Age | Commit message (Collapse) | Author | |
---|---|---|---|
2010-05-21 | Make compile with NBIO = 0 | Marco Peereboom | |
2010-05-20 | New scsi code seems to be stable. Pluck previously identified | Kenneth R Westerback | |
low-hanging splbio/splx pairs that are no longer needed and see if this reveals any hidden scsi flaws. ok dlg@ | |||
2010-05-18 | dont let sys/ioctl.h imply that you get the ioctls in dkio.h. this | David Gwynne | |
gets rid of #include <sys/dkio.h> in sys/ioctl.h and adds #include <sys/dkio.h> to the places that actually want and use the disk ioctls. this became an issue when krw@'s X build failed when he was testing a change to dkio.h. tested by krw@ help from and ok miod@ | |||
2010-03-28 | Completely overhaul optional metadata handling, allowing for multiple | Joel Sing | |
optional metadata per volume and discipline specific optional metadata processing. ok marco@ | |||
2010-03-28 | Avoid potential null dereferences in ioctl paths. | Joel Sing | |
Found by jsg@ using the clang static analyser. ok marco@ | |||
2010-03-28 | Replace a scsi_done() call with sr_scsi_done() which fixes a missing | Joel Sing | |
splx() call. Found by jsg@ using the clang static analyser. ok marco@ | |||
2010-03-27 | Correctly initialise variable to prevent free without malloc in error path. | Joel Sing | |
Found by clang static analyser. | |||
2010-03-26 | Add an ioctl to softraid to allow the boot block and boot loader to be | Joel Sing | |
installed on a softraid volume. This is work in progress but can continue in tree. ok marco@ | |||
2010-03-26 | Add storage for the boot block and boot loader to the softraid metadata. | Joel Sing | |
Also add a new optional metadata type for boot data. This is the first step (of many) towards being able to boot from softraid volumes. WARNING: This version of the softraid metadata is not compatible with previous versions. As a result, any softraid volumes created with older kernels will not assemble. Data on existing softraid volumes should be backed up before upgrading. The volume should then be recreated and the data restored. ok marco@ | |||
2010-03-23 | Change the scsi_cmd function member of scsi_adapter from int to | Kenneth R Westerback | |
void. Use XS_NO_CCB error in the scsi command (xs) to report the NO_CCB condition. Eliminates all SUCCESSFULLY_QUEUED and COMPLETE confusion and untangles the midlayer from the adapter a bit more. Eyes and some fixes by miod@ There may be some compile issues on little used (i.e. I don't have any) drivers but the change is mechanical and thus easy to remedy. ok dlg@ | |||
2010-02-13 | Force chunks into correct order once metadata is attached. This prevents | Joel Sing | |
chunks from being reordered on an existing volume. ok marco@ | |||
2010-02-13 | Unbreak automatic rebuilds from hotspares. | Joel Sing | |
ok marco@ | |||
2010-02-13 | Removed unused variable sml_vn. | Joel Sing | |
ok @marco | |||
2010-02-13 | Pass the number of chunks to sr_meta_attach() rather than recalculating. | Joel Sing | |
ok marco@ | |||
2010-02-08 | Use character raw device to read disklabel during bootprobe, as all | Kenneth R Westerback | |
other devices do. Silences complaints about missing media on removable media devices. ok marco@ jsing@ | |||
2010-01-09 | Zap all setting of ITSDONE in drivers that don't look at it. Nobody | Kenneth R Westerback | |
else cares so it's just noise. Drivers that actually look at ITSDONE are unchanged. ok marco@ (for his files) dlg@ beck@ | |||
2009-12-31 | Add support for key disks. This allows a crypto volume to be constructed | Joel Sing | |
without using a passphrase - instead the encryption mask key is stored on the specified key disk partition (ideally being one on a removable device). This also enables automatic assembly of crypto volumes at boot time. ok marco@ | |||
2009-12-15 | Allow auto assemble flag to be toggled for disciplines that support it. | Joel Sing | |
"Love it!" marco@ | |||
2009-12-15 | Factor out discipline specific create/assemble code. | Joel Sing | |
"in, in, in!" marco@ | |||
2009-12-15 | Explicitly lock the vnode when deleting a volume. This avoids a potential | Joel Sing | |
panic that is caused by vnode aliasing, where checkalias() switches the vnode ops for our vnode. This means that we end up using ufs unlocking rather than specfs unlocking (which is a no-op). ok marco@ | |||
2009-12-07 | Cleanup discipline initialisation. | Joel Sing | |
ok marco@ | |||
2009-12-07 | Define discipline capabilities using a set of flags. | Joel Sing | |
"shiny!!" marco@ | |||
2009-11-24 | Allow the passphrase to be changed on softraid crypto volumes. Ensure that | Joel Sing | |
you backup your data and lock up your pets prior to using this. Tested by todd@ ok marco@ | |||
2009-11-23 | Improve discipline specific ioctl framework and attach to softraid ioctl. | Joel Sing | |
ok marco@ | |||
2009-11-22 | Add a framework for discipline specific ioctls. | Joel Sing | |
ok marco@ | |||
2009-11-22 | Ensure that chunks are not already in use when creating a volume. | Joel Sing | |
ok marco@ | |||
2009-11-15 | Fix memory leak in error case. Found by Parfait. | Joel Sing | |
ok marco@ deraadt@ | |||
2009-10-29 | When probing devices at boot, compare against the full device driver name | Joel Sing | |
rather than the start of the device name. Also, be inclusive with devices probed rather than exclusive - this means we now only look for RAID partitions on sd(4) and wd(4) devices. prodded by deraadt@ ok marco@ deraadt@ | |||
2009-10-28 | Remove FWRITE from closes that aren't opened with FWRITE. | Marco Peereboom | |
spotted and ok jsing | |||
2009-10-13 | Get rid of devact enum, substitute it with an int and coresponding defines. | Paul Irofti | |
This is needed for the addition of further suspend/resume actions. Okay deraadt@, marco@. | |||
2009-09-21 | Fix use before init on wu. | Marco Peereboom | |
Found by jsing, tested and ok jsing. | |||
2009-09-03 | Check the return value of copyin(). | Joel Sing | |
ok marco@ | |||
2009-09-03 | Don't call VOP_CLOSE() with a null pointer, if the VOP_OPEN() call failed. | Joel Sing | |
ok marco@ | |||
2009-08-17 | the null check makes more sense when it's done before deref | Martynas Venckus | |
ok marco@, jsing@ | |||
2009-08-09 | Switch softraid to vnodes to prevent bad things from happening when using | Marco Peereboom | |
d_open/d_close. tested by many, ok jsing, thib, krw | |||
2009-07-31 | Factor out code for checking if a chunk is already in use. | Joel Sing | |
ok marco@ | |||
2009-07-23 | Adding disabled framework for RAID6 | Jordan Hargrave | |
ok marco@ | |||
2009-07-12 | Assert copyright. | Joel Sing | |
suggested by and ok marco@ | |||
2009-07-12 | Fix rebuild percentage so that it ranges from 0-99% rather than 1-100%. | Joel Sing | |
ok marco@ | |||
2009-07-12 | Add support for global hotspares to softraid. | Joel Sing | |
ok marco@ | |||
2009-07-12 | If devices have roamed, complain once then update the metadata, just like | Joel Sing | |
we do for roaming volumes. ok marco@ | |||
2009-07-11 | Ensure that we hold the appropriate lock before calling | Joel Sing | |
sr_ioctl_createraid(). ok marco@ | |||
2009-07-09 | Be sure we don't hit a NULL deref. Pointed out by | Marcus Glocker | |
remco at d-compu dot dyndns dot org. OK marco@ | |||
2009-06-26 | Overhaul sr_boot_assembly(). The original code had a number of problems, | Joel Sing | |
including a "chunk id already in use" bug that occurred when assembling more than one volume at boot. Volumes are now kept in order, as are the chunks which they are assembled from. Duplicated chunk IDs are now handled appropriately, with the chunk that has the most recent ondisk metadata being used. Also tested by nicm@ who got bitten by the "chunk id already in use" bug. ok marco@ | |||
2009-06-24 | Reset the rebuild counter when initiating a rebuild. | Joel Sing | |
ok marco@ | |||
2009-06-24 | Abort the current rebuild if the rebuild chunk disappears. Whilst here also | Joel Sing | |
rename the abort flag so that it is more generalised. ok marco@ | |||
2009-06-23 | Make sure that the metadata level matches the requested assembly level when | Joel Sing | |
creating a volume. ok marco@ | |||
2009-06-23 | Fix some more whitespace issues. | Joel Sing | |
2009-06-19 | Either jordan or jsing left some debug prints behind. tsk tsk. | Marco Peereboom | |
pointed out by todd. | |||
2009-06-18 | Check the set state status rather than always assuming a rebuild has been | Joel Sing | |
requested. Also move the rebuild initialisation code into a separate function whilst here. ok marco@ |