July 2024 - Linux-stable-mirror

[PATCH] media: atomisp: Fix streaming no longer working on BYT / ISP2400 devices

by Hans de Goede

Commit a0821ca14bb8 ("media: atomisp: Remove test pattern generator (TPG) support") broke BYT support because it removed a seemingly unused field from struct sh_css_sp_config and a seemingly unused value from enum ia_css_input_mode. But these are part of the ABI between the kernel and firmware on ISP2400 and this part of the TPG support removal changes broke ISP2400 support. ISP2401 support was not affected because on ISP2401 only a part of struct sh_css_sp_config is used. Restore the removed field and enum value to fix this. Fixes: a0821ca14bb8 ("media: atomisp: Remove test pattern generator (TPG) support") Cc: stable(a)vger.kernel.org Signed-off-by: Hans de Goede <hdegoede(a)redhat.com> --- .../media/atomisp/pci/ia_css_stream_public.h | 8 ++++++-- .../media/atomisp/pci/sh_css_internal.h | 19 ++++++++++++++++--- 2 files changed, 22 insertions(+), 5 deletions(-) diff --git a/drivers/staging/media/atomisp/pci/ia_css_stream_public.h b/drivers/staging/media/atomisp/pci/ia_css_stream_public.h index 961c61288083..aad860e54d3a 100644 --- a/drivers/staging/media/atomisp/pci/ia_css_stream_public.h +++ b/drivers/staging/media/atomisp/pci/ia_css_stream_public.h @@ -27,12 +27,16 @@ #include "ia_css_prbs.h" #include "ia_css_input_port.h" -/* Input modes, these enumerate all supported input modes. - * Note that not all ISP modes support all input modes. +/* + * Input modes, these enumerate all supported input modes. + * This enum is part of the atomisp firmware ABI and must + * NOT be changed! + * Note that not all ISP modes support all input modes. */ enum ia_css_input_mode { IA_CSS_INPUT_MODE_SENSOR, /** data from sensor */ IA_CSS_INPUT_MODE_FIFO, /** data from input-fifo */ + IA_CSS_INPUT_MODE_TPG, /** data from test-pattern generator */ IA_CSS_INPUT_MODE_PRBS, /** data from pseudo-random bit stream */ IA_CSS_INPUT_MODE_MEMORY, /** data from a frame in memory */ IA_CSS_INPUT_MODE_BUFFERED_SENSOR /** data is sent through mipi buffer */ diff --git a/drivers/staging/media/atomisp/pci/sh_css_internal.h b/drivers/staging/media/atomisp/pci/sh_css_internal.h index a2d972ea3fa0..959e7f549641 100644 --- a/drivers/staging/media/atomisp/pci/sh_css_internal.h +++ b/drivers/staging/media/atomisp/pci/sh_css_internal.h @@ -344,7 +344,14 @@ struct sh_css_sp_input_formatter_set { #define IA_CSS_MIPI_SIZE_CHECK_MAX_NOF_ENTRIES_PER_PORT (3) -/* SP configuration information */ +/* + * SP configuration information + * + * This struct is part of the atomisp firmware ABI and is directly copied + * to ISP DRAM by sh_css_store_sp_group_to_ddr() + * + * Do NOT change this struct's layout or remove seemingly unused fields! + */ struct sh_css_sp_config { u8 no_isp_sync; /* Signal host immediately after start */ u8 enable_raw_pool_locking; /** Enable Raw Buffer Locking for HALv3 Support */ @@ -354,6 +361,10 @@ struct sh_css_sp_config { host (true) or when they are passed to the preview/video pipe (false). */ + /* + * Note the fields below are only used on the ISP2400 not on the ISP2401, + * sh_css_store_sp_group_to_ddr() skip copying these when run on the ISP2401. + */ struct { u8 a_changed; u8 b_changed; @@ -363,11 +374,13 @@ struct sh_css_sp_config { } input_formatter; sync_generator_cfg_t sync_gen; + tpg_cfg_t tpg; prbs_cfg_t prbs; input_system_cfg_t input_circuit; u8 input_circuit_cfg_changed; - u32 mipi_sizes_for_check[N_CSI_PORTS][IA_CSS_MIPI_SIZE_CHECK_MAX_NOF_ENTRIES_PER_PORT]; - u8 enable_isys_event_queue; + u32 mipi_sizes_for_check[N_CSI_PORTS][IA_CSS_MIPI_SIZE_CHECK_MAX_NOF_ENTRIES_PER_PORT]; + /* These last 2 fields are used on both the ISP2400 and the ISP2401 */ + u8 enable_isys_event_queue; u8 disable_cont_vf; }; -- 2.45.2

11 months, 1 week

1
0
0 0

[PATCH] thermal/drivers/qcom/lmh: remove false lockdep backtrace

by Dmitry Baryshkov

Annotate LMH IRQs with lockdep classes so that the lockdep doesn't report possible recursive locking issue between LMH and GIC interrupts. For the reference: CPU0 ---- lock(&irq_desc_lock_class); lock(&irq_desc_lock_class); *** DEADLOCK *** Call trace: dump_backtrace+0x98/0xf0 show_stack+0x18/0x24 dump_stack_lvl+0x90/0xd0 dump_stack+0x18/0x24 print_deadlock_bug+0x258/0x348 __lock_acquire+0x1078/0x1f44 lock_acquire+0x1fc/0x32c _raw_spin_lock_irqsave+0x60/0x88 __irq_get_desc_lock+0x58/0x98 enable_irq+0x38/0xa0 lmh_enable_interrupt+0x2c/0x38 irq_enable+0x40/0x8c __irq_startup+0x78/0xa4 irq_startup+0x78/0x168 __enable_irq+0x70/0x7c enable_irq+0x4c/0xa0 qcom_cpufreq_ready+0x20/0x2c cpufreq_online+0x2a8/0x988 cpufreq_add_dev+0x80/0x98 subsys_interface_register+0x104/0x134 cpufreq_register_driver+0x150/0x234 qcom_cpufreq_hw_driver_probe+0x2a8/0x388 platform_probe+0x68/0xc0 really_probe+0xbc/0x298 __driver_probe_device+0x78/0x12c driver_probe_device+0x3c/0x160 __device_attach_driver+0xb8/0x138 bus_for_each_drv+0x84/0xe0 __device_attach+0x9c/0x188 device_initial_probe+0x14/0x20 bus_probe_device+0xac/0xb0 deferred_probe_work_func+0x8c/0xc8 process_one_work+0x20c/0x62c worker_thread+0x1bc/0x36c kthread+0x120/0x124 ret_from_fork+0x10/0x20 Fixes: 53bca371cdf7 ("thermal/drivers/qcom: Add support for LMh driver") Cc: stable(a)vger.kernel.org Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov(a)linaro.org> --- drivers/thermal/qcom/lmh.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/thermal/qcom/lmh.c b/drivers/thermal/qcom/lmh.c index 5225b3621a56..d2d49264cf83 100644 --- a/drivers/thermal/qcom/lmh.c +++ b/drivers/thermal/qcom/lmh.c @@ -73,7 +73,14 @@ static struct irq_chip lmh_irq_chip = { static int lmh_irq_map(struct irq_domain *d, unsigned int irq, irq_hw_number_t hw) { struct lmh_hw_data *lmh_data = d->host_data; + static struct lock_class_key lmh_lock_key; + static struct lock_class_key lmh_request_key; + /* + * This lock class tells lockdep that GPIO irqs are in a different + * category than their parents, so it won't report false recursion. + */ + irq_set_lockdep_class(irq, &lmh_lock_key, &lmh_request_key); irq_set_chip_and_handler(irq, &lmh_irq_chip, handle_simple_irq); irq_set_chip_data(irq, lmh_data); --- base-commit: 797012914d2d031430268fe512af0ccd7d8e46ef change-id: 20240721-lmh-lockdep-88de09e77089 Best regards, -- Dmitry Baryshkov <dmitry.baryshkov(a)linaro.org>

11 months, 1 week

1
0
0 0

[PATCH v2 4.19 4/4] ext4: drop unnecessary journal handle in delalloc write

by WangYuli

From: Zhang Yi <yi.zhang(a)huawei.com> commit cc883236b79297f6266ca6f4e7f24f3fd3c736c1 upstream After we factor out the inline data write procedure from ext4_da_write_end(), we don't need to start journal handle for the cases of both buffer overwrite and append-write. If we need to update i_disksize, mark_inode_dirty() do start handle and update inode buffer. So we could just remove all the journal handle codes in the delalloc write procedure. After this patch, we could get a lot of performance improvement. Below is the Unixbench comparison data test on my machine with 'Intel Xeon Gold 5120' CPU and nvme SSD backend. Test cmd: ./Run -c 56 -i 3 fstime fsbuffer fsdisk Before this patch: System Benchmarks Partial Index BASELINE RESULT INDEX File Copy 1024 bufsize 2000 maxblocks 3960.0 422965.0 1068.1 File Copy 256 bufsize 500 maxblocks 1655.0 105077.0 634.9 File Copy 4096 bufsize 8000 maxblocks 5800.0 1429092.0 2464.0 ====== System Benchmarks Index Score (Partial Only) 1186.6 After this patch: System Benchmarks Partial Index BASELINE RESULT INDEX File Copy 1024 bufsize 2000 maxblocks 3960.0 732716.0 1850.3 File Copy 256 bufsize 500 maxblocks 1655.0 184940.0 1117.5 File Copy 4096 bufsize 8000 maxblocks 5800.0 2427152.0 4184.7 ====== System Benchmarks Index Score (Partial Only) 2053.0 Cc: stable(a)vger.kernel.org Signed-off-by: Zhang Yi <yi.zhang(a)huawei.com> Reviewed-by: Jan Kara <jack(a)suse.cz> Signed-off-by: Theodore Ts'o <tytso(a)mit.edu> Link: https://lore.kernel.org/r/20210716122024.1105856-5-yi.zhang@huawei.com Reviewed-by: Cheng Nie <niecheng1(a)uniontech.com> Signed-off-by: Dandan Zhang <zhangdandan(a)uniontech.com> Signed-off-by: WangYuli <wangyuli(a)uniontech.com> --- fs/ext4/inode.c | 60 +++++-------------------------------------------- 1 file changed, 5 insertions(+), 55 deletions(-) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 1e6753084a00..597c2f49c889 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3032,19 +3032,6 @@ static int ext4_nonda_switch(struct super_block *sb) return 0; } -/* We always reserve for an inode update; the superblock could be there too */ -static int ext4_da_write_credits(struct inode *inode, loff_t pos, unsigned len) -{ - if (likely(ext4_has_feature_large_file(inode->i_sb))) - return 1; - - if (pos + len <= 0x7fffffffULL) - return 1; - - /* We might need to update the superblock to set LARGE_FILE */ - return 2; -} - static int ext4_da_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) @@ -3053,7 +3040,6 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, struct page *page; pgoff_t index; struct inode *inode = mapping->host; - handle_t *handle; if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) return -EIO; @@ -3079,41 +3065,11 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, return 0; } - /* - * grab_cache_page_write_begin() can take a long time if the - * system is thrashing due to memory pressure, or if the page - * is being written back. So grab it first before we start - * the transaction handle. This also allows us to allocate - * the page (if needed) without using GFP_NOFS. - */ -retry_grab: +retry: page = grab_cache_page_write_begin(mapping, index, flags); if (!page) return -ENOMEM; - unlock_page(page); - - /* - * With delayed allocation, we don't log the i_disksize update - * if there is delayed block allocation. But we still need - * to journalling the i_disksize update if writes to the end - * of file which has an already mapped buffer. - */ -retry_journal: - handle = ext4_journal_start(inode, EXT4_HT_WRITE_PAGE, - ext4_da_write_credits(inode, pos, len)); - if (IS_ERR(handle)) { - put_page(page); - return PTR_ERR(handle); - } - lock_page(page); - if (page->mapping != mapping) { - /* The page got truncated from under us */ - unlock_page(page); - put_page(page); - ext4_journal_stop(handle); - goto retry_grab; - } /* In case writeback began while the page was unlocked */ wait_for_stable_page(page); @@ -3125,20 +3081,18 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, #endif if (ret < 0) { unlock_page(page); - ext4_journal_stop(handle); + put_page(page); /* * block_write_begin may have instantiated a few blocks * outside i_size. Trim these off again. Don't need - * i_size_read because we hold i_mutex. + * i_size_read because we hold inode lock. */ if (pos + len > inode->i_size) ext4_truncate_failed_write(inode); if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) - goto retry_journal; - - put_page(page); + goto retry; return ret; } @@ -3175,8 +3129,6 @@ static int ext4_da_write_end(struct file *file, struct page *page, void *fsdata) { struct inode *inode = mapping->host; - int ret; - handle_t *handle = ext4_journal_current_handle(); loff_t new_i_size; unsigned long start, end; int write_mode = (int)(unsigned long)fsdata; @@ -3215,9 +3167,7 @@ static int ext4_da_write_end(struct file *file, ext4_da_should_update_i_disksize(page, end)) ext4_update_i_disksize(inode, new_i_size); - copied = generic_write_end(file, mapping, pos, len, copied, page, fsdata); - ret = ext4_journal_stop(handle); - return ret ? ret : copied; + return generic_write_end(file, mapping, pos, len, copied, page, fsdata); } static void ext4_da_invalidatepage(struct page *page, unsigned int offset, -- 2.43.4

11 months, 1 week

1
0
0 0

[PATCH v2 4.19 3/4] ext4: factor out write end code of inline file

by WangYuli

From: Zhang Yi <yi.zhang(a)huawei.com> commit 6984aef59814fb5c47b0e30c56e101186b5ebf8c upstream Now that the inline_data file write end procedure are falled into the common write end functions, it is not clear. Factor them out and do some cleanup. This patch also drop ext4_da_write_inline_data_end() and switch to use ext4_write_inline_data_end() instead because we also need to do the same error processing if we failed to write data into inline entry. Cc: stable(a)vger.kernel.org Signed-off-by: Zhang Yi <yi.zhang(a)huawei.com> Reviewed-by: Jan Kara <jack(a)suse.cz> Signed-off-by: Theodore Ts'o <tytso(a)mit.edu> Link: https://lore.kernel.org/r/20210716122024.1105856-4-yi.zhang@huawei.com Reviewed-by: Cheng Nie <niecheng1(a)uniontech.com> Signed-off-by: Dandan Zhang <zhangdandan(a)uniontech.com> Signed-off-by: WangYuli <wangyuli(a)uniontech.com> --- fs/ext4/ext4.h | 3 -- fs/ext4/inline.c | 127 ++++++++++++++++++++++++----------------------- fs/ext4/inode.c | 74 +++++++++------------------ 3 files changed, 89 insertions(+), 115 deletions(-) diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index fa2579abea7d..6394c8a3803c 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -3083,9 +3083,6 @@ extern int ext4_da_write_inline_data_begin(struct address_space *mapping, unsigned flags, struct page **pagep, void **fsdata); -extern int ext4_da_write_inline_data_end(struct inode *inode, loff_t pos, - unsigned len, unsigned copied, - struct page *page); extern int ext4_try_add_inline_entry(handle_t *handle, struct ext4_filename *fname, struct inode *dir, struct inode *inode); diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c index de04bd5fb551..eeb696c85a61 100644 --- a/fs/ext4/inline.c +++ b/fs/ext4/inline.c @@ -741,40 +741,82 @@ int ext4_try_to_write_inline_data(struct address_space *mapping, int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len, unsigned copied, struct page *page) { - int ret, no_expand; + handle_t *handle = ext4_journal_current_handle(); + int no_expand; void *kaddr; struct ext4_iloc iloc; + int ret = 0, ret2; if (unlikely(copied < len) && !PageUptodate(page)) - return 0; + copied = 0; - ret = ext4_get_inode_loc(inode, &iloc); - if (ret) { - ext4_std_error(inode->i_sb, ret); - return ret; - } + if (likely(copied)) { + ret = ext4_get_inode_loc(inode, &iloc); + if (ret) { + unlock_page(page); + put_page(page); + ext4_std_error(inode->i_sb, ret); + goto out; + } + ext4_write_lock_xattr(inode, &no_expand); + BUG_ON(!ext4_has_inline_data(inode)); - ext4_write_lock_xattr(inode, &no_expand); - BUG_ON(!ext4_has_inline_data(inode)); + kaddr = kmap_atomic(page); + ext4_write_inline_data(inode, &iloc, kaddr, pos, copied); + kunmap_atomic(kaddr); + SetPageUptodate(page); + /* clear page dirty so that writepages wouldn't work for us. */ + ClearPageDirty(page); - /* - * ei->i_inline_off may have changed since ext4_write_begin() - * called ext4_try_to_write_inline_data() - */ - (void) ext4_find_inline_data_nolock(inode); + /* + * ei->i_inline_off may have changed since ext4_write_begin() + * called ext4_try_to_write_inline_data() + */ + (void) ext4_find_inline_data_nolock(inode); - kaddr = kmap_atomic(page); - ext4_write_inline_data(inode, &iloc, kaddr, pos, copied); - kunmap_atomic(kaddr); - SetPageUptodate(page); - /* clear page dirty so that writepages wouldn't work for us. */ - ClearPageDirty(page); + ext4_write_unlock_xattr(inode, &no_expand); + brelse(iloc.bh); - ext4_write_unlock_xattr(inode, &no_expand); - brelse(iloc.bh); - mark_inode_dirty(inode); + /* + * It's important to update i_size while still holding page + * lock: page writeout could otherwise come in and zero + * beyond i_size. + */ + ext4_update_inode_size(inode, pos + copied); + } + unlock_page(page); + put_page(page); - return copied; + /* + * Don't mark the inode dirty under page lock. First, it unnecessarily + * makes the holding time of page lock longer. Second, it forces lock + * ordering of page lock and transaction start for journaling + * filesystems. + */ + if (likely(copied)) + mark_inode_dirty(inode); +out: + /* + * If we didn't copy as much data as expected, we need to trim back + * size of xattr containing inline data. + */ + if (pos + len > inode->i_size && ext4_can_truncate(inode)) + ext4_orphan_add(handle, inode); + + ret2 = ext4_journal_stop(handle); + if (!ret) + ret = ret2; + if (pos + len > inode->i_size) { + ext4_truncate_failed_write(inode); + /* + * If truncate failed early the inode might still be + * on the orphan list; we need to make sure the inode + * is removed from the orphan list in that case. + */ + if (inode->i_nlink) + ext4_orphan_del(NULL, inode); + } + return ret ? ret : copied; } struct buffer_head * @@ -955,43 +997,6 @@ int ext4_da_write_inline_data_begin(struct address_space *mapping, return ret; } -int ext4_da_write_inline_data_end(struct inode *inode, loff_t pos, - unsigned len, unsigned copied, - struct page *page) -{ - int ret; - - ret = ext4_write_inline_data_end(inode, pos, len, copied, page); - if (ret < 0) { - unlock_page(page); - put_page(page); - return ret; - } - copied = ret; - - /* - * No need to use i_size_read() here, the i_size - * cannot change under us because we hold i_mutex. - * - * But it's important to update i_size while still holding page lock: - * page writeout could otherwise come in and zero beyond i_size. - */ - if (pos+copied > inode->i_size) - i_size_write(inode, pos+copied); - unlock_page(page); - put_page(page); - - /* - * Don't mark the inode dirty under page lock. First, it unnecessarily - * makes the holding time of page lock longer. Second, it forces lock - * ordering of page lock and transaction start for journaling - * filesystems. - */ - mark_inode_dirty(inode); - - return copied; -} - #ifdef INLINE_DIR_DEBUG void ext4_show_inline_dir(struct inode *dir, struct buffer_head *bh, void *inline_start, int inline_size) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 44a715e6aae1..1e6753084a00 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1415,23 +1415,13 @@ static int ext4_write_end(struct file *file, loff_t old_size = inode->i_size; int ret = 0, ret2; int i_size_changed = 0; - int inline_data = ext4_has_inline_data(inode); trace_ext4_write_end(inode, pos, len, copied); - if (inline_data && - ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)) { - ret = ext4_write_inline_data_end(inode, pos, len, - copied, page); - if (ret < 0) { - unlock_page(page); - put_page(page); - goto errout; - } - copied = ret; - ret = 0; - } else - copied = block_write_end(file, mapping, pos, - len, copied, page, fsdata); + + if (ext4_has_inline_data(inode)) + return ext4_write_inline_data_end(inode, pos, len, copied, page); + + copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); /* * it's important to update i_size while still holding page lock: * page writeout could otherwise come in and zero beyond i_size. @@ -1448,10 +1438,9 @@ static int ext4_write_end(struct file *file, * ordering of page lock and transaction start for journaling * filesystems. */ - if (i_size_changed || inline_data) + if (i_size_changed) ext4_mark_inode_dirty(handle, inode); -errout: if (pos + len > inode->i_size && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside @@ -1523,7 +1512,6 @@ static int ext4_journalled_write_end(struct file *file, int partial = 0; unsigned from, to; int size_changed = 0; - int inline_data = ext4_has_inline_data(inode); trace_ext4_journalled_write_end(inode, pos, len, copied); from = pos & (PAGE_SIZE - 1); @@ -1531,17 +1519,10 @@ static int ext4_journalled_write_end(struct file *file, BUG_ON(!ext4_handle_valid(handle)); - if (inline_data) { - ret = ext4_write_inline_data_end(inode, pos, len, - copied, page); - if (ret < 0) { - unlock_page(page); - put_page(page); - goto errout; - } - copied = ret; - ret = 0; - } else if (unlikely(copied < len) && !PageUptodate(page)) { + if (ext4_has_inline_data(inode)) + return ext4_write_inline_data_end(inode, pos, len, copied, page); + + if (unlikely(copied < len) && !PageUptodate(page)) { copied = 0; ext4_journalled_zero_new_buffers(handle, page, from, to); } else { @@ -1563,13 +1544,12 @@ static int ext4_journalled_write_end(struct file *file, if (old_size < pos) pagecache_isize_extended(inode, old_size, pos); - if (size_changed || inline_data) { + if (size_changed) { ret2 = ext4_mark_inode_dirty(handle, inode); if (!ret) ret = ret2; } -errout: if (pos + len > inode->i_size && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside @@ -3195,7 +3175,7 @@ static int ext4_da_write_end(struct file *file, struct page *page, void *fsdata) { struct inode *inode = mapping->host; - int ret = 0, ret2; + int ret; handle_t *handle = ext4_journal_current_handle(); loff_t new_i_size; unsigned long start, end; @@ -3206,6 +3186,12 @@ static int ext4_da_write_end(struct file *file, len, copied, page, fsdata); trace_ext4_da_write_end(inode, pos, len, copied); + + if (write_mode != CONVERT_INLINE_DATA && + ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA) && + ext4_has_inline_data(inode)) + return ext4_write_inline_data_end(inode, pos, len, copied, page); + start = pos & (PAGE_SIZE - 1); end = start + copied - 1; @@ -3225,26 +3211,12 @@ static int ext4_da_write_end(struct file *file, * ext4_da_write_inline_data_end(). */ new_i_size = pos + copied; - if (copied && new_i_size > inode->i_size) { - if (ext4_has_inline_data(inode) || - ext4_da_should_update_i_disksize(page, end)) - ext4_update_i_disksize(inode, new_i_size); - } - - if (write_mode != CONVERT_INLINE_DATA && - ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA) && - ext4_has_inline_data(inode)) - ret = ext4_da_write_inline_data_end(inode, pos, len, copied, - page); - else - ret = generic_write_end(file, mapping, pos, len, copied, - page, fsdata); - - copied = ret; - ret2 = ext4_journal_stop(handle); - if (!ret) - ret = ret2; + if (copied && new_i_size > inode->i_size && + ext4_da_should_update_i_disksize(page, end)) + ext4_update_i_disksize(inode, new_i_size); + copied = generic_write_end(file, mapping, pos, len, copied, page, fsdata); + ret = ext4_journal_stop(handle); return ret ? ret : copied; } -- 2.43.4

11 months, 1 week

1
0
0 0

[PATCH v2 4.19 2/4] ext4: correct the error path of ext4_write_inline_data_end()

by WangYuli

From: Zhang Yi <yi.zhang(a)huawei.com> commit 55ce2f649b9e88111270333a8127e23f4f8f42d7 upstream Current error path of ext4_write_inline_data_end() is not correct. Firstly, it should pass out the error value if ext4_get_inode_loc() return fail, or else it could trigger infinite loop if we inject error here. And then it's better to add inode to orphan list if it return fail in ext4_journal_stop(), otherwise we could not restore inline xattr entry after power failure. Finally, we need to reset the 'ret' value if ext4_write_inline_data_end() return success in ext4_write_end() and ext4_journalled_write_end(), otherwise we could not get the error return value of ext4_journal_stop(). Cc: stable(a)vger.kernel.org Signed-off-by: Zhang Yi <yi.zhang(a)huawei.com> Reviewed-by: Jan Kara <jack(a)suse.cz> Signed-off-by: Theodore Ts'o <tytso(a)mit.edu> Link: https://lore.kernel.org/r/20210716122024.1105856-3-yi.zhang@huawei.com Reviewed-by: Cheng Nie <niecheng1(a)uniontech.com> Signed-off-by: Dandan Zhang <zhangdandan(a)uniontech.com> Signed-off-by: WangYuli <wangyuli(a)uniontech.com> --- fs/ext4/inline.c | 15 +++++---------- fs/ext4/inode.c | 7 +++++-- 2 files changed, 10 insertions(+), 12 deletions(-) diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c index 71bb3cfc5933..de04bd5fb551 100644 --- a/fs/ext4/inline.c +++ b/fs/ext4/inline.c @@ -745,18 +745,13 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len, void *kaddr; struct ext4_iloc iloc; - if (unlikely(copied < len)) { - if (!PageUptodate(page)) { - copied = 0; - goto out; - } - } + if (unlikely(copied < len) && !PageUptodate(page)) + return 0; ret = ext4_get_inode_loc(inode, &iloc); if (ret) { ext4_std_error(inode->i_sb, ret); - copied = 0; - goto out; + return ret; } ext4_write_lock_xattr(inode, &no_expand); @@ -769,7 +764,7 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len, (void) ext4_find_inline_data_nolock(inode); kaddr = kmap_atomic(page); - ext4_write_inline_data(inode, &iloc, kaddr, pos, len); + ext4_write_inline_data(inode, &iloc, kaddr, pos, copied); kunmap_atomic(kaddr); SetPageUptodate(page); /* clear page dirty so that writepages wouldn't work for us. */ @@ -778,7 +773,7 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len, ext4_write_unlock_xattr(inode, &no_expand); brelse(iloc.bh); mark_inode_dirty(inode); -out: + return copied; } diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index d8a8e4ee5ff8..44a715e6aae1 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1428,6 +1428,7 @@ static int ext4_write_end(struct file *file, goto errout; } copied = ret; + ret = 0; } else copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); @@ -1450,13 +1451,14 @@ static int ext4_write_end(struct file *file, if (i_size_changed || inline_data) ext4_mark_inode_dirty(handle, inode); +errout: if (pos + len > inode->i_size && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside * inode->i_size. So truncate them */ ext4_orphan_add(handle, inode); -errout: + ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; @@ -1538,6 +1540,7 @@ static int ext4_journalled_write_end(struct file *file, goto errout; } copied = ret; + ret = 0; } else if (unlikely(copied < len) && !PageUptodate(page)) { copied = 0; ext4_journalled_zero_new_buffers(handle, page, from, to); @@ -1566,6 +1569,7 @@ static int ext4_journalled_write_end(struct file *file, ret = ret2; } +errout: if (pos + len > inode->i_size && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside @@ -1573,7 +1577,6 @@ static int ext4_journalled_write_end(struct file *file, */ ext4_orphan_add(handle, inode); -errout: ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; -- 2.43.4

11 months, 1 week

1
0
0 0

[PATCH 1/4] ext4: check and update i_disksize properly

by WangYuli

From: Zhang Yi <yi.zhang(a)huawei.com> After commit 3da40c7b0898 ("ext4: only call ext4_truncate when size <= isize"), i_disksize could always be updated to i_size in ext4_setattr(), and we could sure that i_disksize <= i_size since holding inode lock and if i_disksize < i_size there are delalloc writes pending in the range upto i_size. If the end of the current write is <= i_size, there's no need to touch i_disksize since writeback will push i_disksize upto i_size eventually. So we can switch to check i_size instead of i_disksize in ext4_da_write_end() when write to the end of the file. we also could remove ext4_mark_inode_dirty() together because we defer inode dirtying to generic_write_end() or ext4_da_write_inline_data_end(). Signed-off-by: Zhang Yi <yi.zhang(a)huawei.com> Reviewed-by: Jan Kara <jack(a)suse.cz> Signed-off-by: Theodore Ts'o <tytso(a)mit.edu> Link: https://lore.kernel.org/r/20210716122024.1105856-2-yi.zhang@huawei.com Reviewed-by: Cheng Nie <niecheng1(a)uniontech.com> Signed-off-by: Dandan Zhang <zhangdandan(a)uniontech.com> Signed-off-by: WangYuli <wangyuli(a)uniontech.com> --- fs/ext4/inode.c | 34 ++++++++++++++++++---------------- 1 file changed, 18 insertions(+), 16 deletions(-) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 646285fbc9fc..d8a8e4ee5ff8 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3207,35 +3207,37 @@ static int ext4_da_write_end(struct file *file, end = start + copied - 1; /* - * generic_write_end() will run mark_inode_dirty() if i_size - * changes. So let's piggyback the i_disksize mark_inode_dirty - * into that. + * Since we are holding inode lock, we are sure i_disksize <= + * i_size. We also know that if i_disksize < i_size, there are + * delalloc writes pending in the range upto i_size. If the end of + * the current write is <= i_size, there's no need to touch + * i_disksize since writeback will push i_disksize upto i_size + * eventually. If the end of the current write is > i_size and + * inside an allocated block (ext4_da_should_update_i_disksize() + * check), we need to update i_disksize here as neither + * ext4_writepage() nor certain ext4_writepages() paths not + * allocating blocks update i_disksize. + * + * Note that we defer inode dirtying to generic_write_end() / + * ext4_da_write_inline_data_end(). */ new_i_size = pos + copied; - if (copied && new_i_size > EXT4_I(inode)->i_disksize) { + if (copied && new_i_size > inode->i_size) { if (ext4_has_inline_data(inode) || - ext4_da_should_update_i_disksize(page, end)) { + ext4_da_should_update_i_disksize(page, end)) ext4_update_i_disksize(inode, new_i_size); - /* We need to mark inode dirty even if - * new_i_size is less that inode->i_size - * bu greater than i_disksize.(hint delalloc) - */ - ext4_mark_inode_dirty(handle, inode); - } } if (write_mode != CONVERT_INLINE_DATA && ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA) && ext4_has_inline_data(inode)) - ret2 = ext4_da_write_inline_data_end(inode, pos, len, copied, + ret = ext4_da_write_inline_data_end(inode, pos, len, copied, page); else - ret2 = generic_write_end(file, mapping, pos, len, copied, + ret = generic_write_end(file, mapping, pos, len, copied, page, fsdata); - copied = ret2; - if (ret2 < 0) - ret = ret2; + copied = ret; ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; -- 2.43.4

11 months, 1 week

2
1
0 0

[PATCH 4/4] ext4: drop unnecessary journal handle in delalloc write

by WangYuli

From: Zhang Yi <yi.zhang(a)huawei.com> After we factor out the inline data write procedure from ext4_da_write_end(), we don't need to start journal handle for the cases of both buffer overwrite and append-write. If we need to update i_disksize, mark_inode_dirty() do start handle and update inode buffer. So we could just remove all the journal handle codes in the delalloc write procedure. After this patch, we could get a lot of performance improvement. Below is the Unixbench comparison data test on my machine with 'Intel Xeon Gold 5120' CPU and nvme SSD backend. Test cmd: ./Run -c 56 -i 3 fstime fsbuffer fsdisk Before this patch: System Benchmarks Partial Index BASELINE RESULT INDEX File Copy 1024 bufsize 2000 maxblocks 3960.0 422965.0 1068.1 File Copy 256 bufsize 500 maxblocks 1655.0 105077.0 634.9 File Copy 4096 bufsize 8000 maxblocks 5800.0 1429092.0 2464.0 ====== System Benchmarks Index Score (Partial Only) 1186.6 After this patch: System Benchmarks Partial Index BASELINE RESULT INDEX File Copy 1024 bufsize 2000 maxblocks 3960.0 732716.0 1850.3 File Copy 256 bufsize 500 maxblocks 1655.0 184940.0 1117.5 File Copy 4096 bufsize 8000 maxblocks 5800.0 2427152.0 4184.7 ====== System Benchmarks Index Score (Partial Only) 2053.0 Signed-off-by: Zhang Yi <yi.zhang(a)huawei.com> Reviewed-by: Jan Kara <jack(a)suse.cz> Signed-off-by: Theodore Ts'o <tytso(a)mit.edu> Link: https://lore.kernel.org/r/20210716122024.1105856-5-yi.zhang@huawei.com Reviewed-by: Cheng Nie <niecheng1(a)uniontech.com> Signed-off-by: Dandan Zhang <zhangdandan(a)uniontech.com> Signed-off-by: WangYuli <wangyuli(a)uniontech.com> --- fs/ext4/inode.c | 60 +++++-------------------------------------------- 1 file changed, 5 insertions(+), 55 deletions(-) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 1e6753084a00..597c2f49c889 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3032,19 +3032,6 @@ static int ext4_nonda_switch(struct super_block *sb) return 0; } -/* We always reserve for an inode update; the superblock could be there too */ -static int ext4_da_write_credits(struct inode *inode, loff_t pos, unsigned len) -{ - if (likely(ext4_has_feature_large_file(inode->i_sb))) - return 1; - - if (pos + len <= 0x7fffffffULL) - return 1; - - /* We might need to update the superblock to set LARGE_FILE */ - return 2; -} - static int ext4_da_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) @@ -3053,7 +3040,6 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, struct page *page; pgoff_t index; struct inode *inode = mapping->host; - handle_t *handle; if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) return -EIO; @@ -3079,41 +3065,11 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, return 0; } - /* - * grab_cache_page_write_begin() can take a long time if the - * system is thrashing due to memory pressure, or if the page - * is being written back. So grab it first before we start - * the transaction handle. This also allows us to allocate - * the page (if needed) without using GFP_NOFS. - */ -retry_grab: +retry: page = grab_cache_page_write_begin(mapping, index, flags); if (!page) return -ENOMEM; - unlock_page(page); - - /* - * With delayed allocation, we don't log the i_disksize update - * if there is delayed block allocation. But we still need - * to journalling the i_disksize update if writes to the end - * of file which has an already mapped buffer. - */ -retry_journal: - handle = ext4_journal_start(inode, EXT4_HT_WRITE_PAGE, - ext4_da_write_credits(inode, pos, len)); - if (IS_ERR(handle)) { - put_page(page); - return PTR_ERR(handle); - } - lock_page(page); - if (page->mapping != mapping) { - /* The page got truncated from under us */ - unlock_page(page); - put_page(page); - ext4_journal_stop(handle); - goto retry_grab; - } /* In case writeback began while the page was unlocked */ wait_for_stable_page(page); @@ -3125,20 +3081,18 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, #endif if (ret < 0) { unlock_page(page); - ext4_journal_stop(handle); + put_page(page); /* * block_write_begin may have instantiated a few blocks * outside i_size. Trim these off again. Don't need - * i_size_read because we hold i_mutex. + * i_size_read because we hold inode lock. */ if (pos + len > inode->i_size) ext4_truncate_failed_write(inode); if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) - goto retry_journal; - - put_page(page); + goto retry; return ret; } @@ -3175,8 +3129,6 @@ static int ext4_da_write_end(struct file *file, struct page *page, void *fsdata) { struct inode *inode = mapping->host; - int ret; - handle_t *handle = ext4_journal_current_handle(); loff_t new_i_size; unsigned long start, end; int write_mode = (int)(unsigned long)fsdata; @@ -3215,9 +3167,7 @@ static int ext4_da_write_end(struct file *file, ext4_da_should_update_i_disksize(page, end)) ext4_update_i_disksize(inode, new_i_size); - copied = generic_write_end(file, mapping, pos, len, copied, page, fsdata); - ret = ext4_journal_stop(handle); - return ret ? ret : copied; + return generic_write_end(file, mapping, pos, len, copied, page, fsdata); } static void ext4_da_invalidatepage(struct page *page, unsigned int offset, -- 2.43.4

11 months, 1 week

1
0
0 0

[PATCH 2/4] ext4: correct the error path of ext4_write_inline_data_end()

by WangYuli

From: Zhang Yi <yi.zhang(a)huawei.com> Current error path of ext4_write_inline_data_end() is not correct. Firstly, it should pass out the error value if ext4_get_inode_loc() return fail, or else it could trigger infinite loop if we inject error here. And then it's better to add inode to orphan list if it return fail in ext4_journal_stop(), otherwise we could not restore inline xattr entry after power failure. Finally, we need to reset the 'ret' value if ext4_write_inline_data_end() return success in ext4_write_end() and ext4_journalled_write_end(), otherwise we could not get the error return value of ext4_journal_stop(). Signed-off-by: Zhang Yi <yi.zhang(a)huawei.com> Reviewed-by: Jan Kara <jack(a)suse.cz> Signed-off-by: Theodore Ts'o <tytso(a)mit.edu> Link: https://lore.kernel.org/r/20210716122024.1105856-3-yi.zhang@huawei.com Reviewed-by: Cheng Nie <niecheng1(a)uniontech.com> Signed-off-by: Dandan Zhang <zhangdandan(a)uniontech.com> Signed-off-by: WangYuli <wangyuli(a)uniontech.com> --- fs/ext4/inline.c | 15 +++++---------- fs/ext4/inode.c | 7 +++++-- 2 files changed, 10 insertions(+), 12 deletions(-) diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c index 71bb3cfc5933..de04bd5fb551 100644 --- a/fs/ext4/inline.c +++ b/fs/ext4/inline.c @@ -745,18 +745,13 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len, void *kaddr; struct ext4_iloc iloc; - if (unlikely(copied < len)) { - if (!PageUptodate(page)) { - copied = 0; - goto out; - } - } + if (unlikely(copied < len) && !PageUptodate(page)) + return 0; ret = ext4_get_inode_loc(inode, &iloc); if (ret) { ext4_std_error(inode->i_sb, ret); - copied = 0; - goto out; + return ret; } ext4_write_lock_xattr(inode, &no_expand); @@ -769,7 +764,7 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len, (void) ext4_find_inline_data_nolock(inode); kaddr = kmap_atomic(page); - ext4_write_inline_data(inode, &iloc, kaddr, pos, len); + ext4_write_inline_data(inode, &iloc, kaddr, pos, copied); kunmap_atomic(kaddr); SetPageUptodate(page); /* clear page dirty so that writepages wouldn't work for us. */ @@ -778,7 +773,7 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len, ext4_write_unlock_xattr(inode, &no_expand); brelse(iloc.bh); mark_inode_dirty(inode); -out: + return copied; } diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index d8a8e4ee5ff8..44a715e6aae1 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1428,6 +1428,7 @@ static int ext4_write_end(struct file *file, goto errout; } copied = ret; + ret = 0; } else copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); @@ -1450,13 +1451,14 @@ static int ext4_write_end(struct file *file, if (i_size_changed || inline_data) ext4_mark_inode_dirty(handle, inode); +errout: if (pos + len > inode->i_size && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside * inode->i_size. So truncate them */ ext4_orphan_add(handle, inode); -errout: + ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; @@ -1538,6 +1540,7 @@ static int ext4_journalled_write_end(struct file *file, goto errout; } copied = ret; + ret = 0; } else if (unlikely(copied < len) && !PageUptodate(page)) { copied = 0; ext4_journalled_zero_new_buffers(handle, page, from, to); @@ -1566,6 +1569,7 @@ static int ext4_journalled_write_end(struct file *file, ret = ret2; } +errout: if (pos + len > inode->i_size && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside @@ -1573,7 +1577,6 @@ static int ext4_journalled_write_end(struct file *file, */ ext4_orphan_add(handle, inode); -errout: ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; -- 2.43.4

11 months, 1 week

1
0
0 0

[PATCH 4.19 0/4] ext4: improve delalloc buffer write performance

by WangYuli

A patchset from linux-5.15 should be backported to 4.19 that can significantly improve ext4 fs read and write performance. Unixbench test results for linux-4.19.318 on Phytium D2000 CPU are shown below. Test cmd: (Phytium D2000 only has 8 cores) ./Run fs -c 8 Before this patch set: File Copy 1024 bufsize 2000 maxblocks 1124181 File Copy 256 bufsize 500 maxblocks 281885 File Copy 4096 bufsize 8000 maxblocks 3383785 File Read 1024 bufsize 2000 maxblocks 8702173 File Read 256 bufsize 500 maxblocks 3869384 File Read 4096 bufsize 8000 maxblocks 13043151 File Write 1024 bufsize 2000 maxblocks 1107185 File Write 256 bufsize 500 maxblocks 270493 File Write 4096 bufsize 8000 maxblocks 4018084 After this patch set: File Copy 1024 bufsize 2000 maxblocks 2026206 File Copy 256 bufsize 500 maxblocks 829534 File Copy 4096 bufsize 8000 maxblocks 4066659 File Read 1024 bufsize 2000 maxblocks 8877219 File Read 256 bufsize 500 maxblocks 3997445 File Read 4096 bufsize 8000 maxblocks 13179885 File Write 1024 bufsize 2000 maxblocks 4256929 File Write 256 bufsize 500 maxblocks 1305320 File Write 4096 bufsize 8000 maxblocks 10721052 We can observe a quantum leap in the test results as a consequence of applying this patchset Link: https://lore.kernel.org/all/20210716122024.1105856-1-yi.zhang@huawei.com/ Original description: This patchset address to improve buffer write performance with delalloc. The first patch reduce the unnecessary update i_disksize, the second two patch refactor the inline data write procedure and also do some small fix, the last patch do improve by remove all unnecessary journal handle in the delalloc write procedure. After this patch set, we could get a lot of performance improvement. Below is the Unixbench comparison data test on my machine with 'Intel Xeon Gold 5120' CPU and nvme SSD backend. Test cmd: ./Run -c 56 -i 3 fstime fsbuffer fsdisk Before this patch set: System Benchmarks Partial Index BASELINE RESULT INDEX File Copy 1024 bufsize 2000 maxblocks 3960.0 422965.0 1068.1 File Copy 256 bufsize 500 maxblocks 1655.0 105077.0 634.9 File Copy 4096 bufsize 8000 maxblocks 5800.0 1429092.0 2464.0 ======== System Benchmarks Index Score (Partial Only) 1186.6 After this patch set: System Benchmarks Partial Index BASELINE RESULT INDEX File Copy 1024 bufsize 2000 maxblocks 3960.0 732716.0 1850.3 File Copy 256 bufsize 500 maxblocks 1655.0 184940.0 1117.5 File Copy 4096 bufsize 8000 maxblocks 5800.0 2427152.0 4184.7 ======== System Benchmarks Index Score (Partial Only) 2053.0 Zhang Yi (4): ext4: check and update i_disksize properly ext4: correct the error path of ext4_write_inline_data_end() ext4: factor out write end code of inline file ext4: drop unnecessary journal handle in delalloc write fs/ext4/ext4.h | 3 - fs/ext4/inline.c | 120 ++++++++++++++++++------------------- fs/ext4/inode.c | 150 ++++++++++++----------------------------------- 3 files changed, 99 insertions(+), 174 deletions(-) -- 2.31.1

11 months, 1 week

1
0
0 0

Re:

by Jackson Jikarjo

Hello, I sent you a message a few hours ago but no reply yet, or you didn't receive it? Kindly read my letter and reply back. I want to make an inquiry Thanks. Dr.Allen Cheng Human Resource Manager | Product Research Assistant FC Industrial Laboratories Ltd

11 months, 1 week

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror July 2024