aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Merge tag 'arc-4.15-rc1' of ↵Linus Torvalds2017-11-258-56/+136
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc Pull ARC updates from Vineet Gupta: - more changes for HS48 cores: supporting MMUv5, detecting new micro-arch gizmos - axs10x platform wiring up reset driver merged in this cycle - ARC perf driver optimizations * tag 'arc-4.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc: ARC: perf: avoid vmalloc backed mmap ARCv2: perf: optimize given that num counters <= 32 ARCv2: perf: tweak overflow interrupt ARC: [plat-axs10x] DTS: Add reset controller node to manage ethernet reset ARCv2: boot log: updates for HS48: dual-issue, ECC, Loop Buffer ARCv2: Accomodate HS48 MMUv5 by relaxing MMU ver checking ARC: [plat-axs10x] auto-select AXS101 or AXS103 given the ISA config
| * ARC: perf: avoid vmalloc backed mmapVineet Gupta2017-11-211-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For non-alising Dcache, vmalloc is not needed. vmalloc triggers additonal D-TLB Misses in the perf interrupt code path making it slightly inefficient as evident from hackbench runs below. | [ARCLinux]# perf stat -e dTLB-load-misses --repeat 5 hackbench | Running with 10*40 (== 400) tasks. | Time: 35.060 | ... | Performance counter stats for 'hackbench' (5 runs): Before: 399235 dTLB-load-misses ( +- 2.08% ) After : 397676 dTLB-load-misses ( +- 2.27% ) Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
| * ARCv2: perf: optimize given that num counters <= 32Vineet Gupta2017-11-211-9/+7
| | | | | | | | | | | | | | | | | | | | | | use ffz primitive which maps to ARCv2 instruction, vs. non atomic __test_and_set_bit It is unlikely if we will even have more than 32 counters, but still add a BUILD_BUG to catch that Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
| * ARCv2: perf: tweak overflow interruptVineet Gupta2017-11-211-10/+14
| | | | | | | | | | | | | | | | | | Current perf ISR loops thru all 32 counters, checking for each if it caused the interrupt. Instead only loop thru counters which actually interrupted (typically 1). Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
| * ARC: [plat-axs10x] DTS: Add reset controller node to manage ethernet resetEugeniy Paltsev2017-11-152-7/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | DW ethernet controller on axs10x hangs sometimes after SW reset. Invoke the newly aded driver (reset-axs10x.c) by adding the DT bits. With this in place, we don't need the open-coded quirk in platform code, so get rid of it as well ! Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com> Signed-off-by: Vineet Gupta <vgupta@synopsys.com> Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com> Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
| * ARCv2: boot log: updates for HS48: dual-issue, ECC, Loop BufferVineet Gupta2017-11-132-5/+71
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Print the hardware support for ECC, Loop Buffer as well as the runtime enabled status Note that unlike the existing boot printing, this one is not read from pre-decoded hardware capabilty info cached in cpuinfo[] struct. Instead we read the AUX regs on the spot and print it, without botherign to save anywhere. There is no point in saving static hardware capabilites in memory when its use is very sporadic and non-performance critical, mainly for /proc/cpuinfo. This gets worse in SMP, given it is per-cpu, and pretty much exactly same across all cpus. So only info needed at runtime (e.g. TLB geometry) needs to be cached in cpuinfo[]. So going fwd we will start converting code to this paradigm. Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
| * ARCv2: Accomodate HS48 MMUv5 by relaxing MMU ver checkingVineet Gupta2017-11-061-24/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | HS48 cpus will have a new MMUv5, although Linux is currently not explicitly supporting the newer features (so remains at V4). The existing software/hardware version check is very tight and causes boot abort. Given that the MMUv5 hardware is backwards compatible, relax the boot check to allow current kernel support level to work with new hardware. Also while at it, move the ancient MMU related code to under ARCompact builds as baseline MMU for HS cpus is v4. Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
| * ARC: [plat-axs10x] auto-select AXS101 or AXS103 given the ISA configVineet Gupta2017-10-291-0/+2
| | | | | | | | | | | | | | | | | | | | | | AXS10x platform has two flavours: - AXS101 (ARC700 cpu, ARCompact ISA) - AXS103 (ARC HS38x cpu, ARCv2 ISA) So depending on the ISA support configured, pre-select the AXS10x variant. Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
* | Merge tag 'kbuild-v4.15-2' of ↵Linus Torvalds2017-11-2529-113/+58
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild Pull more Kbuild updates from Masahiro Yamada: - use 'pwd' instead of '/bin/pwd' for portability - clean up Makefiles - fix ld-option for clang - fix malloc'ed data size in Kconfig - fix parallel building along with coccicheck - fix a minor issue of package building - prompt to use "rpm-pkg" instead of "rpm" - clean up *.i and *.lst patterns by "make clean" * tag 'kbuild-v4.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild: kbuild: drop $(extra-y) from real-objs-y kbuild: clean up *.i and *.lst patterns by make clean kbuild: rpm: prompt to use "rpm-pkg" if "rpm" target is used kbuild: pkg: use --transform option to prefix paths in tar coccinelle: fix parallel build with CHECK=scripts/coccicheck kconfig/symbol.c: use correct pointer type argument for sizeof kbuild: Set KBUILD_CFLAGS before incl. arch Makefile kbuild: remove all dummy assignments to obj- kbuild: create built-in.o automatically if parent directory wants it kbuild: /bin/pwd -> pwd
| * | kbuild: drop $(extra-y) from real-objs-yMasahiro Yamada2017-11-231-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | $(real-objs-y) in only used in scripts/Makefile.build to form "targets", but $(extra-y) is added to "targets" in another line. We do not need to add $(extra-y) twice. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
| * | kbuild: clean up *.i and *.lst patterns by make cleanMasahiro Yamada2017-11-231-2/+2
| | | | | | | | | | | | | | | | | | *.i and *.lst are supported by the single target build. Clean up them. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
| * | kbuild: rpm: prompt to use "rpm-pkg" if "rpm" target is usedMasahiro Yamada2017-11-232-3/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The "rpm" has been kept for backward compatibility since pre-git era. I am planning to remove it after the Linux 4.18 release. Annouce the end of the support, prompting to use "rpm-pkg" instead. If you use "rpm", it will work like "rpm-pkg", but warning messages will be displayed as follows: WARNING: "rpm" target will be removed after Linux 4.18 Please use "rpm-pkg" instead. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
| * | kbuild: pkg: use --transform option to prefix paths in tarMasahiro Yamada2017-11-231-3/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For rpm-pkg and deb-pkg, a source tar file is created. All paths in the archive must be prefixed with the base name of the tar so that everything is contained in the directory when you extract it. Currently, scripts/package/Makefile uses a symlink for that, and removes it after the tar is created. If you terminate the build during the tar creation, the symlink is left over. Then, at the next package build, you will see a warning like follows: ln: '.' and 'kernel-4.14.0+/.' are the same file It is possible to fix it by adding -n (--no-dereference) option to the "ln" command, but a cleaner way is to use --transform option of "tar" command. This option is GNU extension, but it should not hurt to use it in the Linux build system. The 'S' flag is needed to exclude symlinks from the path fixup. Without it, symlinks in the kernel are broken. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
| * | coccinelle: fix parallel build with CHECK=scripts/coccicheckMasahiro Yamada2017-11-231-6/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The command "make -j8 C=1 CHECK=scripts/coccicheck" produces lots of "coccicheck failed" error messages. Julia Lawall explained the Coccinelle behavior as follows: "The problem on the Coccinelle side is that it uses a subdirectory with the name of the semantic patch to store standard output and standard error for the different threads. I didn't want to use a name with the pid, so that one could easily find this information while Coccinelle is running. Normally the subdirectory is cleaned up when Coccinelle completes, so there is only one of them at a time. Maybe it is best to just add the pid. There is the risk that these subdirectories will accumulate if Coccinelle crashes in a way such that they don't get cleaned up, but Coccinelle could print a warning if it detects this case, rather than failing." When scripts/coccicheck is used as CHECK tool and -j option is given to Make, the whole of build process runs in parallel. So, multiple processes try to get access to the same subdirectory. I notice spatch creates the subdirectory only when it runs in parallel (i.e. --jobs <N> is given and <N> is greater than 1). Setting NPROC=1 is a reasonable solution; spatch does not create the subdirectory. Besides, ONLINE=1 mode takes a single file input for each spatch invocation, so there is no reason to parallelize it in the first place. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> Acked-by: Julia Lawall <Julia.Lawall@lip6.fr>
| * | kconfig/symbol.c: use correct pointer type argument for sizeofHeinrich Schuchardt2017-11-231-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | sym_arr is of type struct symbol **. So in malloc we need sizeof(struct symbol *). The problem was indicated by coccinelle. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
| * | kbuild: Set KBUILD_CFLAGS before incl. arch MakefileChris Fries2017-11-231-32/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Set the clang KBUILD_CFLAGS up before including arch/ Makefiles, so that ld-options (etc.) can work correctly. This fixes errors with clang such as ld-options trying to CC against your host architecture, but LD trying to link against your target architecture. Signed-off-by: Chris Fries <cfries@google.com> Signed-off-by: Nick Desaulniers <ndesaulniers@google.com> Reviewed-by: Matthias Kaehlcke <mka@chromium.org> Tested-by: Matthias Kaehlcke <mka@chromium.org> Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
| * | kbuild: remove all dummy assignments to obj-Masahiro Yamada2017-11-1820-58/+0
| | | | | | | | | | | | | | | | | | | | | Now kbuild core scripts create empty built-in.o where necessary. Remove "obj- := dummy.o" tricks. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
| * | kbuild: create built-in.o automatically if parent directory wants itMasahiro Yamada2017-11-182-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | "obj-y += foo/" syntax requires Kbuild to visit the "foo" subdirectory and link built-in.o from that directory. This means foo/Makefile is responsible for creating built-in.o even if there is no object to link (in this case, built-in.o is an empty archive). We have had several fixups like commit 4b024242e8a4 ("kbuild: Fix linking error built-in.o no such file or directory"), then ended up with a complex condition as follows: ifneq ($(strip $(obj-y) $(obj-m) $(obj-) $(subdir-m) $(lib-target)),) builtin-target := $(obj)/built-in.o endif We still have more cases not covered by the above, so we need to add obj- := dummy.o in several places just for creating empty built-in.o. A key point is, the parent Makefile knows whether built-in.o is needed or not. If a subdirectory needs to create built-in.o, its parent can tell the fact when descending. If non-empty $(need-builtin) flag is passed from the parent, built-in.o should be created. $(obj-y) should be still checked to support the single target "%/". All of ugly tricks will go away. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
| * | kbuild: /bin/pwd -> pwdBjørn Forsman2017-11-184-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Most places use pwd and rely on $PATH lookup. Moving the remaining absolute path /bin/pwd users over for consistency. Also, a reason for doing /bin/pwd -> pwd instead of the other way around is because I believe build systems should make little assumptions on host filesystem layout. Case in point, we do this kind of patching already in NixOS. Ref. commit 028568d84da3cfca49f5f846eeeef01441d70451 ("kbuild: revert $(realpath ...) to $(shell cd ... && /bin/pwd)"). Signed-off-by: Bjørn Forsman <bjorn.forsman@gmail.com> Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
* | | Merge tag 'afs-fixes-20171124' of ↵Linus Torvalds2017-11-258-279/+405
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs Pull AFS fixes from David Howells: - Make AFS file locking work again. - Don't write to a page that's being written out, but wait for it to complete. - Do d_drop() and d_add() in the right places. - Put keys on error paths. - Remove some redundant code. * tag 'afs-fixes-20171124' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs: afs: remove redundant assignment of dvnode to itself afs: cell: Remove unnecessary code in afs_lookup_cell afs: Fix signal handling in some file ops afs: Fix some dentry handling in dir ops and missing key_puts afs: Make afs_write_begin() avoid writing to a page that's being stored afs: Fix file locking
| * | | afs: remove redundant assignment of dvnode to itselfColin Ian King2017-11-241-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The assignment of dvnode to itself is redundant and can be removed. Cleans up warning detected by cppcheck: fs/afs/dir.c:975: (warning) Redundant assignment of 'dvnode' to itself. Fixes: d2ddc776a458 ("afs: Overhaul volume and server record caching and fileserver rotation") Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: David Howells <dhowells@redhat.com>
| * | | afs: cell: Remove unnecessary code in afs_lookup_cellGustavo A. R. Silva2017-11-241-6/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Due to recent changes this piece of code is no longer needed. Addresses-Coverity-ID: 1462033 Link: https://lkml.kernel.org/r/4923.1510957307@warthog.procyon.org.uk Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com> Signed-off-by: David Howells <dhowells@redhat.com>
| * | | afs: Fix signal handling in some file opsDavid Howells2017-11-241-0/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | afs_mkdir(), afs_create(), afs_link() and afs_symlink() all need to drop the target dentry if a signal causes the operation to be killed immediately before we try to contact the server. Signed-off-by: David Howells <dhowells@redhat.com>
| * | | afs: Fix some dentry handling in dir ops and missing key_putsDavid Howells2017-11-241-10/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix some of dentry handling in AFS directory ops: (1) Do d_drop() on the new_dentry before assigning a new inode to it in afs_vnode_new_inode(). It's fine to do this before calling afs_iget() because the operation has taken place on the server. (2) Replace d_instantiate()/d_rehash() with d_add(). (3) Don't d_drop() the new_dentry in afs_rename() on error. Also fix afs_link() and afs_rename() to call key_put() on all error paths where the key is taken. Signed-off-by: David Howells <dhowells@redhat.com>
| * | | afs: Make afs_write_begin() avoid writing to a page that's being storedDavid Howells2017-11-241-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Make afs_write_begin() wait for a page that's marked PG_writeback because: (1) We need to avoid interference with the data being stored so that the data on the server ends up in a defined state. (2) page->private is used to track the window of dirty data within a page, but it's also used by the storage code to track what's being written, being cleared by the completion notification. Ownership can't be relinquished by the storage code until completion because it a store fails, the data must be remarked dirty. Tracing shows something like the following (edited): x86_64-linux-gn-15940 [1] afs_page_dirty: vn=ffff8800bef33800 9c75 begin 0-125 kworker/u8:3-114 [2] afs_page_dirty: vn=ffff8800bef33800 9c75 store+ 0-125 x86_64-linux-gn-15940 [1] afs_page_dirty: vn=ffff8800bef33800 9c75 begin 0-2052 kworker/u8:3-114 [2] afs_page_dirty: vn=ffff8800bef33800 9c75 clear 0-2052 kworker/u8:3-114 [2] afs_page_dirty: vn=ffff8800bef33800 9c75 store 0-0 kworker/u8:3-114 [2] afs_page_dirty: vn=ffff8800bef33800 9c75 WARN 0-0 The clear (completion) corresponding to the store+ (store continuation from a previous page) happens between the second begin (afs_write_begin) and the store corresponding to that. This results in the second store not seeing any data to write back, leading to the following warning: WARNING: CPU: 2 PID: 114 at ../fs/afs/write.c:403 afs_write_back_from_locked_page+0x19d/0x76c [kafs] Modules linked in: kafs(E) CPU: 2 PID: 114 Comm: kworker/u8:3 Tainted: G E 4.14.0-fscache+ #242 Hardware name: ASUS All Series/H97-PLUS, BIOS 2306 10/09/2014 Workqueue: writeback wb_workfn (flush-afs-2) task: ffff8800cad72600 task.stack: ffff8800cad44000 RIP: 0010:afs_write_back_from_locked_page+0x19d/0x76c [kafs] RSP: 0018:ffff8800cad47aa0 EFLAGS: 00010246 RAX: 0000000000000001 RBX: ffff8800bef33a20 RCX: 0000000000000000 RDX: 000000000000000f RSI: ffffffff81c5d0e0 RDI: ffff8800cad72e78 RBP: ffff8800d31ea1e8 R08: ffff8800c1358000 R09: ffff8800ca00e400 R10: ffff8800cad47a38 R11: ffff8800c5d9e400 R12: 0000000000000000 R13: ffffea0002d9df00 R14: ffffffffa0023c1c R15: 0000000000007fdf FS: 0000000000000000(0000) GS:ffff8800ca700000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f85ac6c4000 CR3: 0000000001c10001 CR4: 00000000001606e0 Call Trace: ? clear_page_dirty_for_io+0x23a/0x267 afs_writepages_region+0x1be/0x286 [kafs] afs_writepages+0x60/0x127 [kafs] do_writepages+0x36/0x70 __writeback_single_inode+0x12f/0x635 writeback_sb_inodes+0x2cc/0x452 __writeback_inodes_wb+0x68/0x9f wb_writeback+0x208/0x470 ? wb_workfn+0x22b/0x565 wb_workfn+0x22b/0x565 ? worker_thread+0x230/0x2ac process_one_work+0x2cc/0x517 ? worker_thread+0x230/0x2ac worker_thread+0x1d4/0x2ac ? rescuer_thread+0x29b/0x29b kthread+0x15d/0x165 ? kthread_create_on_node+0x3f/0x3f ? call_usermodehelper_exec_async+0x118/0x11f ret_from_fork+0x24/0x30 Signed-off-by: David Howells <dhowells@redhat.com>
| * | | afs: Fix file lockingDavid Howells2017-11-175-262/+385
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix the AFS file locking whereby the use of the big kernel lock (which could be slept with) was replaced by a spinlock (which couldn't). The problem is that the AFS code was doing stuff inside the critical section that might call schedule(), so this is a broken transformation. Fix this by the following means: (1) Use a state machine with a proper state that can only be changed under the spinlock rather than using a collection of bit flags. (2) Cache the key used for the lock and the lock type in the afs_vnode struct so that the manager work function doesn't have to refer to a file_lock struct that's been dequeued. This makes signal handling safer. (4) Move the unlock from afs_do_unlk() to afs_fl_release_private() which means that unlock is achieved in other circumstances too. (5) Unlock the file on the server before taking the next conflicting lock. Also change: (1) Check the permits on a file before actually trying the lock. (2) fsync the file before effecting an explicit unlock operation. We don't fsync if the lock is erased otherwise as we might not be in a context where we can actually do that. Further fixes: (1) Fixed-fileserver address rotation is made to work. It's only used by the locking functions, so couldn't be tested before. Fixes: 72f98e72551f ("locks: turn lock_flocks into a spinlock") Signed-off-by: David Howells <dhowells@redhat.com> cc: jlayton@redhat.com
* | | | Merge tag 'kvm-4.15-2' of git://git.kernel.org/pub/scm/virt/kvm/kvmLinus Torvalds2017-11-2419-158/+819
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull KVM updates from Radim Krčmář: "Trimmed second batch of KVM changes for Linux 4.15: - GICv4 Support for KVM/ARM - re-introduce support for CPUs without virtual NMI (cc stable) and allow testing of KVM without virtual NMI on available CPUs - fix long-standing performance issues with assigned devices on AMD (cc stable)" * tag 'kvm-4.15-2' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (30 commits) kvm: vmx: Allow disabling virtual NMI support kvm: vmx: Reinstate support for CPUs without virtual NMI KVM: SVM: obey guest PAT KVM: arm/arm64: Don't queue VLPIs on INV/INVALL KVM: arm/arm64: Fix GICv4 ITS initialization issues KVM: arm/arm64: GICv4: Theory of operations KVM: arm/arm64: GICv4: Enable VLPI support KVM: arm/arm64: GICv4: Prevent userspace from changing doorbell affinity KVM: arm/arm64: GICv4: Prevent a VM using GICv4 from being saved KVM: arm/arm64: GICv4: Enable virtual cpuif if VLPIs can be delivered KVM: arm/arm64: GICv4: Hook vPE scheduling into vgic flush/sync KVM: arm/arm64: GICv4: Use the doorbell interrupt as an unblocking source KVM: arm/arm64: GICv4: Add doorbell interrupt handling KVM: arm/arm64: GICv4: Use pending_last as a scheduling hint KVM: arm/arm64: GICv4: Handle INVALL applied to a vPE KVM: arm/arm64: GICv4: Propagate property updates to VLPIs KVM: arm/arm64: GICv4: Handle MOVALL applied to a vPE KVM: arm/arm64: GICv4: Handle CLEAR applied to a VLPI KVM: arm/arm64: GICv4: Propagate affinity changes to the physical ITS KVM: arm/arm64: GICv4: Unmap VLPI when freeing an LPI ...
| * | | | kvm: vmx: Allow disabling virtual NMI supportPaolo Bonzini2017-11-171-10/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | To simplify testing of these rarely used code paths, add a module parameter that turns it on. One eventinj.flat test (NMI after iret) fails when loading kvm_intel with vnmi=0. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
| * | | | kvm: vmx: Reinstate support for CPUs without virtual NMIPaolo Bonzini2017-11-171-44/+106
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is more or less a revert of commit 2c82878b0cb3 ("KVM: VMX: require virtual NMI support", 2017-03-27); it turns out that Core 2 Duo machines only had virtual NMIs in some SKUs. The revert is not trivial because in the meanwhile there have been several fixes to nested NMI injection. Therefore, the entire vNMI state is moved to struct loaded_vmcs. Another change compared to before the patch is a simplification here: if (unlikely(!cpu_has_virtual_nmis() && vmx->soft_vnmi_blocked && !(is_guest_mode(vcpu) && nested_cpu_has_virtual_nmis( get_vmcs12(vcpu))))) { The final condition here is always true (because nested_cpu_has_virtual_nmis is always false) and is removed. Fixes: 2c82878b0cb38fd516fd612c67852a6bbf282003 Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1490803 Cc: stable@vger.kernel.org Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
| * | | | KVM: SVM: obey guest PATPaolo Bonzini2017-11-171-0/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For many years some users of assigned devices have reported worse performance on AMD processors with NPT than on AMD without NPT, Intel or bare metal. The reason turned out to be that SVM is discarding the guest PAT setting and uses the default (PA0=PA4=WB, PA1=PA5=WT, PA2=PA6=UC-, PA3=UC). The guest might be using a different setting, and especially might want write combining but isn't getting it (instead getting slow UC or UC- accesses). Thanks a lot to geoff@hostfission.com for noticing the relation to the g_pat setting. The patch has been tested also by a bunch of people on VFIO users forums. Fixes: 709ddebf81cb40e3c36c6109a7892e8b93a09464 Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=196409 Cc: stable@vger.kernel.org Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: David Hildenbrand <david@redhat.com> Tested-by: Nick Sarnie <commendsarnex@gmail.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
| * | | | Merge tag 'kvm-arm-gicv4-for-v4.15' of ↵Paolo Bonzini2017-11-1717-114/+695
| |\ \ \ \ | | |/ / / | |/| | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD GICv4 Support for KVM/ARM for v4.15
| | * | | KVM: arm/arm64: Don't queue VLPIs on INV/INVALLChristoffer Dall2017-11-101-3/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since VLPIs are injected directly by the hardware there's no need to mark these as pending in software and queue them on the AP list. Reviewed-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: Fix GICv4 ITS initialization issuesChristoffer Dall2017-11-103-6/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We should only try to initialize GICv4 data structures on a GICv4 capable system. Move the vgic_supports_direct_msis() check inito vgic_v4_init() so that any KVM VGIC initialization path does not fail on non-GICv4 systems. Also be slightly more strict in the checking of the return value in vgic_its_create, and only error out on negative return values from the vgic_v4_init() function. This is important because the kvm device code only treats negative values as errors and only cleans up in this case. Errornously treating a positive return value as an error from the vgic_v4_init() function can lead to NULL pointer dereferences, as has recently been observed. Acked-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Theory of operationsMarc Zyngier2017-11-101-0/+67
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Yet another braindump so I can free some cells... Acked-by: Christoffer Dall <christoffer.dall@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Enable VLPI supportMarc Zyngier2017-11-102-0/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | All it takes is the has_v4 flag to be set in gic_kvm_info as well as "kvm-arm.vgic_v4_enable=1" being passed on the command line for GICv4 to be enabled in KVM. Acked-by: Christoffer Dall <cdall@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Prevent userspace from changing doorbell affinityMarc Zyngier2017-11-101-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We so far allocate the doorbell interrupts without taking any special measure regarding the affinity of these interrupts. We simply move them around as required when the vcpu gets scheduled on a different CPU. But that's counting without userspace (and the evil irqbalance) that can try and move the VPE interrupt around, causing the ITS code to emit VMOVP commands and remap the doorbell to another redistributor. Worse, this can happen while the vcpu is running, causing all kind of trouble if the VPE is already resident, and we end-up in UNPRED territory. So let's take a definitive action and prevent userspace from messing with us. This is just a matter of adding IRQ_NO_BALANCING to the set of flags we already have, letting the kernel in sole control of the affinity. Acked-by: Christoffer Dall <christoffer.dall@linaro.org> Reviewed-by: Eric Auger <eric.auger@redhat.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Prevent a VM using GICv4 from being savedMarc Zyngier2017-11-102-0/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The GICv4 architecture doesn't make it easy for save/restore to work, as it doesn't give any guarantee that the pending state is written into the pending table. So let's not take any chance, and let's return an error if we encounter any LPI that has the HW bit set. In order for userspace to distinguish this error from other failure modes, use -EACCES as an error code. Reviewed-by: Eric Auger <eric.auger@redhat.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Enable virtual cpuif if VLPIs can be deliveredMarc Zyngier2017-11-101-3/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In order for VLPIs to be delivered to the guest, we must make sure that the virtual cpuif is always enabled, irrespective of the presence of virtual interrupt in the LRs. Acked-by: Christoffer Dall <cdall@linaro.org> Reviewed-by: Eric Auger <eric.auger@redhat.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Hook vPE scheduling into vgic flush/syncMarc Zyngier2017-11-103-0/+45
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The redistributor needs to be told which vPE is about to be run, and tells us whether there is any pending VLPI on exit. Let's add the scheduling calls to the vgic flush/sync functions, allowing the VLPIs to be delivered to the guest. Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Use the doorbell interrupt as an unblocking sourceMarc Zyngier2017-11-103-0/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The doorbell interrupt is only useful if the vcpu is blocked on WFI. In all other cases, recieving a doorbell interrupt is just a waste of cycles. So let's only enable the doorbell if a vcpu is getting blocked, and disable it when it is unblocked. This is very similar to what we're doing for the background timer. Reviewed-by: Christoffer Dall <cdall@linaro.org> Reviewed-by: Eric Auger <eric.auger@redhat.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Add doorbell interrupt handlingMarc Zyngier2017-11-101-0/+48
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When a vPE is not running, a VLPI being made pending results in a doorbell interrupt being delivered. Let's handle this interrupt and update the pending_last flag that indicates that VLPIs are pending. The corresponding vcpu is also kicked into action. Special care is taken to prevent the doorbell from being enabled at request time (this is controlled separately), and to make the disabling on the interrupt non-lazy. Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org> Reviewed-by: Eric Auger <eric.auger@redhat.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Use pending_last as a scheduling hintMarc Zyngier2017-11-101-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When a vPE exits, the pending_last flag is set when there are pending VLPIs stored in the pending table. Similarily, this flag will be set when a doorbell interrupt fires, as it indicates the same condition. Let's update kvm_vgic_vcpu_pending_irq() to account for that flag as well, making a vcpu runnable when set. Acked-by: Christoffer Dall <cdall@linaro.org> Reviewed-by: Eric Auger <eric.auger@redhat.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Handle INVALL applied to a vPEMarc Zyngier2017-11-101-6/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There is no need to perform an INV for each interrupt when updating multiple interrupts. Instead, we can rely on the final VINVALL that gets sent to the ITS to do the work for all of them. Acked-by: Christoffer Dall <cdall@linaro.org> Reviewed-by: Eric Auger <eric.auger@redhat.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Propagate property updates to VLPIsMarc Zyngier2017-11-101-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Upon updating a property, we propagate it all the way to the physical ITS, and ask for an INV command to be executed there. Acked-by: Christoffer Dall <cdall@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Handle MOVALL applied to a vPEMarc Zyngier2017-11-101-9/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current implementation of MOVALL doesn't allow us to call into the core ITS code as we hold a number of spinlocks. Let's try a method used in other parts of the code, were we copy the intids of the candicate interrupts, and then do whatever we need to do with them outside of the critical section. This allows us to move the interrupts one by one, at the expense of a bit of CPU time. Who cares? MOVALL is such a stupid command anyway... Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Handle CLEAR applied to a VLPIMarc Zyngier2017-11-101-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Handling CLEAR is pretty easy. Just ask the ITS driver to clear the corresponding pending bit (which will turn into a CLEAR command on the physical side). Acked-by: Christoffer Dall <cdall@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Propagate affinity changes to the physical ITSMarc Zyngier2017-11-101-1/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When the guest issues an affinity change, we need to tell the physical ITS that we're now targetting a new vcpu. This is done by extracting the current mapping, updating the target, and reapplying the mapping. Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Unmap VLPI when freeing an LPIMarc Zyngier2017-11-101-1/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When freeing an LPI (on a DISCARD command, for example), we need to unmap the VLPI down to the physical ITS level. Acked-by: Christoffer Dall <cdall@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Handle INT command applied to a VLPIMarc Zyngier2017-11-101-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If the guest issues an INT command targetting a VLPI, let's call into the irq_set_irqchip_state() helper to make it pending on the physical side. This works just as well if userspace decides to inject an interrupt using the normal userspace API... Acked-by: Christoffer Dall <cdall@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
| | * | | KVM: arm/arm64: GICv4: Wire mapping/unmapping of VLPIs in VFIO irq bypassMarc Zyngier2017-11-103-2/+116
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Let's use the irq bypass mechanism also used for x86 posted interrupts to intercept the virtual PCIe endpoint configuration and establish our LPI->VLPI mapping. Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
generated by cgit (git 2.34.1) at 2025-02-26 05:14:56 +0000