Revision f085df1be60abf670315c11036261cfaec16b2eb authored by Linus Torvalds on 07 May 2023, 18:32:18 UTC, committed by Linus Torvalds on 07 May 2023, 18:32:18 UTC
Pull perf tool updates from Arnaldo Carvalho de Melo: "Third version of perf tool updates, with the build problems with with using a 'vmlinux.h' generated from the main build fixed, and the bpf skeleton build disabled by default. Build: - Require libtraceevent to build, one can disable it using NO_LIBTRACEEVENT=1. It is required for tools like 'perf sched', 'perf kvm', 'perf trace', etc. libtraceevent is available in most distros so installing 'libtraceevent-devel' should be a one-time event to continue building perf as usual. Using NO_LIBTRACEEVENT=1 produces tooling that is functional and sufficient for lots of users not interested in those libtraceevent dependent features. - Allow Python support in 'perf script' when libtraceevent isn't linked, as not all features requires it, for instance Intel PT does not use tracepoints. - Error if the python interpreter needed for jevents to work isn't available and NO_JEVENTS=1 isn't set, preventing a build without support for JSON vendor events, which is a rare but possible condition. The two check error messages: $(error ERROR: No python interpreter needed for jevents generation. Install python or build with NO_JEVENTS=1.) $(error ERROR: Python interpreter needed for jevents generation too old (older than 3.6). Install a newer python or build with NO_JEVENTS=1.) - Make libbpf 1.0 the minimum required when building with out of tree, distro provided libbpf. - Use libsdtc++'s and LLVM's libcxx's __cxa_demangle, a portable C++ demangler, add 'perf test' entry for it. - Make binutils libraries opt in, as distros disable building with it due to licensing, they were used for C++ demangling, for instance. - Switch libpfm4 to opt-out rather than opt-in, if libpfm-devel (or equivalent) isn't installed, we'll just have a build warning: Makefile.config:1144: libpfm4 not found, disables libpfm4 support. Please install libpfm4-dev - Add a feature test for scandirat(), that is not implemented so far in musl and uclibc, disabling features that need it, such as scanning for tracepoints in /sys/kernel/tracing/events. perf BPF filters: - New feature where BPF can be used to filter samples, for instance: $ sudo ./perf record -e cycles --filter 'period > 1000' true $ sudo ./perf script perf-exec 2273949 546850.708501: 5029 cycles: ffffffff826f9e25 finish_wait+0x5 ([kernel.kallsyms]) perf-exec 2273949 546850.708508: 32409 cycles: ffffffff826f9e25 finish_wait+0x5 ([kernel.kallsyms]) perf-exec 2273949 546850.708526: 143369 cycles: ffffffff82b4cdbf xas_start+0x5f ([kernel.kallsyms]) perf-exec 2273949 546850.708600: 372650 cycles: ffffffff8286b8f7 __pagevec_lru_add+0x117 ([kernel.kallsyms]) perf-exec 2273949 546850.708791: 482953 cycles: ffffffff829190de __mod_memcg_lruvec_state+0x4e ([kernel.kallsyms]) true 2273949 546850.709036: 501985 cycles: ffffffff828add7c tlb_gather_mmu+0x4c ([kernel.kallsyms]) true 2273949 546850.709292: 503065 cycles: 7f2446d97c03 _dl_map_object_deps+0x973 (/usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2) - In addition to 'period' (PERF_SAMPLE_PERIOD), the other PERF_SAMPLE_ can be used for filtering, and also some other sample accessible values, from tools/perf/Documentation/perf-record.txt: Essentially the BPF filter expression is: <term> <operator> <value> (("," | "||") <term> <operator> <value>)* The <term> can be one of: ip, id, tid, pid, cpu, time, addr, period, txn, weight, phys_addr, code_pgsz, data_pgsz, weight1, weight2, weight3, ins_lat, retire_lat, p_stage_cyc, mem_op, mem_lvl, mem_snoop, mem_remote, mem_lock, mem_dtlb, mem_blk, mem_hops The <operator> can be one of: ==, !=, >, >=, <, <=, & The <value> can be one of: <number> (for any term) na, load, store, pfetch, exec (for mem_op) l1, l2, l3, l4, cxl, io, any_cache, lfb, ram, pmem (for mem_lvl) na, none, hit, miss, hitm, fwd, peer (for mem_snoop) remote (for mem_remote) na, locked (for mem_locked) na, l1_hit, l1_miss, l2_hit, l2_miss, any_hit, any_miss, walk, fault (for mem_dtlb) na, by_data, by_addr (for mem_blk) hops0, hops1, hops2, hops3 (for mem_hops) perf lock contention: - Show lock type with address. - Track and show mmap_lock, siglock and per-cpu rq_lock with address. This is done for mmap_lock by following the current->mm pointer: $ sudo ./perf lock con -abl -- sleep 10 contended total wait max wait avg wait address symbol ... 16344 312.30 ms 2.22 ms 19.11 us ffff8cc702595640 17686 310.08 ms 1.49 ms 17.53 us ffff8cc7025952c0 3 84.14 ms 45.79 ms 28.05 ms ffff8cc78114c478 mmap_lock 3557 76.80 ms 68.75 us 21.59 us ffff8cc77ca3af58 1 68.27 ms 68.27 ms 68.27 ms ffff8cda745dfd70 9 54.53 ms 7.96 ms 6.06 ms ffff8cc7642a48b8 mmap_lock 14629 44.01 ms 60.00 us 3.01 us ffff8cc7625f9ca0 3481 42.63 ms 140.71 us 12.24 us ffffffff937906ac vmap_area_lock 16194 38.73 ms 42.15 us 2.39 us ffff8cd397cbc560 11 38.44 ms 10.39 ms 3.49 ms ffff8ccd6d12fbb8 mmap_lock 1 5.43 ms 5.43 ms 5.43 ms ffff8cd70018f0d8 1674 5.38 ms 422.93 us 3.21 us ffffffff92e06080 tasklist_lock 581 4.51 ms 130.68 us 7.75 us ffff8cc9b1259058 5 3.52 ms 1.27 ms 703.23 us ffff8cc754510070 112 3.47 ms 56.47 us 31.02 us ffff8ccee38b3120 381 3.31 ms 73.44 us 8.69 us ffffffff93790690 purge_vmap_area_lock 255 3.19 ms 36.35 us 12.49 us ffff8d053ce30c80 - Update default map size to 16384. - Allocate single letter option -M for --map-nr-entries, as it is proving being frequently used. - Fix struct rq lock access for older kernels with BPF's CO-RE (Compile once, run everywhere). - Fix problems found with MSAn. perf report/top: - Add inline information when using --call-graph=fp or lbr, as was already done to the --call-graph=dwarf callchain mode. - Improve the 'srcfile' sort key performance by really using an optimization introduced in 6.2 for the 'srcline' sort key that avoids calling addr2line for comparision with each sample. perf sched: - Make 'perf sched latency/map/replay' to use "sched:sched_waking" instead of "sched:sched_waking", consistent with 'perf record' since d566a9c2d482 ("perf sched: Prefer sched_waking event when it exists"). perf ftrace: - Make system wide the default target for latency subcommand, run the following command then generate some network traffic and press control+C: # perf ftrace latency -T __kfree_skb ^C DURATION | COUNT | GRAPH | 0 - 1 us | 27 | ############# | 1 - 2 us | 22 | ########### | 2 - 4 us | 8 | #### | 4 - 8 us | 5 | ## | 8 - 16 us | 24 | ############ | 16 - 32 us | 2 | # | 32 - 64 us | 1 | | 64 - 128 us | 0 | | 128 - 256 us | 0 | | 256 - 512 us | 0 | | 512 - 1024 us | 0 | | 1 - 2 ms | 0 | | 2 - 4 ms | 0 | | 4 - 8 ms | 0 | | 8 - 16 ms | 0 | | 16 - 32 ms | 0 | | 32 - 64 ms | 0 | | 64 - 128 ms | 0 | | 128 - 256 ms | 0 | | 256 - 512 ms | 0 | | 512 - 1024 ms | 0 | | 1 - ... s | 0 | | # perf top: - Add --branch-history (LBR: Last Branch Record) option, just like already available for 'perf record'. - Fix segfault in thread__comm_len() where thread->comm was being used outside thread->comm_lock. perf annotate: - Allow configuring objdump and addr2line in ~/.perfconfig., so that you can use alternative binaries, such as llvm's. perf kvm: - Add TUI mode for 'perf kvm stat report'. Reference counting: - Add reference count checking infrastructure to check for use after free, done to the 'cpumap', 'namespaces', 'maps' and 'map' structs, more to come. To build with it use -DREFCNT_CHECKING=1 in the make command line to build tools/perf. Documented at: https://perf.wiki.kernel.org/index.php/Reference_Count_Checking - The above caught, for instance, fix, present in this series: - Fix maps use after put in 'perf test "Share thread maps"': 'maps' is copied from leader, but the leader is put on line 79 and then 'maps' is used to read the reference count below - so a use after put, with the put of maps happening within thread__put. Fixed by reversing the order of puts so that the leader is put last. - Also several fixes were made to places where reference counts were not being held. - Make this one of the tests in 'make -C tools/perf build-test' to regularly build test it and to make sure no direct access to the reference counted structs are made, doing that via accessors to check the validity of the struct pointer. ARM64: - Fix 'perf report' segfault when filtering coresight traces by sparse lists of CPUs. - Add support for 'simd' as a sort field for 'perf report', to show ARM's NEON SIMD's predicate flags: "partial" and "empty". arm64 vendor events: - Add N1 metrics. Intel vendor events: - Add graniterapids, grandridge and sierraforrest events. - Refresh events for: alderlake, aldernaken, broadwell, broadwellde, broadwellx, cascadelakx, haswell, haswellx, icelake, icelakex, jaketown, meteorlake, knightslanding, sandybridge, sapphirerapids, silvermont, skylake, tigerlake and westmereep-dp - Refresh metrics for alderlake-n, broadwell, broadwellde, broadwellx, haswell, haswellx, icelakex, ivybridge, ivytown and skylakex. perf stat: - Implement --topdown using JSON metrics. - Add TopdownL1 JSON metric as a default if present, but disable it for now for some Intel hybrid architectures, a series of patches addressing this is being reviewed and will be submitted for v6.5. - Use metrics for --smi-cost. - Update topdown documentation. Vendor events (JSON) infrastructure: - Add support for computing and printing metric threshold values. For instance, here is one found in thesapphirerapids json file: { "BriefDescription": "Percentage of cycles spent in System Management Interrupts.", "MetricExpr": "((msr@aperf@ - cycles) / msr@aperf@ if msr@smi@ > 0 else 0)", "MetricGroup": "smi", "MetricName": "smi_cycles", "MetricThreshold": "smi_cycles > 0.1", "ScaleUnit": "100%" }, - Test parsing metric thresholds with the fake PMU in 'perf test pmu-events'. - Support for printing metric thresholds in 'perf list'. - Add --metric-no-threshold option to 'perf stat'. - Add rand (reverse and) and has_pmem (optane memory) support to metrics. - Sort list of input files to avoid depending on the order from readdir() helping in obtaining reproducible builds. S/390: - Add common metrics: - CPI (cycles per instruction), prbstate (ratio of instructions executed in problem state compared to total number of instructions), l1mp (Level one instruction and data cache misses per 100 instructions). - Add cache metrics for z13, z14, z15 and z16. - Add metric for TLB and cache. ARM: - Add raw decoding for SPE (Statistical Profiling Extension) v1.3 MTE (Memory Tagging Extension) and MOPS (Memory Operations) load/store. Intel PT hardware tracing: - Add event type names UINTR (User interrupt delivered) and UIRET (Exiting from user interrupt routine), documented in table 32-50 "CFE Packet Type and Vector Fields Details" in the Intel Processor Trace chapter of The Intel SDM Volume 3 version 078. - Add support for new branch instructions ERETS and ERETU. - Fix CYC timestamps after standalone CBR ARM CoreSight hardware tracing: - Allow user to override timestamp and contextid settings. - Fix segfault in dso lookup. - Fix timeless decode mode detection. - Add separate decode paths for timeless and per-thread modes. auxtrace: - Fix address filter entire kernel size. Miscellaneous: - Fix use-after-free and unaligned bugs in the PLT handling routines. - Use zfree() to reduce chances of use after free. - Add missing 0x prefix for addresses printed in hexadecimal in 'perf probe'. - Suppress massive unsupported target platform errors in the unwind code. - Fix return incorrect build_id size in elf_read_build_id(). - Fix 'perf scripts intel-pt-events.py' IPC output for Python 2 . - Add missing new parameter in kfree_skb tracepoint to the python scripts using it. - Add 'perf bench syscall fork' benchmark. - Add support for printing PERF_MEM_LVLNUM_UNC (Uncached access) in 'perf mem'. - Fix wrong size expectation for perf test 'Setup struct perf_event_attr' caused by the patch adding perf_event_attr::config3. - Fix some spelling mistakes" * tag 'perf-tools-for-v6.4-3-2023-05-06' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux: (365 commits) Revert "perf build: Make BUILD_BPF_SKEL default, rename to NO_BPF_SKEL" Revert "perf build: Warn for BPF skeletons if endian mismatches" perf metrics: Fix SEGV with --for-each-cgroup perf bpf skels: Stop using vmlinux.h generated from BTF, use subset of used structs + CO-RE perf stat: Separate bperf from bpf_profiler perf test record+probe_libc_inet_pton: Fix call chain match on x86_64 perf test record+probe_libc_inet_pton: Fix call chain match on s390 perf tracepoint: Fix memory leak in is_valid_tracepoint() perf cs-etm: Add fix for coresight trace for any range of CPUs perf build: Fix unescaped # in perf build-test perf unwind: Suppress massive unsupported target platform errors perf script: Add new parameter in kfree_skb tracepoint to the python scripts using it perf script: Print raw ip instead of binary offset for callchain perf symbols: Fix return incorrect build_id size in elf_read_build_id() perf list: Modify the warning message about scandirat(3) perf list: Fix memory leaks in print_tracepoint_events() perf lock contention: Rework offset calculation with BPF CO-RE perf lock contention: Fix struct rq lock access perf stat: Disable TopdownL1 on hybrid perf stat: Avoid SEGV on counter->name ...
mpage.c
// SPDX-License-Identifier: GPL-2.0
/*
* fs/mpage.c
*
* Copyright (C) 2002, Linus Torvalds.
*
* Contains functions related to preparing and submitting BIOs which contain
* multiple pagecache pages.
*
* 15May2002 Andrew Morton
* Initial version
* 27Jun2002 axboe@suse.de
* use bio_add_page() to build bio's just the right size
*/
#include <linux/kernel.h>
#include <linux/export.h>
#include <linux/mm.h>
#include <linux/kdev_t.h>
#include <linux/gfp.h>
#include <linux/bio.h>
#include <linux/fs.h>
#include <linux/buffer_head.h>
#include <linux/blkdev.h>
#include <linux/highmem.h>
#include <linux/prefetch.h>
#include <linux/mpage.h>
#include <linux/mm_inline.h>
#include <linux/writeback.h>
#include <linux/backing-dev.h>
#include <linux/pagevec.h>
#include "internal.h"
/*
* I/O completion handler for multipage BIOs.
*
* The mpage code never puts partial pages into a BIO (except for end-of-file).
* If a page does not map to a contiguous run of blocks then it simply falls
* back to block_read_full_folio().
*
* Why is this? If a page's completion depends on a number of different BIOs
* which can complete in any order (or at the same time) then determining the
* status of that page is hard. See end_buffer_async_read() for the details.
* There is no point in duplicating all that complexity.
*/
static void mpage_read_end_io(struct bio *bio)
{
struct folio_iter fi;
int err = blk_status_to_errno(bio->bi_status);
bio_for_each_folio_all(fi, bio) {
if (err)
folio_set_error(fi.folio);
else
folio_mark_uptodate(fi.folio);
folio_unlock(fi.folio);
}
bio_put(bio);
}
static void mpage_write_end_io(struct bio *bio)
{
struct folio_iter fi;
int err = blk_status_to_errno(bio->bi_status);
bio_for_each_folio_all(fi, bio) {
if (err) {
folio_set_error(fi.folio);
mapping_set_error(fi.folio->mapping, err);
}
folio_end_writeback(fi.folio);
}
bio_put(bio);
}
static struct bio *mpage_bio_submit_read(struct bio *bio)
{
bio->bi_end_io = mpage_read_end_io;
guard_bio_eod(bio);
submit_bio(bio);
return NULL;
}
static struct bio *mpage_bio_submit_write(struct bio *bio)
{
bio->bi_end_io = mpage_write_end_io;
guard_bio_eod(bio);
submit_bio(bio);
return NULL;
}
/*
* support function for mpage_readahead. The fs supplied get_block might
* return an up to date buffer. This is used to map that buffer into
* the page, which allows read_folio to avoid triggering a duplicate call
* to get_block.
*
* The idea is to avoid adding buffers to pages that don't already have
* them. So when the buffer is up to date and the page size == block size,
* this marks the page up to date instead of adding new buffers.
*/
static void map_buffer_to_folio(struct folio *folio, struct buffer_head *bh,
int page_block)
{
struct inode *inode = folio->mapping->host;
struct buffer_head *page_bh, *head;
int block = 0;
head = folio_buffers(folio);
if (!head) {
/*
* don't make any buffers if there is only one buffer on
* the folio and the folio just needs to be set up to date
*/
if (inode->i_blkbits == PAGE_SHIFT &&
buffer_uptodate(bh)) {
folio_mark_uptodate(folio);
return;
}
create_empty_buffers(&folio->page, i_blocksize(inode), 0);
head = folio_buffers(folio);
}
page_bh = head;
do {
if (block == page_block) {
page_bh->b_state = bh->b_state;
page_bh->b_bdev = bh->b_bdev;
page_bh->b_blocknr = bh->b_blocknr;
break;
}
page_bh = page_bh->b_this_page;
block++;
} while (page_bh != head);
}
struct mpage_readpage_args {
struct bio *bio;
struct folio *folio;
unsigned int nr_pages;
bool is_readahead;
sector_t last_block_in_bio;
struct buffer_head map_bh;
unsigned long first_logical_block;
get_block_t *get_block;
};
/*
* This is the worker routine which does all the work of mapping the disk
* blocks and constructs largest possible bios, submits them for IO if the
* blocks are not contiguous on the disk.
*
* We pass a buffer_head back and forth and use its buffer_mapped() flag to
* represent the validity of its disk mapping and to decide when to do the next
* get_block() call.
*/
static struct bio *do_mpage_readpage(struct mpage_readpage_args *args)
{
struct folio *folio = args->folio;
struct inode *inode = folio->mapping->host;
const unsigned blkbits = inode->i_blkbits;
const unsigned blocks_per_page = PAGE_SIZE >> blkbits;
const unsigned blocksize = 1 << blkbits;
struct buffer_head *map_bh = &args->map_bh;
sector_t block_in_file;
sector_t last_block;
sector_t last_block_in_file;
sector_t blocks[MAX_BUF_PER_PAGE];
unsigned page_block;
unsigned first_hole = blocks_per_page;
struct block_device *bdev = NULL;
int length;
int fully_mapped = 1;
blk_opf_t opf = REQ_OP_READ;
unsigned nblocks;
unsigned relative_block;
gfp_t gfp = mapping_gfp_constraint(folio->mapping, GFP_KERNEL);
/* MAX_BUF_PER_PAGE, for example */
VM_BUG_ON_FOLIO(folio_test_large(folio), folio);
if (args->is_readahead) {
opf |= REQ_RAHEAD;
gfp |= __GFP_NORETRY | __GFP_NOWARN;
}
if (folio_buffers(folio))
goto confused;
block_in_file = (sector_t)folio->index << (PAGE_SHIFT - blkbits);
last_block = block_in_file + args->nr_pages * blocks_per_page;
last_block_in_file = (i_size_read(inode) + blocksize - 1) >> blkbits;
if (last_block > last_block_in_file)
last_block = last_block_in_file;
page_block = 0;
/*
* Map blocks using the result from the previous get_blocks call first.
*/
nblocks = map_bh->b_size >> blkbits;
if (buffer_mapped(map_bh) &&
block_in_file > args->first_logical_block &&
block_in_file < (args->first_logical_block + nblocks)) {
unsigned map_offset = block_in_file - args->first_logical_block;
unsigned last = nblocks - map_offset;
for (relative_block = 0; ; relative_block++) {
if (relative_block == last) {
clear_buffer_mapped(map_bh);
break;
}
if (page_block == blocks_per_page)
break;
blocks[page_block] = map_bh->b_blocknr + map_offset +
relative_block;
page_block++;
block_in_file++;
}
bdev = map_bh->b_bdev;
}
/*
* Then do more get_blocks calls until we are done with this folio.
*/
map_bh->b_folio = folio;
while (page_block < blocks_per_page) {
map_bh->b_state = 0;
map_bh->b_size = 0;
if (block_in_file < last_block) {
map_bh->b_size = (last_block-block_in_file) << blkbits;
if (args->get_block(inode, block_in_file, map_bh, 0))
goto confused;
args->first_logical_block = block_in_file;
}
if (!buffer_mapped(map_bh)) {
fully_mapped = 0;
if (first_hole == blocks_per_page)
first_hole = page_block;
page_block++;
block_in_file++;
continue;
}
/* some filesystems will copy data into the page during
* the get_block call, in which case we don't want to
* read it again. map_buffer_to_folio copies the data
* we just collected from get_block into the folio's buffers
* so read_folio doesn't have to repeat the get_block call
*/
if (buffer_uptodate(map_bh)) {
map_buffer_to_folio(folio, map_bh, page_block);
goto confused;
}
if (first_hole != blocks_per_page)
goto confused; /* hole -> non-hole */
/* Contiguous blocks? */
if (page_block && blocks[page_block-1] != map_bh->b_blocknr-1)
goto confused;
nblocks = map_bh->b_size >> blkbits;
for (relative_block = 0; ; relative_block++) {
if (relative_block == nblocks) {
clear_buffer_mapped(map_bh);
break;
} else if (page_block == blocks_per_page)
break;
blocks[page_block] = map_bh->b_blocknr+relative_block;
page_block++;
block_in_file++;
}
bdev = map_bh->b_bdev;
}
if (first_hole != blocks_per_page) {
folio_zero_segment(folio, first_hole << blkbits, PAGE_SIZE);
if (first_hole == 0) {
folio_mark_uptodate(folio);
folio_unlock(folio);
goto out;
}
} else if (fully_mapped) {
folio_set_mappedtodisk(folio);
}
/*
* This folio will go to BIO. Do we need to send this BIO off first?
*/
if (args->bio && (args->last_block_in_bio != blocks[0] - 1))
args->bio = mpage_bio_submit_read(args->bio);
alloc_new:
if (args->bio == NULL) {
args->bio = bio_alloc(bdev, bio_max_segs(args->nr_pages), opf,
gfp);
if (args->bio == NULL)
goto confused;
args->bio->bi_iter.bi_sector = blocks[0] << (blkbits - 9);
}
length = first_hole << blkbits;
if (!bio_add_folio(args->bio, folio, length, 0)) {
args->bio = mpage_bio_submit_read(args->bio);
goto alloc_new;
}
relative_block = block_in_file - args->first_logical_block;
nblocks = map_bh->b_size >> blkbits;
if ((buffer_boundary(map_bh) && relative_block == nblocks) ||
(first_hole != blocks_per_page))
args->bio = mpage_bio_submit_read(args->bio);
else
args->last_block_in_bio = blocks[blocks_per_page - 1];
out:
return args->bio;
confused:
if (args->bio)
args->bio = mpage_bio_submit_read(args->bio);
if (!folio_test_uptodate(folio))
block_read_full_folio(folio, args->get_block);
else
folio_unlock(folio);
goto out;
}
/**
* mpage_readahead - start reads against pages
* @rac: Describes which pages to read.
* @get_block: The filesystem's block mapper function.
*
* This function walks the pages and the blocks within each page, building and
* emitting large BIOs.
*
* If anything unusual happens, such as:
*
* - encountering a page which has buffers
* - encountering a page which has a non-hole after a hole
* - encountering a page with non-contiguous blocks
*
* then this code just gives up and calls the buffer_head-based read function.
* It does handle a page which has holes at the end - that is a common case:
* the end-of-file on blocksize < PAGE_SIZE setups.
*
* BH_Boundary explanation:
*
* There is a problem. The mpage read code assembles several pages, gets all
* their disk mappings, and then submits them all. That's fine, but obtaining
* the disk mappings may require I/O. Reads of indirect blocks, for example.
*
* So an mpage read of the first 16 blocks of an ext2 file will cause I/O to be
* submitted in the following order:
*
* 12 0 1 2 3 4 5 6 7 8 9 10 11 13 14 15 16
*
* because the indirect block has to be read to get the mappings of blocks
* 13,14,15,16. Obviously, this impacts performance.
*
* So what we do it to allow the filesystem's get_block() function to set
* BH_Boundary when it maps block 11. BH_Boundary says: mapping of the block
* after this one will require I/O against a block which is probably close to
* this one. So you should push what I/O you have currently accumulated.
*
* This all causes the disk requests to be issued in the correct order.
*/
void mpage_readahead(struct readahead_control *rac, get_block_t get_block)
{
struct folio *folio;
struct mpage_readpage_args args = {
.get_block = get_block,
.is_readahead = true,
};
while ((folio = readahead_folio(rac))) {
prefetchw(&folio->flags);
args.folio = folio;
args.nr_pages = readahead_count(rac);
args.bio = do_mpage_readpage(&args);
}
if (args.bio)
mpage_bio_submit_read(args.bio);
}
EXPORT_SYMBOL(mpage_readahead);
/*
* This isn't called much at all
*/
int mpage_read_folio(struct folio *folio, get_block_t get_block)
{
struct mpage_readpage_args args = {
.folio = folio,
.nr_pages = 1,
.get_block = get_block,
};
args.bio = do_mpage_readpage(&args);
if (args.bio)
mpage_bio_submit_read(args.bio);
return 0;
}
EXPORT_SYMBOL(mpage_read_folio);
/*
* Writing is not so simple.
*
* If the page has buffers then they will be used for obtaining the disk
* mapping. We only support pages which are fully mapped-and-dirty, with a
* special case for pages which are unmapped at the end: end-of-file.
*
* If the page has no buffers (preferred) then the page is mapped here.
*
* If all blocks are found to be contiguous then the page can go into the
* BIO. Otherwise fall back to the mapping's writepage().
*
* FIXME: This code wants an estimate of how many pages are still to be
* written, so it can intelligently allocate a suitably-sized BIO. For now,
* just allocate full-size (16-page) BIOs.
*/
struct mpage_data {
struct bio *bio;
sector_t last_block_in_bio;
get_block_t *get_block;
};
/*
* We have our BIO, so we can now mark the buffers clean. Make
* sure to only clean buffers which we know we'll be writing.
*/
static void clean_buffers(struct page *page, unsigned first_unmapped)
{
unsigned buffer_counter = 0;
struct buffer_head *bh, *head;
if (!page_has_buffers(page))
return;
head = page_buffers(page);
bh = head;
do {
if (buffer_counter++ == first_unmapped)
break;
clear_buffer_dirty(bh);
bh = bh->b_this_page;
} while (bh != head);
/*
* we cannot drop the bh if the page is not uptodate or a concurrent
* read_folio would fail to serialize with the bh and it would read from
* disk before we reach the platter.
*/
if (buffer_heads_over_limit && PageUptodate(page))
try_to_free_buffers(page_folio(page));
}
/*
* For situations where we want to clean all buffers attached to a page.
* We don't need to calculate how many buffers are attached to the page,
* we just need to specify a number larger than the maximum number of buffers.
*/
void clean_page_buffers(struct page *page)
{
clean_buffers(page, ~0U);
}
static int __mpage_writepage(struct folio *folio, struct writeback_control *wbc,
void *data)
{
struct mpage_data *mpd = data;
struct bio *bio = mpd->bio;
struct address_space *mapping = folio->mapping;
struct inode *inode = mapping->host;
const unsigned blkbits = inode->i_blkbits;
const unsigned blocks_per_page = PAGE_SIZE >> blkbits;
sector_t last_block;
sector_t block_in_file;
sector_t blocks[MAX_BUF_PER_PAGE];
unsigned page_block;
unsigned first_unmapped = blocks_per_page;
struct block_device *bdev = NULL;
int boundary = 0;
sector_t boundary_block = 0;
struct block_device *boundary_bdev = NULL;
size_t length;
struct buffer_head map_bh;
loff_t i_size = i_size_read(inode);
int ret = 0;
struct buffer_head *head = folio_buffers(folio);
if (head) {
struct buffer_head *bh = head;
/* If they're all mapped and dirty, do it */
page_block = 0;
do {
BUG_ON(buffer_locked(bh));
if (!buffer_mapped(bh)) {
/*
* unmapped dirty buffers are created by
* block_dirty_folio -> mmapped data
*/
if (buffer_dirty(bh))
goto confused;
if (first_unmapped == blocks_per_page)
first_unmapped = page_block;
continue;
}
if (first_unmapped != blocks_per_page)
goto confused; /* hole -> non-hole */
if (!buffer_dirty(bh) || !buffer_uptodate(bh))
goto confused;
if (page_block) {
if (bh->b_blocknr != blocks[page_block-1] + 1)
goto confused;
}
blocks[page_block++] = bh->b_blocknr;
boundary = buffer_boundary(bh);
if (boundary) {
boundary_block = bh->b_blocknr;
boundary_bdev = bh->b_bdev;
}
bdev = bh->b_bdev;
} while ((bh = bh->b_this_page) != head);
if (first_unmapped)
goto page_is_mapped;
/*
* Page has buffers, but they are all unmapped. The page was
* created by pagein or read over a hole which was handled by
* block_read_full_folio(). If this address_space is also
* using mpage_readahead then this can rarely happen.
*/
goto confused;
}
/*
* The page has no buffers: map it to disk
*/
BUG_ON(!folio_test_uptodate(folio));
block_in_file = (sector_t)folio->index << (PAGE_SHIFT - blkbits);
/*
* Whole page beyond EOF? Skip allocating blocks to avoid leaking
* space.
*/
if (block_in_file >= (i_size + (1 << blkbits) - 1) >> blkbits)
goto page_is_mapped;
last_block = (i_size - 1) >> blkbits;
map_bh.b_folio = folio;
for (page_block = 0; page_block < blocks_per_page; ) {
map_bh.b_state = 0;
map_bh.b_size = 1 << blkbits;
if (mpd->get_block(inode, block_in_file, &map_bh, 1))
goto confused;
if (!buffer_mapped(&map_bh))
goto confused;
if (buffer_new(&map_bh))
clean_bdev_bh_alias(&map_bh);
if (buffer_boundary(&map_bh)) {
boundary_block = map_bh.b_blocknr;
boundary_bdev = map_bh.b_bdev;
}
if (page_block) {
if (map_bh.b_blocknr != blocks[page_block-1] + 1)
goto confused;
}
blocks[page_block++] = map_bh.b_blocknr;
boundary = buffer_boundary(&map_bh);
bdev = map_bh.b_bdev;
if (block_in_file == last_block)
break;
block_in_file++;
}
BUG_ON(page_block == 0);
first_unmapped = page_block;
page_is_mapped:
/* Don't bother writing beyond EOF, truncate will discard the folio */
if (folio_pos(folio) >= i_size)
goto confused;
length = folio_size(folio);
if (folio_pos(folio) + length > i_size) {
/*
* The page straddles i_size. It must be zeroed out on each
* and every writepage invocation because it may be mmapped.
* "A file is mapped in multiples of the page size. For a file
* that is not a multiple of the page size, the remaining memory
* is zeroed when mapped, and writes to that region are not
* written out to the file."
*/
length = i_size - folio_pos(folio);
folio_zero_segment(folio, length, folio_size(folio));
}
/*
* This page will go to BIO. Do we need to send this BIO off first?
*/
if (bio && mpd->last_block_in_bio != blocks[0] - 1)
bio = mpage_bio_submit_write(bio);
alloc_new:
if (bio == NULL) {
bio = bio_alloc(bdev, BIO_MAX_VECS,
REQ_OP_WRITE | wbc_to_write_flags(wbc),
GFP_NOFS);
bio->bi_iter.bi_sector = blocks[0] << (blkbits - 9);
wbc_init_bio(wbc, bio);
}
/*
* Must try to add the page before marking the buffer clean or
* the confused fail path above (OOM) will be very confused when
* it finds all bh marked clean (i.e. it will not write anything)
*/
wbc_account_cgroup_owner(wbc, &folio->page, folio_size(folio));
length = first_unmapped << blkbits;
if (!bio_add_folio(bio, folio, length, 0)) {
bio = mpage_bio_submit_write(bio);
goto alloc_new;
}
clean_buffers(&folio->page, first_unmapped);
BUG_ON(folio_test_writeback(folio));
folio_start_writeback(folio);
folio_unlock(folio);
if (boundary || (first_unmapped != blocks_per_page)) {
bio = mpage_bio_submit_write(bio);
if (boundary_block) {
write_boundary_block(boundary_bdev,
boundary_block, 1 << blkbits);
}
} else {
mpd->last_block_in_bio = blocks[blocks_per_page - 1];
}
goto out;
confused:
if (bio)
bio = mpage_bio_submit_write(bio);
/*
* The caller has a ref on the inode, so *mapping is stable
*/
ret = block_write_full_page(&folio->page, mpd->get_block, wbc);
mapping_set_error(mapping, ret);
out:
mpd->bio = bio;
return ret;
}
/**
* mpage_writepages - walk the list of dirty pages of the given address space & writepage() all of them
* @mapping: address space structure to write
* @wbc: subtract the number of written pages from *@wbc->nr_to_write
* @get_block: the filesystem's block mapper function.
*
* This is a library function, which implements the writepages()
* address_space_operation.
*/
int
mpage_writepages(struct address_space *mapping,
struct writeback_control *wbc, get_block_t get_block)
{
struct mpage_data mpd = {
.get_block = get_block,
};
struct blk_plug plug;
int ret;
blk_start_plug(&plug);
ret = write_cache_pages(mapping, wbc, __mpage_writepage, &mpd);
if (mpd.bio)
mpage_bio_submit_write(mpd.bio);
blk_finish_plug(&plug);
return ret;
}
EXPORT_SYMBOL(mpage_writepages);
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...