| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2008-03-24 | Change XXgetdisklabel() to the 'normal' four parameter model to | Kenneth R Westerback | |
| prepare for adding missing DIOC* ioctl's. No functional change. | |||
| 2007-10-15 | more remove unneeded declarations that shadows existing vars; ok by many. | Federico G. Schwindt | |
| 2007-09-29 | make safe vnds work on block devices. | Felix Kronlage | |
| from Piotr Durlej, tested by henning@, krw@, thib@ and myself ok thib@, grunk@, krw@ | |||
| 2007-09-12 | There is no such word as `detatch'. | Michael Knudsen | |
| ok mglocker | |||
| 2007-09-11 | KNF | Gilles Chehade | |
| prompted and "much better" by marco@, ok pyr@ | |||
| 2007-09-08 | more M_ZERO changes | Gilles Chehade | |
| ok pyr@ and krw@ | |||
| 2007-06-20 | b_cylinder does not need to be set on the callpath down into drivers. | Theo de Raadt | |
| cpu_disklabel can go away, since nothing anymore needs to use it; ok miod | |||
| 2007-06-08 | all drivers should spoof version 1 labels | Theo de Raadt | |
| 2007-06-06 | now that all partition size/offsets are potentially 64-bit, change the | Theo de Raadt | |
| type of all variables to daddr64_t. this includes the APIs for XXsize() and XXdump(), all range checks inside bio drivers, internal variables for disklabel handling, and even uvm's swap offsets. re-read numerous times by otto, miod, krw, thib to look for errors | |||
| 2007-06-05 | use six new macros to access & store the 48-bit disklabel fields related | Theo de Raadt | |
| to size. tested on almost all machines, double checked by miod and krw next comes the type handling surrounding these values | |||
| 2007-06-01 | Don't initialize d_partitions[RAW_DISK] just before calling | Kenneth R Westerback | |
| readdisklabel(), since all readdisklabel()'s do that already. ok deraadt@ | |||
| 2007-05-12 | Use VNDLABELDEV() and not DISKLABELDEV() in writedisklabel() since | Kenneth R Westerback | |
| VNDLABELDEV() is used in readdisklabel(). VNDLABELDEV() preserves the vndsimple() property encoded in the device. | |||
| 2007-04-29 | Replace expansions of DISKLABELDEV() with DISKLABELDEV(). Shorter, and | Kenneth R Westerback | |
| more consistant. No change to code. ok miod@ | |||
| 2007-03-25 | include rwlock.h ourselves, better than depending on it getting picked | Ted Unangst | |
| up elsewhere | |||
| 2007-03-25 | use a rwlock instead of the homegrown approach | Ted Unangst | |
| ok art pedro thib | |||
| 2007-02-26 | Don't enforce RLIMIT_FSIZE on vnd(4) I/O operations, okay deraadt@ | Pedro Martelletto | |
| 2007-02-21 | Don't use arbitrary 128 bytes as size of the blf key array if we know exactly | Alexander von Gernler | |
| how much blowfish takes at max, which is 72 bytes. Also define a constant for this in the include file, suggested by ray@. ok pedro@ thib@ tedu@ | |||
| 2007-02-01 | Don't throttle and serialize I/O operations, okay miod@ krw@ | Pedro Martelletto | |
| Fixes kern/5347 | |||
| 2007-01-13 | Touch of clean up: | Thordur I. Bjornsson | |
| s/DEBUG/VNDDEBUG/ and introduce DNPRINTF, fold all the #ifdef DEBUG <stuff> #endif into DNPRINTFs. Use DTYPE_VND for the disklabel d_type member, dont calcluate d_secpercyl from d_ntrack * d_nsectors, just set it too 100 directly since we set d_ntrack and d_nsectors to 1 and 100 respectivly. testing and OK krw@ | |||
| 2006-12-24 | ansify, no binary change, from Bret Lambert | Pedro Martelletto | |
| 2006-10-05 | Always allow read-only opens on (s)vnd devices, despite the | Thordur I. Bjornsson | |
| type of the first open (svnd vs vnd). ok pedro@ | |||
| 2006-10-03 | Introduce daddr64_t and use it for physical block numbers | Pedro Martelletto | |
| Okay weingart@, "I'm game with putting my name on it" dlg@ | |||
| 2006-09-20 | Make vnd(4) work on read-only file systems, from Paul Stoeber, okay tedu@ | Pedro Martelletto | |
| 2006-08-13 | fix a panic when trying to fdisk vnd devices (not svnd) wich | Thordur I. Bjornsson | |
| are back by weird images (such as the ones qemu creates). proplem found the hard way by msf@ ok mickey@, pedro@ | |||
| 2006-08-12 | Setting d_secsize to DEV_BSIZE (or 1 << DEV_BSHIFT, or 512) and then | Kenneth R Westerback | |
| setting RAW_PART's p_size to d_secperunit * (d_secsize / DEV_BSIZE) is a waste of a few ops. And p_size should be in sectors anyway. Just set RAW_PART's p_size to d_secperunit to make usage consistant across the tree. Should be a no-op. | |||
| 2006-03-15 | Nuke dk_establish(), no longer used. | Miod Vallat | |
| 2006-03-04 | Disk interrupts can cause buffers to be returned to the vnd pool, so | Pedro Martelletto | |
| splbio() is required when allocating. Fixes PR kern/5041, okay tedu@. | |||
| 2006-01-22 | b_cylin -> b_cylinder; no functional change. | Miod Vallat | |
| 2005-12-29 | Correctly bounds check transfers passed to vndstrategy() | Pedro Martelletto | |
| 2005-07-20 | change sc_keyctx from void * to blf_ctx *. there's no point for it to be | Ted Unangst | |
| void and means we can't correctly use sizeof() on the target. error report from frank denis in pr4290. | |||
| 2005-04-19 | use pool for struct vndbug; tested by many and pedro@ ok | Michael Shalayeff | |
| 2005-04-12 | simplify vndopen() a bit. | Joris Vink | |
| ok pedro@ | |||
| 2005-01-05 | copyoutstr doesn't work so well with the kernel addresses | Ted Unangst | |
| 2004-06-30 | since we're not using 'len' for anything, nuke it and pass null to | Pedro Martelletto | |
| copyin/copyout. suggested by millert@, ok'd by him and deraadt@ | |||
| 2004-06-30 | don't strlcpy() to user-space, use copyoutstr() instead. ok deraadt@ | Pedro Martelletto | |
| 2004-06-28 | You *cannot* strlcpy from userland. Pedro, go read this and find out what | Theo de Raadt | |
| you did wrong. Bit in other changes that were backed out. | |||
| 2004-06-27 | back out crud that breaks the macppc | Theo de Raadt | |
| 2004-06-20 | add the VNDIOCGET ioctl to vnode disks, ok millert@ | Pedro Martelletto | |
| 2004-06-19 | save curproc for further reference, otherwise we may end up in lockmgr() | Pedro Martelletto | |
| with no process context. ok miod@. closes pr 3214. | |||
| 2004-05-22 | Do not truncate intermediate results computing aiov.iov_len in vndsetcred(); | Otto Moerbeek | |
| print size_t as unsigned long long. ok tedu@ miod@ | |||
| 2004-03-04 | 3 buglets. | Ted Unangst | |
| if keylen < 0, bad things happen. large keys are useless, limit size and use stack for temp storage bzero of keyctx is unrelated to keylen, wipe whole thing. ok markus tdeval | |||
| 2004-03-03 | Make sure than svnd read their disklabel as svnd, not vnd devices. | Miod Vallat | |
| ok tedu@ | |||
| 2004-02-15 | knf | Ted Unangst | |
| 2004-02-15 | new arg to disk_unbusy, to record separate read/write statistics. | Ted Unangst | |
| looked at by various, testing henning@ mcbride@ dan weeks mostly from netbsd via Pedro Martelletto <pbastos@rdc.puc-rio.br> | |||
| 2003-10-17 | vndunit() wasn't quite right after last change, limiting vnds to 16. | Ted Unangst | |
| apply fix from tom cosgrove to get the full range. | |||
| 2003-10-11 | bump minor gap for svnd devices, so more vnds are allowed. | Ted Unangst | |
| this change REQUIRES running MAKEDEV again after booting with a new kernel. with a hint from tom cosgrove. ok deraadt@ millert@ | |||
| 2003-08-15 | change arguments to suser. suser now takes the process, and a flags | Ted Unangst | |
| argument. old cred only calls user suser_ucred. this will allow future work to more flexibly implement the idea of a root process. looks like something i saw in freebsd, but a little different. use of suser_ucred vs suser in file system code should be looked at again, for the moment semantics remain unchanged. review and input from art@ testing and further review miod@ | |||
| 2003-06-02 | Remove the advertising clause in the UCB license which Berkeley | Todd C. Miller | |
| rescinded 22 July 1999. Proofed by myself and Theo. | |||
| 2003-04-19 | Some string cleanup: sprintf -> snprintf and magic numbers to sizeof | Kenneth R Westerback | |
| (remember: d_[type|pack]name do not need terminating null). Take the opportunity to introduce some paranoia and check the device name lengths to make sure they fit, with appropriate errors if not. ok tedu@ tdeval@ | |||
| 2003-04-06 | The return value for bounds_check_with_label() is not set in stone, and it | Miod Vallat | |
| turns out that some arches will return (-1) in case of error, while others will return (0). Until we make our minds on this, make sure that callers of this function properly handle 0 or (-1) as failure. ok krw@ | |||
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |