Re: [f2fs-dev] Weird EROFS data corruption
Hi Juhyung, On 2023/12/5 22:43, Juhyung Park wrote: On Tue, Dec 5, 2023 at 11:34 PM Gao Xiang wrote: ... I'm still analyzing this behavior as well as the root cause and I will also try to get a recent cloud server with FSRM myself to find more clues. Down the rabbit hole we go... Let me know if you have trouble getting an instance with FSRM. I'll see what I can do. I've sent out a fix to address this, please help check: https://lore.kernel.org/r/20231206030758.3760521-1-hsiang...@linux.alibaba.com Thanks, Gao Xiang ___ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Re: [f2fs-dev] [PATCH 4/4 v2] f2fs: let's finish or reset zones all the time
In order to limit # of open zones, let's finish or reset zones given # of valid blocks per section and its zone condition. Reviewed-by: Daeho Jeong Signed-off-by: Jaegeuk Kim --- - remove unnecessary wp_block fs/f2fs/segment.c | 75 +++ 1 file changed, 17 insertions(+), 58 deletions(-) diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 9081c9af977a..007ebb107236 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -4870,82 +4870,43 @@ static int check_zone_write_pointer(struct f2fs_sb_info *sbi, struct f2fs_dev_info *fdev, struct blk_zone *zone) { - unsigned int wp_segno, wp_blkoff, zone_secno, zone_segno, segno; - block_t zone_block, wp_block, last_valid_block; + unsigned int zone_segno; + block_t zone_block, valid_block_cnt; unsigned int log_sectors_per_block = sbi->log_blocksize - SECTOR_SHIFT; - int i, s, b, ret; - struct seg_entry *se; + int ret; if (zone->type != BLK_ZONE_TYPE_SEQWRITE_REQ) return 0; - wp_block = fdev->start_blk + (zone->wp >> log_sectors_per_block); - wp_segno = GET_SEGNO(sbi, wp_block); - wp_blkoff = wp_block - START_BLOCK(sbi, wp_segno); zone_block = fdev->start_blk + (zone->start >> log_sectors_per_block); zone_segno = GET_SEGNO(sbi, zone_block); - zone_secno = GET_SEC_FROM_SEG(sbi, zone_segno); - - if (zone_segno >= MAIN_SEGS(sbi)) - return 0; /* * Skip check of zones cursegs point to, since * fix_curseg_write_pointer() checks them. */ - for (i = 0; i < NO_CHECK_TYPE; i++) - if (zone_secno == GET_SEC_FROM_SEG(sbi, - CURSEG_I(sbi, i)->segno)) - return 0; + if (zone_segno >= MAIN_SEGS(sbi) || + IS_CURSEC(sbi, GET_SEC_FROM_SEG(sbi, zone_segno))) + return 0; /* -* Get last valid block of the zone. +* Get # of valid block of the zone. */ - last_valid_block = zone_block - 1; - for (s = sbi->segs_per_sec - 1; s >= 0; s--) { - segno = zone_segno + s; - se = get_seg_entry(sbi, segno); - for (b = sbi->blocks_per_seg - 1; b >= 0; b--) - if (f2fs_test_bit(b, se->cur_valid_map)) { - last_valid_block = START_BLOCK(sbi, segno) + b; - break; - } - if (last_valid_block >= zone_block) - break; - } + valid_block_cnt = get_valid_blocks(sbi, zone_segno, true); - /* -* When safely unmounted in the previous mount, we can trust write -* pointers. Otherwise, finish zones. -*/ - if (is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) { - /* -* The write pointer matches with the valid blocks or -* already points to the end of the zone. -*/ - if ((last_valid_block + 1 == wp_block) || - (zone->wp == zone->start + zone->len)) - return 0; - } + if ((!valid_block_cnt && zone->cond == BLK_ZONE_COND_EMPTY) || + (valid_block_cnt && zone->cond == BLK_ZONE_COND_FULL)) + return 0; - if (last_valid_block + 1 == zone_block) { - if (is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) { - /* -* If there is no valid block in the zone and if write -* pointer is not at zone start, reset the write -* pointer. -*/ - f2fs_notice(sbi, - "Zone without valid block has non-zero write " - "pointer. Reset the write pointer: wp[0x%x,0x%x]", - wp_segno, wp_blkoff); - } + if (!valid_block_cnt) { + f2fs_notice(sbi, "Zone without valid block has non-zero write " + "pointer. Reset the write pointer: cond[0x%x]", + zone->cond); ret = __f2fs_issue_discard_zone(sbi, fdev->bdev, zone_block, zone->len >> log_sectors_per_block); if (ret) f2fs_err(sbi, "Discard zone failed: %s (errno=%d)", fdev->path, ret); - return ret; } @@ -4957,10 +4918,8 @@ static int check_zone_write_pointer(struct f2fs_sb_info *sbi, * selected for write operation until it get discarded. */ f2fs_notice(sbi, "Valid blocks are not aligned with write " - "pointer: valid block[0x%x,0x%x] wp[0x%x,0x%x]", -
Re: [f2fs-dev] Weird EROFS data corruption
On Tue, Dec 5, 2023 at 11:34 PM Gao Xiang wrote: > > > > On 2023/12/5 22:23, Juhyung Park wrote: > > Hi Gao, > > > > On Tue, Dec 5, 2023 at 4:32 PM Gao Xiang > > wrote: > >> > >> Hi Juhyung, > >> > >> On 2023/12/4 11:41, Juhyung Park wrote: > >> > >> ... > >>> > > - Could you share the full message about the output of `lscpu`? > >>> > >>> Sure: > >>> > >>> Architecture:x86_64 > >>> CPU op-mode(s):32-bit, 64-bit > >>> Address sizes: 39 bits physical, 48 bits virtual > >>> Byte Order:Little Endian > >>> CPU(s): 8 > >>> On-line CPU(s) list: 0-7 > >>> Vendor ID: GenuineIntel > >>> BIOS Vendor ID:Intel(R) Corporation > >>> Model name:11th Gen Intel(R) Core(TM) i7-1185G7 @ 3.00GHz > >>> BIOS Model name: 11th Gen Intel(R) Core(TM) i7-1185G7 @ 3.00GHz > >>> None CPU > >>> @ 3.0GHz > >>> BIOS CPU family: 198 > >>> CPU family: 6 > >>> Model: 140 > >>> Thread(s) per core: 2 > >>> Core(s) per socket: 4 > >>> Socket(s): 1 > >>> Stepping:1 > >>> CPU(s) scaling MHz: 60% > >>> CPU max MHz: 4800. > >>> CPU min MHz: 400. > >>> BogoMIPS:5990.40 > >>> Flags: fpu vme de pse tsc msr pae mce cx8 apic sep > >>> mtrr pge mc > >>>a cmov pat pse36 clflush dts acpi mmx fxsr sse > >>> sse2 ss > >>>ht tm pbe syscall nx pdpe1gb rdtscp lm > >>> constant_tsc art > >>> arch_perfmon pebs bts rep_good nopl xtopology > >>> nonstop_ > >>>tsc cpuid aperfmperf tsc_known_freq pni > >>> pclmulqdq dtes6 > >>>4 monitor ds_cpl vmx smx est tm2 ssse3 sdbg > >>> fma cx16 xt > >>>pr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt > >>> tsc_dead > >>>line_timer aes xsave avx f16c rdrand lahf_lm > >>> abm 3dnowp > >>>refetch cpuid_fault epb cat_l2 cdp_l2 ssbd > >>> ibrs ibpb st > >>>ibp ibrs_enhanced tpr_shadow flexpriority ept > >>> vpid ept_ > >>>ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 > >>> erms invpcid > >>> rdt_a avx512f avx512dq rdseed adx smap > >>> avx512ifma clfl > >>>ushopt clwb intel_pt avx512cd sha_ni avx512bw > >>> avx512vl > >>>xsaveopt xsavec xgetbv1 xsaves > >>> split_lock_detect dtherm > >>> ida arat pln pts hwp hwp_notify > >>> hwp_act_window hwp_epp > >>> hwp_pkg_req vnmi avx512vbmi umip pku ospke > >>> avx512_vbmi > >>>2 gfni vaes vpclmulqdq avx512_vnni > >>> avx512_bitalg tme av > >>>x512_vpopcntdq rdpid movdiri movdir64b fsrm > >>> avx512_vp2i > >> > >> Sigh, I've been thinking. Here FSRM is the most significant difference > >> between > >> our environments, could you only try the following diff to see if there's > >> any > >> difference anymore? (without the previous disable patch.) > >> > >> diff --git a/arch/x86/lib/memmove_64.S b/arch/x86/lib/memmove_64.S > >> index 1b60ae81ecd8..1b52a913233c 100644 > >> --- a/arch/x86/lib/memmove_64.S > >> +++ b/arch/x86/lib/memmove_64.S > >> @@ -41,9 +41,7 @@ SYM_FUNC_START(__memmove) > >>#define CHECK_LEN cmp $0x20, %rdx; jb 1f > >>#define MEMMOVE_BYTES movq %rdx, %rcx; rep movsb; RET > >>.Lmemmove_begin_forward: > >> - ALTERNATIVE_2 __stringify(CHECK_LEN), \ > >> - __stringify(CHECK_LEN; MEMMOVE_BYTES), > >> X86_FEATURE_ERMS, \ > >> - __stringify(MEMMOVE_BYTES), X86_FEATURE_FSRM > >> + CHECK_LEN > >> > >> /* > >> * movsq instruction have many startup latency > > > > Yup, that also seems to fix it. > > Are we looking at a potential memmove issue? > > I'm still analyzing this behavior as well as the root cause and > I will also try to get a recent cloud server with FSRM myself > to find more clues. Down the rabbit hole we go... Let me know if you have trouble getting an instance with FSRM. I'll see what I can do. > > Thanks, > Gao Xiang ___ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Re: [f2fs-dev] Weird EROFS data corruption
On 2023/12/5 22:23, Juhyung Park wrote: Hi Gao, On Tue, Dec 5, 2023 at 4:32 PM Gao Xiang wrote: Hi Juhyung, On 2023/12/4 11:41, Juhyung Park wrote: ... - Could you share the full message about the output of `lscpu`? Sure: Architecture:x86_64 CPU op-mode(s):32-bit, 64-bit Address sizes: 39 bits physical, 48 bits virtual Byte Order:Little Endian CPU(s): 8 On-line CPU(s) list: 0-7 Vendor ID: GenuineIntel BIOS Vendor ID:Intel(R) Corporation Model name:11th Gen Intel(R) Core(TM) i7-1185G7 @ 3.00GHz BIOS Model name: 11th Gen Intel(R) Core(TM) i7-1185G7 @ 3.00GHz None CPU @ 3.0GHz BIOS CPU family: 198 CPU family: 6 Model: 140 Thread(s) per core: 2 Core(s) per socket: 4 Socket(s): 1 Stepping:1 CPU(s) scaling MHz: 60% CPU max MHz: 4800. CPU min MHz: 400. BogoMIPS:5990.40 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mc a cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_ tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes6 4 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xt pr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_dead line_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowp refetch cpuid_fault epb cat_l2 cdp_l2 ssbd ibrs ibpb st ibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdt_a avx512f avx512dq rdseed adx smap avx512ifma clfl ushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves split_lock_detect dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp hwp_pkg_req vnmi avx512vbmi umip pku ospke avx512_vbmi 2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme av x512_vpopcntdq rdpid movdiri movdir64b fsrm avx512_vp2i Sigh, I've been thinking. Here FSRM is the most significant difference between our environments, could you only try the following diff to see if there's any difference anymore? (without the previous disable patch.) diff --git a/arch/x86/lib/memmove_64.S b/arch/x86/lib/memmove_64.S index 1b60ae81ecd8..1b52a913233c 100644 --- a/arch/x86/lib/memmove_64.S +++ b/arch/x86/lib/memmove_64.S @@ -41,9 +41,7 @@ SYM_FUNC_START(__memmove) #define CHECK_LEN cmp $0x20, %rdx; jb 1f #define MEMMOVE_BYTES movq %rdx, %rcx; rep movsb; RET .Lmemmove_begin_forward: - ALTERNATIVE_2 __stringify(CHECK_LEN), \ - __stringify(CHECK_LEN; MEMMOVE_BYTES), X86_FEATURE_ERMS, \ - __stringify(MEMMOVE_BYTES), X86_FEATURE_FSRM + CHECK_LEN /* * movsq instruction have many startup latency Yup, that also seems to fix it. Are we looking at a potential memmove issue? I'm still analyzing this behavior as well as the root cause and I will also try to get a recent cloud server with FSRM myself to find more clues. Thanks, Gao Xiang ___ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Re: [f2fs-dev] Weird EROFS data corruption
Hi Gao, On Tue, Dec 5, 2023 at 4:32 PM Gao Xiang wrote: > > Hi Juhyung, > > On 2023/12/4 11:41, Juhyung Park wrote: > > ... > > > >> > >> - Could you share the full message about the output of `lscpu`? > > > > Sure: > > > > Architecture:x86_64 > >CPU op-mode(s):32-bit, 64-bit > >Address sizes: 39 bits physical, 48 bits virtual > >Byte Order:Little Endian > > CPU(s): 8 > >On-line CPU(s) list: 0-7 > > Vendor ID: GenuineIntel > >BIOS Vendor ID:Intel(R) Corporation > >Model name:11th Gen Intel(R) Core(TM) i7-1185G7 @ 3.00GHz > > BIOS Model name: 11th Gen Intel(R) Core(TM) i7-1185G7 @ 3.00GHz > > None CPU > >@ 3.0GHz > > BIOS CPU family: 198 > > CPU family: 6 > > Model: 140 > > Thread(s) per core: 2 > > Core(s) per socket: 4 > > Socket(s): 1 > > Stepping:1 > > CPU(s) scaling MHz: 60% > > CPU max MHz: 4800. > > CPU min MHz: 400. > > BogoMIPS:5990.40 > > Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr > > pge mc > > a cmov pat pse36 clflush dts acpi mmx fxsr sse > > sse2 ss > > ht tm pbe syscall nx pdpe1gb rdtscp lm > > constant_tsc art > >arch_perfmon pebs bts rep_good nopl xtopology > > nonstop_ > > tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq > > dtes6 > > 4 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma > > cx16 xt > > pr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt > > tsc_dead > > line_timer aes xsave avx f16c rdrand lahf_lm abm > > 3dnowp > > refetch cpuid_fault epb cat_l2 cdp_l2 ssbd ibrs > > ibpb st > > ibp ibrs_enhanced tpr_shadow flexpriority ept > > vpid ept_ > > ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms > > invpcid > >rdt_a avx512f avx512dq rdseed adx smap > > avx512ifma clfl > > ushopt clwb intel_pt avx512cd sha_ni avx512bw > > avx512vl > > xsaveopt xsavec xgetbv1 xsaves split_lock_detect > > dtherm > >ida arat pln pts hwp hwp_notify hwp_act_window > > hwp_epp > >hwp_pkg_req vnmi avx512vbmi umip pku ospke > > avx512_vbmi > > 2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg > > tme av > > x512_vpopcntdq rdpid movdiri movdir64b fsrm > > avx512_vp2i > > Sigh, I've been thinking. Here FSRM is the most significant difference > between > our environments, could you only try the following diff to see if there's any > difference anymore? (without the previous disable patch.) > > diff --git a/arch/x86/lib/memmove_64.S b/arch/x86/lib/memmove_64.S > index 1b60ae81ecd8..1b52a913233c 100644 > --- a/arch/x86/lib/memmove_64.S > +++ b/arch/x86/lib/memmove_64.S > @@ -41,9 +41,7 @@ SYM_FUNC_START(__memmove) > #define CHECK_LEN cmp $0x20, %rdx; jb 1f > #define MEMMOVE_BYTES movq %rdx, %rcx; rep movsb; RET > .Lmemmove_begin_forward: > - ALTERNATIVE_2 __stringify(CHECK_LEN), \ > - __stringify(CHECK_LEN; MEMMOVE_BYTES), > X86_FEATURE_ERMS, \ > - __stringify(MEMMOVE_BYTES), X86_FEATURE_FSRM > + CHECK_LEN > > /* > * movsq instruction have many startup latency Yup, that also seems to fix it. Are we looking at a potential memmove issue? > > Thanks, > Gao Xiang ___ Linux-f2fs-devel mailing list Linux-f2fs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel