diff options
Diffstat (limited to 'Documentation')
| -rw-r--r-- | Documentation/SubmittingPatches | 38 | ||||
| -rw-r--r-- | Documentation/devicetree/bindings/i2c/i2c-efm32.txt | 4 | ||||
| -rw-r--r-- | Documentation/filesystems/cifs/AUTHORS | 1 | ||||
| -rw-r--r-- | Documentation/filesystems/cifs/TODO | 97 | ||||
| -rw-r--r-- | Documentation/i2c/busses/i2c-i801 | 1 | ||||
| -rw-r--r-- | Documentation/i2c/i2c-stub | 23 | ||||
| -rw-r--r-- | Documentation/ioctl/00-INDEX | 2 | ||||
| -rw-r--r-- | Documentation/ioctl/botching-up-ioctls.txt | 219 | ||||
| -rw-r--r-- | Documentation/laptops/freefall.c | 6 |
9 files changed, 310 insertions, 81 deletions
diff --git a/Documentation/SubmittingPatches b/Documentation/SubmittingPatches index dcadffcab2d..0a523c9a5ff 100644 --- a/Documentation/SubmittingPatches +++ b/Documentation/SubmittingPatches @@ -84,18 +84,42 @@ is another popular alternative. 2) Describe your changes. -Describe the technical detail of the change(s) your patch includes. - -Be as specific as possible. The WORST descriptions possible include -things like "update driver X", "bug fix for driver X", or "this patch -includes updates for subsystem X. Please apply." +Describe your problem. Whether your patch is a one-line bug fix or +5000 lines of a new feature, there must be an underlying problem that +motivated you to do this work. Convince the reviewer that there is a +problem worth fixing and that it makes sense for them to read past the +first paragraph. + +Describe user-visible impact. Straight up crashes and lockups are +pretty convincing, but not all bugs are that blatant. Even if the +problem was spotted during code review, describe the impact you think +it can have on users. Keep in mind that the majority of Linux +installations run kernels from secondary stable trees or +vendor/product-specific trees that cherry-pick only specific patches +from upstream, so include anything that could help route your change +downstream: provoking circumstances, excerpts from dmesg, crash +descriptions, performance regressions, latency spikes, lockups, etc. + +Quantify optimizations and trade-offs. If you claim improvements in +performance, memory consumption, stack footprint, or binary size, +include numbers that back them up. But also describe non-obvious +costs. Optimizations usually aren't free but trade-offs between CPU, +memory, and readability; or, when it comes to heuristics, between +different workloads. Describe the expected downsides of your +optimization so that the reviewer can weigh costs against benefits. + +Once the problem is established, describe what you are actually doing +about it in technical detail. It's important to describe the change +in plain English for the reviewer to verify that the code is behaving +as you intend it to. The maintainer will thank you if you write your patch description in a form which can be easily pulled into Linux's source code management system, git, as a "commit log". See #15, below. -If your description starts to get long, that's a sign that you probably -need to split up your patch. See #3, next. +Solve only one problem per patch. If your description starts to get +long, that's a sign that you probably need to split up your patch. +See #3, next. When you submit or resubmit a patch or patch series, include the complete patch description and justification for it. Don't just diff --git a/Documentation/devicetree/bindings/i2c/i2c-efm32.txt b/Documentation/devicetree/bindings/i2c/i2c-efm32.txt index fc15ac51943..50b25c3da18 100644 --- a/Documentation/devicetree/bindings/i2c/i2c-efm32.txt +++ b/Documentation/devicetree/bindings/i2c/i2c-efm32.txt @@ -10,7 +10,7 @@ Required properties : Recommended properties : - clock-frequency : maximal I2C bus clock frequency in Hz. - - efm32,location : Decides the location of the USART I/O pins. + - energymicro,location : Decides the location of the USART I/O pins. Allowed range : [0 .. 6] Example: @@ -23,7 +23,7 @@ Example: clocks = <&cmu clk_HFPERCLKI2C0>; clock-frequency = <100000>; status = "ok"; - efm32,location = <3>; + energymicro,location = <3>; eeprom@50 { compatible = "microchip,24c02"; diff --git a/Documentation/filesystems/cifs/AUTHORS b/Documentation/filesystems/cifs/AUTHORS index ca4a67a0bb1..c98800df677 100644 --- a/Documentation/filesystems/cifs/AUTHORS +++ b/Documentation/filesystems/cifs/AUTHORS @@ -40,6 +40,7 @@ Gunter Kukkukk (testing and suggestions for support of old servers) Igor Mammedov (DFS support) Jeff Layton (many, many fixes, as well as great work on the cifs Kerberos code) Scott Lovenberg +Pavel Shilovsky (for great work adding SMB2 support, and various SMB3 features) Test case and Bug Report contributors ------------------------------------- diff --git a/Documentation/filesystems/cifs/TODO b/Documentation/filesystems/cifs/TODO index 355abcdcda9..066ffddc396 100644 --- a/Documentation/filesystems/cifs/TODO +++ b/Documentation/filesystems/cifs/TODO @@ -1,4 +1,4 @@ -Version 1.53 May 20, 2008 +Version 2.03 August 1, 2014 A Partial List of Missing Features ================================== @@ -7,63 +7,49 @@ Contributions are welcome. There are plenty of opportunities for visible, important contributions to this module. Here is a partial list of the known problems and missing features: -a) Support for SecurityDescriptors(Windows/CIFS ACLs) for chmod/chgrp/chown -so that these operations can be supported to Windows servers +a) SMB3 (and SMB3.02) missing optional features: + - RDMA + - multichannel (started) + - directory leases (improved metadata caching) + - T10 copy offload (copy chunk is only mechanism supported) + - encrypted shares -b) Mapping POSIX ACLs (and eventually NFSv4 ACLs) to CIFS -SecurityDescriptors +b) improved sparse file support -c) Better pam/winbind integration (e.g. to handle uid mapping -better) - -d) Cleanup now unneeded SessSetup code in -fs/cifs/connect.c and add back in NTLMSSP code if any servers -need it - -e) fix NTLMv2 signing when two mounts with different users to same -server. - -f) Directory entry caching relies on a 1 second timer, rather than +c) Directory entry caching relies on a 1 second timer, rather than using FindNotify or equivalent. - (started) -g) quota support (needs minor kernel change since quota calls +d) quota support (needs minor kernel change since quota calls to make it to network filesystems or deviceless filesystems) -h) investigate sync behavior (including syncpage) and check -for proper behavior of intr/nointr - -i) improve support for very old servers (OS/2 and Win9x for example) +e) improve support for very old servers (OS/2 and Win9x for example) Including support for changing the time remotely (utimes command). -j) hook lower into the sockets api (as NFS/SunRPC does) to avoid the +f) hook lower into the sockets api (as NFS/SunRPC does) to avoid the extra copy in/out of the socket buffers in some cases. -k) Better optimize open (and pathbased setfilesize) to reduce the +g) Better optimize open (and pathbased setfilesize) to reduce the oplock breaks coming from windows srv. Piggyback identical file opens on top of each other by incrementing reference count rather than resending (helps reduce server resource utilization and avoid spurious oplock breaks). -l) Improve performance of readpages by sending more than one read -at a time when 8 pages or more are requested. In conjuntion -add support for async_cifs_readpages. - -m) Add support for storing symlink info to Windows servers +h) Add support for storing symlink info to Windows servers in the Extended Attribute format their SFU clients would recognize. -n) Finish fcntl D_NOTIFY support so kde and gnome file list windows +i) Finish inotify support so kde and gnome file list windows will autorefresh (partially complete by Asser). Needs minor kernel vfs change to support removing D_NOTIFY on a file. -o) Add GUI tool to configure /proc/fs/cifs settings and for display of +j) Add GUI tool to configure /proc/fs/cifs settings and for display of the CIFS statistics (started) -p) implement support for security and trusted categories of xattrs +k) implement support for security and trusted categories of xattrs (requires minor protocol extension) to enable better support for SELINUX -q) Implement O_DIRECT flag on open (already supported on mount) +l) Implement O_DIRECT flag on open (already supported on mount) -r) Create UID mapping facility so server UIDs can be mapped on a per +m) Create UID mapping facility so server UIDs can be mapped on a per mount or a per server basis to client UIDs or nobody if no mapping exists. This is helpful when Unix extensions are negotiated to allow better permission checking when UIDs differ on the server @@ -71,28 +57,29 @@ and client. Add new protocol request to the CIFS protocol standard for asking the server for the corresponding name of a particular uid. -s) Add support for CIFS Unix and also the newer POSIX extensions to the -server side for Samba 4. +n) DOS attrs - returned as pseudo-xattr in Samba format (check VFAT and NTFS for this too) + +o) mount check for unmatched uids -t) In support for OS/2 (LANMAN 1.2 and LANMAN2.1 based SMB servers) -need to add ability to set time to server (utimes command) +p) Add support for new vfs entry point for fallocate -u) DOS attrs - returned as pseudo-xattr in Samba format (check VFAT and NTFS for this too) +q) Add tools to take advantage of cifs/smb3 specific ioctls and features +such as "CopyChunk" (fast server side file copy) -v) mount check for unmatched uids +r) encrypted file support -w) Add support for new vfs entry point for fallocate +s) improved stats gathering, tools (perhaps integration with nfsometer?) -x) Fix Samba 3 server to handle Linux kernel aio so dbench with lots of -processes can proceed better in parallel (on the server) +t) allow setting more NTFS/SMB3 file attributes remotely (currently limited to compressed +file attribute via chflags) -y) Fix Samba 3 to handle reads/writes over 127K (and remove the cifs mount -restriction of wsize max being 127K) +u) mount helper GUI (to simplify the various configuration options on mount) -KNOWN BUGS (updated April 24, 2007) + +KNOWN BUGS ==================================== See http://bugzilla.samba.org - search on product "CifsVFS" for -current bug list. +current bug list. Also check http://bugzilla.kernel.org (Product = File System, Component = CIFS) 1) existing symbolic links (Windows reparse points) are recognized but can not be created remotely. They are implemented for Samba and those that @@ -100,30 +87,18 @@ support the CIFS Unix extensions, although earlier versions of Samba overly restrict the pathnames. 2) follow_link and readdir code does not follow dfs junctions but recognizes them -3) create of new files to FAT partitions on Windows servers can -succeed but still return access denied (appears to be Windows -server not cifs client problem) and has not been reproduced recently. -NTFS partitions do not have this problem. -4) Unix/POSIX capabilities are reset after reconnection, and affect -a few fields in the tree connection but we do do not know which -superblocks to apply these changes to. We should probably walk -the list of superblocks to set these. Also need to check the -flags on the second mount to the same share, and see if we -can do the same trick that NFS does to remount duplicate shares. Misc testing to do ================== 1) check out max path names and max path name components against various server types. Try nested symlinks (8 deep). Return max path name in stat -f information -2) Modify file portion of ltp so it can run against a mounted network -share and run it against cifs vfs in automated fashion. +2) Improve xfstest's cifs enablement and adapt xfstests where needed to test +cifs better 3) Additional performance testing and optimization using iozone and similar - there are some easy changes that can be done to parallelize sequential writes, and when signing is disabled to request larger read sizes (larger than negotiated size) and send larger write sizes to modern servers. -4) More exhaustively test against less common servers. More testing -against Windows 9x, Windows ME servers. - +4) More exhaustively test against less common servers diff --git a/Documentation/i2c/busses/i2c-i801 b/Documentation/i2c/busses/i2c-i801 index adf5e33e831..e9c803ea306 100644 --- a/Documentation/i2c/busses/i2c-i801 +++ b/Documentation/i2c/busses/i2c-i801 @@ -25,6 +25,7 @@ Supported adapters: * Intel Avoton (SOC) * Intel Wellsburg (PCH) * Intel Coleto Creek (PCH) + * Intel Wildcat Point (PCH) * Intel Wildcat Point-LP (PCH) * Intel BayTrail (SOC) Datasheets: Publicly available at the Intel website diff --git a/Documentation/i2c/i2c-stub b/Documentation/i2c/i2c-stub index fa4b669c166..a16924fbd28 100644 --- a/Documentation/i2c/i2c-stub +++ b/Documentation/i2c/i2c-stub @@ -2,9 +2,9 @@ MODULE: i2c-stub DESCRIPTION: -This module is a very simple fake I2C/SMBus driver. It implements five +This module is a very simple fake I2C/SMBus driver. It implements six types of SMBus commands: write quick, (r/w) byte, (r/w) byte data, (r/w) -word data, and (r/w) I2C block data. +word data, (r/w) I2C block data, and (r/w) SMBus block data. You need to provide chip addresses as a module parameter when loading this driver, which will then only react to SMBus commands to these addresses. @@ -19,6 +19,14 @@ A pointer register with auto-increment is implemented for all byte operations. This allows for continuous byte reads like those supported by EEPROMs, among others. +SMBus block command support is disabled by default, and must be enabled +explicitly by setting the respective bits (0x03000000) in the functionality +module parameter. + +SMBus block commands must be written to configure an SMBus command for +SMBus block operations. Writes can be partial. Block read commands always +return the number of bytes selected with the largest write so far. + The typical use-case is like this: 1. load this module 2. use i2cset (from the i2c-tools project) to pre-load some data @@ -39,15 +47,18 @@ unsigned long functionality: value 0x1f0000 would only enable the quick, byte and byte data commands. +u8 bank_reg[10] +u8 bank_mask[10] +u8 bank_start[10] +u8 bank_end[10]: + Optional bank settings. They tell which bits in which register + select the active bank, as well as the range of banked registers. + CAVEATS: If your target driver polls some byte or word waiting for it to change, the stub could lock it up. Use i2cset to unlock it. -If the hardware for your driver has banked registers (e.g. Winbond sensors -chips) this module will not work well - although it could be extended to -support that pretty easily. - If you spam it hard enough, printk can be lossy. This module really wants something like relayfs. diff --git a/Documentation/ioctl/00-INDEX b/Documentation/ioctl/00-INDEX index d2fe4d4729e..c1a92578795 100644 --- a/Documentation/ioctl/00-INDEX +++ b/Documentation/ioctl/00-INDEX @@ -1,5 +1,7 @@ 00-INDEX - this file +botching-up-ioctls.txt + - how to avoid botching up ioctls cdrom.txt - summary of CDROM ioctl calls hdio.txt diff --git a/Documentation/ioctl/botching-up-ioctls.txt b/Documentation/ioctl/botching-up-ioctls.txt new file mode 100644 index 00000000000..45fe78c5801 --- /dev/null +++ b/Documentation/ioctl/botching-up-ioctls.txt @@ -0,0 +1,219 @@ +(How to avoid) Botching up ioctls +================================= + +From: http://blog.ffwll.ch/2013/11/botching-up-ioctls.html + +By: Daniel Vetter, Copyright © 2013 Intel Corporation + +One clear insight kernel graphics hackers gained in the past few years is that +trying to come up with a unified interface to manage the execution units and +memory on completely different GPUs is a futile effort. So nowadays every +driver has its own set of ioctls to allocate memory and submit work to the GPU. +Which is nice, since there's no more insanity in the form of fake-generic, but +actually only used once interfaces. But the clear downside is that there's much +more potential to screw things up. + +To avoid repeating all the same mistakes again I've written up some of the +lessons learned while botching the job for the drm/i915 driver. Most of these +only cover technicalities and not the big-picture issues like what the command +submission ioctl exactly should look like. Learning these lessons is probably +something every GPU driver has to do on its own. + + +Prerequisites +------------- + +First the prerequisites. Without these you have already failed, because you +will need to add a a 32-bit compat layer: + + * Only use fixed sized integers. To avoid conflicts with typedefs in userspace + the kernel has special types like __u32, __s64. Use them. + + * Align everything to the natural size and use explicit padding. 32-bit + platforms don't necessarily align 64-bit values to 64-bit boundaries, but + 64-bit platforms do. So we always need padding to the natural size to get + this right. + + * Pad the entire struct to a multiple of 64-bits - the structure size will + otherwise differ on 32-bit versus 64-bit. Having a different structure size + hurts when passing arrays of structures to the kernel, or if the kernel + checks the structure size, which e.g. the drm core does. + + * Pointers are __u64, cast from/to a uintprt_t on the userspace side and + from/to a void __user * in the kernel. Try really hard not to delay this + conversion or worse, fiddle the raw __u64 through your code since that + diminishes the checking tools like sparse can provide. + + +Basics +------ + +With the joys of writing a compat layer avoided we can take a look at the basic +fumbles. Neglecting these will make backward and forward compatibility a real +pain. And since getting things wrong on the first attempt is guaranteed you +will have a second iteration or at least an extension for any given interface. + + * Have a clear way for userspace to figure out whether your new ioctl or ioctl + extension is supported on a given kernel. If you can't rely on old kernels + rejecting the new flags/modes or ioctls (since doing that was botched in the + past) then you need a driver feature flag or revision number somewhere. + + * Have a plan for extending ioctls with new flags or new fields at the end of + the structure. The drm core checks the passed-in size for each ioctl call + and zero-extends any mismatches between kernel and userspace. That helps, + but isn't a complete solution since newer userspace on older kernels won't + notice that the newly added fields at the end get ignored. So this still + needs a new driver feature flags. + + * Check all unused fields and flags and all the padding for whether it's 0, + and reject the ioctl if that's not the case. Otherwise your nice plan for + future extensions is going right down the gutters since someone will submit + an ioctl struct with random stack garbage in the yet unused parts. Which + then bakes in the ABI that those fields can never be used for anything else + but garbage. + + * Have simple testcases for all of the above. + + +Fun with Error Paths +-------------------- + +Nowadays we don't have any excuse left any more for drm drivers being neat +little root exploits. This means we both need full input validation and solid +error handling paths - GPUs will die eventually in the oddmost corner cases +anyway: + + * The ioctl must check for array overflows. Also it needs to check for + over/underflows and clamping issues of integer values in general. The usual + example is sprite positioning values fed directly into the hardware with the + hardware just having 12 bits or so. Works nicely until some odd display + server doesn't bother with clamping itself and the cursor wraps around the + screen. + + * Have simple testcases for every input validation failure case in your ioctl. + Check that the error code matches your expectations. And finally make sure + that you only test for one single error path in each subtest by submitting + otherwise perfectly valid data. Without this an earlier check might reject + the ioctl already and shadow the codepath you actually want to test, hiding + bugs and regressions. + + * Make all your ioctls restartable. First X really loves signals and second + this will allow you to test 90% of all error handling paths by just + interrupting your main test suite constantly with signals. Thanks to X's + love for signal you'll get an excellent base coverage of all your error + paths pretty much for free for graphics drivers. Also, be consistent with + how you handle ioctl restarting - e.g. drm has a tiny drmIoctl helper in its + userspace library. The i915 driver botched this with the set_tiling ioctl, + now we're stuck forever with some arcane semantics in both the kernel and + userspace. + + * If you can't make a given codepath restartable make a stuck task at least + killable. GPUs just die and your users won't like you more if you hang their + entire box (by means of an unkillable X process). If the state recovery is + still too tricky have a timeout or hangcheck safety net as a last-ditch + effort in case the hardware has gone bananas. + + * Have testcases for the really tricky corner cases in your error recovery code + - it's way too easy to create a deadlock between your hangcheck code and + waiters. + + +Time, Waiting and Missing it +---------------------------- + +GPUs do most everything asynchronously, so we have a need to time operations and +wait for oustanding ones. This is really tricky business; at the moment none of +the ioctls supported by the drm/i915 get this fully right, which means there's +still tons more lessons to learn here. + + * Use CLOCK_MONOTONIC as your reference time, always. It's what alsa, drm and + v4l use by default nowadays. But let userspace know which timestamps are + derived from different clock domains like your main system clock (provided + by the kernel) or some independent hardware counter somewhere else. Clocks + will mismatch if you look close enough, but if performance measuring tools + have this information they can at least compensate. If your userspace can + get at the raw values of some clocks (e.g. through in-command-stream + performance counter sampling instructions) consider exposing those also. + + * Use __s64 seconds plus __u64 nanoseconds to specify time. It's not the most + convenient time specification, but it's mostly the standard. + + * Check that input time values are normalized and reject them if not. Note + that the kernel native struct ktime has a signed integer for both seconds + and nanoseconds, so beware here. + + * For timeouts, use absolute times. If you're a good fellow and made your + ioctl restartable relative timeouts tend to be too coarse and can + indefinitely extend your wait time due to rounding on each restart. + Especially if your reference clock is something really slow like the display + frame counter. With a spec laywer hat on this isn't a bug since timeouts can + always be extended - but users will surely hate you if their neat animations + starts to stutter due to this. + + * Consider ditching any synchronous wait ioctls with timeouts and just deliver + an asynchronous event on a pollable file descriptor. It fits much better + into event driven applications' main loop. + + * Have testcases for corner-cases, especially whether the return values for + already-completed events, successful waits and timed-out waits are all sane + and suiting to your needs. + + +Leaking Resources, Not +---------------------- + +A full-blown drm driver essentially implements a little OS, but specialized to +the given GPU platforms. This means a driver needs to expose tons of handles +for different objects and other resources to userspace. Doing that right +entails its own little set of pitfalls: + + * Always attach the lifetime of your dynamically created resources to the + lifetime of a file descriptor. Consider using a 1:1 mapping if your resource + needs to be shared across processes - fd-passing over unix domain sockets + also simplifies lifetime management for userspace. + + * Always have O_CLOEXEC support. + + * Ensure that you have sufficient insulation between different clients. By + default pick a private per-fd namespace which forces any sharing to be done + explictly. Only go with a more global per-device namespace if the objects + are truly device-unique. One counterexample in the drm modeset interfaces is + that the per-device modeset objects like connectors share a namespace with + framebuffer objects, which mostly are not shared at all. A separate + namespace, private by default, for framebuffers would have been more + suitable. + + * Think about uniqueness requirements for userspace handles. E.g. for most drm + drivers it's a userspace bug to submit the same object twice in the same + command submission ioctl. But then if objects are shareable userspace needs + to know whether it has seen an imported object from a different process + already or not. I haven't tried this myself yet due to lack of a new class + of objects, but consider using inode numbers on your shared file descriptors + as unique identifiers - it's how real files are told apart, too. + Unfortunately this requires a full-blown virtual filesystem in the kernel. + + +Last, but not Least +------------------- + +Not every problem needs a new ioctl: + + * Think hard whether you really want a driver-private interface. Of course + it's much quicker to push a driver-private interface than engaging in + lengthy discussions for a more generic solution. And occasionally doing a + private interface to spearhead a new concept is what's required. But in the + end, once the generic interface comes around you'll end up maintainer two + interfaces. Indefinitely. + + * Consider other interfaces than ioctls. A sysfs attribute is much better for + per-device settings, or for child objects with fairly static lifetimes (like + output connectors in drm with all the detection override attributes). Or + maybe only your testsuite needs this interface, and then debugfs with its + disclaimer of not having a stable ABI would be better. + +Finally, the name of the game is to get it right on the first attempt, since if +your driver proves popular and your hardware platforms long-lived then you'll +be stuck with a given ioctl essentially forever. You can try to deprecate +horrible ioctls on newer iterations of your hardware, but generally it takes +years to accomplish this. And then again years until the last user able to +complain about regressions disappears, too. diff --git a/Documentation/laptops/freefall.c b/Documentation/laptops/freefall.c index d4173186555..5e44b20b184 100644 --- a/Documentation/laptops/freefall.c +++ b/Documentation/laptops/freefall.c @@ -29,16 +29,12 @@ static const char app_name[] = "FREE FALL"; static int set_unload_heads_path(char *device) { - char devname[64]; - if (strlen(device) <= 5 || strncmp(device, "/dev/", 5) != 0) return -EINVAL; - strncpy(devname, device + 5, sizeof(devname) - 1); - devname[sizeof(devname) - 1] = '\0'; strncpy(device_path, device, sizeof(device_path) - 1); snprintf(unload_heads_path, sizeof(unload_heads_path) - 1, - "/sys/block/%s/device/unload_heads", devname); + "/sys/block/%s/device/unload_heads", device+5); return 0; } |