- Linux-stable-mirror - lists.linaro.org

[PATCH][v3] mtd: rawnand: fsl_ifc: Fix nand waitfunc return value

by Jagdish Gediya

As per the IFC hardware manual, Most significant 2 bytes in nand_fsr register are the outcome of NAND READ STATUS command. So status value need to be shifted and aligned as per the nand framework requirement. Fixes: 82771882d960 ("NAND Machine support for Integrated Flash Controller") Cc: stable(a)vger.kernel.org # v3.18+ Signed-off-by: Jagdish Gediya <jagdish.gediya(a)nxp.com> Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha(a)nxp.com> --- Changes for v2: Incorporated comments from Boris Brezillon - Added fixes tag Changes for v3: Incorporated comments from Boris Brezillon drivers/mtd/nand/fsl_ifc_nand.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/mtd/nand/fsl_ifc_nand.c b/drivers/mtd/nand/fsl_ifc_nand.c index 9390cbd..16104ca 100644 --- a/drivers/mtd/nand/fsl_ifc_nand.c +++ b/drivers/mtd/nand/fsl_ifc_nand.c @@ -629,6 +629,7 @@ static int fsl_ifc_wait(struct mtd_info *mtd, struct nand_chip *chip) struct fsl_ifc_ctrl *ctrl = priv->ctrl; struct fsl_ifc_runtime __iomem *ifc = ctrl->rregs; u32 nand_fsr; + int status; /* Use READ_STATUS command, but wait for the device to be ready */ ifc_out32((IFC_FIR_OP_CW0 << IFC_NAND_FIR0_OP0_SHIFT) | @@ -643,12 +644,12 @@ static int fsl_ifc_wait(struct mtd_info *mtd, struct nand_chip *chip) fsl_ifc_run_command(mtd); nand_fsr = ifc_in32(&ifc->ifc_nand.nand_fsr); - + status = nand_fsr >> 24; /* * The chip always seems to report that it is * write-protected, even when it is not. */ - return nand_fsr | NAND_STATUS_WP; + return status | NAND_STATUS_WP; } /* -- 1.9.1

7 years, 9 months

2
1
0 0

[PATCH] mtd: rawnand: fsl_ifc: Read ECCSTAT0 and ECCSTAT1 registers for IFC 2.0

by Jagdish Gediya

Due to missing information in Hardware manual, current implementation doesn't read ECCSTAT0 and ECCSTAT1 registers for IFC 2.0. Add support to read ECCSTAT0 and ECCSTAT1 registers during ecccheck for IFC 2.0. Fixes: 656441478ed5 ("mtd: nand: ifc: Fix location of eccstat registers for IFC V1.0") Cc: stable(a)vger.kernel.org # v3.18+ Signed-off-by: Jagdish Gediya <jagdish.gediya(a)nxp.com> Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha(a)nxp.com> --- drivers/mtd/nand/fsl_ifc_nand.c | 6 +----- include/linux/fsl_ifc.h | 6 +----- 2 files changed, 2 insertions(+), 10 deletions(-) diff --git a/drivers/mtd/nand/fsl_ifc_nand.c b/drivers/mtd/nand/fsl_ifc_nand.c index ed64759..1513298 100644 --- a/drivers/mtd/nand/fsl_ifc_nand.c +++ b/drivers/mtd/nand/fsl_ifc_nand.c @@ -227,11 +227,7 @@ static void fsl_ifc_run_command(struct mtd_info *mtd) int sector_end = sector_start + chip->ecc.steps - 1; __be32 *eccstat_regs; - if (ctrl->version >= FSL_IFC_VERSION_2_0_0) - eccstat_regs = ifc->ifc_nand.v2_nand_eccstat; - else - eccstat_regs = ifc->ifc_nand.v1_nand_eccstat; - + eccstat_regs = ifc->ifc_nand.nand_eccstat; eccstat = ifc_in32(&eccstat_regs[sector_start / 4]); for (i = sector_start; i <= sector_end; i++) { diff --git a/include/linux/fsl_ifc.h b/include/linux/fsl_ifc.h index c332f0a..3fdfede 100644 --- a/include/linux/fsl_ifc.h +++ b/include/linux/fsl_ifc.h @@ -734,11 +734,7 @@ struct fsl_ifc_nand { u32 res19[0x10]; __be32 nand_fsr; u32 res20; - /* The V1 nand_eccstat is actually 4 words that overlaps the - * V2 nand_eccstat. - */ - __be32 v1_nand_eccstat[2]; - __be32 v2_nand_eccstat[6]; + __be32 nand_eccstat[8]; u32 res21[0x1c]; __be32 nanndcr; u32 res22[0x2]; -- 1.9.1

7 years, 9 months

2
1
0 0

[PATCH][v5] mtd: rawnand: fsl_ifc: Fix eccstat array overflow for IFC ver >= 2.0.0

by Jagdish Gediya

Number of ECC status registers i.e. (ECCSTATx) has been increased in IFC version 2.0.0 due to increase in SRAM size. This is causing eccstat array to over flow. So, replace eccstat array with u32 variable to make it fail-safe and independent of number of ECC status registers or SRAM size. Fixes: bccb06c353af ("mtd: nand: ifc: update bufnum mask for ver >= 2.0.0") Cc: stable(a)vger.kernel.org # 3.18+ Signed-off-by: Prabhakar Kushwaha <prabhakar.kushwaha(a)nxp.com> Signed-off-by: Jagdish Gediya <jagdish.gediya(a)nxp.com> --- Changes for v2: Incorporated comments from Miquel Raynal and Boris Brezillon - Updated patch subject - Remove usage of eccstat array - Added Cc: stable(a)vger.kernel.org Changes for v3: Incorporated comments from Boris Brezillon - Added fixes tag Changes for v4: Incorporated comments from Boris Brezillon Changes for v5: Incorporated comments from Boris Brezillon drivers/mtd/nand/fsl_ifc_nand.c | 23 ++++++++++------------- 1 file changed, 10 insertions(+), 13 deletions(-) diff --git a/drivers/mtd/nand/fsl_ifc_nand.c b/drivers/mtd/nand/fsl_ifc_nand.c index 4872a7b..f0b4ecd 100644 --- a/drivers/mtd/nand/fsl_ifc_nand.c +++ b/drivers/mtd/nand/fsl_ifc_nand.c @@ -173,14 +173,9 @@ static void set_addr(struct mtd_info *mtd, int column, int page_addr, int oob) /* returns nonzero if entire page is blank */ static int check_read_ecc(struct mtd_info *mtd, struct fsl_ifc_ctrl *ctrl, - u32 *eccstat, unsigned int bufnum) + u32 eccstat, unsigned int bufnum) { - u32 reg = eccstat[bufnum / 4]; - int errors; - - errors = (reg >> ((3 - bufnum % 4) * 8)) & 15; - - return errors; + return (eccstat >> ((3 - bufnum % 4) * 8)) & 15; } /* @@ -193,7 +188,7 @@ static void fsl_ifc_run_command(struct mtd_info *mtd) struct fsl_ifc_ctrl *ctrl = priv->ctrl; struct fsl_ifc_nand_ctrl *nctrl = ifc_nand_ctrl; struct fsl_ifc_runtime __iomem *ifc = ctrl->rregs; - u32 eccstat[4]; + u32 eccstat; int i; /* set the chip select for NAND Transaction */ @@ -228,8 +223,8 @@ static void fsl_ifc_run_command(struct mtd_info *mtd) if (nctrl->eccread) { int errors; int bufnum = nctrl->page & priv->bufnum_mask; - int sector = bufnum * chip->ecc.steps; - int sector_end = sector + chip->ecc.steps - 1; + int sector_start = bufnum * chip->ecc.steps; + int sector_end = sector_start + chip->ecc.steps - 1; __be32 *eccstat_regs; if (ctrl->version >= FSL_IFC_VERSION_2_0_0) @@ -237,10 +232,12 @@ static void fsl_ifc_run_command(struct mtd_info *mtd) else eccstat_regs = ifc->ifc_nand.v1_nand_eccstat; - for (i = sector / 4; i <= sector_end / 4; i++) - eccstat[i] = ifc_in32(&eccstat_regs[i]); + eccstat = ifc_in32(&eccstat_regs[sector_start / 4]); + + for (i = sector_start; i <= sector_end; i++) { + if ((i != sector_start) && !(i % 4)) + eccstat = ifc_in32(&eccstat_regs[i / 4]); - for (i = sector; i <= sector_end; i++) { errors = check_read_ecc(mtd, ctrl, eccstat, i); if (errors == 15) { -- 1.9.1

7 years, 9 months

2
2
0 0

[PATCH] drm/amdgpu: Add an ATPX quirk for hybrid laptop

by Alex Deucher

_PR3 doesn't seem to work properly, use ATPX instead. Bug: https://bugs.freedesktop.org/show_bug.cgi?id=104064 Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Cc: stable(a)vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c index c53095b3b0fb..1ae5ae8c45a4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c @@ -569,6 +569,7 @@ static const struct amdgpu_px_quirk amdgpu_px_quirk_list[] = { { 0x1002, 0x6900, 0x1002, 0x0124, AMDGPU_PX_QUIRK_FORCE_ATPX }, { 0x1002, 0x6900, 0x1028, 0x0812, AMDGPU_PX_QUIRK_FORCE_ATPX }, { 0x1002, 0x6900, 0x1028, 0x0813, AMDGPU_PX_QUIRK_FORCE_ATPX }, + { 0x1002, 0x67DF, 0x1028, 0x0774, AMDGPU_PX_QUIRK_FORCE_ATPX }, { 0, 0, 0, 0, 0 }, }; -- 2.13.6

7 years, 9 months

2
1
0 0

[PATCH 05/15] mm/hmm: hmm_pfns_bad() was accessing wrong struct

by jglisse＠redhat.com

From: Jérôme Glisse <jglisse(a)redhat.com> The private field of mm_walk struct point to an hmm_vma_walk struct and not to the hmm_range struct desired. Fix to get proper struct pointer. Signed-off-by: Jérôme Glisse <jglisse(a)redhat.com> Cc: stable(a)vger.kernel.org Cc: Evgeny Baskakov <ebaskakov(a)nvidia.com> Cc: Ralph Campbell <rcampbell(a)nvidia.com> Cc: Mark Hairgrove <mhairgrove(a)nvidia.com> Cc: John Hubbard <jhubbard(a)nvidia.com> --- mm/hmm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/mm/hmm.c b/mm/hmm.c index 2d00769e8985..812a66997627 100644 --- a/mm/hmm.c +++ b/mm/hmm.c @@ -336,7 +336,8 @@ static int hmm_pfns_bad(unsigned long addr, unsigned long end, struct mm_walk *walk) { - struct hmm_range *range = walk->private; + struct hmm_vma_walk *hmm_vma_walk = walk->private; + struct hmm_range *range = hmm_vma_walk->range; hmm_pfn_t *pfns = range->pfns; unsigned long i; -- 2.14.3

7 years, 9 months

1
0
0 0

[PATCH 02/15] mm/hmm: fix header file if/else/endif maze v2

by jglisse＠redhat.com

From: Jérôme Glisse <jglisse(a)redhat.com> The #if/#else/#endif for IS_ENABLED(CONFIG_HMM) were wrong. Because of this after multiple include there was multiple definition of both hmm_mm_init() and hmm_mm_destroy() leading to build failure if HMM was enabled (CONFIG_HMM set). Changed since v1: - Fix the maze when CONFIG_HMM is disabled not just when it is disabled. This fix bot build failure. - Improved commit message. Signed-off-by: Jérôme Glisse <jglisse(a)redhat.com> Acked-by: Balbir Singh <bsingharora(a)gmail.com> Cc: stable(a)vger.kernel.org Cc: Andrew Morton <akpm(a)linux-foundation.org> Cc: Ralph Campbell <rcampbell(a)nvidia.com> Cc: John Hubbard <jhubbard(a)nvidia.com> Cc: Evgeny Baskakov <ebaskakov(a)nvidia.com> --- include/linux/hmm.h | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) diff --git a/include/linux/hmm.h b/include/linux/hmm.h index 325017ad9311..36dd21fe5caf 100644 --- a/include/linux/hmm.h +++ b/include/linux/hmm.h @@ -498,23 +498,16 @@ struct hmm_device { struct hmm_device *hmm_device_new(void *drvdata); void hmm_device_put(struct hmm_device *hmm_device); #endif /* CONFIG_DEVICE_PRIVATE || CONFIG_DEVICE_PUBLIC */ -#endif /* IS_ENABLED(CONFIG_HMM) */ /* Below are for HMM internal use only! Not to be used by device driver! */ -#if IS_ENABLED(CONFIG_HMM_MIRROR) void hmm_mm_destroy(struct mm_struct *mm); static inline void hmm_mm_init(struct mm_struct *mm) { mm->hmm = NULL; } -#else /* IS_ENABLED(CONFIG_HMM_MIRROR) */ -static inline void hmm_mm_destroy(struct mm_struct *mm) {} -static inline void hmm_mm_init(struct mm_struct *mm) {} -#endif /* IS_ENABLED(CONFIG_HMM_MIRROR) */ - - #else /* IS_ENABLED(CONFIG_HMM) */ static inline void hmm_mm_destroy(struct mm_struct *mm) {} static inline void hmm_mm_init(struct mm_struct *mm) {} +#endif /* IS_ENABLED(CONFIG_HMM) */ #endif /* LINUX_HMM_H */ -- 2.14.3

7 years, 9 months

1
0
0 0

[PATCH] ACPI / video: Add a quirk to force acpi-video backlight on Samsung 670Z5E

by Hans de Goede

Just like many other Samsung models, the 670Z5E needs to use the acpi-video backlight interface rather then the native one for backlight control to work, add a quirk for this. Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1557060 Cc: stable(a)vger.kernel.org Signed-off-by: Hans de Goede <hdegoede(a)redhat.com> --- drivers/acpi/video_detect.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c index 601e5d372887..43587ac680e4 100644 --- a/drivers/acpi/video_detect.c +++ b/drivers/acpi/video_detect.c @@ -219,6 +219,15 @@ static const struct dmi_system_id video_detect_dmi_table[] = { "3570R/370R/470R/450R/510R/4450RV"), }, }, + { + /* https://bugzilla.redhat.com/show_bug.cgi?id=1557060 */ + .callback = video_detect_force_video, + .ident = "SAMSUNG 670Z5E", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "SAMSUNG ELECTRONICS CO., LTD."), + DMI_MATCH(DMI_PRODUCT_NAME, "670Z5E"), + }, + }, { /* https://bugzilla.redhat.com/show_bug.cgi?id=1094948 */ .callback = video_detect_force_video, -- 2.14.3

7 years, 9 months

2
1
0 0

[patch 12/13] mm/vmscan: wake up flushers for legacy cgroups too

by akpm＠linux-foundation.org

From: Andrey Ryabinin <aryabinin(a)virtuozzo.com> Subject: mm/vmscan: wake up flushers for legacy cgroups too Commit 726d061fbd36 ("mm: vmscan: kick flushers when we encounter dirty pages on the LRU") added flusher invocation to shrink_inactive_list() when many dirty pages on the LRU are encountered. However, shrink_inactive_list() doesn't wake up flushers for legacy cgroup reclaim, so the next commit bbef938429f5 ("mm: vmscan: remove old flusher wakeup from direct reclaim path") removed the only source of flusher's wake up in legacy mem cgroup reclaim path. This leads to premature OOM if there is too many dirty pages in cgroup: # mkdir /sys/fs/cgroup/memory/test # echo $$ > /sys/fs/cgroup/memory/test/tasks # echo 50M > /sys/fs/cgroup/memory/test/memory.limit_in_bytes # dd if=/dev/zero of=tmp_file bs=1M count=100 Killed dd invoked oom-killer: gfp_mask=0x14000c0(GFP_KERNEL), nodemask=(null), order=0, oom_score_adj=0 Call Trace: dump_stack+0x46/0x65 dump_header+0x6b/0x2ac oom_kill_process+0x21c/0x4a0 out_of_memory+0x2a5/0x4b0 mem_cgroup_out_of_memory+0x3b/0x60 mem_cgroup_oom_synchronize+0x2ed/0x330 pagefault_out_of_memory+0x24/0x54 __do_page_fault+0x521/0x540 page_fault+0x45/0x50 Task in /test killed as a result of limit of /test memory: usage 51200kB, limit 51200kB, failcnt 73 memory+swap: usage 51200kB, limit 9007199254740988kB, failcnt 0 kmem: usage 296kB, limit 9007199254740988kB, failcnt 0 Memory cgroup stats for /test: cache:49632KB rss:1056KB rss_huge:0KB shmem:0KB mapped_file:0KB dirty:49500KB writeback:0KB swap:0KB inactive_anon:0KB active_anon:1168KB inactive_file:24760KB active_file:24960KB unevictable:0KB Memory cgroup out of memory: Kill process 3861 (bash) score 88 or sacrifice child Killed process 3876 (dd) total-vm:8484kB, anon-rss:1052kB, file-rss:1720kB, shmem-rss:0kB oom_reaper: reaped process 3876 (dd), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB Wake up flushers in legacy cgroup reclaim too. Link: http://lkml.kernel.org/r/20180315164553.17856-1-aryabinin@virtuozzo.com Fixes: bbef938429f5 ("mm: vmscan: remove old flusher wakeup from direct reclaim path") Signed-off-by: Andrey Ryabinin <aryabinin(a)virtuozzo.com> Tested-by: Shakeel Butt <shakeelb(a)google.com> Acked-by: Michal Hocko <mhocko(a)suse.cz> Cc: Mel Gorman <mgorman(a)techsingularity.net> Cc: Tejun Heo <tj(a)kernel.org> Cc: Johannes Weiner <hannes(a)cmpxchg.org> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <> --- mm/vmscan.c | 31 ++++++++++++++++--------------- 1 file changed, 16 insertions(+), 15 deletions(-) diff -puN mm/vmscan.c~mm-vmscan-wake-up-flushers-for-legacy-cgroups-too mm/vmscan.c --- a/mm/vmscan.c~mm-vmscan-wake-up-flushers-for-legacy-cgroups-too +++ a/mm/vmscan.c @@ -1780,6 +1780,20 @@ shrink_inactive_list(unsigned long nr_to set_bit(PGDAT_WRITEBACK, &pgdat->flags); /* + * If dirty pages are scanned that are not queued for IO, it + * implies that flushers are not doing their job. This can + * happen when memory pressure pushes dirty pages to the end of + * the LRU before the dirty limits are breached and the dirty + * data has expired. It can also happen when the proportion of + * dirty pages grows not through writes but through memory + * pressure reclaiming all the clean cache. And in some cases, + * the flushers simply cannot keep up with the allocation + * rate. Nudge the flusher threads in case they are asleep. + */ + if (stat.nr_unqueued_dirty == nr_taken) + wakeup_flusher_threads(WB_REASON_VMSCAN); + + /* * Legacy memcg will stall in page writeback so avoid forcibly * stalling here. */ @@ -1791,22 +1805,9 @@ shrink_inactive_list(unsigned long nr_to if (stat.nr_dirty && stat.nr_dirty == stat.nr_congested) set_bit(PGDAT_CONGESTED, &pgdat->flags); - /* - * If dirty pages are scanned that are not queued for IO, it - * implies that flushers are not doing their job. This can - * happen when memory pressure pushes dirty pages to the end of - * the LRU before the dirty limits are breached and the dirty - * data has expired. It can also happen when the proportion of - * dirty pages grows not through writes but through memory - * pressure reclaiming all the clean cache. And in some cases, - * the flushers simply cannot keep up with the allocation - * rate. Nudge the flusher threads in case they are asleep, but - * also allow kswapd to start writing pages during reclaim. - */ - if (stat.nr_unqueued_dirty == nr_taken) { - wakeup_flusher_threads(WB_REASON_VMSCAN); + /* Allow kswapd to start writing pages during reclaim. */ + if (stat.nr_unqueued_dirty == nr_taken) set_bit(PGDAT_DIRTY, &pgdat->flags); - } /* * If kswapd scans pages marked marked for immediate _

7 years, 9 months

1
0
0 0

[patch 11/13] Revert "mm: page_alloc: skip over regions of invalid pfns where possible"

by akpm＠linux-foundation.org

From: Daniel Vacek <neelx(a)redhat.com> Subject: Revert "mm: page_alloc: skip over regions of invalid pfns where possible" This reverts b92df1de5d289c0b ("mm: page_alloc: skip over regions of invalid pfns where possible"). The commit is meant to be a boot init speed up skipping the loop in memmap_init_zone() for invalid pfns. But given some specific memory mapping on x86_64 (or more generally theoretically anywhere but on arm with CONFIG_HAVE_ARCH_PFN_VALID) the implementation also skips valid pfns which is plain wrong and causes 'kernel BUG at mm/page_alloc.c:1389!' crash> log | grep -e BUG -e RIP -e Call.Trace -e move_freepages_block -e rmqueue -e freelist -A1 kernel BUG at mm/page_alloc.c:1389! invalid opcode: 0000 [#1] SMP -- RIP: 0010:[<ffffffff8118833e>] [<ffffffff8118833e>] move_freepages+0x15e/0x160 RSP: 0018:ffff88054d727688 EFLAGS: 00010087 -- Call Trace: [<ffffffff811883b3>] move_freepages_block+0x73/0x80 [<ffffffff81189e63>] __rmqueue+0x263/0x460 [<ffffffff8118c781>] get_page_from_freelist+0x7e1/0x9e0 [<ffffffff8118caf6>] __alloc_pages_nodemask+0x176/0x420 -- RIP [<ffffffff8118833e>] move_freepages+0x15e/0x160 RSP <ffff88054d727688> crash> page_init_bug -v | grep RAM <struct resource 0xffff88067fffd2f8> 1000 - 9bfff System RAM (620.00 KiB) <struct resource 0xffff88067fffd3a0> 100000 - 430bffff System RAM ( 1.05 GiB = 1071.75 MiB = 1097472.00 KiB) <struct resource 0xffff88067fffd410> 4b0c8000 - 4bf9cfff System RAM ( 14.83 MiB = 15188.00 KiB) <struct resource 0xffff88067fffd480> 4bfac000 - 646b1fff System RAM (391.02 MiB = 400408.00 KiB) <struct resource 0xffff88067fffd560> 7b788000 - 7b7fffff System RAM (480.00 KiB) <struct resource 0xffff88067fffd640> 100000000 - 67fffffff System RAM ( 22.00 GiB) crash> page_init_bug | head -6 <struct resource 0xffff88067fffd560> 7b788000 - 7b7fffff System RAM (480.00 KiB) <struct page 0xffffea0001ede200> 1fffff00000000 0 <struct pglist_data 0xffff88047ffd9000> 1 <struct zone 0xffff88047ffd9800> DMA32 4096 1048575 <struct page 0xffffea0001ede200> 505736 505344 <struct page 0xffffea0001ed8000> 505855 <struct page 0xffffea0001edffc0> <struct page 0xffffea0001ed8000> 0 0 <struct pglist_data 0xffff88047ffd9000> 0 <struct zone 0xffff88047ffd9000> DMA 1 4095 <struct page 0xffffea0001edffc0> 1fffff00000400 0 <struct pglist_data 0xffff88047ffd9000> 1 <struct zone 0xffff88047ffd9800> DMA32 4096 1048575 BUG, zones differ! crash> kmem -p 77fff000 78000000 7b5ff000 7b600000 7b787000 7b788000 PAGE PHYSICAL MAPPING INDEX CNT FLAGS ffffea0001e00000 78000000 0 0 0 0 ffffea0001ed7fc0 7b5ff000 0 0 0 0 ffffea0001ed8000 7b600000 0 0 0 0 <<<< ffffea0001ede1c0 7b787000 0 0 0 0 ffffea0001ede200 7b788000 0 0 1 1fffff00000000 Link: http://lkml.kernel.org/r/20180316143855.29838-1-neelx@redhat.com Fixes: b92df1de5d28 ("mm: page_alloc: skip over regions of invalid pfns where possible") Signed-off-by: Daniel Vacek <neelx(a)redhat.com> Acked-by: Ard Biesheuvel <ard.biesheuvel(a)linaro.org> Acked-by: Michal Hocko <mhocko(a)suse.com> Reviewed-by: Andrew Morton <akpm(a)linux-foundation.org> Cc: Vlastimil Babka <vbabka(a)suse.cz> Cc: Mel Gorman <mgorman(a)techsingularity.net> Cc: Pavel Tatashin <pasha.tatashin(a)oracle.com> Cc: Paul Burton <paul.burton(a)imgtec.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <> --- include/linux/memblock.h | 1 - mm/memblock.c | 28 ---------------------------- mm/page_alloc.c | 11 +---------- 3 files changed, 1 insertion(+), 39 deletions(-) diff -puN include/linux/memblock.h~revert-mm-page_alloc-skip-over-regions-of-invalid-pfns-where-possible include/linux/memblock.h --- a/include/linux/memblock.h~revert-mm-page_alloc-skip-over-regions-of-invalid-pfns-where-possible +++ a/include/linux/memblock.h @@ -187,7 +187,6 @@ int memblock_search_pfn_nid(unsigned lon unsigned long *end_pfn); void __next_mem_pfn_range(int *idx, int nid, unsigned long *out_start_pfn, unsigned long *out_end_pfn, int *out_nid); -unsigned long memblock_next_valid_pfn(unsigned long pfn, unsigned long max_pfn); /** * for_each_mem_pfn_range - early memory pfn range iterator diff -puN mm/memblock.c~revert-mm-page_alloc-skip-over-regions-of-invalid-pfns-where-possible mm/memblock.c --- a/mm/memblock.c~revert-mm-page_alloc-skip-over-regions-of-invalid-pfns-where-possible +++ a/mm/memblock.c @@ -1101,34 +1101,6 @@ void __init_memblock __next_mem_pfn_rang *out_nid = r->nid; } -unsigned long __init_memblock memblock_next_valid_pfn(unsigned long pfn, - unsigned long max_pfn) -{ - struct memblock_type *type = &memblock.memory; - unsigned int right = type->cnt; - unsigned int mid, left = 0; - phys_addr_t addr = PFN_PHYS(++pfn); - - do { - mid = (right + left) / 2; - - if (addr < type->regions[mid].base) - right = mid; - else if (addr >= (type->regions[mid].base + - type->regions[mid].size)) - left = mid + 1; - else { - /* addr is within the region, so pfn is valid */ - return pfn; - } - } while (left < right); - - if (right == type->cnt) - return -1UL; - else - return PHYS_PFN(type->regions[right].base); -} - /** * memblock_set_node - set node ID on memblock regions * @base: base of area to set node ID for diff -puN mm/page_alloc.c~revert-mm-page_alloc-skip-over-regions-of-invalid-pfns-where-possible mm/page_alloc.c --- a/mm/page_alloc.c~revert-mm-page_alloc-skip-over-regions-of-invalid-pfns-where-possible +++ a/mm/page_alloc.c @@ -5356,17 +5356,8 @@ void __meminit memmap_init_zone(unsigned if (context != MEMMAP_EARLY) goto not_early; - if (!early_pfn_valid(pfn)) { -#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP - /* - * Skip to the pfn preceding the next valid one (or - * end_pfn), such that we hit a valid pfn (or end_pfn) - * on our next iteration of the loop. - */ - pfn = memblock_next_valid_pfn(pfn, end_pfn) - 1; -#endif + if (!early_pfn_valid(pfn)) continue; - } if (!early_pfn_in_nid(pfn, nid)) continue; if (!update_defer_init(pgdat, pfn, end_pfn, &nr_initialised)) _

7 years, 9 months

1
0
0 0

[patch 10/13] mm/shmem: do not wait for lock_page() in shmem_unused_huge_shrink()

by akpm＠linux-foundation.org

From: "Kirill A. Shutemov" <kirill.shutemov(a)linux.intel.com> Subject: mm/shmem: do not wait for lock_page() in shmem_unused_huge_shrink() shmem_unused_huge_shrink() gets called from reclaim path. Waiting for page lock may lead to deadlock there. There was a bug report that may be attributed to this: http://lkml.kernel.org/r/alpine.LRH.2.11.1801242349220.30642@mail.ewheeler.… Replace lock_page() with trylock_page() and skip the page if we failed to lock it. We will get to the page on the next scan. We can test for the PageTransHuge() outside the page lock as we only need protection against splitting the page under us. Holding pin oni the page is enough for this. Link: http://lkml.kernel.org/r/20180316210830.43738-1-kirill.shutemov@linux.intel… Fixes: 779750d20b93 ("shmem: split huge pages beyond i_size under memory pressure") Signed-off-by: Kirill A. Shutemov <kirill.shutemov(a)linux.intel.com> Reported-by: Eric Wheeler <linux-mm(a)lists.ewheeler.net> Acked-by: Michal Hocko <mhocko(a)suse.com> Reviewed-by: Andrew Morton <akpm(a)linux-foundation.org> Cc: Tetsuo Handa <penguin-kernel(a)I-love.SAKURA.ne.jp> Cc: Hugh Dickins <hughd(a)google.com> Cc: <stable(a)vger.kernel.org> [4.8+] Signed-off-by: Andrew Morton <> --- mm/shmem.c | 31 ++++++++++++++++++++----------- 1 file changed, 20 insertions(+), 11 deletions(-) diff -puN mm/shmem.c~mm-shmem-do-not-wait-for-lock_page-in-shmem_unused_huge_shrink mm/shmem.c --- a/mm/shmem.c~mm-shmem-do-not-wait-for-lock_page-in-shmem_unused_huge_shrink +++ a/mm/shmem.c @@ -493,36 +493,45 @@ next: info = list_entry(pos, struct shmem_inode_info, shrinklist); inode = &info->vfs_inode; - if (nr_to_split && split >= nr_to_split) { - iput(inode); - continue; - } + if (nr_to_split && split >= nr_to_split) + goto leave; - page = find_lock_page(inode->i_mapping, + page = find_get_page(inode->i_mapping, (inode->i_size & HPAGE_PMD_MASK) >> PAGE_SHIFT); if (!page) goto drop; + /* No huge page at the end of the file: nothing to split */ if (!PageTransHuge(page)) { - unlock_page(page); put_page(page); goto drop; } + /* + * Leave the inode on the list if we failed to lock + * the page at this time. + * + * Waiting for the lock may lead to deadlock in the + * reclaim path. + */ + if (!trylock_page(page)) { + put_page(page); + goto leave; + } + ret = split_huge_page(page); unlock_page(page); put_page(page); - if (ret) { - /* split failed: leave it on the list */ - iput(inode); - continue; - } + /* If split failed leave the inode on the list */ + if (ret) + goto leave; split++; drop: list_del_init(&info->shrinklist); removed++; +leave: iput(inode); } _

7 years, 9 months

1
0
0 0

[patch 09/13] mm/thp: do not wait for lock_page() in deferred_split_scan()

by akpm＠linux-foundation.org

From: "Kirill A. Shutemov" <kirill.shutemov(a)linux.intel.com> Subject: mm/thp: do not wait for lock_page() in deferred_split_scan() deferred_split_scan() gets called from reclaim path. Waiting for page lock may lead to deadlock there. Replace lock_page() with trylock_page() and skip the page if we failed to lock it. We will get to the page on the next scan. Link: http://lkml.kernel.org/r/20180315150747.31945-1-kirill.shutemov@linux.intel… Fixes: 9a982250f773 ("thp: introduce deferred_split_huge_page()") Signed-off-by: Kirill A. Shutemov <kirill.shutemov(a)linux.intel.com> Acked-by: Michal Hocko <mhocko(a)suse.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <> --- mm/huge_memory.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff -puN mm/huge_memory.c~mm-thp-do-not-wait-for-lock_page-in-deferred_split_scan mm/huge_memory.c --- a/mm/huge_memory.c~mm-thp-do-not-wait-for-lock_page-in-deferred_split_scan +++ a/mm/huge_memory.c @@ -2783,11 +2783,13 @@ static unsigned long deferred_split_scan list_for_each_safe(pos, next, &list) { page = list_entry((void *)pos, struct page, mapping); - lock_page(page); + if (!trylock_page(page)) + goto next; /* split_huge_page() removes page from list on success */ if (!split_huge_page(page)) split++; unlock_page(page); +next: put_page(page); } _

7 years, 9 months

1
0
0 0

[patch 08/13] mm/khugepaged.c: convert VM_BUG_ON() to collapse fail

by akpm＠linux-foundation.org

From: "Kirill A. Shutemov" <kirill.shutemov(a)linux.intel.com> Subject: mm/khugepaged.c: convert VM_BUG_ON() to collapse fail khugepaged is not yet able to convert PTE-mapped huge pages back to PMD mapped. We do not collapse such pages. See check khugepaged_scan_pmd(). But if between khugepaged_scan_pmd() and __collapse_huge_page_isolate() somebody managed to instantiate THP in the range and then split the PMD back to PTEs we would have a problem -- VM_BUG_ON_PAGE(PageCompound(page)) will get triggered. It's possible since we drop mmap_sem during collapse to re-take for write. Replace the VM_BUG_ON() with graceful collapse fail. Link: http://lkml.kernel.org/r/20180315152353.27989-1-kirill.shutemov@linux.intel… Fixes: b1caa957ae6d ("khugepaged: ignore pmd tables with THP mapped with ptes") Signed-off-by: Kirill A. Shutemov <kirill.shutemov(a)linux.intel.com> Cc: Laura Abbott <labbott(a)redhat.com> Cc: Jerome Marchand <jmarchan(a)redhat.com> Cc: Vlastimil Babka <vbabka(a)suse.cz> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <> --- mm/khugepaged.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff -puN mm/khugepaged.c~mm-khugepaged-convert-vm_bug_on-to-collapse-fail mm/khugepaged.c --- a/mm/khugepaged.c~mm-khugepaged-convert-vm_bug_on-to-collapse-fail +++ a/mm/khugepaged.c @@ -530,7 +530,12 @@ static int __collapse_huge_page_isolate( goto out; } - VM_BUG_ON_PAGE(PageCompound(page), page); + /* TODO: teach khugepaged to collapse THP mapped with pte */ + if (PageCompound(page)) { + result = SCAN_PAGE_COMPOUND; + goto out; + } + VM_BUG_ON_PAGE(!PageAnon(page), page); /* _

7 years, 9 months

1
0
0 0

[patch 07/13] x86/mm: implement free pmd/pte page interfaces

by akpm＠linux-foundation.org

From: Toshi Kani <toshi.kani(a)hpe.com> Subject: x86/mm: implement free pmd/pte page interfaces Implement pud_free_pmd_page() and pmd_free_pte_page() on x86, which clear a given pud/pmd entry and free up lower level page table(s). Address range associated with the pud/pmd entry must have been purged by INVLPG. Link: http://lkml.kernel.org/r/20180314180155.19492-3-toshi.kani@hpe.com Fixes: e61ce6ade404e ("mm: change ioremap to set up huge I/O mappings") Signed-off-by: Toshi Kani <toshi.kani(a)hpe.com> Reported-by: Lei Li <lious.lilei(a)hisilicon.com> Cc: Michal Hocko <mhocko(a)suse.com> Cc: Thomas Gleixner <tglx(a)linutronix.de> Cc: Ingo Molnar <mingo(a)redhat.com> Cc: "H. Peter Anvin" <hpa(a)zytor.com> Cc: Borislav Petkov <bp(a)suse.de> Cc: Matthew Wilcox <willy(a)infradead.org> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <> --- arch/x86/mm/pgtable.c | 28 ++++++++++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-) diff -puN arch/x86/mm/pgtable.c~x86-mm-implement-free-pmd-pte-page-interfaces arch/x86/mm/pgtable.c --- a/arch/x86/mm/pgtable.c~x86-mm-implement-free-pmd-pte-page-interfaces +++ a/arch/x86/mm/pgtable.c @@ -712,7 +712,22 @@ int pmd_clear_huge(pmd_t *pmd) */ int pud_free_pmd_page(pud_t *pud) { - return pud_none(*pud); + pmd_t *pmd; + int i; + + if (pud_none(*pud)) + return 1; + + pmd = (pmd_t *)pud_page_vaddr(*pud); + + for (i = 0; i < PTRS_PER_PMD; i++) + if (!pmd_free_pte_page(&pmd[i])) + return 0; + + pud_clear(pud); + free_page((unsigned long)pmd); + + return 1; } /** @@ -724,6 +739,15 @@ int pud_free_pmd_page(pud_t *pud) */ int pmd_free_pte_page(pmd_t *pmd) { - return pmd_none(*pmd); + pte_t *pte; + + if (pmd_none(*pmd)) + return 1; + + pte = (pte_t *)pmd_page_vaddr(*pmd); + pmd_clear(pmd); + free_page((unsigned long)pte); + + return 1; } #endif /* CONFIG_HAVE_ARCH_HUGE_VMAP */ _

7 years, 9 months

1
0
0 0

[patch 05/13] h8300: remove extraneous __BIG_ENDIAN definition

by akpm＠linux-foundation.org

From: Arnd Bergmann <arnd(a)arndb.de> Subject: h8300: remove extraneous __BIG_ENDIAN definition A bugfix I did earlier caused a build regression on h8300, which defines the __BIG_ENDIAN macro in a slightly different way than the generic code: arch/h8300/include/asm/byteorder.h:5:0: warning: "__BIG_ENDIAN" redefined We don't need to define it here, as the same macro is already provided by the linux/byteorder/big_endian.h, and that version does not conflict. While this is a v4.16 regression, my earlier patch also got backported to the 4.14 and 4.15 stable kernels, so we need the fixup there as well. Link: http://lkml.kernel.org/r/20180313120752.2645129-1-arnd@arndb.de Fixes: 101110f6271c ("Kbuild: always define endianess in kconfig.h") Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> Cc: Yoshinori Sato <ysato(a)users.sourceforge.jp> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <> --- arch/h8300/include/asm/byteorder.h | 1 - 1 file changed, 1 deletion(-) diff -puN arch/h8300/include/asm/byteorder.h~h8300-remove-extranous-__big_endian-definition arch/h8300/include/asm/byteorder.h --- a/arch/h8300/include/asm/byteorder.h~h8300-remove-extranous-__big_endian-definition +++ a/arch/h8300/include/asm/byteorder.h @@ -2,7 +2,6 @@ #ifndef __H8300_BYTEORDER_H__ #define __H8300_BYTEORDER_H__ -#define __BIG_ENDIAN __ORDER_BIG_ENDIAN__ #include <linux/byteorder/big_endian.h> #endif _

7 years, 9 months

1
0
0 0

[patch 04/13] hugetlbfs: check for pgoff value overflow

by akpm＠linux-foundation.org

From: Mike Kravetz <mike.kravetz(a)oracle.com> Subject: hugetlbfs: check for pgoff value overflow A vma with vm_pgoff large enough to overflow a loff_t type when converted to a byte offset can be passed via the remap_file_pages system call. The hugetlbfs mmap routine uses the byte offset to calculate reservations and file size. A sequence such as: mmap(0x20a00000, 0x600000, 0, 0x66033, -1, 0); remap_file_pages(0x20a00000, 0x600000, 0, 0x20000000000000, 0); will result in the following when task exits/file closed, kernel BUG at mm/hugetlb.c:749! Call Trace: hugetlbfs_evict_inode+0x2f/0x40 evict+0xcb/0x190 __dentry_kill+0xcb/0x150 __fput+0x164/0x1e0 task_work_run+0x84/0xa0 exit_to_usermode_loop+0x7d/0x80 do_syscall_64+0x18b/0x190 entry_SYSCALL_64_after_hwframe+0x3d/0xa2 The overflowed pgoff value causes hugetlbfs to try to set up a mapping with a negative range (end < start) that leaves invalid state which causes the BUG. The previous overflow fix to this code was incomplete and did not take the remap_file_pages system call into account. [mike.kravetz(a)oracle.com: v3] Link: http://lkml.kernel.org/r/20180309002726.7248-1-mike.kravetz@oracle.com [akpm(a)linux-foundation.org: include mmdebug.h] [akpm(a)linux-foundation.org: fix -ve left shift count on sh] Link: http://lkml.kernel.org/r/20180308210502.15952-1-mike.kravetz@oracle.com Fixes: 045c7a3f53d9 ("hugetlbfs: fix offset overflow in hugetlbfs mmap") Signed-off-by: Mike Kravetz <mike.kravetz(a)oracle.com> Reported-by: Nic Losby <blurbdust(a)gmail.com> Acked-by: Michal Hocko <mhocko(a)suse.com> Cc: "Kirill A . Shutemov" <kirill.shutemov(a)linux.intel.com> Cc: Yisheng Xie <xieyisheng1(a)huawei.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <> --- fs/hugetlbfs/inode.c | 17 ++++++++++++++--- mm/hugetlb.c | 7 +++++++ 2 files changed, 21 insertions(+), 3 deletions(-) diff -puN fs/hugetlbfs/inode.c~hugetlbfs-check-for-pgoff-value-overflow fs/hugetlbfs/inode.c --- a/fs/hugetlbfs/inode.c~hugetlbfs-check-for-pgoff-value-overflow +++ a/fs/hugetlbfs/inode.c @@ -108,6 +108,16 @@ static void huge_pagevec_release(struct pagevec_reinit(pvec); } +/* + * Mask used when checking the page offset value passed in via system + * calls. This value will be converted to a loff_t which is signed. + * Therefore, we want to check the upper PAGE_SHIFT + 1 bits of the + * value. The extra bit (- 1 in the shift value) is to take the sign + * bit into account. + */ +#define PGOFF_LOFFT_MAX \ + (((1UL << (PAGE_SHIFT + 1)) - 1) << (BITS_PER_LONG - (PAGE_SHIFT + 1))) + static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) { struct inode *inode = file_inode(file); @@ -127,12 +137,13 @@ static int hugetlbfs_file_mmap(struct fi vma->vm_ops = &hugetlb_vm_ops; /* - * Offset passed to mmap (before page shift) could have been - * negative when represented as a (l)off_t. + * page based offset in vm_pgoff could be sufficiently large to + * overflow a (l)off_t when converted to byte offset. */ - if (((loff_t)vma->vm_pgoff << PAGE_SHIFT) < 0) + if (vma->vm_pgoff & PGOFF_LOFFT_MAX) return -EINVAL; + /* must be huge page aligned */ if (vma->vm_pgoff & (~huge_page_mask(h) >> PAGE_SHIFT)) return -EINVAL; diff -puN mm/hugetlb.c~hugetlbfs-check-for-pgoff-value-overflow mm/hugetlb.c --- a/mm/hugetlb.c~hugetlbfs-check-for-pgoff-value-overflow +++ a/mm/hugetlb.c @@ -18,6 +18,7 @@ #include <linux/bootmem.h> #include <linux/sysfs.h> #include <linux/slab.h> +#include <linux/mmdebug.h> #include <linux/sched/signal.h> #include <linux/rmap.h> #include <linux/string_helpers.h> @@ -4374,6 +4375,12 @@ int hugetlb_reserve_pages(struct inode * struct resv_map *resv_map; long gbl_reserve; + /* This should never happen */ + if (from > to) { + VM_WARN(1, "%s called with a negative range\n", __func__); + return -EINVAL; + } + /* * Only apply hugepage reservation if asked. At fault time, an * attempt will be made for VM_NORESERVE to allocate a page _

7 years, 9 months

1
0
0 0

[patch 03/13] lockdep: fix fs_reclaim warning

by akpm＠linux-foundation.org

From: Tetsuo Handa <penguin-kernel(a)I-love.SAKURA.ne.jp> Subject: lockdep: fix fs_reclaim warning Dave Jones reported fs_reclaim lockdep warnings. ============================================ WARNING: possible recursive locking detected 4.15.0-rc9-backup-debug+ #1 Not tainted -------------------------------------------- sshd/24800 is trying to acquire lock: (fs_reclaim){+.+.}, at: [<0000000084f438c2>] fs_reclaim_acquire.part.102+0x5/0x30 but task is already holding lock: (fs_reclaim){+.+.}, at: [<0000000084f438c2>] fs_reclaim_acquire.part.102+0x5/0x30 other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(fs_reclaim); lock(fs_reclaim); *** DEADLOCK *** May be due to missing lock nesting notation 2 locks held by sshd/24800: #0: (sk_lock-AF_INET6){+.+.}, at: [<000000001a069652>] tcp_sendmsg+0x19/0x40 #1: (fs_reclaim){+.+.}, at: [<0000000084f438c2>] fs_reclaim_acquire.part.102+0x5/0x30 stack backtrace: CPU: 3 PID: 24800 Comm: sshd Not tainted 4.15.0-rc9-backup-debug+ #1 Call Trace: dump_stack+0xbc/0x13f __lock_acquire+0xa09/0x2040 lock_acquire+0x12e/0x350 fs_reclaim_acquire.part.102+0x29/0x30 kmem_cache_alloc+0x3d/0x2c0 alloc_extent_state+0xa7/0x410 __clear_extent_bit+0x3ea/0x570 try_release_extent_mapping+0x21a/0x260 __btrfs_releasepage+0xb0/0x1c0 btrfs_releasepage+0x161/0x170 try_to_release_page+0x162/0x1c0 shrink_page_list+0x1d5a/0x2fb0 shrink_inactive_list+0x451/0x940 shrink_node_memcg.constprop.88+0x4c9/0x5e0 shrink_node+0x12d/0x260 try_to_free_pages+0x418/0xaf0 __alloc_pages_slowpath+0x976/0x1790 __alloc_pages_nodemask+0x52c/0x5c0 new_slab+0x374/0x3f0 ___slab_alloc.constprop.81+0x47e/0x5a0 __slab_alloc.constprop.80+0x32/0x60 __kmalloc_track_caller+0x267/0x310 __kmalloc_reserve.isra.40+0x29/0x80 __alloc_skb+0xee/0x390 sk_stream_alloc_skb+0xb8/0x340 tcp_sendmsg_locked+0x8e6/0x1d30 tcp_sendmsg+0x27/0x40 inet_sendmsg+0xd0/0x310 sock_write_iter+0x17a/0x240 __vfs_write+0x2ab/0x380 vfs_write+0xfb/0x260 SyS_write+0xb6/0x140 do_syscall_64+0x1e5/0xc05 entry_SYSCALL64_slow_path+0x25/0x25 This warning is caused by commit d92a8cfcb37ecd13 ("locking/lockdep: Rework FS_RECLAIM annotation") which replaced lockdep_set_current_reclaim_state()/ lockdep_clear_current_reclaim_state() in __perform_reclaim() and lockdep_trace_alloc() in slab_pre_alloc_hook() with fs_reclaim_acquire()/ fs_reclaim_release(). Since __kmalloc_reserve() from __alloc_skb() adds __GFP_NOMEMALLOC | __GFP_NOWARN to gfp_mask, and all reclaim path simply propagates __GFP_NOMEMALLOC, fs_reclaim_acquire() in slab_pre_alloc_hook() is trying to grab the 'fake' lock again when __perform_reclaim() already grabbed the 'fake' lock. The /* this guy won't enter reclaim */ if ((current->flags & PF_MEMALLOC) && !(gfp_mask & __GFP_NOMEMALLOC)) return false; test which causes slab_pre_alloc_hook() to try to grab the 'fake' lock was added by commit cf40bd16fdad42c0 ("lockdep: annotate reclaim context (__GFP_NOFS)"). But that test is outdated because PF_MEMALLOC thread won't enter reclaim regardless of __GFP_NOMEMALLOC after commit 341ce06f69abfafa ("page allocator: calculate the alloc_flags for allocation only once") added the PF_MEMALLOC safeguard ( /* Avoid recursion of direct reclaim */ if (p->flags & PF_MEMALLOC) goto nopage; in __alloc_pages_slowpath()). Thus, let's fix outdated test by removing __GFP_NOMEMALLOC test and allow __need_fs_reclaim() to return false. Link: http://lkml.kernel.org/r/201802280650.FJC73911.FOSOMLJVFFQtHO@I-love.SAKURA… Fixes: d92a8cfcb37ecd13 ("locking/lockdep: Rework FS_RECLAIM annotation") Signed-off-by: Tetsuo Handa <penguin-kernel(a)I-love.SAKURA.ne.jp> Reported-by: Dave Jones <davej(a)codemonkey.org.uk> Tested-by: Dave Jones <davej(a)codemonkey.org.uk> Cc: Peter Zijlstra <peterz(a)infradead.org> Cc: Nick Piggin <npiggin(a)gmail.com> Cc: Ingo Molnar <mingo(a)elte.hu> Cc: Nikolay Borisov <nborisov(a)suse.com> Cc: Michal Hocko <mhocko(a)kernel.org> Cc: <stable(a)vger.kernel.org> [4.14+] Signed-off-by: Andrew Morton <> --- mm/page_alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff -puN mm/page_alloc.c~lockdep-fix-fs_reclaim-warning mm/page_alloc.c --- a/mm/page_alloc.c~lockdep-fix-fs_reclaim-warning +++ a/mm/page_alloc.c @@ -3596,7 +3596,7 @@ static bool __need_fs_reclaim(gfp_t gfp_ return false; /* this guy won't enter reclaim */ - if ((current->flags & PF_MEMALLOC) && !(gfp_mask & __GFP_NOMEMALLOC)) + if (current->flags & PF_MEMALLOC) return false; /* We're only interested __GFP_FS allocations for now */ _

7 years, 9 months

1
0
0 0

Patch "RDMA/vmw_pvrdma: Fix usage of user response structures in ABI file" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled RDMA/vmw_pvrdma: Fix usage of user response structures in ABI file to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: rdma-vmw_pvrdma-fix-usage-of-user-response-structures-in-abi-file.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 1f5a6c47aabc4606f91ad2e6ef71a1ff1924101c Mon Sep 17 00:00:00 2001 From: Adit Ranadive <aditr(a)vmware.com> Date: Thu, 15 Feb 2018 12:36:46 -0800 Subject: RDMA/vmw_pvrdma: Fix usage of user response structures in ABI file From: Adit Ranadive <aditr(a)vmware.com> commit 1f5a6c47aabc4606f91ad2e6ef71a1ff1924101c upstream. This ensures that we return the right structures back to userspace. Otherwise, it looks like the reserved fields in the response structures in userspace might have uninitialized data in them. Fixes: 8b10ba783c9d ("RDMA/vmw_pvrdma: Add shared receive queue support") Fixes: 29c8d9eba550 ("IB: Add vmw_pvrdma driver") Suggested-by: Jason Gunthorpe <jgg(a)mellanox.com> Reviewed-by: Bryan Tan <bryantan(a)vmware.com> Reviewed-by: Aditya Sarwade <asarwade(a)vmware.com> Reviewed-by: Jorgen Hansen <jhansen(a)vmware.com> Signed-off-by: Adit Ranadive <aditr(a)vmware.com> Signed-off-by: Jason Gunthorpe <jgg(a)mellanox.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c | 4 +++- drivers/infiniband/hw/vmw_pvrdma/pvrdma_srq.c | 4 +++- drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c | 4 +++- 3 files changed, 9 insertions(+), 3 deletions(-) --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c @@ -114,6 +114,7 @@ struct ib_cq *pvrdma_create_cq(struct ib union pvrdma_cmd_resp rsp; struct pvrdma_cmd_create_cq *cmd = &req.create_cq; struct pvrdma_cmd_create_cq_resp *resp = &rsp.create_cq_resp; + struct pvrdma_create_cq_resp cq_resp = {0}; struct pvrdma_create_cq ucmd; BUILD_BUG_ON(sizeof(struct pvrdma_cqe) != 64); @@ -198,6 +199,7 @@ struct ib_cq *pvrdma_create_cq(struct ib cq->ibcq.cqe = resp->cqe; cq->cq_handle = resp->cq_handle; + cq_resp.cqn = resp->cq_handle; spin_lock_irqsave(&dev->cq_tbl_lock, flags); dev->cq_tbl[cq->cq_handle % dev->dsr->caps.max_cq] = cq; spin_unlock_irqrestore(&dev->cq_tbl_lock, flags); @@ -206,7 +208,7 @@ struct ib_cq *pvrdma_create_cq(struct ib cq->uar = &(to_vucontext(context)->uar); /* Copy udata back. */ - if (ib_copy_to_udata(udata, &cq->cq_handle, sizeof(__u32))) { + if (ib_copy_to_udata(udata, &cq_resp, sizeof(cq_resp))) { dev_warn(&dev->pdev->dev, "failed to copy back udata\n"); pvrdma_destroy_cq(&cq->ibcq); --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_srq.c +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_srq.c @@ -113,6 +113,7 @@ struct ib_srq *pvrdma_create_srq(struct union pvrdma_cmd_resp rsp; struct pvrdma_cmd_create_srq *cmd = &req.create_srq; struct pvrdma_cmd_create_srq_resp *resp = &rsp.create_srq_resp; + struct pvrdma_create_srq_resp srq_resp = {0}; struct pvrdma_create_srq ucmd; unsigned long flags; int ret; @@ -204,12 +205,13 @@ struct ib_srq *pvrdma_create_srq(struct } srq->srq_handle = resp->srqn; + srq_resp.srqn = resp->srqn; spin_lock_irqsave(&dev->srq_tbl_lock, flags); dev->srq_tbl[srq->srq_handle % dev->dsr->caps.max_srq] = srq; spin_unlock_irqrestore(&dev->srq_tbl_lock, flags); /* Copy udata back. */ - if (ib_copy_to_udata(udata, &srq->srq_handle, sizeof(__u32))) { + if (ib_copy_to_udata(udata, &srq_resp, sizeof(srq_resp))) { dev_warn(&dev->pdev->dev, "failed to copy back udata\n"); pvrdma_destroy_srq(&srq->ibsrq); return ERR_PTR(-EINVAL); --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c @@ -447,6 +447,7 @@ struct ib_pd *pvrdma_alloc_pd(struct ib_ union pvrdma_cmd_resp rsp; struct pvrdma_cmd_create_pd *cmd = &req.create_pd; struct pvrdma_cmd_create_pd_resp *resp = &rsp.create_pd_resp; + struct pvrdma_alloc_pd_resp pd_resp = {0}; int ret; void *ptr; @@ -475,9 +476,10 @@ struct ib_pd *pvrdma_alloc_pd(struct ib_ pd->privileged = !context; pd->pd_handle = resp->pd_handle; pd->pdn = resp->pd_handle; + pd_resp.pdn = resp->pd_handle; if (context) { - if (ib_copy_to_udata(udata, &pd->pdn, sizeof(__u32))) { + if (ib_copy_to_udata(udata, &pd_resp, sizeof(pd_resp))) { dev_warn(&dev->pdev->dev, "failed to copy back protection domain\n"); pvrdma_dealloc_pd(&pd->ibpd); Patches currently in stable-queue which might be from aditr(a)vmware.com are queue-4.15/rdma-vmw_pvrdma-fix-usage-of-user-response-structures-in-abi-file.patch

7 years, 9 months

1
0
0 0

Patch "RDMA/vmw_pvrdma: Fix usage of user response structures in ABI file" has been added to the 4.14-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled RDMA/vmw_pvrdma: Fix usage of user response structures in ABI file to the 4.14-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: rdma-vmw_pvrdma-fix-usage-of-user-response-structures-in-abi-file.patch and it can be found in the queue-4.14 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 1f5a6c47aabc4606f91ad2e6ef71a1ff1924101c Mon Sep 17 00:00:00 2001 From: Adit Ranadive <aditr(a)vmware.com> Date: Thu, 15 Feb 2018 12:36:46 -0800 Subject: RDMA/vmw_pvrdma: Fix usage of user response structures in ABI file From: Adit Ranadive <aditr(a)vmware.com> commit 1f5a6c47aabc4606f91ad2e6ef71a1ff1924101c upstream. This ensures that we return the right structures back to userspace. Otherwise, it looks like the reserved fields in the response structures in userspace might have uninitialized data in them. Fixes: 8b10ba783c9d ("RDMA/vmw_pvrdma: Add shared receive queue support") Fixes: 29c8d9eba550 ("IB: Add vmw_pvrdma driver") Suggested-by: Jason Gunthorpe <jgg(a)mellanox.com> Reviewed-by: Bryan Tan <bryantan(a)vmware.com> Reviewed-by: Aditya Sarwade <asarwade(a)vmware.com> Reviewed-by: Jorgen Hansen <jhansen(a)vmware.com> Signed-off-by: Adit Ranadive <aditr(a)vmware.com> Signed-off-by: Jason Gunthorpe <jgg(a)mellanox.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c | 4 +++- drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-) --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_cq.c @@ -114,6 +114,7 @@ struct ib_cq *pvrdma_create_cq(struct ib union pvrdma_cmd_resp rsp; struct pvrdma_cmd_create_cq *cmd = &req.create_cq; struct pvrdma_cmd_create_cq_resp *resp = &rsp.create_cq_resp; + struct pvrdma_create_cq_resp cq_resp = {0}; struct pvrdma_create_cq ucmd; BUILD_BUG_ON(sizeof(struct pvrdma_cqe) != 64); @@ -198,6 +199,7 @@ struct ib_cq *pvrdma_create_cq(struct ib cq->ibcq.cqe = resp->cqe; cq->cq_handle = resp->cq_handle; + cq_resp.cqn = resp->cq_handle; spin_lock_irqsave(&dev->cq_tbl_lock, flags); dev->cq_tbl[cq->cq_handle % dev->dsr->caps.max_cq] = cq; spin_unlock_irqrestore(&dev->cq_tbl_lock, flags); @@ -206,7 +208,7 @@ struct ib_cq *pvrdma_create_cq(struct ib cq->uar = &(to_vucontext(context)->uar); /* Copy udata back. */ - if (ib_copy_to_udata(udata, &cq->cq_handle, sizeof(__u32))) { + if (ib_copy_to_udata(udata, &cq_resp, sizeof(cq_resp))) { dev_warn(&dev->pdev->dev, "failed to copy back udata\n"); pvrdma_destroy_cq(&cq->ibcq); --- a/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c +++ b/drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c @@ -444,6 +444,7 @@ struct ib_pd *pvrdma_alloc_pd(struct ib_ union pvrdma_cmd_resp rsp; struct pvrdma_cmd_create_pd *cmd = &req.create_pd; struct pvrdma_cmd_create_pd_resp *resp = &rsp.create_pd_resp; + struct pvrdma_alloc_pd_resp pd_resp = {0}; int ret; void *ptr; @@ -472,9 +473,10 @@ struct ib_pd *pvrdma_alloc_pd(struct ib_ pd->privileged = !context; pd->pd_handle = resp->pd_handle; pd->pdn = resp->pd_handle; + pd_resp.pdn = resp->pd_handle; if (context) { - if (ib_copy_to_udata(udata, &pd->pdn, sizeof(__u32))) { + if (ib_copy_to_udata(udata, &pd_resp, sizeof(pd_resp))) { dev_warn(&dev->pdev->dev, "failed to copy back protection domain\n"); pvrdma_dealloc_pd(&pd->ibpd); Patches currently in stable-queue which might be from aditr(a)vmware.com are queue-4.14/rdma-vmw_pvrdma-fix-usage-of-user-response-structures-in-abi-file.patch

7 years, 9 months

1
0
0 0

Patch "kbuild: fix linker feature test macros when cross compiling with Clang" has been added to the 4.14-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled kbuild: fix linker feature test macros when cross compiling with Clang to the 4.14-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: kbuild-fix-linker-feature-test-macros-when-cross-compiling-with-clang.patch and it can be found in the queue-4.14 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 86a9df597cdd564d2d29c65897bcad42519e3678 Mon Sep 17 00:00:00 2001 From: Nick Desaulniers <ndesaulniers(a)google.com> Date: Mon, 6 Nov 2017 10:47:54 -0800 Subject: kbuild: fix linker feature test macros when cross compiling with Clang From: Nick Desaulniers <ndesaulniers(a)google.com> commit 86a9df597cdd564d2d29c65897bcad42519e3678 upstream. I was not seeing my linker flags getting added when using ld-option when cross compiling with Clang. Upon investigation, this seems to be due to a difference in how GCC vs Clang handle cross compilation. GCC is configured at build time to support one backend, that is implicit when compiling. Clang is explicit via the use of `-target <triple>` and ships with all supported backends by default. GNU Make feature test macros that compile then link will always fail when cross compiling with Clang unless Clang's triple is passed along to the compiler. For example: $ clang -x c /dev/null -c -o temp.o $ aarch64-linux-android/bin/ld -E temp.o aarch64-linux-android/bin/ld: unknown architecture of input file `temp.o' is incompatible with aarch64 output aarch64-linux-android/bin/ld: warning: cannot find entry symbol _start; defaulting to 0000000000400078 $ echo $? 1 $ clang -target aarch64-linux-android- -x c /dev/null -c -o temp.o $ aarch64-linux-android/bin/ld -E temp.o aarch64-linux-android/bin/ld: warning: cannot find entry symbol _start; defaulting to 00000000004002e4 $ echo $? 0 This causes conditional checks that invoke $(CC) without the target triple, then $(LD) on the result, to always fail. Suggested-by: Masahiro Yamada <yamada.masahiro(a)socionext.com> Signed-off-by: Nick Desaulniers <ndesaulniers(a)google.com> Reviewed-by: Matthias Kaehlcke <mka(a)chromium.org> Signed-off-by: Masahiro Yamada <yamada.masahiro(a)socionext.com> Signed-off-by: Greg Hackmann <ghackmann(a)google.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- scripts/Kbuild.include | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) --- a/scripts/Kbuild.include +++ b/scripts/Kbuild.include @@ -160,12 +160,13 @@ cc-if-fullversion = $(shell [ $(cc-fullv # cc-ldoption # Usage: ldflags += $(call cc-ldoption, -Wl$(comma)--hash-style=both) cc-ldoption = $(call try-run,\ - $(CC) $(1) -nostdlib -x c /dev/null -o "$$TMP",$(1),$(2)) + $(CC) $(1) $(KBUILD_CPPFLAGS) $(CC_OPTION_CFLAGS) -nostdlib -x c /dev/null -o "$$TMP",$(1),$(2)) # ld-option # Usage: LDFLAGS += $(call ld-option, -X) ld-option = $(call try-run,\ - $(CC) -x c /dev/null -c -o "$$TMPO" ; $(LD) $(1) "$$TMPO" -o "$$TMP",$(1),$(2)) + $(CC) $(KBUILD_CPPFLAGS) $(CC_OPTION_CFLAGS) -x c /dev/null -c -o "$$TMPO"; \ + $(LD) $(LDFLAGS) $(1) "$$TMPO" -o "$$TMP",$(1),$(2)) # ar-option # Usage: KBUILD_ARFLAGS := $(call ar-option,D) Patches currently in stable-queue which might be from ndesaulniers(a)google.com are queue-4.14/kbuild-fix-linker-feature-test-macros-when-cross-compiling-with-clang.patch

7 years, 9 months

1
0
0 0

Patch "usb: gadget: f_hid: fix: Move IN request allocation to set_alt()" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled usb: gadget: f_hid: fix: Move IN request allocation to set_alt() to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: usb-gadget-f_hid-fix-move-in-request-allocation-to-set_alt.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 749494b6bdbbaf0899aa1c62a1ad74cd747bce47 Mon Sep 17 00:00:00 2001 From: Krzysztof Opasiak <kopasiak90(a)gmail.com> Date: Tue, 24 Jan 2017 03:27:24 +0100 Subject: usb: gadget: f_hid: fix: Move IN request allocation to set_alt() From: Krzysztof Opasiak <kopasiak90(a)gmail.com> commit 749494b6bdbbaf0899aa1c62a1ad74cd747bce47 upstream. Since commit: ba1582f22231 ("usb: gadget: f_hid: use alloc_ep_req()") we cannot allocate any requests in bind() as we check if we should align request buffer based on endpoint descriptor which is assigned in set_alt(). Allocating request in bind() function causes a NULL pointer dereference. This commit moves allocation of IN request from bind() to set_alt() to prevent this issue. Fixes: ba1582f22231 ("usb: gadget: f_hid: use alloc_ep_req()") Cc: stable(a)vger.kernel.org Tested-by: David Lechner <david(a)lechnology.com> Signed-off-by: Krzysztof Opasiak <k.opasiak(a)samsung.com> Signed-off-by: Felipe Balbi <felipe.balbi(a)linux.intel.com> Cc: Bin Liu <b-liu(a)ti.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/usb/gadget/function/f_hid.c | 89 +++++++++++++++++++++++++++--------- 1 file changed, 67 insertions(+), 22 deletions(-) --- a/drivers/usb/gadget/function/f_hid.c +++ b/drivers/usb/gadget/function/f_hid.c @@ -284,6 +284,7 @@ static ssize_t f_hidg_write(struct file size_t count, loff_t *offp) { struct f_hidg *hidg = file->private_data; + struct usb_request *req; unsigned long flags; ssize_t status = -ENOMEM; @@ -293,7 +294,7 @@ static ssize_t f_hidg_write(struct file spin_lock_irqsave(&hidg->write_spinlock, flags); #define WRITE_COND (!hidg->write_pending) - +try_again: /* write queue */ while (!WRITE_COND) { spin_unlock_irqrestore(&hidg->write_spinlock, flags); @@ -308,6 +309,7 @@ static ssize_t f_hidg_write(struct file } hidg->write_pending = 1; + req = hidg->req; count = min_t(unsigned, count, hidg->report_length); spin_unlock_irqrestore(&hidg->write_spinlock, flags); @@ -320,24 +322,38 @@ static ssize_t f_hidg_write(struct file goto release_write_pending; } - hidg->req->status = 0; - hidg->req->zero = 0; - hidg->req->length = count; - hidg->req->complete = f_hidg_req_complete; - hidg->req->context = hidg; + spin_lock_irqsave(&hidg->write_spinlock, flags); + + /* we our function has been disabled by host */ + if (!hidg->req) { + free_ep_req(hidg->in_ep, hidg->req); + /* + * TODO + * Should we fail with error here? + */ + goto try_again; + } + + req->status = 0; + req->zero = 0; + req->length = count; + req->complete = f_hidg_req_complete; + req->context = hidg; status = usb_ep_queue(hidg->in_ep, hidg->req, GFP_ATOMIC); if (status < 0) { ERROR(hidg->func.config->cdev, "usb_ep_queue error on int endpoint %zd\n", status); - goto release_write_pending; + goto release_write_pending_unlocked; } else { status = count; } + spin_unlock_irqrestore(&hidg->write_spinlock, flags); return status; release_write_pending: spin_lock_irqsave(&hidg->write_spinlock, flags); +release_write_pending_unlocked: hidg->write_pending = 0; spin_unlock_irqrestore(&hidg->write_spinlock, flags); @@ -541,12 +557,23 @@ static void hidg_disable(struct usb_func kfree(list); } spin_unlock_irqrestore(&hidg->read_spinlock, flags); + + spin_lock_irqsave(&hidg->write_spinlock, flags); + if (!hidg->write_pending) { + free_ep_req(hidg->in_ep, hidg->req); + hidg->write_pending = 1; + } + + hidg->req = NULL; + spin_unlock_irqrestore(&hidg->write_spinlock, flags); } static int hidg_set_alt(struct usb_function *f, unsigned intf, unsigned alt) { struct usb_composite_dev *cdev = f->config->cdev; struct f_hidg *hidg = func_to_hidg(f); + struct usb_request *req_in = NULL; + unsigned long flags; int i, status = 0; VDBG(cdev, "hidg_set_alt intf:%d alt:%d\n", intf, alt); @@ -567,6 +594,12 @@ static int hidg_set_alt(struct usb_funct goto fail; } hidg->in_ep->driver_data = hidg; + + req_in = hidg_alloc_ep_req(hidg->in_ep, hidg->report_length); + if (!req_in) { + status = -ENOMEM; + goto disable_ep_in; + } } @@ -578,12 +611,12 @@ static int hidg_set_alt(struct usb_funct hidg->out_ep); if (status) { ERROR(cdev, "config_ep_by_speed FAILED!\n"); - goto fail; + goto free_req_in; } status = usb_ep_enable(hidg->out_ep); if (status < 0) { ERROR(cdev, "Enable OUT endpoint FAILED!\n"); - goto fail; + goto free_req_in; } hidg->out_ep->driver_data = hidg; @@ -599,17 +632,37 @@ static int hidg_set_alt(struct usb_funct req->context = hidg; status = usb_ep_queue(hidg->out_ep, req, GFP_ATOMIC); - if (status) + if (status) { ERROR(cdev, "%s queue req --> %d\n", hidg->out_ep->name, status); + free_ep_req(hidg->out_ep, req); + } } else { - usb_ep_disable(hidg->out_ep); status = -ENOMEM; - goto fail; + goto disable_out_ep; } } } + if (hidg->in_ep != NULL) { + spin_lock_irqsave(&hidg->write_spinlock, flags); + hidg->req = req_in; + hidg->write_pending = 0; + spin_unlock_irqrestore(&hidg->write_spinlock, flags); + + wake_up(&hidg->write_queue); + } + return 0; +disable_out_ep: + usb_ep_disable(hidg->out_ep); +free_req_in: + if (req_in) + free_ep_req(hidg->in_ep, req_in); + +disable_ep_in: + if (hidg->in_ep) + usb_ep_disable(hidg->in_ep); + fail: return status; } @@ -658,12 +711,6 @@ static int hidg_bind(struct usb_configur goto fail; hidg->out_ep = ep; - /* preallocate request and buffer */ - status = -ENOMEM; - hidg->req = alloc_ep_req(hidg->in_ep, hidg->report_length); - if (!hidg->req) - goto fail; - /* set descriptor dynamic values */ hidg_interface_desc.bInterfaceSubClass = hidg->bInterfaceSubClass; hidg_interface_desc.bInterfaceProtocol = hidg->bInterfaceProtocol; @@ -690,6 +737,8 @@ static int hidg_bind(struct usb_configur goto fail; spin_lock_init(&hidg->write_spinlock); + hidg->write_pending = 1; + hidg->req = NULL; spin_lock_init(&hidg->read_spinlock); init_waitqueue_head(&hidg->write_queue); init_waitqueue_head(&hidg->read_queue); @@ -954,10 +1003,6 @@ static void hidg_unbind(struct usb_confi device_destroy(hidg_class, MKDEV(major, hidg->minor)); cdev_del(&hidg->cdev); - /* disable/free request and end point */ - usb_ep_disable(hidg->in_ep); - free_ep_req(hidg->in_ep, hidg->req); - usb_free_all_descriptors(f); } Patches currently in stable-queue which might be from kopasiak90(a)gmail.com are queue-4.9/usb-gadget-f_hid-fix-move-in-request-allocation-to-set_alt.patch

7 years, 9 months

1
0
0 0

[PATCH] usb: gadget: f_hid: fix: Move IN request allocation to set_alt()

by Bin Liu

From: Krzysztof Opasiak <kopasiak90(a)gmail.com> commit 749494b6bdbbaf0899aa1c62a1ad74cd747bce47 upstream. Since commit: ba1582f22231 ("usb: gadget: f_hid: use alloc_ep_req()") we cannot allocate any requests in bind() as we check if we should align request buffer based on endpoint descriptor which is assigned in set_alt(). Allocating request in bind() function causes a NULL pointer dereference. This commit moves allocation of IN request from bind() to set_alt() to prevent this issue. Fixes: ba1582f22231 ("usb: gadget: f_hid: use alloc_ep_req()") Tested-by: David Lechner <david(a)lechnology.com> Signed-off-by: Krzysztof Opasiak <k.opasiak(a)samsung.com> Signed-off-by: Felipe Balbi <felipe.balbi(a)linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- This backport is for v4.9 only. drivers/usb/gadget/function/f_hid.c | 89 ++++++++++++++++++++++++++++--------- 1 file changed, 67 insertions(+), 22 deletions(-) diff --git a/drivers/usb/gadget/function/f_hid.c b/drivers/usb/gadget/function/f_hid.c index b6d4b484c51a..5815120c0402 100644 --- a/drivers/usb/gadget/function/f_hid.c +++ b/drivers/usb/gadget/function/f_hid.c @@ -284,6 +284,7 @@ static ssize_t f_hidg_write(struct file *file, const char __user *buffer, size_t count, loff_t *offp) { struct f_hidg *hidg = file->private_data; + struct usb_request *req; unsigned long flags; ssize_t status = -ENOMEM; @@ -293,7 +294,7 @@ static ssize_t f_hidg_write(struct file *file, const char __user *buffer, spin_lock_irqsave(&hidg->write_spinlock, flags); #define WRITE_COND (!hidg->write_pending) - +try_again: /* write queue */ while (!WRITE_COND) { spin_unlock_irqrestore(&hidg->write_spinlock, flags); @@ -308,6 +309,7 @@ static ssize_t f_hidg_write(struct file *file, const char __user *buffer, } hidg->write_pending = 1; + req = hidg->req; count = min_t(unsigned, count, hidg->report_length); spin_unlock_irqrestore(&hidg->write_spinlock, flags); @@ -320,24 +322,38 @@ static ssize_t f_hidg_write(struct file *file, const char __user *buffer, goto release_write_pending; } - hidg->req->status = 0; - hidg->req->zero = 0; - hidg->req->length = count; - hidg->req->complete = f_hidg_req_complete; - hidg->req->context = hidg; + spin_lock_irqsave(&hidg->write_spinlock, flags); + + /* we our function has been disabled by host */ + if (!hidg->req) { + free_ep_req(hidg->in_ep, hidg->req); + /* + * TODO + * Should we fail with error here? + */ + goto try_again; + } + + req->status = 0; + req->zero = 0; + req->length = count; + req->complete = f_hidg_req_complete; + req->context = hidg; status = usb_ep_queue(hidg->in_ep, hidg->req, GFP_ATOMIC); if (status < 0) { ERROR(hidg->func.config->cdev, "usb_ep_queue error on int endpoint %zd\n", status); - goto release_write_pending; + goto release_write_pending_unlocked; } else { status = count; } + spin_unlock_irqrestore(&hidg->write_spinlock, flags); return status; release_write_pending: spin_lock_irqsave(&hidg->write_spinlock, flags); +release_write_pending_unlocked: hidg->write_pending = 0; spin_unlock_irqrestore(&hidg->write_spinlock, flags); @@ -541,12 +557,23 @@ static void hidg_disable(struct usb_function *f) kfree(list); } spin_unlock_irqrestore(&hidg->read_spinlock, flags); + + spin_lock_irqsave(&hidg->write_spinlock, flags); + if (!hidg->write_pending) { + free_ep_req(hidg->in_ep, hidg->req); + hidg->write_pending = 1; + } + + hidg->req = NULL; + spin_unlock_irqrestore(&hidg->write_spinlock, flags); } static int hidg_set_alt(struct usb_function *f, unsigned intf, unsigned alt) { struct usb_composite_dev *cdev = f->config->cdev; struct f_hidg *hidg = func_to_hidg(f); + struct usb_request *req_in = NULL; + unsigned long flags; int i, status = 0; VDBG(cdev, "hidg_set_alt intf:%d alt:%d\n", intf, alt); @@ -567,6 +594,12 @@ static int hidg_set_alt(struct usb_function *f, unsigned intf, unsigned alt) goto fail; } hidg->in_ep->driver_data = hidg; + + req_in = hidg_alloc_ep_req(hidg->in_ep, hidg->report_length); + if (!req_in) { + status = -ENOMEM; + goto disable_ep_in; + } } @@ -578,12 +611,12 @@ static int hidg_set_alt(struct usb_function *f, unsigned intf, unsigned alt) hidg->out_ep); if (status) { ERROR(cdev, "config_ep_by_speed FAILED!\n"); - goto fail; + goto free_req_in; } status = usb_ep_enable(hidg->out_ep); if (status < 0) { ERROR(cdev, "Enable OUT endpoint FAILED!\n"); - goto fail; + goto free_req_in; } hidg->out_ep->driver_data = hidg; @@ -599,17 +632,37 @@ static int hidg_set_alt(struct usb_function *f, unsigned intf, unsigned alt) req->context = hidg; status = usb_ep_queue(hidg->out_ep, req, GFP_ATOMIC); - if (status) + if (status) { ERROR(cdev, "%s queue req --> %d\n", hidg->out_ep->name, status); + free_ep_req(hidg->out_ep, req); + } } else { - usb_ep_disable(hidg->out_ep); status = -ENOMEM; - goto fail; + goto disable_out_ep; } } } + if (hidg->in_ep != NULL) { + spin_lock_irqsave(&hidg->write_spinlock, flags); + hidg->req = req_in; + hidg->write_pending = 0; + spin_unlock_irqrestore(&hidg->write_spinlock, flags); + + wake_up(&hidg->write_queue); + } + return 0; +disable_out_ep: + usb_ep_disable(hidg->out_ep); +free_req_in: + if (req_in) + free_ep_req(hidg->in_ep, req_in); + +disable_ep_in: + if (hidg->in_ep) + usb_ep_disable(hidg->in_ep); + fail: return status; } @@ -658,12 +711,6 @@ static int hidg_bind(struct usb_configuration *c, struct usb_function *f) goto fail; hidg->out_ep = ep; - /* preallocate request and buffer */ - status = -ENOMEM; - hidg->req = alloc_ep_req(hidg->in_ep, hidg->report_length); - if (!hidg->req) - goto fail; - /* set descriptor dynamic values */ hidg_interface_desc.bInterfaceSubClass = hidg->bInterfaceSubClass; hidg_interface_desc.bInterfaceProtocol = hidg->bInterfaceProtocol; @@ -690,6 +737,8 @@ static int hidg_bind(struct usb_configuration *c, struct usb_function *f) goto fail; spin_lock_init(&hidg->write_spinlock); + hidg->write_pending = 1; + hidg->req = NULL; spin_lock_init(&hidg->read_spinlock); init_waitqueue_head(&hidg->write_queue); init_waitqueue_head(&hidg->read_queue); @@ -954,10 +1003,6 @@ static void hidg_unbind(struct usb_configuration *c, struct usb_function *f) device_destroy(hidg_class, MKDEV(major, hidg->minor)); cdev_del(&hidg->cdev); - /* disable/free request and end point */ - usb_ep_disable(hidg->in_ep); - free_ep_req(hidg->in_ep, hidg->req); - usb_free_all_descriptors(f); } -- 1.9.1

7 years, 9 months

2
1
0 0

[PATCH 4.4 000/108] 4.4.116-stable review

by Greg Kroah-Hartman

This is the start of the stable review cycle for the 4.4.116 release. There are 108 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know. Responses should be made by Sat Feb 17 15:11:36 UTC 2018. Anything received after that time might be too late. The whole patch series can be found in one patch at: kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.4.116-rc1.gz or in the git tree and branch at: git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-4.4.y and the diffstat can be found below. thanks, greg k-h ------------- Pseudo-Shortlog of commits: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Linux 4.4.116-rc1 Steven Rostedt (VMware) <rostedt(a)goodmis.org> ftrace: Remove incorrect setting of glob search field Eric W. Biederman <ebiederm(a)xmission.com> mn10300/misalignment: Use SIGSEGV SEGV_MAPERR to report a failed user copy Amir Goldstein <amir73il(a)gmail.com> ovl: fix failure to fsync lower dir Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> ACPI: sbshc: remove raw pointer from printk() message Keith Busch <keith.busch(a)intel.com> nvme: Fix managing degraded controllers Nikolay Borisov <nborisov(a)suse.com> btrfs: Handle btrfs_set_extent_delalloc failure in fixup worker Bart Van Assche <bart.vanassche(a)wdc.com> pktcdvd: Fix pkt_setup_dev() error path James Hogan <jhogan(a)kernel.org> EDAC, octeon: Fix an uninitialized variable warning Max Filippov <jcmvbkbc(a)gmail.com> xtensa: fix futex_atomic_cmpxchg_inatomic Mikulas Patocka <mpatocka(a)redhat.com> alpha: fix reboot on Avanti platform Mikulas Patocka <mpatocka(a)redhat.com> alpha: fix crash if pthread_create races with signal delivery Eric W. Biederman <ebiederm(a)xmission.com> signal/sh: Ensure si_signo is initialized in do_divide_error Eric W. Biederman <ebiederm(a)xmission.com> signal/openrisc: Fix do_unaligned_access to send the proper signal Hans de Goede <hdegoede(a)redhat.com> Bluetooth: btusb: Restore QCA Rome suspend/resume fix with a "rewritten" version Kai-Heng Feng <kai.heng.feng(a)canonical.com> Revert "Bluetooth: btusb: fix QCA Rome suspend/resume" Hans de Goede <hdegoede(a)redhat.com> Bluetooth: btsdio: Do not bind to non-removable BCM43341 Hans de Goede <hdegoede(a)redhat.com> HID: quirks: Fix keyboard + touchpad on Toshiba Click Mini not working Rasmus Villemoes <linux(a)rasmusvillemoes.dk> kernel/async.c: revert "async: simplify lowest_in_progress()" Mauro Carvalho Chehab <mchehab(a)osg.samsung.com> media: cxusb, dib0700: ignore XC2028_I2C_FLUSH Mauro Carvalho Chehab <mchehab(a)s-opensource.com> media: ts2020: avoid integer overflows on 32 bit machines Martin Kaiser <martin(a)kaiser.cx> watchdog: imx2_wdt: restore previous timeout after suspend+resume Liran Alon <liran.alon(a)oracle.com> KVM: nVMX: Fix races when sending nested PI while dest enters/leaves L2 Marc Zyngier <marc.zyngier(a)arm.com> arm: KVM: Fix SMCCC handling of unimplemented SMC/HVC calls Horia Geantă <horia.geanta(a)nxp.com> crypto: caam - fix endless loop when DECO acquire fails Daniel Mentz <danielmentz(a)google.com> media: v4l2-compat-ioctl32.c: refactor compat ioctl32 logic Hans Verkuil <hans.verkuil(a)cisco.com> media: v4l2-compat-ioctl32.c: don't copy back the result for certain errors Hans Verkuil <hans.verkuil(a)cisco.com> media: v4l2-compat-ioctl32.c: drop pr_info for unknown buffer type Hans Verkuil <hans.verkuil(a)cisco.com> media: v4l2-compat-ioctl32.c: copy clip list in put_v4l2_window32 Daniel Mentz <danielmentz(a)google.com> media: v4l2-compat-ioctl32: Copy v4l2_window->global_alpha Hans Verkuil <hansverk(a)cisco.com> media: v4l2-compat-ioctl32.c: make ctrl_is_pointer work for subdevs Hans Verkuil <hans.verkuil(a)cisco.com> media: v4l2-compat-ioctl32.c: fix ctrl_is_pointer Hans Verkuil <hans.verkuil(a)cisco.com> media: v4l2-compat-ioctl32.c: copy m.userptr in put_v4l2_plane32 Hans Verkuil <hans.verkuil(a)cisco.com> media: v4l2-compat-ioctl32.c: avoid sizeof(type) Hans Verkuil <hans.verkuil(a)cisco.com> media: v4l2-compat-ioctl32.c: move 'helper' functions to __get/put_v4l2_format32 Hans Verkuil <hans.verkuil(a)cisco.com> media: v4l2-compat-ioctl32.c: fix the indentation Hans Verkuil <hans.verkuil(a)cisco.com> media: v4l2-compat-ioctl32.c: add missing VIDIOC_PREPARE_BUF Ricardo Ribalda Delgado <ricardo.ribalda(a)gmail.com> vb2: V4L2_BUF_FLAG_DONE is set after DQBUF Hans Verkuil <hans.verkuil(a)cisco.com> media: v4l2-ioctl.c: don't copy back the result for -ENOTTY Cong Wang <xiyou.wangcong(a)gmail.com> nsfs: mark dentry with DCACHE_RCUACCESS Eric Biggers <ebiggers(a)google.com> crypto: poly1305 - remove ->setkey() method Eric Biggers <ebiggers(a)google.com> crypto: cryptd - pass through absence of ->setkey() Eric Biggers <ebiggers(a)google.com> crypto: hash - introduce crypto_hash_alg_has_setkey() Mika Westerberg <mika.westerberg(a)linux.intel.com> ahci: Add Intel Cannon Lake PCH-H PCI ID Hans de Goede <hdegoede(a)redhat.com> ahci: Add PCI ids for Intel Bay Trail, Cherry Trail and Apollo Lake AHCI Hans de Goede <hdegoede(a)redhat.com> ahci: Annotate PCI ids for mobile Intel chipsets as such Ivan Vecera <ivecera(a)redhat.com> kernfs: fix regression in kernfs_fop_write caused by wrong type Eric Biggers <ebiggers(a)google.com> NFS: reject request for id_legacy key without auxdata J. Bruce Fields <bfields(a)redhat.com> NFS: commit direct writes even if they fail partially Trond Myklebust <trond.myklebust(a)primarydata.com> NFS: Add a cond_resched() to nfs_commit_release_pages() Scott Mayhew <smayhew(a)redhat.com> nfs/pnfs: fix nfs_direct_req ref leak when i/o falls back to the mds Bradley Bolen <bradleybolen(a)gmail.com> ubi: block: Fix locking for idr_alloc/idr_remove Miquel Raynal <miquel.raynal(a)free-electrons.com> mtd: nand: sunxi: Fix ECC strength choice Miquel Raynal <miquel.raynal(a)free-electrons.com> mtd: nand: Fix nand_do_read_oob() return value Kamal Dasu <kdasu.kdev(a)gmail.com> mtd: nand: brcmnand: Disable prefetch by default Arnd Bergmann <arnd(a)arndb.de> mtd: cfi: convert inline functions to macros Malcolm Priestley <tvboxspy(a)gmail.com> media: dvb-usb-v2: lmedm04: move ts2020 attach to dm04_lme2510_tuner Malcolm Priestley <tvboxspy(a)gmail.com> media: dvb-usb-v2: lmedm04: Improve logic checking of warm start Mohamed Ghannam <simo.ghannam(a)gmail.com> dccp: CVE-2017-8824: use-after-free in DCCP code Steven Rostedt (VMware) <rostedt(a)goodmis.org> sched/rt: Up the root domain ref count when passing it around via IPIs Steven Rostedt (VMware) <rostedt(a)goodmis.org> sched/rt: Use container_of() to get root domain in rto_push_irq_work_func() Petr Cvek <petr.cvek(a)tul.cz> usb: gadget: uvc: Missing files for configfs interface Thomas Gleixner <tglx(a)linutronix.de> posix-timer: Properly check sigevent->sigev_notify Eric W. Biederman <ebiederm(a)xmission.com> netfilter: nf_queue: Make the queue_handler pernet Hugh Dickins <hughd(a)google.com> kaiser: fix compile error without vsyscall Eric Biggers <ebiggers(a)google.com> x86/kaiser: fix build error with KASAN && !FUNCTION_GRAPH_TRACER Yang Shunyong <shunyong.yang(a)hxt-semitech.com> dmaengine: dmatest: fix container_of member in dmatest_callback Aurelien Aptel <aaptel(a)suse.com> CIFS: zero sensitive data when freeing Daniel N Pettersson <danielnp(a)axis.com> cifs: Fix autonegotiate security settings mismatch Matthew Wilcox <mawilcox(a)microsoft.com> cifs: Fix missing put_xid in cifs_file_strict_mmap Michal Suchanek <msuchanek(a)suse.de> powerpc/pseries: include linux/types.h in asm/hvcall.h Borislav Petkov <bp(a)suse.de> x86/microcode: Do the family check first Borislav Petkov <bp(a)suse.de> x86/microcode/AMD: Do not load when running on a hypervisor Robert Baronescu <robert.baronescu(a)nxp.com> crypto: tcrypt - fix S/G table for test_aead_speed() Al Viro <viro(a)zeniv.linux.org.uk> don't put symlink bodies in pagecache into highmem Eric Biggers <ebiggers(a)google.com> KEYS: encrypted: fix buffer overread in valid_master_desc() Jesse Chan <jc(a)linux.com> media: soc_camera: soc_scale_crop: add missing MODULE_DESCRIPTION/AUTHOR/LICENSE Jason Wang <jasowang(a)redhat.com> vhost_net: stop device during reset owner Li RongQing <lirongqing(a)baidu.com> tcp: release sk_frag.page in tcp_disconnect Chunhao Lin <hau(a)realtek.com> r8169: fix RTL8168EP take too long to complete driver initialization. Junxiao Bi <junxiao.bi(a)oracle.com> qlcnic: fix deadlock bug Eric Dumazet <edumazet(a)google.com> net: igmp: add a missing rcu locking section Nikolay Aleksandrov <nikolay(a)cumulusnetworks.com> ip6mr: fix stale iterator Josh Poimboeuf <jpoimboe(a)redhat.com> x86/asm: Fix inline asm call constraints for GCC 4.4 Laurent Pinchart <laurent.pinchart+renesas(a)ideasonboard.com> drm: rcar-du: Fix race condition when disabling planes at CRTC stop Laurent Pinchart <laurent.pinchart+renesas(a)ideasonboard.com> drm: rcar-du: Use the VBK interrupt for vblank events Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> ASoC: rsnd: avoid duplicate free_irq() Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> ASoC: rsnd: don't call free_irq() on Parent SSI Julian Scheel <julian(a)jusst.de> ASoC: simple-card: Fix misleading error message Matthias Hintzmann <matthias.dev(a)gmx.de> net: cdc_ncm: initialize drvflags before usage Shuah Khan <shuahkh(a)osg.samsung.com> usbip: fix 3eee23c3ec14 tcp_socket address still in the status file Shuah Khan <shuahkh(a)osg.samsung.com> usbip: vhci_hcd: clear just the USB_PORT_STAT_POWER bit Jesse Chan <jc(a)linux.com> ASoC: pcm512x: add missing MODULE_DESCRIPTION/AUTHOR/LICENSE Michael Ellerman <mpe(a)ellerman.id.au> powerpc/64s: Allow control of RFI flush via debugfs Michael Ellerman <mpe(a)ellerman.id.au> powerpc/64s: Wire up cpu_show_meltdown() Oliver O'Halloran <oohall(a)gmail.com> powerpc/powernv: Check device-tree for RFI flush settings Michael Neuling <mikey(a)neuling.org> powerpc/pseries: Query hypervisor for RFI flush settings Michael Ellerman <mpe(a)ellerman.id.au> powerpc/64s: Support disabling RFI flush with no_rfi_flush and nopti Michael Ellerman <mpe(a)ellerman.id.au> powerpc/64s: Add support for RFI flush of L1-D cache Nicholas Piggin <npiggin(a)gmail.com> powerpc/64s: Convert slb_miss_common to use RFI_TO_USER/KERNEL Nicholas Piggin <npiggin(a)gmail.com> powerpc/64: Convert the syscall exit path to use RFI_TO_USER/KERNEL Nicholas Piggin <npiggin(a)gmail.com> powerpc/64: Convert fast_exception_return to use RFI_TO_USER/KERNEL Nicholas Piggin <npiggin(a)gmail.com> powerpc/64s: Simple RFI macro conversions Nicholas Piggin <npiggin(a)gmail.com> powerpc/64: Add macros for annotating the destination of rfid/hrfid Michael Neuling <mikey(a)neuling.org> powerpc/pseries: Add H_GET_CPU_CHARACTERISTICS flags & wrapper Alan Modra <amodra(a)gmail.com> powerpc: Simplify module TOC handling Benjamin Herrenschmidt <benh(a)kernel.crashing.org> powerpc: Fix VSX enabling/flushing to also test MSR_FP and MSR_VEC Oliver O'Halloran <oohall(a)gmail.com> powerpc/64: Fix flush_(d|i)cache_range() called from modules Naveen N. Rao <naveen.n.rao(a)linux.vnet.ibm.com> powerpc/bpf/jit: Disable classic BPF JIT on ppc64le ------------- Diffstat: Makefile | 4 +- arch/alpha/kernel/pci_impl.h | 3 +- arch/alpha/kernel/process.c | 3 +- arch/arm/kvm/handle_exit.c | 13 +- arch/mn10300/mm/misalignment.c | 2 +- arch/openrisc/kernel/traps.c | 10 +- arch/powerpc/Kconfig | 3 +- arch/powerpc/include/asm/exception-64e.h | 6 + arch/powerpc/include/asm/exception-64s.h | 55 +- arch/powerpc/include/asm/feature-fixups.h | 15 + arch/powerpc/include/asm/hvcall.h | 18 + arch/powerpc/include/asm/paca.h | 10 + arch/powerpc/include/asm/plpar_wrappers.h | 14 + arch/powerpc/include/asm/ppc_asm.h | 12 + arch/powerpc/include/asm/setup.h | 13 + arch/powerpc/kernel/asm-offsets.c | 4 + arch/powerpc/kernel/entry_64.S | 44 +- arch/powerpc/kernel/exceptions-64s.S | 126 ++- arch/powerpc/kernel/misc_64.S | 32 +- arch/powerpc/kernel/module_64.c | 12 +- arch/powerpc/kernel/process.c | 5 +- arch/powerpc/kernel/setup_64.c | 139 +++ arch/powerpc/kernel/vmlinux.lds.S | 9 + arch/powerpc/kvm/book3s_hv_rmhandlers.S | 7 +- arch/powerpc/kvm/book3s_rmhandlers.S | 7 +- arch/powerpc/kvm/book3s_segment.S | 4 +- arch/powerpc/lib/feature-fixups.c | 42 + arch/powerpc/platforms/powernv/setup.c | 50 + arch/powerpc/platforms/pseries/setup.c | 37 +- arch/sh/kernel/traps_32.c | 3 +- arch/x86/crypto/poly1305_glue.c | 1 - arch/x86/include/asm/asm.h | 4 +- arch/x86/include/asm/vsyscall.h | 2 +- arch/x86/kernel/cpu/microcode/core.c | 47 +- arch/x86/kvm/vmx.c | 5 +- arch/x86/mm/kaiser.c | 2 +- arch/xtensa/include/asm/futex.h | 23 +- crypto/ahash.c | 11 + crypto/cryptd.c | 3 +- crypto/poly1305_generic.c | 17 +- crypto/tcrypt.c | 6 +- drivers/acpi/sbshc.c | 4 +- drivers/ata/ahci.c | 37 +- drivers/block/pktcdvd.c | 4 +- drivers/bluetooth/btsdio.c | 9 + drivers/bluetooth/btusb.c | 19 +- drivers/crypto/caam/ctrl.c | 8 +- drivers/dma/dmatest.c | 2 +- drivers/edac/octeon_edac-lmc.c | 1 + drivers/gpu/drm/rcar-du/rcar_du_crtc.c | 56 +- drivers/gpu/drm/rcar-du/rcar_du_crtc.h | 8 + drivers/hid/hid-core.c | 12 +- drivers/media/dvb-frontends/ts2020.c | 4 +- drivers/media/platform/soc_camera/soc_scale_crop.c | 4 + drivers/media/usb/dvb-usb-v2/lmedm04.c | 39 +- drivers/media/usb/dvb-usb/cxusb.c | 2 + drivers/media/usb/dvb-usb/dib0700_devices.c | 1 + drivers/media/v4l2-core/v4l2-compat-ioctl32.c | 1023 ++++++++++++-------- drivers/media/v4l2-core/v4l2-ioctl.c | 5 +- drivers/media/v4l2-core/videobuf2-v4l2.c | 6 + drivers/mtd/nand/brcmnand/brcmnand.c | 13 +- drivers/mtd/nand/nand_base.c | 5 +- drivers/mtd/nand/sunxi_nand.c | 8 +- drivers/mtd/ubi/block.c | 42 +- .../net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c | 18 +- drivers/net/ethernet/realtek/r8169.c | 4 +- drivers/net/usb/cdc_ncm.c | 6 +- drivers/nvme/host/pci.c | 13 +- drivers/usb/gadget/function/uvc_configfs.c | 16 +- drivers/usb/usbip/vhci_hcd.c | 2 +- drivers/usb/usbip/vhci_sysfs.c | 7 +- drivers/vhost/net.c | 1 + drivers/watchdog/imx2_wdt.c | 20 +- fs/btrfs/inode.c | 10 +- fs/cifs/cifsencrypt.c | 3 +- fs/cifs/connect.c | 6 +- fs/cifs/file.c | 26 +- fs/cifs/misc.c | 14 +- fs/cifs/smb2pdu.c | 3 +- fs/ext4/inode.c | 1 + fs/ext4/namei.c | 1 + fs/ext4/symlink.c | 10 +- fs/f2fs/inode.c | 1 + fs/f2fs/namei.c | 5 +- fs/inode.c | 6 + fs/kernfs/file.c | 2 +- fs/nfs/direct.c | 4 +- fs/nfs/nfs4idmap.c | 6 +- fs/nfs/pnfs.c | 4 +- fs/nfs/write.c | 2 + fs/nsfs.c | 1 + fs/overlayfs/readdir.c | 6 +- include/crypto/internal/hash.h | 2 + include/crypto/poly1305.h | 2 - include/linux/fs.h | 1 + include/linux/mtd/map.h | 130 ++- include/net/netfilter/nf_queue.h | 4 +- include/net/netns/netfilter.h | 2 + kernel/async.c | 20 +- kernel/sched/core.c | 13 + kernel/sched/rt.c | 24 +- kernel/sched/sched.h | 2 + kernel/time/posix-timers.c | 34 +- kernel/trace/ftrace.c | 1 - net/dccp/proto.c | 5 + net/ipv4/igmp.c | 4 + net/ipv4/tcp.c | 6 + net/ipv6/ip6mr.c | 1 + net/netfilter/nf_queue.c | 17 +- net/netfilter/nfnetlink_queue.c | 18 +- scripts/mod/modpost.c | 3 +- security/keys/encrypted-keys/encrypted.c | 31 +- sound/soc/codecs/pcm512x-spi.c | 4 + sound/soc/generic/simple-card.c | 8 +- sound/soc/sh/rcar/rsnd.h | 2 + sound/soc/sh/rcar/ssi.c | 5 + 116 files changed, 1821 insertions(+), 851 deletions(-)

7 years, 9 months

9
124
0 0

Patch "RDMA/ucma: Fix access to non-initialized CM_ID object" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled RDMA/ucma: Fix access to non-initialized CM_ID object to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: rdma-ucma-fix-access-to-non-initialized-cm_id-object.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 7688f2c3bbf55e52388e37ac5d63ca471a7712e1 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky <leonro(a)mellanox.com> Date: Tue, 13 Mar 2018 11:43:23 +0200 Subject: RDMA/ucma: Fix access to non-initialized CM_ID object From: Leon Romanovsky <leonro(a)mellanox.com> commit 7688f2c3bbf55e52388e37ac5d63ca471a7712e1 upstream. The attempt to join multicast group without ensuring that CMA device exists will lead to the following crash reported by syzkaller. [ 64.076794] BUG: KASAN: null-ptr-deref in rdma_join_multicast+0x26e/0x12c0 [ 64.076797] Read of size 8 at addr 00000000000000b0 by task join/691 [ 64.076797] [ 64.076800] CPU: 1 PID: 691 Comm: join Not tainted 4.16.0-rc1-00219-gb97853b65b93 #23 [ 64.076802] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.0-0-g63451fca13-prebuilt.qemu-proj4 [ 64.076803] Call Trace: [ 64.076809] dump_stack+0x5c/0x77 [ 64.076817] kasan_report+0x163/0x380 [ 64.085859] ? rdma_join_multicast+0x26e/0x12c0 [ 64.086634] rdma_join_multicast+0x26e/0x12c0 [ 64.087370] ? rdma_disconnect+0xf0/0xf0 [ 64.088579] ? __radix_tree_replace+0xc3/0x110 [ 64.089132] ? node_tag_clear+0x81/0xb0 [ 64.089606] ? idr_alloc_u32+0x12e/0x1a0 [ 64.090517] ? __fprop_inc_percpu_max+0x150/0x150 [ 64.091768] ? tracing_record_taskinfo+0x10/0xc0 [ 64.092340] ? idr_alloc+0x76/0xc0 [ 64.092951] ? idr_alloc_u32+0x1a0/0x1a0 [ 64.093632] ? ucma_process_join+0x23d/0x460 [ 64.094510] ucma_process_join+0x23d/0x460 [ 64.095199] ? ucma_migrate_id+0x440/0x440 [ 64.095696] ? futex_wake+0x10b/0x2a0 [ 64.096159] ucma_join_multicast+0x88/0xe0 [ 64.096660] ? ucma_process_join+0x460/0x460 [ 64.097540] ? _copy_from_user+0x5e/0x90 [ 64.098017] ucma_write+0x174/0x1f0 [ 64.098640] ? ucma_resolve_route+0xf0/0xf0 [ 64.099343] ? rb_erase_cached+0x6c7/0x7f0 [ 64.099839] __vfs_write+0xc4/0x350 [ 64.100622] ? perf_syscall_enter+0xe4/0x5f0 [ 64.101335] ? kernel_read+0xa0/0xa0 [ 64.103525] ? perf_sched_cb_inc+0xc0/0xc0 [ 64.105510] ? syscall_exit_register+0x2a0/0x2a0 [ 64.107359] ? __switch_to+0x351/0x640 [ 64.109285] ? fsnotify+0x899/0x8f0 [ 64.111610] ? fsnotify_unmount_inodes+0x170/0x170 [ 64.113876] ? __fsnotify_update_child_dentry_flags+0x30/0x30 [ 64.115813] ? ring_buffer_record_is_on+0xd/0x20 [ 64.117824] ? __fget+0xa8/0xf0 [ 64.119869] vfs_write+0xf7/0x280 [ 64.122001] SyS_write+0xa1/0x120 [ 64.124213] ? SyS_read+0x120/0x120 [ 64.126644] ? SyS_read+0x120/0x120 [ 64.128563] do_syscall_64+0xeb/0x250 [ 64.130732] entry_SYSCALL_64_after_hwframe+0x21/0x86 [ 64.132984] RIP: 0033:0x7f5c994ade99 [ 64.135699] RSP: 002b:00007f5c99b97d98 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 [ 64.138740] RAX: ffffffffffffffda RBX: 00000000200001e4 RCX: 00007f5c994ade99 [ 64.141056] RDX: 00000000000000a0 RSI: 00000000200001c0 RDI: 0000000000000015 [ 64.143536] RBP: 00007f5c99b97ec0 R08: 0000000000000000 R09: 0000000000000000 [ 64.146017] R10: 0000000000000000 R11: 0000000000000246 R12: 00007f5c99b97fc0 [ 64.148608] R13: 0000000000000000 R14: 00007fff660e1c40 R15: 00007f5c99b989c0 [ 64.151060] [ 64.153703] Disabling lock debugging due to kernel taint [ 64.156032] BUG: unable to handle kernel NULL pointer dereference at 00000000000000b0 [ 64.159066] IP: rdma_join_multicast+0x26e/0x12c0 [ 64.161451] PGD 80000001d0298067 P4D 80000001d0298067 PUD 1dea39067 PMD 0 [ 64.164442] Oops: 0000 [#1] SMP KASAN PTI [ 64.166817] CPU: 1 PID: 691 Comm: join Tainted: G B 4.16.0-rc1-00219-gb97853b65b93 #23 [ 64.170004] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.0-0-g63451fca13-prebuilt.qemu-proj4 [ 64.174985] RIP: 0010:rdma_join_multicast+0x26e/0x12c0 [ 64.177246] RSP: 0018:ffff8801c8207860 EFLAGS: 00010282 [ 64.179901] RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffffff94789522 [ 64.183344] RDX: 1ffffffff2d50fa5 RSI: 0000000000000297 RDI: 0000000000000297 [ 64.186237] RBP: ffff8801c8207a50 R08: 0000000000000000 R09: ffffed0039040ea7 [ 64.189328] R10: 0000000000000001 R11: ffffed0039040ea6 R12: 0000000000000000 [ 64.192634] R13: 0000000000000000 R14: ffff8801e2022800 R15: ffff8801d4ac2400 [ 64.196105] FS: 00007f5c99b98700(0000) GS:ffff8801e5d00000(0000) knlGS:0000000000000000 [ 64.199211] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 64.202046] CR2: 00000000000000b0 CR3: 00000001d1c48004 CR4: 00000000003606a0 [ 64.205032] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 64.208221] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 64.211554] Call Trace: [ 64.213464] ? rdma_disconnect+0xf0/0xf0 [ 64.216124] ? __radix_tree_replace+0xc3/0x110 [ 64.219337] ? node_tag_clear+0x81/0xb0 [ 64.222140] ? idr_alloc_u32+0x12e/0x1a0 [ 64.224422] ? __fprop_inc_percpu_max+0x150/0x150 [ 64.226588] ? tracing_record_taskinfo+0x10/0xc0 [ 64.229763] ? idr_alloc+0x76/0xc0 [ 64.232186] ? idr_alloc_u32+0x1a0/0x1a0 [ 64.234505] ? ucma_process_join+0x23d/0x460 [ 64.237024] ucma_process_join+0x23d/0x460 [ 64.240076] ? ucma_migrate_id+0x440/0x440 [ 64.243284] ? futex_wake+0x10b/0x2a0 [ 64.245302] ucma_join_multicast+0x88/0xe0 [ 64.247783] ? ucma_process_join+0x460/0x460 [ 64.250841] ? _copy_from_user+0x5e/0x90 [ 64.253878] ucma_write+0x174/0x1f0 [ 64.257008] ? ucma_resolve_route+0xf0/0xf0 [ 64.259877] ? rb_erase_cached+0x6c7/0x7f0 [ 64.262746] __vfs_write+0xc4/0x350 [ 64.265537] ? perf_syscall_enter+0xe4/0x5f0 [ 64.267792] ? kernel_read+0xa0/0xa0 [ 64.270358] ? perf_sched_cb_inc+0xc0/0xc0 [ 64.272575] ? syscall_exit_register+0x2a0/0x2a0 [ 64.275367] ? __switch_to+0x351/0x640 [ 64.277700] ? fsnotify+0x899/0x8f0 [ 64.280530] ? fsnotify_unmount_inodes+0x170/0x170 [ 64.283156] ? __fsnotify_update_child_dentry_flags+0x30/0x30 [ 64.286182] ? ring_buffer_record_is_on+0xd/0x20 [ 64.288749] ? __fget+0xa8/0xf0 [ 64.291136] vfs_write+0xf7/0x280 [ 64.292972] SyS_write+0xa1/0x120 [ 64.294965] ? SyS_read+0x120/0x120 [ 64.297474] ? SyS_read+0x120/0x120 [ 64.299751] do_syscall_64+0xeb/0x250 [ 64.301826] entry_SYSCALL_64_after_hwframe+0x21/0x86 [ 64.304352] RIP: 0033:0x7f5c994ade99 [ 64.306711] RSP: 002b:00007f5c99b97d98 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 [ 64.309577] RAX: ffffffffffffffda RBX: 00000000200001e4 RCX: 00007f5c994ade99 [ 64.312334] RDX: 00000000000000a0 RSI: 00000000200001c0 RDI: 0000000000000015 [ 64.315783] RBP: 00007f5c99b97ec0 R08: 0000000000000000 R09: 0000000000000000 [ 64.318365] R10: 0000000000000000 R11: 0000000000000246 R12: 00007f5c99b97fc0 [ 64.320980] R13: 0000000000000000 R14: 00007fff660e1c40 R15: 00007f5c99b989c0 [ 64.323515] Code: e8 e8 79 08 ff 4c 89 ff 45 0f b6 a7 b8 01 00 00 e8 68 7c 08 ff 49 8b 1f 4d 89 e5 49 c1 e4 04 48 8 [ 64.330753] RIP: rdma_join_multicast+0x26e/0x12c0 RSP: ffff8801c8207860 [ 64.332979] CR2: 00000000000000b0 [ 64.335550] ---[ end trace 0c00c17a408849c1 ]--- Reported-by: <syzbot+e6aba77967bd72cbc9d6(a)syzkaller.appspotmail.com> Fixes: c8f6a362bf3e ("RDMA/cma: Add multicast communication support") Signed-off-by: Leon Romanovsky <leonro(a)mellanox.com> Reviewed-by: Sean Hefty <sean.hefty(a)intel.com> Signed-off-by: Doug Ledford <dledford(a)redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/infiniband/core/cma.c | 3 +++ 1 file changed, 3 insertions(+) --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -4039,6 +4039,9 @@ int rdma_join_multicast(struct rdma_cm_i struct cma_multicast *mc; int ret; + if (!id->device) + return -EINVAL; + id_priv = container_of(id, struct rdma_id_private, id); if (!cma_comp(id_priv, RDMA_CM_ADDR_BOUND) && !cma_comp(id_priv, RDMA_CM_ADDR_RESOLVED)) Patches currently in stable-queue which might be from leonro(a)mellanox.com are queue-4.9/rdma-ucma-don-t-allow-join-attempts-for-unsupported-af-family.patch queue-4.9/rdma-ucma-fix-access-to-non-initialized-cm_id-object.patch queue-4.9/ib-rdmavt-restore-irqs-on-error-path-in-rvt_create_ah.patch

7 years, 9 months

1
0
0 0

Patch "RDMA/ucma: Don't allow join attempts for unsupported AF family" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled RDMA/ucma: Don't allow join attempts for unsupported AF family to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: rdma-ucma-don-t-allow-join-attempts-for-unsupported-af-family.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 0c81ffc60d5280991773d17e84bda605387148b1 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky <leonro(a)mellanox.com> Date: Tue, 13 Mar 2018 18:37:27 +0200 Subject: RDMA/ucma: Don't allow join attempts for unsupported AF family From: Leon Romanovsky <leonro(a)mellanox.com> commit 0c81ffc60d5280991773d17e84bda605387148b1 upstream. Users can provide garbage while calling to ucma_join_ip_multicast(), it will indirectly cause to rdma_addr_size() return 0, making the call to ucma_process_join(), which had the right checks, but it is better to check the input as early as possible. The following crash from syzkaller revealed it. kernel BUG at lib/string.c:1052! invalid opcode: 0000 [#1] SMP KASAN Dumping ftrace buffer: (ftrace buffer empty) Modules linked in: CPU: 0 PID: 4113 Comm: syz-executor0 Not tainted 4.16.0-rc5+ #261 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:fortify_panic+0x13/0x20 lib/string.c:1051 RSP: 0018:ffff8801ca81f8f0 EFLAGS: 00010286 RAX: 0000000000000022 RBX: 1ffff10039503f23 RCX: 0000000000000000 RDX: 0000000000000022 RSI: 1ffff10039503ed3 RDI: ffffed0039503f12 RBP: ffff8801ca81f8f0 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000006 R11: 0000000000000000 R12: ffff8801ca81f998 R13: ffff8801ca81f938 R14: ffff8801ca81fa58 R15: 000000000000fa00 FS: 0000000000000000(0000) GS:ffff8801db200000(0063) knlGS:000000000a12a900 CS: 0010 DS: 002b ES: 002b CR0: 0000000080050033 CR2: 0000000008138024 CR3: 00000001cbb58004 CR4: 00000000001606f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: memcpy include/linux/string.h:344 [inline] ucma_join_ip_multicast+0x36b/0x3b0 drivers/infiniband/core/ucma.c:1421 ucma_write+0x2d6/0x3d0 drivers/infiniband/core/ucma.c:1633 __vfs_write+0xef/0x970 fs/read_write.c:480 vfs_write+0x189/0x510 fs/read_write.c:544 SYSC_write fs/read_write.c:589 [inline] SyS_write+0xef/0x220 fs/read_write.c:581 do_syscall_32_irqs_on arch/x86/entry/common.c:330 [inline] do_fast_syscall_32+0x3ec/0xf9f arch/x86/entry/common.c:392 entry_SYSENTER_compat+0x70/0x7f arch/x86/entry/entry_64_compat.S:139 RIP: 0023:0xf7f9ec99 RSP: 002b:00000000ff8172cc EFLAGS: 00000282 ORIG_RAX: 0000000000000004 RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 0000000020000100 RDX: 0000000000000063 RSI: 0000000000000000 RDI: 0000000000000000 RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 Code: 08 5b 41 5c 41 5d 41 5e 41 5f 5d c3 0f 0b 48 89 df e8 42 2c e3 fb eb de 55 48 89 fe 48 c7 c7 80 75 98 86 48 89 e5 e8 85 95 94 fb <0f> 0b 90 90 90 90 90 90 90 90 90 90 90 55 48 89 e5 41 57 41 56 RIP: fortify_panic+0x13/0x20 lib/string.c:1051 RSP: ffff8801ca81f8f0 Fixes: 5bc2b7b397b0 ("RDMA/ucma: Allow user space to specify AF_IB when joining multicast") Reported-by: <syzbot+2287ac532caa81900a4e(a)syzkaller.appspotmail.com> Signed-off-by: Leon Romanovsky <leonro(a)mellanox.com> Reviewed-by: Sean Hefty <sean.hefty(a)intel.com> Signed-off-by: Doug Ledford <dledford(a)redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/infiniband/core/ucma.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) --- a/drivers/infiniband/core/ucma.c +++ b/drivers/infiniband/core/ucma.c @@ -1330,7 +1330,7 @@ static ssize_t ucma_process_join(struct return -ENOSPC; addr = (struct sockaddr *) &cmd->addr; - if (!cmd->addr_size || (cmd->addr_size != rdma_addr_size(addr))) + if (cmd->addr_size != rdma_addr_size(addr)) return -EINVAL; if (cmd->join_flags == RDMA_MC_JOIN_FLAG_FULLMEMBER) @@ -1398,6 +1398,9 @@ static ssize_t ucma_join_ip_multicast(st join_cmd.uid = cmd.uid; join_cmd.id = cmd.id; join_cmd.addr_size = rdma_addr_size((struct sockaddr *) &cmd.addr); + if (!join_cmd.addr_size) + return -EINVAL; + join_cmd.join_flags = RDMA_MC_JOIN_FLAG_FULLMEMBER; memcpy(&join_cmd.addr, &cmd.addr, join_cmd.addr_size); @@ -1413,6 +1416,9 @@ static ssize_t ucma_join_multicast(struc if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; + if (!rdma_addr_size((struct sockaddr *)&cmd.addr)) + return -EINVAL; + return ucma_process_join(file, &cmd, out_len); } Patches currently in stable-queue which might be from leonro(a)mellanox.com are queue-4.9/rdma-ucma-don-t-allow-join-attempts-for-unsupported-af-family.patch queue-4.9/rdma-ucma-fix-access-to-non-initialized-cm_id-object.patch queue-4.9/ib-rdmavt-restore-irqs-on-error-path-in-rvt_create_ah.patch

7 years, 9 months

1
0
0 0

Patch "IB/mlx5: Fix out-of-bounds read in create_raw_packet_qp_rq" has been added to the 4.9-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled IB/mlx5: Fix out-of-bounds read in create_raw_packet_qp_rq to the 4.9-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: ib-mlx5-fix-out-of-bounds-read-in-create_raw_packet_qp_rq.patch and it can be found in the queue-4.9 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 2c292dbb398ee46fc1343daf6c3cf9715a75688e Mon Sep 17 00:00:00 2001 From: Boris Pismenny <borisp(a)mellanox.com> Date: Thu, 8 Mar 2018 15:51:40 +0200 Subject: IB/mlx5: Fix out-of-bounds read in create_raw_packet_qp_rq From: Boris Pismenny <borisp(a)mellanox.com> commit 2c292dbb398ee46fc1343daf6c3cf9715a75688e upstream. Add a check for the length of the qpin structure to prevent out-of-bounds reads BUG: KASAN: slab-out-of-bounds in create_raw_packet_qp+0x114c/0x15e2 Read of size 8192 at addr ffff880066b99290 by task syz-executor3/549 CPU: 3 PID: 549 Comm: syz-executor3 Not tainted 4.15.0-rc2+ #27 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5-0-ge51488c-20140602_164612-nilsson.home.kraxel.org 04/01/2014 Call Trace: dump_stack+0x8d/0xd4 print_address_description+0x73/0x290 kasan_report+0x25c/0x370 ? create_raw_packet_qp+0x114c/0x15e2 memcpy+0x1f/0x50 create_raw_packet_qp+0x114c/0x15e2 ? create_raw_packet_qp_tis.isra.28+0x13d/0x13d ? lock_acquire+0x370/0x370 create_qp_common+0x2245/0x3b50 ? destroy_qp_user.isra.47+0x100/0x100 ? kasan_kmalloc+0x13d/0x170 ? sched_clock_cpu+0x18/0x180 ? fs_reclaim_acquire.part.15+0x5/0x30 ? __lock_acquire+0xa11/0x1da0 ? sched_clock_cpu+0x18/0x180 ? kmem_cache_alloc_trace+0x17e/0x310 ? mlx5_ib_create_qp+0x30e/0x17b0 mlx5_ib_create_qp+0x33d/0x17b0 ? sched_clock_cpu+0x18/0x180 ? create_qp_common+0x3b50/0x3b50 ? lock_acquire+0x370/0x370 ? __radix_tree_lookup+0x180/0x220 ? uverbs_try_lock_object+0x68/0xc0 ? rdma_lookup_get_uobject+0x114/0x240 create_qp.isra.5+0xce4/0x1e20 ? ib_uverbs_ex_create_cq_cb+0xa0/0xa0 ? copy_ah_attr_from_uverbs.isra.2+0xa00/0xa00 ? ib_uverbs_cq_event_handler+0x160/0x160 ? __might_fault+0x17c/0x1c0 ib_uverbs_create_qp+0x21b/0x2a0 ? ib_uverbs_destroy_cq+0x2e0/0x2e0 ib_uverbs_write+0x55a/0xad0 ? ib_uverbs_destroy_cq+0x2e0/0x2e0 ? ib_uverbs_destroy_cq+0x2e0/0x2e0 ? ib_uverbs_open+0x760/0x760 ? futex_wake+0x147/0x410 ? check_prev_add+0x1680/0x1680 ? do_futex+0x3d3/0xa60 ? sched_clock_cpu+0x18/0x180 __vfs_write+0xf7/0x5c0 ? ib_uverbs_open+0x760/0x760 ? kernel_read+0x110/0x110 ? lock_acquire+0x370/0x370 ? __fget+0x264/0x3b0 vfs_write+0x18a/0x460 SyS_write+0xc7/0x1a0 ? SyS_read+0x1a0/0x1a0 ? trace_hardirqs_on_thunk+0x1a/0x1c entry_SYSCALL_64_fastpath+0x18/0x85 RIP: 0033:0x4477b9 RSP: 002b:00007f1822cadc18 EFLAGS: 00000292 ORIG_RAX: 0000000000000001 RAX: ffffffffffffffda RBX: 0000000000000005 RCX: 00000000004477b9 RDX: 0000000000000070 RSI: 000000002000a000 RDI: 0000000000000005 RBP: 0000000000708000 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000292 R12: 00000000ffffffff R13: 0000000000005d70 R14: 00000000006e6e30 R15: 0000000020010ff0 Allocated by task 549: __kmalloc+0x15e/0x340 kvmalloc_node+0xa1/0xd0 create_user_qp.isra.46+0xd42/0x1610 create_qp_common+0x2e63/0x3b50 mlx5_ib_create_qp+0x33d/0x17b0 create_qp.isra.5+0xce4/0x1e20 ib_uverbs_create_qp+0x21b/0x2a0 ib_uverbs_write+0x55a/0xad0 __vfs_write+0xf7/0x5c0 vfs_write+0x18a/0x460 SyS_write+0xc7/0x1a0 entry_SYSCALL_64_fastpath+0x18/0x85 Freed by task 368: kfree+0xeb/0x2f0 kernfs_fop_release+0x140/0x180 __fput+0x266/0x700 task_work_run+0x104/0x180 exit_to_usermode_loop+0xf7/0x110 syscall_return_slowpath+0x298/0x370 entry_SYSCALL_64_fastpath+0x83/0x85 The buggy address belongs to the object at ffff880066b99180 which belongs to the cache kmalloc-512 of size 512 The buggy address is located 272 bytes inside of 512-byte region [ffff880066b99180, ffff880066b99380) The buggy address belongs to the page: page:000000006040eedd count:1 mapcount:0 mapping: (null) index:0x0 compound_mapcount: 0 flags: 0x4000000000008100(slab|head) raw: 4000000000008100 0000000000000000 0000000000000000 0000000180190019 raw: ffffea00019a7500 0000000b0000000b ffff88006c403080 0000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffff880066b99180: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ffff880066b99200: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 >ffff880066b99280: 00 00 fc fc fc fc fc fc fc fc fc fc fc fc fc fc ^ ffff880066b99300: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ffff880066b99380: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc Cc: syzkaller <syzkaller(a)googlegroups.com> Fixes: 0fb2ed66a14c ("IB/mlx5: Add create and destroy functionality for Raw Packet QP") Signed-off-by: Boris Pismenny <borisp(a)mellanox.com> Signed-off-by: Leon Romanovsky <leon(a)kernel.org> Signed-off-by: Doug Ledford <dledford(a)redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/infiniband/hw/mlx5/qp.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -1130,7 +1130,7 @@ static void destroy_raw_packet_qp_sq(str ib_umem_release(sq->ubuffer.umem); } -static int get_rq_pas_size(void *qpc) +static size_t get_rq_pas_size(void *qpc) { u32 log_page_size = MLX5_GET(qpc, qpc, log_page_size) + 12; u32 log_rq_stride = MLX5_GET(qpc, qpc, log_rq_stride); @@ -1146,7 +1146,8 @@ static int get_rq_pas_size(void *qpc) } static int create_raw_packet_qp_rq(struct mlx5_ib_dev *dev, - struct mlx5_ib_rq *rq, void *qpin) + struct mlx5_ib_rq *rq, void *qpin, + size_t qpinlen) { struct mlx5_ib_qp *mqp = rq->base.container_mibqp; __be64 *pas; @@ -1155,9 +1156,12 @@ static int create_raw_packet_qp_rq(struc void *rqc; void *wq; void *qpc = MLX5_ADDR_OF(create_qp_in, qpin, qpc); - int inlen; + size_t rq_pas_size = get_rq_pas_size(qpc); + size_t inlen; int err; - u32 rq_pas_size = get_rq_pas_size(qpc); + + if (qpinlen < rq_pas_size + MLX5_BYTE_OFF(create_qp_in, pas)) + return -EINVAL; inlen = MLX5_ST_SZ_BYTES(create_rq_in) + rq_pas_size; in = mlx5_vzalloc(inlen); @@ -1235,7 +1239,7 @@ static void destroy_raw_packet_qp_tir(st } static int create_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp, - u32 *in, + u32 *in, size_t inlen, struct ib_pd *pd) { struct mlx5_ib_raw_packet_qp *raw_packet_qp = &qp->raw_packet_qp; @@ -1262,7 +1266,7 @@ static int create_raw_packet_qp(struct m if (qp->rq.wqe_cnt) { rq->base.container_mibqp = qp; - err = create_raw_packet_qp_rq(dev, rq, in); + err = create_raw_packet_qp_rq(dev, rq, in, inlen); if (err) goto err_destroy_sq; @@ -1753,10 +1757,15 @@ static int create_qp_common(struct mlx5_ qp->flags |= MLX5_IB_QP_LSO; } + if (inlen < 0) { + err = -EINVAL; + goto err; + } + if (init_attr->qp_type == IB_QPT_RAW_PACKET) { qp->raw_packet_qp.sq.ubuffer.buf_addr = ucmd.sq_buf_addr; raw_packet_qp_copy_info(qp, &qp->raw_packet_qp); - err = create_raw_packet_qp(dev, qp, in, pd); + err = create_raw_packet_qp(dev, qp, in, inlen, pd); } else { err = mlx5_core_create_qp(dev->mdev, &base->mqp, in, inlen); } Patches currently in stable-queue which might be from borisp(a)mellanox.com are queue-4.9/ib-mlx5-fix-integer-overflows-in-mlx5_ib_create_srq.patch queue-4.9/ib-mlx5-fix-out-of-bounds-read-in-create_raw_packet_qp_rq.patch

7 years, 9 months

1
0
0 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror