to USER_ENDPT, not only because that is a more fitting name (it should never be used for I/O after all), but also in order to intentionally break any old system source code outside the base system. If this patch breaks your code, fixing it is fairly simple: - DL_ENDPT should be replaced with m_source; - IO_ENDPT should be replaced with m_source when used for safecopies; - IO_ENDPT should be replaced with USER_ENDPT for any other use, e.g. when setting REP_ENDPT, matching requests in CANCEL calls, getting DEV_SELECT flags, and retrieving of the real user process's endpoint in DEV_OPEN. The changes in this patch are binary backward compatible.
2011-04-11 19:35:05 +02:00
dev_mess.USER_ENDPT = ioproc;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
dev_mess.IO_GRANT = (char *) gid;
/* This R_BIT/W_BIT check taken from suspend()/unpause()
* logic. Mode is expected in the COUNT field.
*/
dev_mess.COUNT = 0;
if(call_nr == READ) dev_mess.COUNT = R_BIT;
else if(call_nr == WRITE) dev_mess.COUNT = W_BIT;
dev_mess.DEVICE = (dev >> MINOR) & BYTE;
(*dp->dmap_io)(dp->dmap_driver, &dev_mess);
/* Should do something about EINTR -> EAGAIN mapping */
}
Initial revision
2005-04-21 16:53:53 +02:00
return(SUSPEND);
}
}
FS support for grant-based i/o. For character device i/o, FS does a so-called 'magic' grant to let the driver copy from or to user space. As this is done in FS address space, the driver is told to do this in FS address space. The redirection to the right user process then happens at copy-time in the kernel, using the FS grant table. This also happens for DEV_READ and DEV_WRITE on block devices. For other block device i/o, which happens from/to FS buffers, FS does a 'direct' grant to its own address space for the driver. After the i/o returns, this access has to be K-I-L-L-E-D, revoked. Sometimes this is after a SUSPEND and DEV_REVIVE, in which case the revoking happens in pipe.c. This conversion happens in safe_io_conversion() in device.c, called by dev_io and dev_bio. FS has to pre-allocate its own space for these grant tables. This happens in main.c.
2006-06-20 12:12:09 +02:00
/* No suspend, or cancelled suspend, so I/O is over and can be cleaned up. */
if(safe) safe_io_cleanup(gid, gids, vec_grants);
Initial revision
2005-04-21 16:53:53 +02:00
return(dev_mess.REP_STATUS);
}
/*===========================================================================*
* gen_opcl *
*===========================================================================*/
Remove the types Dev_t, _mnx_Gui, _mnx_Uid, and similar. Use ANSI-style function declarations where necessary.
2010-04-13 12:58:41 +02:00
PUBLIC int gen_opcl(
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
int op, /* operation, (B)DEV_OPEN or (B)DEV_CLOSE */
Remove the types Dev_t, _mnx_Gui, _mnx_Uid, and similar. Use ANSI-style function declarations where necessary.
2010-04-13 12:58:41 +02:00
dev_t dev, /* device to open or close */
int proc_e, /* process to open/close for */
int flags /* mode bits and flags */
)
Initial revision
2005-04-21 16:53:53 +02:00
{
/* Called from the dmap struct in table.c on opens & closes of special files.*/
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
int r, is_bdev;
Initial revision
2005-04-21 16:53:53 +02:00
struct dmap *dp;
message dev_mess;
/* Determine task dmap. */
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
dp = &dmap[major(dev)];
Initial revision
2005-04-21 16:53:53 +02:00
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
is_bdev = IS_BDEV_RQ(op);
if (is_bdev) {
memset(&dev_mess, 0, sizeof(dev_mess));
dev_mess.m_type = op;
dev_mess.BDEV_MINOR = minor(dev);
dev_mess.BDEV_ACCESS = flags;
dev_mess.BDEV_ID = 0;
} else {
dev_mess.m_type = op;
dev_mess.DEVICE = minor(dev);
dev_mess.USER_ENDPT = proc_e;
dev_mess.COUNT = flags;
}
Initial revision
2005-04-21 16:53:53 +02:00
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
if (dp->dmap_driver == NONE) {
printf("FS: gen_opcl: no driver for dev %x\n", dev);
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
return(ENXIO);
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
}
Initial revision
2005-04-21 16:53:53 +02:00
/* Call the task. */
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
r= (*dp->dmap_io)(dp->dmap_driver, &dev_mess);
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (r != OK) return(r);
Initial revision
2005-04-21 16:53:53 +02:00
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
if (is_bdev)
return(dev_mess.BDEV_STATUS);
else
return(dev_mess.REP_STATUS);
Initial revision
2005-04-21 16:53:53 +02:00
}
/*===========================================================================*
* tty_opcl *
*===========================================================================*/
Remove the types Dev_t, _mnx_Gui, _mnx_Uid, and similar. Use ANSI-style function declarations where necessary.
2010-04-13 12:58:41 +02:00
PUBLIC int tty_opcl(
int op, /* operation, DEV_OPEN or DEV_CLOSE */
dev_t dev, /* device to open or close */
int proc_e, /* process to open/close for */
int flags /* mode bits and flags */
)
Initial revision
2005-04-21 16:53:53 +02:00
{
/* This procedure is called from the dmap struct on tty open/close. */
int r;
register struct fproc *rfp;
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
assert(!IS_BDEV_RQ(op));
Initial revision
2005-04-21 16:53:53 +02:00
/* Add O_NOCTTY to the flags if this process is not a session leader, or
* if it already has a controlling tty, or if it is someone elses
* controlling tty.
*/
if (!fp->fp_sesldr || fp->fp_tty != 0) {
flags |= O_NOCTTY;
} else {
for (rfp = &fproc[0]; rfp < &fproc[NR_PROCS]; rfp++) {
FS: . loops checked for PID_FREE . exit broken down in exit and cleanup functions; when reboot happens, cleanup is done but not exit (as processes have not actually exited), this keeps drivers working . fixed a few uninitialized and unused variables scripts: . new packaging system
2006-03-15 16:34:12 +01:00
if(rfp->fp_pid == PID_FREE) continue;
Initial revision
2005-04-21 16:53:53 +02:00
if (rfp->fp_tty == dev) flags |= O_NOCTTY;
}
}
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
r = gen_opcl(op, dev, proc_e, flags);
Initial revision
2005-04-21 16:53:53 +02:00
/* Did this call make the tty the controlling tty? */
if (r == 1) {
fp->fp_tty = dev;
r = OK;
}
return(r);
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
Initial revision
2005-04-21 16:53:53 +02:00
/*===========================================================================*
* ctty_opcl *
*===========================================================================*/
Remove the types Dev_t, _mnx_Gui, _mnx_Uid, and similar. Use ANSI-style function declarations where necessary.
2010-04-13 12:58:41 +02:00
PUBLIC int ctty_opcl(
int op, /* operation, DEV_OPEN or DEV_CLOSE */
dev_t dev, /* device to open or close */
int proc_e, /* process to open/close for */
int flags /* mode bits and flags */
)
Initial revision
2005-04-21 16:53:53 +02:00
{
/* This procedure is called from the dmap struct in table.c on opening/closing
* /dev/tty, the magic device that translates to the controlling tty.
*/
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
assert(!IS_BDEV_RQ(op));
Initial revision
2005-04-21 16:53:53 +02:00
return(fp->fp_tty == 0 ? ENXIO : OK);
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
Initial revision
2005-04-21 16:53:53 +02:00
/*===========================================================================*
New interface between PM and FS.
2006-05-11 16:57:23 +02:00
* pm_setsid *
Initial revision
2005-04-21 16:53:53 +02:00
*===========================================================================*/
New interface between PM and FS.
2006-05-11 16:57:23 +02:00
PUBLIC void pm_setsid(proc_e)
int proc_e;
Initial revision
2005-04-21 16:53:53 +02:00
{
/* Perform the FS side of the SETSID call, i.e. get rid of the controlling
* terminal of a process, and make the process a session leader.
*/
register struct fproc *rfp;
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
int slot;
Initial revision
2005-04-21 16:53:53 +02:00
/* Make the process a session leader with no controlling tty. */
New interface between PM and FS.
2006-05-11 16:57:23 +02:00
okendpt(proc_e, &slot);
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
rfp = &fproc[slot];
Initial revision
2005-04-21 16:53:53 +02:00
rfp->fp_sesldr = TRUE;
rfp->fp_tty = 0;
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
Initial revision
2005-04-21 16:53:53 +02:00
/*===========================================================================*
* do_ioctl *
*===========================================================================*/
PUBLIC int do_ioctl()
{
/* Perform the ioctl(ls_fd, request, argx) system call (uses m2 fmt). */
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
int suspend_reopen;
Initial revision
2005-04-21 16:53:53 +02:00
struct filp *f;
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
register struct vnode *vp;
Initial revision
2005-04-21 16:53:53 +02:00
dev_t dev;
Use of all NIL_* defines converted to NULL
2010-05-10 15:26:00 +02:00
if ((f = get_filp(m_in.ls_fd)) == NULL) return(err_code);
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
vp = f->filp_vno; /* get vnode pointer */
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if ((vp->v_mode & I_TYPE) != I_CHAR_SPECIAL &&
(vp->v_mode & I_TYPE) != I_BLOCK_SPECIAL) return(ENOTTY);
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
suspend_reopen= (f->filp_state != FS_NORMAL);
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
dev = (dev_t) vp->v_sdev;
Initial revision
2005-04-21 16:53:53 +02:00
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
if ((vp->v_mode & I_TYPE) == I_BLOCK_SPECIAL)
return bdev_ioctl(dev, who_e, m_in.REQUEST, m_in.ADDRESS);
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
return dev_io(VFS_DEV_IOCTL, dev, who_e, m_in.ADDRESS, cvu64(0),
m_in.REQUEST, f->filp_flags, suspend_reopen);
Initial revision
2005-04-21 16:53:53 +02:00
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
Initial revision
2005-04-21 16:53:53 +02:00
/*===========================================================================*
* gen_io *
*===========================================================================*/
dmap_io now returns a status. map_driver no longer calls dev_up.
2006-03-15 14:37:20 +01:00
PUBLIC int gen_io(task_nr, mess_ptr)
Initial revision
2005-04-21 16:53:53 +02:00
int task_nr; /* which task to call */
message *mess_ptr; /* pointer to message for task */
{
/* All file system I/O ultimately comes down to I/O on major/minor device
* pairs. These lead to calls on the following routines via the dmap table.
*/
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
int r, status, proc_e, is_bdev;
Initial revision
2005-04-21 16:53:53 +02:00
Prepare VFS to support back calls from PFS. For security reasons and to support file descriptor passing, PFS does some back calls to VFS. For example, to verify the validity of a path provided by a process and to tell VFS it must copy file descriptors from one process to another.
2010-08-30 15:44:07 +02:00
if(task_nr == SYSTEM) {
printf("VFS: sending %d to SYSTEM\n", mess_ptr->m_type);
}
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
is_bdev = IS_BDEV_RQ(mess_ptr->m_type);
if (!is_bdev) proc_e = mess_ptr->USER_ENDPT;
Prepare VFS to support back calls from PFS. For security reasons and to support file descriptor passing, PFS does some back calls to VFS. For example, to verify the validity of a path provided by a process and to tell VFS it must copy file descriptors from one process to another.
2010-08-30 15:44:07 +02:00
for (;;) {
r = sendrec(task_nr, mess_ptr);
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
if(r == OK) {
if (is_bdev)
status = mess_ptr->BDEV_STATUS;
else
status = mess_ptr->REP_STATUS;
if (status == ERESTART) r = EDEADEPT;
}
Prepare VFS to support back calls from PFS. For security reasons and to support file descriptor passing, PFS does some back calls to VFS. For example, to verify the validity of a path provided by a process and to tell VFS it must copy file descriptors from one process to another.
2010-08-30 15:44:07 +02:00
if (r != OK) {
if (r == EDEADSRCDST || r == EDEADEPT) {
printf("fs: dead driver %d\n", task_nr);
dmap_unmap_by_endpt(task_nr);
return(r);
}
if (r == ELOCKED) {
printf("fs: ELOCKED talking to %d\n", task_nr);
return(r);
}
panic("call_task: can't send/receive: %d", r);
Initial revision
2005-04-21 16:53:53 +02:00
}
Prepare VFS to support back calls from PFS. For security reasons and to support file descriptor passing, PFS does some back calls to VFS. For example, to verify the validity of a path provided by a process and to tell VFS it must copy file descriptors from one process to another.
2010-08-30 15:44:07 +02:00
/* Did the process we did the sendrec() for get a result? */
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
if (!is_bdev &&
mess_ptr->REP_ENDPT != proc_e && VFS_PROC_NR != proc_e) {
Prepare VFS to support back calls from PFS. For security reasons and to support file descriptor passing, PFS does some back calls to VFS. For example, to verify the validity of a path provided by a process and to tell VFS it must copy file descriptors from one process to another.
2010-08-30 15:44:07 +02:00
printf("fs: strange device reply from %d, type = %d, "
"proc = %d (not %d) (2) ignored\n", mess_ptr->m_source,
mess_ptr->m_type, proc_e, mess_ptr->REP_ENDPT);
return(EIO);
Initial revision
2005-04-21 16:53:53 +02:00
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
Prepare VFS to support back calls from PFS. For security reasons and to support file descriptor passing, PFS does some back calls to VFS. For example, to verify the validity of a path provided by a process and to tell VFS it must copy file descriptors from one process to another.
2010-08-30 15:44:07 +02:00
if (mess_ptr->m_type == TASK_REPLY ||
IS_DEV_RS(mess_ptr->m_type) ||
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
IS_BDEV_RS(mess_ptr->m_type) ||
Prepare VFS to support back calls from PFS. For security reasons and to support file descriptor passing, PFS does some back calls to VFS. For example, to verify the validity of a path provided by a process and to tell VFS it must copy file descriptors from one process to another.
2010-08-30 15:44:07 +02:00
mess_ptr->m_type <= 0) {
break; /* reply */
} else {
nested_dev_call(mess_ptr);
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
}
dmap_io now returns a status. map_driver no longer calls dev_up.
2006-03-15 14:37:20 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
return(OK);
Initial revision
2005-04-21 16:53:53 +02:00
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
/*===========================================================================*
* asyn_io *
*===========================================================================*/
PUBLIC int asyn_io(task_nr, mess_ptr)
int task_nr; /* which task to call */
message *mess_ptr; /* pointer to message for task */
{
/* All file system I/O ultimately comes down to I/O on major/minor device
* pairs. These lead to calls on the following routines via the dmap table.
*/
Remove useless variables and the computations on them.
2010-02-19 11:00:32 +01:00
int r;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
assert(!IS_BDEV_RQ(mess_ptr->m_type));
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
r = asynsend(task_nr, mess_ptr);
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
if (r != OK) panic("asyn_io: asynsend failed: %d", r);
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
/* Fake a SUSPEND */
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
mess_ptr->REP_STATUS = SUSPEND;
return(OK);
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
Initial revision
2005-04-21 16:53:53 +02:00
/*===========================================================================*
* ctty_io *
*===========================================================================*/
dmap_io now returns a status. map_driver no longer calls dev_up.
2006-03-15 14:37:20 +01:00
PUBLIC int ctty_io(task_nr, mess_ptr)
Initial revision
2005-04-21 16:53:53 +02:00
int task_nr; /* not used - for compatibility with dmap_t */
message *mess_ptr; /* pointer to message for task */
{
/* This routine is only called for one device, namely /dev/tty. Its job
* is to change the message to use the controlling terminal, instead of the
* major/minor pair for /dev/tty itself.
*/
struct dmap *dp;
if (fp->fp_tty == 0) {
/* No controlling tty present anymore, return an I/O error. */
mess_ptr->REP_STATUS = EIO;
} else {
/* Substitute the controlling terminal device. */
dp = &dmap[(fp->fp_tty >> MAJOR) & BYTE];
mess_ptr->DEVICE = (fp->fp_tty >> MINOR) & BYTE;
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (dp->dmap_driver == NONE) {
printf("FS: ctty_io: no driver for dev\n");
return(EIO);
}
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
if(isokendpt(dp->dmap_driver, &dummyproc) != OK) {
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
printf("FS: ctty_io: old driver %d\n", dp->dmap_driver);
return(EIO);
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
}
Initial revision
2005-04-21 16:53:53 +02:00
(*dp->dmap_io)(dp->dmap_driver, mess_ptr);
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
return(OK);
Initial revision
2005-04-21 16:53:53 +02:00
}
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
Initial revision
2005-04-21 16:53:53 +02:00
/*===========================================================================*
* no_dev *
*===========================================================================*/
Remove the types Dev_t, _mnx_Gui, _mnx_Uid, and similar. Use ANSI-style function declarations where necessary.
2010-04-13 12:58:41 +02:00
PUBLIC int no_dev(
int UNUSED(op), /* operation, DEV_OPEN or DEV_CLOSE */
dev_t UNUSED(dev), /* device to open or close */
int UNUSED(proc), /* process to open/close for */
int UNUSED(flags) /* mode bits and flags */
)
Initial revision
2005-04-21 16:53:53 +02:00
{
/* Called when opening a nonexistent device. */
return(ENODEV);
}
FS: . unmap device drivers from dmap when PM signals they are dead . new null-io function (no_dev_io) to fill in for io functions of unmapped drivers . driver (process number) of unmapped drivers is NONE instead of 0 (a valid process number) IS: . print mutable flag of dmap table too FS changes require sync() to be done 'manually' (currently by reboot/shutdown) at shutdown time; could be caught by SIGTERM in the future.
2005-10-05 17:38:15 +02:00
/*===========================================================================*
* no_dev_io *
*===========================================================================*/
dmap_io now returns a status. map_driver no longer calls dev_up.
2006-03-15 14:37:20 +01:00
PUBLIC int no_dev_io(int proc, message *m)
FS: . unmap device drivers from dmap when PM signals they are dead . new null-io function (no_dev_io) to fill in for io functions of unmapped drivers . driver (process number) of unmapped drivers is NONE instead of 0 (a valid process number) IS: . print mutable flag of dmap table too FS changes require sync() to be done 'manually' (currently by reboot/shutdown) at shutdown time; could be caught by SIGTERM in the future.
2005-10-05 17:38:15 +02:00
{
/* Called when doing i/o on a nonexistent device. */
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
printf("VFS: I/O on unmapped device number\n");
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
return(EIO);
FS: . unmap device drivers from dmap when PM signals they are dead . new null-io function (no_dev_io) to fill in for io functions of unmapped drivers . driver (process number) of unmapped drivers is NONE instead of 0 (a valid process number) IS: . print mutable flag of dmap table too FS changes require sync() to be done 'manually' (currently by reboot/shutdown) at shutdown time; could be caught by SIGTERM in the future.
2005-10-05 17:38:15 +02:00
}
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
Initial revision
2005-04-21 16:53:53 +02:00
/*===========================================================================*
* clone_opcl *
*===========================================================================*/
Remove the types Dev_t, _mnx_Gui, _mnx_Uid, and similar. Use ANSI-style function declarations where necessary.
2010-04-13 12:58:41 +02:00
PUBLIC int clone_opcl(
int op, /* operation, DEV_OPEN or DEV_CLOSE */
dev_t dev, /* device to open or close */
int proc_e, /* process to open/close for */
int flags /* mode bits and flags */
)
Initial revision
2005-04-21 16:53:53 +02:00
{
/* Some devices need special processing upon open. Such a device is "cloned",
* i.e. on a succesful open it is replaced by a new device with a new unique
* minor device number. This new device number identifies a new object (such
* as a new network connection) that has been allocated within a task.
*/
struct dmap *dp;
dmap_io now returns a status. map_driver no longer calls dev_up.
2006-03-15 14:37:20 +01:00
int r, minor;
Initial revision
2005-04-21 16:53:53 +02:00
message dev_mess;
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
assert(!IS_BDEV_RQ(op));
Initial revision
2005-04-21 16:53:53 +02:00
/* Determine task dmap. */
dp = &dmap[(dev >> MAJOR) & BYTE];
minor = (dev >> MINOR) & BYTE;
dev_mess.m_type = op;
dev_mess.DEVICE = minor;
Server/driver protocols: no longer allow third-party copies. Before safecopies, the IO_ENDPT and DL_ENDPT message fields were needed to know which actual process to copy data from/to, as that process may not always be the caller. Now that we have full safecopy support, these fields have become useless for that purpose: the owner of the grant is *always* the caller. Allowing the caller to supply another endpoint is in fact dangerous, because the callee may then end up using a grant from a third party. One could call this a variant of the confused deputy problem. From now on, safecopy calls should always use the caller's endpoint as grant owner. This fully obsoletes the DL_ENDPT field in the inet/ethernet protocol. IO_ENDPT has other uses besides identifying the grant owner though. This patch renames IO_ENDPT to USER_ENDPT, not only because that is a more fitting name (it should never be used for I/O after all), but also in order to intentionally break any old system source code outside the base system. If this patch breaks your code, fixing it is fairly simple: - DL_ENDPT should be replaced with m_source; - IO_ENDPT should be replaced with m_source when used for safecopies; - IO_ENDPT should be replaced with USER_ENDPT for any other use, e.g. when setting REP_ENDPT, matching requests in CANCEL calls, getting DEV_SELECT flags, and retrieving of the real user process's endpoint in DEV_OPEN. The changes in this patch are binary backward compatible.
2011-04-11 19:35:05 +02:00
dev_mess.USER_ENDPT = proc_e;
Initial revision
2005-04-21 16:53:53 +02:00
dev_mess.COUNT = flags;
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
if (dp->dmap_driver == NONE) {
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
printf("VFS clone_opcl: no driver for dev %x\n", dev);
return(ENXIO);
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
}
if(isokendpt(dp->dmap_driver, &dummyproc) != OK) {
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
printf("VFS clone_opcl: bad driver endpoint for dev %x (%d)\n", dev,
dp->dmap_driver);
return(ENXIO);
endpoint-aware conversion of servers. 'who', indicating caller number in pm and fs and some other servers, has been removed in favour of 'who_e' (endpoint) and 'who_p' (proc nr.). In both PM and FS, isokendpt() convert endpoints to process slot numbers, returning OK if it was a valid and consistent endpoint number. okendpt() does the same but panic()s if it doesn't succeed. (In PM, this is pm_isok..) pm and fs keep their own records of process endpoints in their proc tables, which are needed to make kernel calls about those processes. message field names have changed. fs drivers are endpoints. fs now doesn't try to get out of driver deadlock, as the protocol isn't supposed to let that happen any more. (A warning is printed if ELOCKED is detected though.) fproc[].fp_task (indicating which driver the process is suspended on) became an int. PM and FS now get endpoint numbers of initial boot processes from the kernel. These happen to be the same as the old proc numbers, to let user processes reach them with the old numbers, but FS and PM don't know that. All new processes after INIT, even after the generation number wraps around, get endpoint numbers with generation 1 and higher, so the first instances of the boot processes are the only processes ever to have endpoint numbers in the old proc number range. More return code checks of sys_* functions have been added. IS has become endpoint-aware. Ditched the 'text' and 'data' fields in the kernel dump (which show locations, not sizes, so aren't terribly useful) in favour of the endpoint number. Proc number is still visible. Some other dumps (e.g. dmap, rs) show endpoint numbers now too which got the formatting changed. PM reading segments using rw_seg() has changed - it uses other fields in the message now instead of encoding the segment and process number and fd in the fd field. For that it uses _read_pm() and _write_pm() which to _taskcall()s directly in pm/misc.c. PM now sys_exit()s itself on panic(), instead of sys_abort(). RS also talks in endpoints instead of process numbers.
2006-03-03 11:20:58 +01:00
}
Initial revision
2005-04-21 16:53:53 +02:00
/* Call the task. */
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
r = (*dp->dmap_io)(dp->dmap_driver, &dev_mess);
if (r != OK) return(r);
Initial revision
2005-04-21 16:53:53 +02:00
if (op == DEV_OPEN && dev_mess.REP_STATUS >= 0) {
if (dev_mess.REP_STATUS != minor) {
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
struct vnode *vp;
struct node_details res;
New VFS interface
2007-08-07 14:52:47 +02:00
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
/* A new minor device number has been returned.
Move allocation of temporary inodes for cloned character special devices from MFS to PFS.
2010-03-30 17:00:09 +02:00
* Request PFS to create a temporary device file to hold it.
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
*/
/* Device number of the new device. */
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
dev = (dev & ~(BYTE << MINOR)) | (dev_mess.REP_STATUS << MINOR);
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
/* Issue request */
Move allocation of temporary inodes for cloned character special devices from MFS to PFS.
2010-03-30 17:00:09 +02:00
r = req_newnode(PFS_PROC_NR, fp->fp_effuid, fp->fp_effgid,
New VFS interface
2007-08-07 14:52:47 +02:00
ALL_MODES | I_CHAR_SPECIAL, dev, &res);
if (r != OK) {
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
(void) clone_opcl(DEV_CLOSE, dev, proc_e, 0);
return r;
}
/* Drop old node and use the new values */
vp = fp->fp_filp[m_in.fd]->filp_vno;
put_vnode(vp);
Use of all NIL_* defines converted to NULL
2010-05-10 15:26:00 +02:00
if ((vp = get_free_vnode()) == NULL)
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
vp = fp->fp_filp[m_in.fd]->filp_vno;
vp->v_fs_e = res.fs_e;
Use of all NIL_* defines converted to NULL
2010-05-10 15:26:00 +02:00
vp->v_vmnt = NULL;
Move allocation of temporary inodes for cloned character special devices from MFS to PFS.
2010-03-30 17:00:09 +02:00
vp->v_dev = NO_DEV;
New VFS interface
2007-08-07 14:52:47 +02:00
vp->v_fs_e = res.fs_e;
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
vp->v_inode_nr = res.inode_nr;
vp->v_mode = res.fmode;
vp->v_sdev = dev;
VFS cleanup (mostly open).
2007-01-05 17:36:55 +01:00
vp->v_fs_count = 1;
vp->v_ref_count = 1;
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
fp->fp_filp[m_in.fd]->filp_vno = vp;
Initial revision
2005-04-21 16:53:53 +02:00
}
dev_mess.REP_STATUS = OK;
}
return(dev_mess.REP_STATUS);
}
FS understanding revives and selects as alerts (new notifies)
2005-07-27 15:08:52 +02:00
New VFS interface
2007-08-07 14:52:47 +02:00
Two 'dynamic driver' features in FS: . When drivers disappear that have pending select()s, wake up those user processes with EAGAIN so that they can retry their select() and won't hang forever on it. . When drivers re-appear and are mapped into the dmap, run through the list of mounted filesystems and re-dev_open() every one (for partition tables and such). This can't happen before the driver has exec()ced itself, so processes that have fork()ed but not exec()ced yet are marked as DMAP_BABY in the dmap table if they are dmapped before they are execced. If that happens, the above procedure happens after the exec(). If the exec() happens before the dmapping, it (the dev_open()ing) happens right away.
2005-10-20 21:39:32 +02:00
/*===========================================================================*
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
* bdev_up *
Two 'dynamic driver' features in FS: . When drivers disappear that have pending select()s, wake up those user processes with EAGAIN so that they can retry their select() and won't hang forever on it. . When drivers re-appear and are mapped into the dmap, run through the list of mounted filesystems and re-dev_open() every one (for partition tables and such). This can't happen before the driver has exec()ced itself, so processes that have fork()ed but not exec()ced yet are marked as DMAP_BABY in the dmap table if they are dmapped before they are execced. If that happens, the above procedure happens after the exec(). If the exec() happens before the dmapping, it (the dev_open()ing) happens right away.
2005-10-20 21:39:32 +02:00
*===========================================================================*/
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
PUBLIC void bdev_up(int maj)
Two 'dynamic driver' features in FS: . When drivers disappear that have pending select()s, wake up those user processes with EAGAIN so that they can retry their select() and won't hang forever on it. . When drivers re-appear and are mapped into the dmap, run through the list of mounted filesystems and re-dev_open() every one (for partition tables and such). This can't happen before the driver has exec()ced itself, so processes that have fork()ed but not exec()ced yet are marked as DMAP_BABY in the dmap table if they are dmapped before they are execced. If that happens, the above procedure happens after the exec(). If the exec() happens before the dmapping, it (the dev_open()ing) happens right away.
2005-10-20 21:39:32 +02:00
{
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
/* A new block device driver has been mapped in. This may affect both mounted
* file systems and open block-special files.
*/
VFS/FS: REQ_NEW_DRIVER now provides a label
2011-11-30 19:05:26 +01:00
int r, found, bits;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
struct filp *fp;
struct vmnt *vmp;
struct vnode *vp;
VFS/FS: REQ_NEW_DRIVER now provides a label
2011-11-30 19:05:26 +01:00
char *label;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
VFS/FS: REQ_NEW_DRIVER now provides a label
2011-11-30 19:05:26 +01:00
label = dmap[maj].dmap_label;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
VFS/FS: REQ_NEW_DRIVER now provides a label
2011-11-30 19:05:26 +01:00
/* Tell each affected mounted file system about the new driver. This code
VFS: bugfixes for handling block-special files: - on driver restarts, reopen devices on a per-file basis, not per-mount - do not assume that there is just one vnode per block-special device - update block-special files in the uncommon mounting success paths, too - upon mount, sync but also invalidate affected buffers on the root FS - upon unmount, check whether a vnode is in use before updating it
2011-03-25 11:56:43 +01:00
* is currently useless, as driver endpoints do not change across restarts.
*/
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
for (vmp = &vmnt[0]; vmp < &vmnt[NR_MNTS]; ++vmp) {
VFS: bugfixes for handling block-special files: - on driver restarts, reopen devices on a per-file basis, not per-mount - do not assume that there is just one vnode per block-special device - update block-special files in the uncommon mounting success paths, too - upon mount, sync but also invalidate affected buffers on the root FS - upon unmount, check whether a vnode is in use before updating it
2011-03-25 11:56:43 +01:00
if (major(vmp->m_dev) != maj) continue;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
VFS/FS: REQ_NEW_DRIVER now provides a label
2011-11-30 19:05:26 +01:00
/* Send the driver label to the mounted file system. */
if (OK != req_newdriver(vmp->m_fs_e, vmp->m_dev, label))
printf("VFSdev_up: error sending new driver label to %d\n",
vmp->m_fs_e);
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
}
Merge of VFS by Balasz Gerofi with Minix trunk.
2006-10-25 15:40:36 +02:00
VFS: bugfixes for handling block-special files: - on driver restarts, reopen devices on a per-file basis, not per-mount - do not assume that there is just one vnode per block-special device - update block-special files in the uncommon mounting success paths, too - upon mount, sync but also invalidate affected buffers on the root FS - upon unmount, check whether a vnode is in use before updating it
2011-03-25 11:56:43 +01:00
/* For each block-special file that was previously opened on the affected
* device, we need to reopen it on the new driver.
*/
found = 0;
for (fp = filp; fp < &filp[NR_FILPS]; fp++) {
if(fp->filp_count < 1 || !(vp = fp->filp_vno)) continue;
if(major(vp->v_sdev) != maj) continue;
if(!S_ISBLK(vp->v_mode)) continue;
/* Reopen the device on the driver, once per filp. */
vfs/avfs: map O_ACCMODE to R_BIT|W_BIT on recovery
2011-11-24 13:57:36 +01:00
bits = mode_map[fp->filp_mode & O_ACCMODE];
if ((r = bdev_open(vp->v_sdev, bits)) != OK)
VFS: bugfixes for handling block-special files: - on driver restarts, reopen devices on a per-file basis, not per-mount - do not assume that there is just one vnode per block-special device - update block-special files in the uncommon mounting success paths, too - upon mount, sync but also invalidate affected buffers on the root FS - upon unmount, check whether a vnode is in use before updating it
2011-03-25 11:56:43 +01:00
printf("VFS: mounted dev %d/%d re-open failed: %d.\n",
maj, minor(vp->v_sdev), r);
found = 1;
}
/* If any block-special file was open for this major at all, also inform the
VFS/FS: REQ_NEW_DRIVER now provides a label
2011-11-30 19:05:26 +01:00
* root file system about the endpoint update of the driver. We do this even
* if the block-special file is linked to another mounted file system, merely
VFS: bugfixes for handling block-special files: - on driver restarts, reopen devices on a per-file basis, not per-mount - do not assume that there is just one vnode per block-special device - update block-special files in the uncommon mounting success paths, too - upon mount, sync but also invalidate affected buffers on the root FS - upon unmount, check whether a vnode is in use before updating it
2011-03-25 11:56:43 +01:00
* because it is more work to check for that case.
*/
if (found) {
VFS/FS: REQ_NEW_DRIVER now provides a label
2011-11-30 19:05:26 +01:00
if (OK != req_newdriver(ROOT_FS_E, makedev(maj, 0), label))
printf("VFSdev_up: error sending new driver label to %d\n",
ROOT_FS_E);
VFS: bugfixes for handling block-special files: - on driver restarts, reopen devices on a per-file basis, not per-mount - do not assume that there is just one vnode per block-special device - update block-special files in the uncommon mounting success paths, too - upon mount, sync but also invalidate affected buffers on the root FS - upon unmount, check whether a vnode is in use before updating it
2011-03-25 11:56:43 +01:00
}
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
}
VFS: bugfixes for handling block-special files: - on driver restarts, reopen devices on a per-file basis, not per-mount - do not assume that there is just one vnode per block-special device - update block-special files in the uncommon mounting success paths, too - upon mount, sync but also invalidate affected buffers on the root FS - upon unmount, check whether a vnode is in use before updating it
2011-03-25 11:56:43 +01:00
Split block/character protocols and libdriver This patch separates the character and block driver communication protocols. The old character protocol remains the same, but a new block protocol is introduced. The libdriver library is replaced by two new libraries: libchardriver and libblockdriver. Their exposed API, and drivers that use them, have been updated accordingly. Together, libbdev and libblockdriver now completely abstract away the message format used by the block protocol. As the memory driver is both a character and a block device driver, it now implements its own message loop. The most important semantic change made to the block protocol is that it is no longer possible to return both partial results and an error for a single transfer. This simplifies the interaction between the caller and the driver, as the I/O vector no longer needs to be copied back. Also, drivers are now no longer supposed to decide based on the layout of the I/O vector when a transfer should be cut short. Put simply, transfers are now supposed to either succeed completely, or result in an error. After this patch, the state of the various pieces is as follows: - block protocol: stable - libbdev API: stable for synchronous communication - libblockdriver API: needs slight revision (the drvlib/partition API in particular; the threading API will also change shortly) - character protocol: needs cleanup - libchardriver API: needs cleanup accordingly - driver restarts: largely unsupported until endpoint changes are reintroduced As a side effect, this patch eliminates several bugs, hacks, and gcc -Wall and -W warnings all over the place. It probably introduces a few new ones, too. Update warning: this patch changes the protocol between MFS and disk drivers, so in order to use old/new images, the MFS from the ramdisk must be used to mount all file systems.
2011-11-22 13:27:53 +01:00
/*===========================================================================*
* cdev_up *
*===========================================================================*/
PUBLIC void cdev_up(int maj)
{
/* A new character device driver has been mapped in.
*/
int needs_reopen, fd_nr;
struct filp *fp;
struct fproc *rfp;
struct vnode *vp;
/* Look for processes that are suspened in an OPEN call. Set SUSP_REOPEN
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
* to indicate that this process was suspended before the call to dev_up.
*/
for (rfp = &fproc[0]; rfp < &fproc[NR_PROCS]; rfp++) {
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if(rfp->fp_pid == PID_FREE) continue;
if(rfp->fp_blocked_on != FP_BLOCKED_ON_DOPEN) continue;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
Removed dependency of vfs on NR_TASKS macro - all macros in consts.h that depend on NR_TASKS replaced by a FP_BLOCKED_ON_* - fp_suspended removed and replaced by fp_blocked_on. Testing whether a process is supended is qeual to testing whether fp_blocked_on is FP_BLOCKED_ON_NONE or not - fp_task is valid only if fp_blocked_on == FP_BLOCKED_ON_OTHER - no need of special values that do not colide with valid and special endpoints since they are not used as endpoints anymore - suspend only takes FP_BLOCKED_ON_* values not endpoints anymore - suspend(task) replaced by wait_for(task) which sets fp_task so we remember who are we waiting for and suspend sets fp_blocked_on to FP_BLOCKED_ON_OTHER to signal that we are waiting for some other process - some functions should take endpoint_t instead of int, fixed
2009-09-22 23:48:26 +02:00
printf("dev_up: found process in FP_BLOCKED_ON_DOPEN, fd %d\n",
vfs - split fp_fd field into fd + callnr fields
2010-07-22 16:55:28 +02:00
rfp->fp_block_fd);
fd_nr = rfp->fp_block_fd;
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
fp = rfp->fp_filp[fd_nr];
vp = fp->filp_vno;
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
if (!vp) panic("restart_reopen: no vp");
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
if ((vp->v_mode & I_TYPE) != I_CHAR_SPECIAL) continue;
if (((vp->v_sdev >> MAJOR) & BYTE) != maj) continue;
rfp->fp_flags |= SUSP_REOPEN;
}
needs_reopen= FALSE;
for (fp = filp; fp < &filp[NR_FILPS]; fp++) {
if(fp->filp_count < 1 || !(vp = fp->filp_vno)) continue;
if(((vp->v_sdev >> MAJOR) & BYTE) != maj) continue;
VFS: bugfixes for handling block-special files: - on driver restarts, reopen devices on a per-file basis, not per-mount - do not assume that there is just one vnode per block-special device - update block-special files in the uncommon mounting success paths, too - upon mount, sync but also invalidate affected buffers on the root FS - upon unmount, check whether a vnode is in use before updating it
2011-03-25 11:56:43 +01:00
if(!S_ISCHR(vp->v_mode)) continue;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
fp->filp_state = FS_NEEDS_REOPEN;
needs_reopen = TRUE;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
}
if (needs_reopen)
restart_reopen(maj);
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
/*===========================================================================*
* restart_reopen *
*===========================================================================*/
PRIVATE void restart_reopen(maj)
int maj;
{
int n, r, minor, fd_nr;
endpoint_t driver_e;
struct vnode *vp;
struct filp *fp;
struct fproc *rfp;
for (fp = filp; fp < &filp[NR_FILPS]; fp++) {
if (fp->filp_count < 1 || !(vp = fp->filp_vno)) continue;
if (fp->filp_state != FS_NEEDS_REOPEN) continue;
if (((vp->v_sdev >> MAJOR) & BYTE) != maj) continue;
if ((vp->v_mode & I_TYPE) != I_CHAR_SPECIAL) continue;
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
minor = ((vp->v_sdev >> MINOR) & BYTE);
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (!(fp->filp_flags & O_REOPEN)) {
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
/* File descriptor is to be closed when driver restarts. */
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
n = invalidate(fp);
if (n != fp->filp_count) {
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
printf("VFS: warning: invalidate/count "
"discrepancy (%d, %d)\n", n, fp->filp_count);
}
fp->filp_count = 0;
continue;
}
r = dev_reopen(vp->v_sdev, fp-filp, vp->v_mode & (R_BIT|W_BIT));
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (r == OK) return;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
/* Device could not be reopened. Invalidate all filps on that device.*/
n = invalidate(fp);
if (n != fp->filp_count) {
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
printf("VFS: warning: invalidate/count "
"discrepancy (%d, %d)\n", n, fp->filp_count);
}
fp->filp_count = 0;
printf("VFS: file on dev %d/%d re-open failed: %d; "
"invalidated %d fd's.\n", maj, minor, r, n);
}
/* Nothing more to re-open. Restart suspended processes */
driver_e= dmap[maj].dmap_driver;
for (rfp = &fproc[0]; rfp < &fproc[NR_PROCS]; rfp++) {
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if(rfp->fp_pid == PID_FREE) continue;
Removed dependency of vfs on NR_TASKS macro - all macros in consts.h that depend on NR_TASKS replaced by a FP_BLOCKED_ON_* - fp_suspended removed and replaced by fp_blocked_on. Testing whether a process is supended is qeual to testing whether fp_blocked_on is FP_BLOCKED_ON_NONE or not - fp_task is valid only if fp_blocked_on == FP_BLOCKED_ON_OTHER - no need of special values that do not colide with valid and special endpoints since they are not used as endpoints anymore - suspend only takes FP_BLOCKED_ON_* values not endpoints anymore - suspend(task) replaced by wait_for(task) which sets fp_task so we remember who are we waiting for and suspend sets fp_blocked_on to FP_BLOCKED_ON_OTHER to signal that we are waiting for some other process - some functions should take endpoint_t instead of int, fixed
2009-09-22 23:48:26 +02:00
if(rfp->fp_blocked_on == FP_BLOCKED_ON_OTHER &&
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
rfp->fp_task == driver_e && (rfp->fp_flags & SUSP_REOPEN)) {
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
rfp->fp_flags &= ~SUSP_REOPEN;
Removed dependency of vfs on NR_TASKS macro - all macros in consts.h that depend on NR_TASKS replaced by a FP_BLOCKED_ON_* - fp_suspended removed and replaced by fp_blocked_on. Testing whether a process is supended is qeual to testing whether fp_blocked_on is FP_BLOCKED_ON_NONE or not - fp_task is valid only if fp_blocked_on == FP_BLOCKED_ON_OTHER - no need of special values that do not colide with valid and special endpoints since they are not used as endpoints anymore - suspend only takes FP_BLOCKED_ON_* values not endpoints anymore - suspend(task) replaced by wait_for(task) which sets fp_task so we remember who are we waiting for and suspend sets fp_blocked_on to FP_BLOCKED_ON_OTHER to signal that we are waiting for some other process - some functions should take endpoint_t instead of int, fixed
2009-09-22 23:48:26 +02:00
rfp->fp_blocked_on = FP_BLOCKED_ON_NONE;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
reply(rfp->fp_endpoint, ERESTART);
}
}
/* Look for processes that are suspened in an OPEN call */
for (rfp = &fproc[0]; rfp < &fproc[NR_PROCS]; rfp++) {
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (rfp->fp_pid == PID_FREE) continue;
Removed dependency of vfs on NR_TASKS macro - all macros in consts.h that depend on NR_TASKS replaced by a FP_BLOCKED_ON_* - fp_suspended removed and replaced by fp_blocked_on. Testing whether a process is supended is qeual to testing whether fp_blocked_on is FP_BLOCKED_ON_NONE or not - fp_task is valid only if fp_blocked_on == FP_BLOCKED_ON_OTHER - no need of special values that do not colide with valid and special endpoints since they are not used as endpoints anymore - suspend only takes FP_BLOCKED_ON_* values not endpoints anymore - suspend(task) replaced by wait_for(task) which sets fp_task so we remember who are we waiting for and suspend sets fp_blocked_on to FP_BLOCKED_ON_OTHER to signal that we are waiting for some other process - some functions should take endpoint_t instead of int, fixed
2009-09-22 23:48:26 +02:00
if (rfp->fp_blocked_on == FP_BLOCKED_ON_DOPEN ||
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
!(rfp->fp_flags & SUSP_REOPEN)) continue;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
Removed dependency of vfs on NR_TASKS macro - all macros in consts.h that depend on NR_TASKS replaced by a FP_BLOCKED_ON_* - fp_suspended removed and replaced by fp_blocked_on. Testing whether a process is supended is qeual to testing whether fp_blocked_on is FP_BLOCKED_ON_NONE or not - fp_task is valid only if fp_blocked_on == FP_BLOCKED_ON_OTHER - no need of special values that do not colide with valid and special endpoints since they are not used as endpoints anymore - suspend only takes FP_BLOCKED_ON_* values not endpoints anymore - suspend(task) replaced by wait_for(task) which sets fp_task so we remember who are we waiting for and suspend sets fp_blocked_on to FP_BLOCKED_ON_OTHER to signal that we are waiting for some other process - some functions should take endpoint_t instead of int, fixed
2009-09-22 23:48:26 +02:00
printf("restart_reopen: found process in FP_BLOCKED_ON_DOPEN, fd %d\n",
vfs - split fp_fd field into fd + callnr fields
2010-07-22 16:55:28 +02:00
rfp->fp_block_fd);
fd_nr = rfp->fp_block_fd;
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
fp = rfp->fp_filp[fd_nr];
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (!fp) {
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
/* Open failed, and automatic reopen was not requested */
Removed dependency of vfs on NR_TASKS macro - all macros in consts.h that depend on NR_TASKS replaced by a FP_BLOCKED_ON_* - fp_suspended removed and replaced by fp_blocked_on. Testing whether a process is supended is qeual to testing whether fp_blocked_on is FP_BLOCKED_ON_NONE or not - fp_task is valid only if fp_blocked_on == FP_BLOCKED_ON_OTHER - no need of special values that do not colide with valid and special endpoints since they are not used as endpoints anymore - suspend only takes FP_BLOCKED_ON_* values not endpoints anymore - suspend(task) replaced by wait_for(task) which sets fp_task so we remember who are we waiting for and suspend sets fp_blocked_on to FP_BLOCKED_ON_OTHER to signal that we are waiting for some other process - some functions should take endpoint_t instead of int, fixed
2009-09-22 23:48:26 +02:00
rfp->fp_blocked_on = FP_BLOCKED_ON_NONE;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
FD_CLR(fd_nr, &rfp->fp_filp_inuse);
reply(rfp->fp_endpoint, EIO);
continue;
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
vp = fp->filp_vno;
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
if (!vp) panic("restart_reopen: no vp");
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
if ((vp->v_mode & I_TYPE) != I_CHAR_SPECIAL) continue;
if (((vp->v_sdev >> MAJOR) & BYTE) != maj) continue;
Removed dependency of vfs on NR_TASKS macro - all macros in consts.h that depend on NR_TASKS replaced by a FP_BLOCKED_ON_* - fp_suspended removed and replaced by fp_blocked_on. Testing whether a process is supended is qeual to testing whether fp_blocked_on is FP_BLOCKED_ON_NONE or not - fp_task is valid only if fp_blocked_on == FP_BLOCKED_ON_OTHER - no need of special values that do not colide with valid and special endpoints since they are not used as endpoints anymore - suspend only takes FP_BLOCKED_ON_* values not endpoints anymore - suspend(task) replaced by wait_for(task) which sets fp_task so we remember who are we waiting for and suspend sets fp_blocked_on to FP_BLOCKED_ON_OTHER to signal that we are waiting for some other process - some functions should take endpoint_t instead of int, fixed
2009-09-22 23:48:26 +02:00
rfp->fp_blocked_on = FP_BLOCKED_ON_NONE;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
reply(rfp->fp_endpoint, fd_nr);
}
}
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
/*===========================================================================*
* reopen_reply *
*===========================================================================*/
PUBLIC void reopen_reply()
{
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
endpoint_t driver_e;
int filp_no, status, maj;
struct filp *fp;
struct vnode *vp;
struct dmap *dp;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
driver_e = m_in.m_source;
filp_no = m_in.REP_ENDPT;
status = m_in.REP_STATUS;
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (filp_no < 0 || filp_no >= NR_FILPS) {
printf("reopen_reply: bad filp number %d from driver %d\n", filp_no,
driver_e);
return;
}
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
fp = &filp[filp_no];
if (fp->filp_count < 1) {
printf("reopen_reply: filp number %d not inuse (from driver %d)\n",
filp_no, driver_e);
return;
}
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
vp = fp->filp_vno;
if (!vp) {
printf("reopen_reply: no vnode for filp number %d (from driver %d)\n",
filp_no, driver_e);
return;
}
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if (fp->filp_state != FS_NEEDS_REOPEN) {
printf("reopen_reply: bad state %d for filp number %d"
" (from driver %d)\n", fp->filp_state, filp_no, driver_e);
return;
}
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
if ((vp->v_mode & I_TYPE) != I_CHAR_SPECIAL) {
printf("reopen_reply: bad mode 0%o for filp number %d"
" (from driver %d)\n", vp->v_mode, filp_no, driver_e);
return;
}
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
- Introduce support for sticky bit. - Revise VFS-FS protocol and update VFS/MFS/ISOFS accordingly. - Clean up MFS by removing old, dead code (backwards compatibility is broken by the new VFS-FS protocol, anyway) and rewrite other parts. Also, make sure all functions have proper banners and prototypes. - VFS should always provide a (syntactically) valid path to the FS; no need for the FS to do sanity checks when leaving/entering mount points. - Fix several bugs in MFS: - Several path lookup bugs in MFS. - A link can be too big for the path buffer. - A mountpoint can become inaccessible when the creation of a new inode fails, because the inode already exists and is a mountpoint. - Introduce support for supplemental groups. - Add test 46 to test supplemental group functionality (and removed obsolete suppl. tests from test 2). - Clean up VFS (not everything is done yet). - ISOFS now opens device read-only. This makes the -r flag in the mount command unnecessary (but will still report to be mounted read-write). - Introduce PipeFS. PipeFS is a new FS that handles all anonymous and named pipes. However, named pipes still reside on the (M)FS, as they are part of the file system on disk. To make this work VFS now has a concept of 'mapped' inodes, which causes read, write, truncate and stat requests to be redirected to the mapped FS, and all other requests to the original FS.
2009-12-20 21:27:14 +01:00
maj = ((vp->v_sdev >> MAJOR) & BYTE);
dp = &dmap[maj];
if (dp->dmap_driver != driver_e) {
printf("reopen_reply: bad major %d for filp number %d "
"(from driver %d, current driver is %d)\n", maj, filp_no,
driver_e, dp->dmap_driver);
return;
}
if (status == OK) {
fp->filp_state= FS_NORMAL;
} else {
printf("reopen_reply: should handle error status\n");
return;
}
restart_reopen(maj);
Asynchronous communication with character specials.
2008-02-22 16:41:07 +01:00
}
Reference in a new issue Copy permalink