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

[PATCH v3 2/5] scsi: fnic: Fix crash in fnic_wq_cmpl_handler when FDMI times out

by Karan Tilak Kumar

When both the RHBA and RPA FDMI requests time out, fnic reuses a frame to send ABTS for each of them. On send completion, this causes an attempt to free the same frame twice that leads to a crash. Fix crash by allocating separate frames for RHBA and RPA, and modify ABTS logic accordingly. Tested by checking MDS for FDMI information. Tested by using instrumented driver to: Drop PLOGI response Drop RHBA response Drop RPA response Drop RHBA and RPA response Drop PLOGI response + ABTS response Drop RHBA response + ABTS response Drop RPA response + ABTS response Drop RHBA and RPA response + ABTS response for both of them Fixes: 09c1e6ab4ab2 ("scsi: fnic: Add and integrate support for FDMI") Reviewed-by: Sesidhar Baddela <sebaddel(a)cisco.com> Reviewed-by: Arulprabhu Ponnusamy <arulponn(a)cisco.com> Reviewed-by: Gian Carlo Boffa <gcboffa(a)cisco.com> Tested-by: Arun Easi <aeasi(a)cisco.com> Co-developed-by: Arun Easi <aeasi(a)cisco.com> Signed-off-by: Arun Easi <aeasi(a)cisco.com> Tested-by: Karan Tilak Kumar <kartilak(a)cisco.com> Cc: <stable(a)vger.kernel.org> # 6.14.x Please see patch description Signed-off-by: Karan Tilak Kumar <kartilak(a)cisco.com> --- drivers/scsi/fnic/fdls_disc.c | 113 +++++++++++++++++++++++++--------- drivers/scsi/fnic/fnic_fdls.h | 1 + 2 files changed, 86 insertions(+), 28 deletions(-) diff --git a/drivers/scsi/fnic/fdls_disc.c b/drivers/scsi/fnic/fdls_disc.c index c2b6f4eb338e..0ee1b74967b9 100644 --- a/drivers/scsi/fnic/fdls_disc.c +++ b/drivers/scsi/fnic/fdls_disc.c @@ -763,47 +763,69 @@ static void fdls_send_fabric_abts(struct fnic_iport_s *iport) iport->fabric.timer_pending = 1; } -static void fdls_send_fdmi_abts(struct fnic_iport_s *iport) +static uint8_t *fdls_alloc_init_fdmi_abts_frame(struct fnic_iport_s *iport, + uint16_t oxid) { - uint8_t *frame; + struct fc_frame_header *pfdmi_abts; uint8_t d_id[3]; + uint8_t *frame; struct fnic *fnic = iport->fnic; - struct fc_frame_header *pfabric_abts; - unsigned long fdmi_tov; - uint16_t oxid; - uint16_t frame_size = FNIC_ETH_FCOE_HDRS_OFFSET + - sizeof(struct fc_frame_header); frame = fdls_alloc_frame(iport); if (frame == NULL) { FNIC_FCS_DBG(KERN_ERR, fnic->host, fnic->fnic_num, "Failed to allocate frame to send FDMI ABTS"); - return; + return NULL; } - pfabric_abts = (struct fc_frame_header *) (frame + FNIC_ETH_FCOE_HDRS_OFFSET); + pfdmi_abts = (struct fc_frame_header *) (frame + FNIC_ETH_FCOE_HDRS_OFFSET); fdls_init_fabric_abts_frame(frame, iport); hton24(d_id, FC_FID_MGMT_SERV); - FNIC_STD_SET_D_ID(*pfabric_abts, d_id); + FNIC_STD_SET_D_ID(*pfdmi_abts, d_id); + FNIC_STD_SET_OX_ID(*pfdmi_abts, oxid); + + return frame; +} + +static void fdls_send_fdmi_abts(struct fnic_iport_s *iport) +{ + uint8_t *frame; + unsigned long fdmi_tov; + uint16_t frame_size = FNIC_ETH_FCOE_HDRS_OFFSET + + sizeof(struct fc_frame_header); if (iport->fabric.fdmi_pending & FDLS_FDMI_PLOGI_PENDING) { - oxid = iport->active_oxid_fdmi_plogi; - FNIC_STD_SET_OX_ID(*pfabric_abts, oxid); + frame = fdls_alloc_init_fdmi_abts_frame(iport, + iport->active_oxid_fdmi_plogi); + if (frame == NULL) + return; + fnic_send_fcoe_frame(iport, frame, frame_size); } else { if (iport->fabric.fdmi_pending & FDLS_FDMI_REG_HBA_PENDING) { - oxid = iport->active_oxid_fdmi_rhba; - FNIC_STD_SET_OX_ID(*pfabric_abts, oxid); + frame = fdls_alloc_init_fdmi_abts_frame(iport, + iport->active_oxid_fdmi_rhba); + if (frame == NULL) + return; + fnic_send_fcoe_frame(iport, frame, frame_size); } if (iport->fabric.fdmi_pending & FDLS_FDMI_RPA_PENDING) { - oxid = iport->active_oxid_fdmi_rpa; - FNIC_STD_SET_OX_ID(*pfabric_abts, oxid); + frame = fdls_alloc_init_fdmi_abts_frame(iport, + iport->active_oxid_fdmi_rpa); + if (frame == NULL) { + if (iport->fabric.fdmi_pending & FDLS_FDMI_REG_HBA_PENDING) + goto arm_timer; + else + return; + } + fnic_send_fcoe_frame(iport, frame, frame_size); } } +arm_timer: fdmi_tov = jiffies + msecs_to_jiffies(2 * iport->e_d_tov); mod_timer(&iport->fabric.fdmi_timer, round_jiffies(fdmi_tov)); iport->fabric.fdmi_pending |= FDLS_FDMI_ABORT_PENDING; @@ -2244,6 +2266,21 @@ void fdls_fabric_timer_callback(struct timer_list *t) spin_unlock_irqrestore(&fnic->fnic_lock, flags); } +void fdls_fdmi_retry_plogi(struct fnic_iport_s *iport) +{ + struct fnic *fnic = iport->fnic; + + iport->fabric.fdmi_pending = 0; + /* If max retries not exhausted, start over from fdmi plogi */ + if (iport->fabric.fdmi_retry < FDLS_FDMI_MAX_RETRY) { + iport->fabric.fdmi_retry++; + FNIC_FCS_DBG(KERN_INFO, fnic->host, fnic->fnic_num, + "Retry FDMI PLOGI. FDMI retry: %d", + iport->fabric.fdmi_retry); + fdls_send_fdmi_plogi(iport); + } +} + void fdls_fdmi_timer_callback(struct timer_list *t) { struct fnic_fdls_fabric_s *fabric = from_timer(fabric, t, fdmi_timer); @@ -2289,14 +2326,7 @@ void fdls_fdmi_timer_callback(struct timer_list *t) FNIC_FCS_DBG(KERN_INFO, fnic->host, fnic->fnic_num, "fdmi timer callback : 0x%x\n", iport->fabric.fdmi_pending); - iport->fabric.fdmi_pending = 0; - /* If max retries not exhaused, start over from fdmi plogi */ - if (iport->fabric.fdmi_retry < FDLS_FDMI_MAX_RETRY) { - iport->fabric.fdmi_retry++; - FNIC_FCS_DBG(KERN_INFO, fnic->host, fnic->fnic_num, - "retry fdmi timer %d", iport->fabric.fdmi_retry); - fdls_send_fdmi_plogi(iport); - } + fdls_fdmi_retry_plogi(iport); FNIC_FCS_DBG(KERN_INFO, fnic->host, fnic->fnic_num, "fdmi timer callback : 0x%x\n", iport->fabric.fdmi_pending); spin_unlock_irqrestore(&fnic->fnic_lock, flags); @@ -3714,11 +3744,32 @@ static void fdls_process_fdmi_abts_rsp(struct fnic_iport_s *iport, switch (FNIC_FRAME_TYPE(oxid)) { case FNIC_FRAME_TYPE_FDMI_PLOGI: fdls_free_oxid(iport, oxid, &iport->active_oxid_fdmi_plogi); + + iport->fabric.fdmi_pending &= ~FDLS_FDMI_PLOGI_PENDING; + iport->fabric.fdmi_pending &= ~FDLS_FDMI_ABORT_PENDING; break; case FNIC_FRAME_TYPE_FDMI_RHBA: + iport->fabric.fdmi_pending &= ~FDLS_FDMI_REG_HBA_PENDING; + + /* If RPA is still pending, don't turn off ABORT PENDING. + * We count on the timer to detect the ABTS timeout and take + * corrective action. + */ + if (!(iport->fabric.fdmi_pending & FDLS_FDMI_RPA_PENDING)) + iport->fabric.fdmi_pending &= ~FDLS_FDMI_ABORT_PENDING; + fdls_free_oxid(iport, oxid, &iport->active_oxid_fdmi_rhba); break; case FNIC_FRAME_TYPE_FDMI_RPA: + iport->fabric.fdmi_pending &= ~FDLS_FDMI_RPA_PENDING; + + /* If RHBA is still pending, don't turn off ABORT PENDING. + * We count on the timer to detect the ABTS timeout and take + * corrective action. + */ + if (!(iport->fabric.fdmi_pending & FDLS_FDMI_REG_HBA_PENDING)) + iport->fabric.fdmi_pending &= ~FDLS_FDMI_ABORT_PENDING; + fdls_free_oxid(iport, oxid, &iport->active_oxid_fdmi_rpa); break; default: @@ -3728,10 +3779,16 @@ static void fdls_process_fdmi_abts_rsp(struct fnic_iport_s *iport, break; } - timer_delete_sync(&iport->fabric.fdmi_timer); - iport->fabric.fdmi_pending &= ~FDLS_FDMI_ABORT_PENDING; - - fdls_send_fdmi_plogi(iport); + /* + * Only if ABORT PENDING is off, delete the timer, and if no other + * operations are pending, retry FDMI. + * Otherwise, let the timer pop and take the appropriate action. + */ + if (!(iport->fabric.fdmi_pending & FDLS_FDMI_ABORT_PENDING)) { + timer_delete_sync(&iport->fabric.fdmi_timer); + if (!iport->fabric.fdmi_pending) + fdls_fdmi_retry_plogi(iport); + } } static void diff --git a/drivers/scsi/fnic/fnic_fdls.h b/drivers/scsi/fnic/fnic_fdls.h index 8e610b65ad57..531d0b37e450 100644 --- a/drivers/scsi/fnic/fnic_fdls.h +++ b/drivers/scsi/fnic/fnic_fdls.h @@ -394,6 +394,7 @@ void fdls_send_tport_abts(struct fnic_iport_s *iport, bool fdls_delete_tport(struct fnic_iport_s *iport, struct fnic_tport_s *tport); void fdls_fdmi_timer_callback(struct timer_list *t); +void fdls_fdmi_retry_plogi(struct fnic_iport_s *iport); /* fnic_fcs.c */ void fnic_fdls_init(struct fnic *fnic, int usefip); -- 2.47.1

3 months

3
3
0 0

[PATCH net] net_sched: sch_sfq: reject invalid perturb period

by Eric Dumazet

Gerrard Tai reported that SFQ perturb_period has no range check yet, and this can be used to trigger a race condition fixed in a separate patch. We want to make sure ctl->perturb_period * HZ will not overflow and is positive. Tested: tc qd add dev lo root sfq perturb -10 # negative value : error Error: sch_sfq: invalid perturb period. tc qd add dev lo root sfq perturb 1000000000 # too big : error Error: sch_sfq: invalid perturb period. tc qd add dev lo root sfq perturb 2000000 # acceptable value tc -s -d qd sh dev lo qdisc sfq 8005: root refcnt 2 limit 127p quantum 64Kb depth 127 flows 128 divisor 1024 perturb 2000000sec Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: Gerrard Tai <gerrard.tai(a)starlabs.sg> Signed-off-by: Eric Dumazet <edumazet(a)google.com> Cc: stable(a)vger.kernel.org --- net/sched/sch_sfq.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 77fa02f2bfcd56a36815199aa2e7987943ea226f..a8cca549b5a2eb2407949560c2b6b658fb7a581f 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -656,6 +656,14 @@ static int sfq_change(struct Qdisc *sch, struct nlattr *opt, NL_SET_ERR_MSG_MOD(extack, "invalid quantum"); return -EINVAL; } + + if (ctl->perturb_period < 0 || + ctl->perturb_period > INT_MAX / HZ) { + NL_SET_ERR_MSG_MOD(extack, "invalid perturb period"); + return -EINVAL; + } + perturb_period = ctl->perturb_period * HZ; + if (ctl_v1 && !red_check_params(ctl_v1->qth_min, ctl_v1->qth_max, ctl_v1->Wlog, ctl_v1->Scell_log, NULL)) return -EINVAL; @@ -672,14 +680,12 @@ static int sfq_change(struct Qdisc *sch, struct nlattr *opt, headdrop = q->headdrop; maxdepth = q->maxdepth; maxflows = q->maxflows; - perturb_period = q->perturb_period; quantum = q->quantum; flags = q->flags; /* update and validate configuration */ if (ctl->quantum) quantum = ctl->quantum; - perturb_period = ctl->perturb_period * HZ; if (ctl->flows) maxflows = min_t(u32, ctl->flows, SFQ_MAX_FLOWS); if (ctl->divisor) { -- 2.50.0.rc0.642.g800a2b2222-goog

3 months

3
2
0 0

FAILED: patch "[PATCH] iio: dac: ad3552r-common: fix ad3541/2r ranges" failed to apply to 6.12-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.12-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.12.y git checkout FETCH_HEAD git cherry-pick -x 1e758b613212b6964518a67939535910b5aee831 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025021010-antarctic-untried-a72b@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 1e758b613212b6964518a67939535910b5aee831 Mon Sep 17 00:00:00 2001 From: Angelo Dureghello <adureghello(a)baylibre.com> Date: Wed, 8 Jan 2025 18:29:15 +0100 Subject: [PATCH] iio: dac: ad3552r-common: fix ad3541/2r ranges Fix ad3541/2r voltage ranges to be as per ad3542r datasheet, rev. C, table 38 (page 57). The wrong ad354xr ranges was generating erroneous Vpp output. In more details: - fix wrong number of ranges, they are 5 ranges, not 6, - remove non-existent 0-3V range, - adjust order, since ad3552r_find_range() get a wrong index, producing a wrong Vpp as output. Retested all the ranges on real hardware, EVALAD3542RFMCZ: adi,output-range-microvolt (fdt): <(000000) (2500000)>; ok (Rfbx1, switch 10) <(000000) (5000000)>; ok (Rfbx1, switch 10) <(000000) (10000000)>; ok (Rfbx1, switch 10) <(-5000000) (5000000)>; ok (Rfbx2, switch +/- 5) <(-2500000) (7500000)>; ok (Rfbx2, switch -2.5/7.5) Fixes: 8f2b54824b28 ("drivers:iio:dac: Add AD3552R driver support") Signed-off-by: Angelo Dureghello <adureghello(a)baylibre.com> Reviewed-by: David Lechner <dlechner(a)baylibre.com> Link: https://patch.msgid.link/20250108-wip-bl-ad3552r-axi-v0-iio-testing-carlos-… Cc: <Stable(a)vger.kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron(a)huawei.com> diff --git a/drivers/iio/dac/ad3552r-common.c b/drivers/iio/dac/ad3552r-common.c index 0f495df2e5ce..03e0864f5084 100644 --- a/drivers/iio/dac/ad3552r-common.c +++ b/drivers/iio/dac/ad3552r-common.c @@ -22,11 +22,10 @@ EXPORT_SYMBOL_NS_GPL(ad3552r_ch_ranges, "IIO_AD3552R"); const s32 ad3542r_ch_ranges[AD3542R_MAX_RANGES][2] = { [AD3542R_CH_OUTPUT_RANGE_0__2P5V] = { 0, 2500 }, - [AD3542R_CH_OUTPUT_RANGE_0__3V] = { 0, 3000 }, [AD3542R_CH_OUTPUT_RANGE_0__5V] = { 0, 5000 }, [AD3542R_CH_OUTPUT_RANGE_0__10V] = { 0, 10000 }, - [AD3542R_CH_OUTPUT_RANGE_NEG_2P5__7P5V] = { -2500, 7500 }, - [AD3542R_CH_OUTPUT_RANGE_NEG_5__5V] = { -5000, 5000 } + [AD3542R_CH_OUTPUT_RANGE_NEG_5__5V] = { -5000, 5000 }, + [AD3542R_CH_OUTPUT_RANGE_NEG_2P5__7P5V] = { -2500, 7500 } }; EXPORT_SYMBOL_NS_GPL(ad3542r_ch_ranges, "IIO_AD3552R"); diff --git a/drivers/iio/dac/ad3552r.h b/drivers/iio/dac/ad3552r.h index fd5a3dfd1d1c..4b5581039ae9 100644 --- a/drivers/iio/dac/ad3552r.h +++ b/drivers/iio/dac/ad3552r.h @@ -131,7 +131,7 @@ #define AD3552R_CH1_ACTIVE BIT(1) #define AD3552R_MAX_RANGES 5 -#define AD3542R_MAX_RANGES 6 +#define AD3542R_MAX_RANGES 5 #define AD3552R_QUAD_SPI 2 extern const s32 ad3552r_ch_ranges[AD3552R_MAX_RANGES][2]; @@ -189,16 +189,14 @@ enum ad3552r_ch_vref_select { enum ad3542r_ch_output_range { /* Range from 0 V to 2.5 V. Requires Rfb1x connection */ AD3542R_CH_OUTPUT_RANGE_0__2P5V, - /* Range from 0 V to 3 V. Requires Rfb1x connection */ - AD3542R_CH_OUTPUT_RANGE_0__3V, /* Range from 0 V to 5 V. Requires Rfb1x connection */ AD3542R_CH_OUTPUT_RANGE_0__5V, /* Range from 0 V to 10 V. Requires Rfb2x connection */ AD3542R_CH_OUTPUT_RANGE_0__10V, - /* Range from -2.5 V to 7.5 V. Requires Rfb2x connection */ - AD3542R_CH_OUTPUT_RANGE_NEG_2P5__7P5V, /* Range from -5 V to 5 V. Requires Rfb2x connection */ AD3542R_CH_OUTPUT_RANGE_NEG_5__5V, + /* Range from -2.5 V to 7.5 V. Requires Rfb2x connection */ + AD3542R_CH_OUTPUT_RANGE_NEG_2P5__7P5V, }; enum ad3552r_ch_output_range {

3 months

3
4
0 0

[PATCH v2 1/3] mm/huge_memory: don't ignore queried cachemode in vmf_insert_pfn_pud()

by David Hildenbrand

We setup the cache mode but ... don't forward the updated pgprot to insert_pfn_pud(). Only a problem on x86-64 PAT when mapping PFNs using PUDs that require a special cachemode. Fix it by using the proper pgprot where the cachemode was setup. Identified by code inspection. Fixes: 7b806d229ef1 ("mm: remove vmf_insert_pfn_xxx_prot() for huge page-table entries") Cc: <stable(a)vger.kernel.org> Signed-off-by: David Hildenbrand <david(a)redhat.com> --- mm/huge_memory.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/mm/huge_memory.c b/mm/huge_memory.c index d3e66136e41a3..49b98082c5401 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1516,10 +1516,9 @@ static pud_t maybe_pud_mkwrite(pud_t pud, struct vm_area_struct *vma) } static void insert_pfn_pud(struct vm_area_struct *vma, unsigned long addr, - pud_t *pud, pfn_t pfn, bool write) + pud_t *pud, pfn_t pfn, pgprot_t prot, bool write) { struct mm_struct *mm = vma->vm_mm; - pgprot_t prot = vma->vm_page_prot; pud_t entry; if (!pud_none(*pud)) { @@ -1581,7 +1580,7 @@ vm_fault_t vmf_insert_pfn_pud(struct vm_fault *vmf, pfn_t pfn, bool write) pfnmap_setup_cachemode_pfn(pfn_t_to_pfn(pfn), &pgprot); ptl = pud_lock(vma->vm_mm, vmf->pud); - insert_pfn_pud(vma, addr, vmf->pud, pfn, write); + insert_pfn_pud(vma, addr, vmf->pud, pfn, pgprot, write); spin_unlock(ptl); return VM_FAULT_NOPAGE; @@ -1625,7 +1624,7 @@ vm_fault_t vmf_insert_folio_pud(struct vm_fault *vmf, struct folio *folio, add_mm_counter(mm, mm_counter_file(folio), HPAGE_PUD_NR); } insert_pfn_pud(vma, addr, vmf->pud, pfn_to_pfn_t(folio_pfn(folio)), - write); + vma->vm_page_prot, write); spin_unlock(ptl); return VM_FAULT_NOPAGE; -- 2.49.0

3 months

5
9
0 0

[PATCH v3] arm64: Restrict pagetable teardown to avoid false warning

by Dev Jain

Commit 9c006972c3fe removes the pxd_present() checks because the caller checks pxd_present(). But, in case of vmap_try_huge_pud(), the caller only checks pud_present(); pud_free_pmd_page() recurses on each pmd through pmd_free_pte_page(), wherein the pmd may be none. Thus it is possible to hit a warning in the latter, since pmd_none => !pmd_table(). Thus, add a pmd_present() check in pud_free_pmd_page(). This problem was found by code inspection. Fixes: 9c006972c3fe (arm64: mmu: drop pXd_present() checks from pXd_free_pYd_table()) Cc: <stable(a)vger.kernel.org> Reported-by: Ryan Roberts <ryan.roberts(a)arm.com> Acked-by: David Hildenbrand <david(a)redhat.com> Signed-off-by: Dev Jain <dev.jain(a)arm.com> --- This patch is based on 6.15-rc6. v2->v3: - Use pmdp_get() v1->v2: - Enforce check in caller arch/arm64/mm/mmu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c index ea6695d53fb9..5a9bf291c649 100644 --- a/arch/arm64/mm/mmu.c +++ b/arch/arm64/mm/mmu.c @@ -1286,7 +1286,8 @@ int pud_free_pmd_page(pud_t *pudp, unsigned long addr) next = addr; end = addr + PUD_SIZE; do { - pmd_free_pte_page(pmdp, next); + if (pmd_present(pmdp_get(pmdp))) + pmd_free_pte_page(pmdp, next); } while (pmdp++, next += PMD_SIZE, next != end); pud_clear(pudp); -- 2.30.2

3 months

5
7
0 0

[PATCH v2 0/2] Samsung Exynos 7870 DECON driver support

by Kaustabh Chakraborty

This patch series aims at adding support for Exynos7870's DECON in the Exynos7 DECON driver. It introduces a driver data struct so that support for DECON on other SoCs can be added to it in the future. It also fixes a few bugs in the driver, such as functions receiving bad pointers. Tested on Samsung Galaxy J7 Prime (samsung-on7xelte), Samsung Galaxy A2 Core (samsung-a2corelte), and Samsung Galaxy J6 (samsung-j6lte). Signed-off-by: Kaustabh Chakraborty <kauschluss(a)disroot.org> --- Changes in v2: - Add a new commit to prevent an occasional panic under circumstances. - Rewrite and redo [v1 2/6] to be a more sensible commit. - Link to v1: https://lore.kernel.org/r/20240919-exynosdrm-decon-v1-0-6c5861c1cb04@disroo… --- Kaustabh Chakraborty (2): drm/exynos: exynos7_drm_decon: fix call of decon_commit() drm/exynos: exynos7_drm_decon: add vblank check in IRQ handling drivers/gpu/drm/exynos/exynos7_drm_decon.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) --- base-commit: 0bb71d301869446810a0b13d3da290bd455d7c78 change-id: 20240917-exynosdrm-decon-4c228dd1d2bf Best regards, -- Kaustabh Chakraborty <kauschluss(a)disroot.org>

3 months

1
2
0 0

[PATCH] net: pfcp: fix typo in message_priority field name

by RubenKelevra

Fix 'message_priprity' typo to 'message_priority' in big endian bitfield definition. This typo breaks compilation on big endian architectures. Fixes: 6dd514f48110 ("pfcp: always set pfcp metadata") Cc: stable(a)vger.kernel.org # commit 6dd514f48110 ("pfcp: always set pfcp metadata") Signed-off-by: RubenKelevra <rubenkelevra(a)gmail.com> --- include/net/pfcp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/net/pfcp.h b/include/net/pfcp.h index af14f970b80e1..639553797d3e4 100644 --- a/include/net/pfcp.h +++ b/include/net/pfcp.h @@ -45,7 +45,7 @@ struct pfcphdr_session { reserved:4; #elif defined(__BIG_ENDIAN_BITFIELD) u8 reserved:4, - message_priprity:4; + message_priority:4; #else #error "Please fix <asm/byteorder>" #endif -- 2.49.0

3 months

4
4
0 0

[PATCH v2] s390/pkey: prevent overflow in size calculation for memdup_user()

by Fedor Pchelkin

Number of apqn target list entries contained in 'nr_apqns' variable is determined by userspace via an ioctl call so the result of the product in calculation of size passed to memdup_user() may overflow. In this case the actual size of the allocated area and the value describing it won't be in sync leading to various types of unpredictable behaviour later. Use a proper memdup_array_user() helper which returns an error if an overflow is detected. Note that it is different from when nr_apqns is initially zero - that case is considered valid and should be handled in subsequent pkey_handler implementations. Found by Linux Verification Center (linuxtesting.org). Fixes: f2bbc96e7cfa ("s390/pkey: add CCA AES cipher key support") Cc: stable(a)vger.kernel.org Signed-off-by: Fedor Pchelkin <pchelkin(a)ispras.ru> --- v2: use memdup_array_user() helper (Heiko Carstens) drivers/s390/crypto/pkey_api.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/s390/crypto/pkey_api.c b/drivers/s390/crypto/pkey_api.c index cef60770f68b..b3fcdcae379e 100644 --- a/drivers/s390/crypto/pkey_api.c +++ b/drivers/s390/crypto/pkey_api.c @@ -86,7 +86,7 @@ static void *_copy_apqns_from_user(void __user *uapqns, size_t nr_apqns) if (!uapqns || nr_apqns == 0) return NULL; - return memdup_user(uapqns, nr_apqns * sizeof(struct pkey_apqn)); + return memdup_array_user(uapqns, nr_apqns, sizeof(struct pkey_apqn)); } static int pkey_ioctl_genseck(struct pkey_genseck __user *ugs) -- 2.49.0

3 months

4
3
0 0

Qualcomm Snapdragon <SM6375>

by Alan Ye

The Qualcomm SM6375 processor is a 7nm process SoC for the mid-range market with the following features: CPU: Eight-core design, including high-performance Kryo 670 core and efficient Kryo 265 core, optimized performance and energy efficiency. GPU: Equipped with Adreno 642L GPU, supporting high-quality graphics and gaming experience. AI Engine: Integrated Qualcomm AI engine to enhance intelligent features such as voice recognition and image processing. Connectivity: Supports modern wireless standards such as 5G, Wi-Fi 6 and Bluetooth 5.2. Multimedia: Supports 4K video encoding and decoding Mainly used in mid-to-high-end smartphones, tablets and some IoT devices, suitable for users who need to balance cost performance and performance. .# Part Number Manufacturer Date Code Quantity Unit Price Lead Time Condition (PCS) USD/Each one 1 SM-6375-1-PSP837-TR-00-0-AB QUALCOMM 2023+ 12000pcs US$18.00/pcs 7days New & original - stock 2 PM-6375-0-FOWNSP144-TR-01-0；TR-01-1 QUALCOMM 2023+ 12000pcs US$1.00/pcs 3 PMR-735A-0-WLNSP48-TR-05-0,TR-05-1 QUALCOMM 2023+ 12000pcs US$0.85/pcs 4 PMK-8003-0-FOWPSP36-TR-01-0 QUALCOMM 2023+ 12000pcs US$0.24/pcs 5 SDR-735-0-PSP219B-TR-01-0；TR-01-1 QUALCOMM 2023+ 12000pcs US$2.50/pcs 6 WCD-9370-0-WLPSP55-TR-01-0;TR-01-4 QUALCOMM 2023+ 12000pcs US$0.50/pcs 7 WCN-3988-0-82BWLPSP-TR-00-0 QUALCOMM 2023+ 12000pcs US$3.50/pcs 8 QET-6105-0-WLNSP24B-TR-00-1 QUALCOMM 2023+ 12000pcs US$1.20/pcs 9 QET4101-0-12WLNSP-TR-00-0 QUALCOMM 2022+ 12000pcs US$0.21/pcs These materials are sold as a set for $28/usd, and are guaranteed to be authentic. If you need other Qualcomm materials, please feel free to contact me Stay in tune with product evolutions—tap . Keep Receiving Notices Feel like taking a break? Select Configure Your Mailing.

3 months

1
0
0 0

[PATCH] wifi: mt76: mt7925: fix invalid array index in ssid assignment during hw scan

by Mingyen Hsieh

From: Michael Lo <michael.lo(a)mediatek.com> Update the destination index to use 'n_ssids', which is incremented only when a valid SSID is present. Previously, both mt76_connac_mcu_hw_scan() and mt7925_mcu_hw_scan() used the loop index 'i' for the destination array, potentially leaving gaps if any source SSIDs had zero length. Cc: stable(a)vger.kernel.org Fixes: c948b5da6bbe ("wifi: mt76: mt7925: add Mediatek Wi-Fi7 driver for mt7925 chips") Signed-off-by: Michael Lo <michael.lo(a)mediatek.com> Signed-off-by: Ming Yen Hsieh <mingyen.hsieh(a)mediatek.com> --- drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 4 ++-- drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c index db85a9d984c5..660c8df89910 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c @@ -1740,8 +1740,8 @@ int mt76_connac_mcu_hw_scan(struct mt76_phy *phy, struct ieee80211_vif *vif, if (!sreq->ssids[i].ssid_len) continue; - req->ssids[i].ssid_len = cpu_to_le32(sreq->ssids[i].ssid_len); - memcpy(req->ssids[i].ssid, sreq->ssids[i].ssid, + req->ssids[n_ssids].ssid_len = cpu_to_le32(sreq->ssids[i].ssid_len); + memcpy(req->ssids[n_ssids].ssid, sreq->ssids[i].ssid, sreq->ssids[i].ssid_len); n_ssids++; } diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c index 2bd506a4208c..66bac3047b2b 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c @@ -3178,8 +3178,8 @@ int mt7925_mcu_hw_scan(struct mt76_phy *phy, struct ieee80211_vif *vif, if (i > MT7925_RNR_SCAN_MAX_BSSIDS) break; - ssid->ssids[i].ssid_len = cpu_to_le32(sreq->ssids[i].ssid_len); - memcpy(ssid->ssids[i].ssid, sreq->ssids[i].ssid, + ssid->ssids[n_ssids].ssid_len = cpu_to_le32(sreq->ssids[i].ssid_len); + memcpy(ssid->ssids[n_ssids].ssid, sreq->ssids[i].ssid, sreq->ssids[i].ssid_len); n_ssids++; } -- 2.34.1

3 months

1
0
0 0

[PATCH] wifi: mt76: mt7925: fix the wrong config for tx interrupt

by Mingyen Hsieh

From: Ming Yen Hsieh <mingyen.hsieh(a)mediatek.com> MT_INT_TX_DONE_MCU_WM may cause tx interrupt to be mishandled during a reset failure, leading to the reset process failing. By using MT_INT_TX_DONE_MCU instead of MT_INT_TX_DONE_MCU_WM, the handling of tx interrupt is improved. Cc: stable(a)vger.kernel.org Fixes: c948b5da6bbe ("wifi: mt76: mt7925: add Mediatek Wi-Fi7 driver for mt7925 chips") Signed-off-by: Ming Yen Hsieh <mingyen.hsieh(a)mediatek.com> --- drivers/net/wireless/mediatek/mt76/mt7925/regs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/regs.h b/drivers/net/wireless/mediatek/mt76/mt7925/regs.h index 547489092c29..341987e47f67 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7925/regs.h +++ b/drivers/net/wireless/mediatek/mt76/mt7925/regs.h @@ -58,7 +58,7 @@ #define MT_INT_TX_DONE_MCU (MT_INT_TX_DONE_MCU_WM | \ MT_INT_TX_DONE_FWDL) -#define MT_INT_TX_DONE_ALL (MT_INT_TX_DONE_MCU_WM | \ +#define MT_INT_TX_DONE_ALL (MT_INT_TX_DONE_MCU | \ MT_INT_TX_DONE_BAND0 | \ GENMASK(18, 4)) -- 2.34.1

3 months

1
0
0 0

[merged mm-hotfixes-stable] drivers-rapidio-rio_cmc-prevent-possible-used-uninitialized.patch removed from -mm tree

by Andrew Morton

The quilt patch titled Subject: drivers/rapidio/rio_cm.c: prevent possible heap overwrite has been removed from the -mm tree. Its filename was drivers-rapidio-rio_cmc-prevent-possible-used-uninitialized.patch This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm ------------------------------------------------------ From: Andrew Morton <akpm(a)linux-foundation.org> Subject: drivers/rapidio/rio_cm.c: prevent possible heap overwrite Date: Sat Jun 7 05:43:18 PM PDT 2025 In riocm_cdev_ioctl(RIO_CM_CHAN_SEND) -> cm_chan_msg_send() -> riocm_ch_send() cm_chan_msg_send() checks that userspace didn't send too much data but riocm_ch_send() failed to check that userspace sent sufficient data. The result is that riocm_ch_send() can write to fields in the rio_ch_chan_hdr which were outside the bounds of the space which cm_chan_msg_send() allocated. Address this by teaching riocm_ch_send() to check that the entire rio_ch_chan_hdr was copied in from userspace. Reported-by: maher azz <maherazz04(a)gmail.com> Cc: Matt Porter <mporter(a)kernel.crashing.org> Cc: Alexandre Bounine <alex.bou9(a)gmail.com> Cc: Linus Torvalds <torvalds(a)linuxfoundation.org> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- drivers/rapidio/rio_cm.c | 3 +++ 1 file changed, 3 insertions(+) --- a/drivers/rapidio/rio_cm.c~drivers-rapidio-rio_cmc-prevent-possible-used-uninitialized +++ a/drivers/rapidio/rio_cm.c @@ -783,6 +783,9 @@ static int riocm_ch_send(u16 ch_id, void if (buf == NULL || ch_id == 0 || len == 0 || len > RIO_MAX_MSG_SIZE) return -EINVAL; + if (len < sizeof(struct rio_ch_chan_hdr)) + return -EINVAL; /* insufficient data from user */ + ch = riocm_get_channel(ch_id); if (!ch) { riocm_error("%s(%d) ch_%d not found", current->comm, _ Patches currently in -mm which might be from akpm(a)linux-foundation.org are mm-add-mmap_prepare-compatibility-layer-for-nested-file-systems-fix.patch

3 months

1
0
0 0

[merged mm-hotfixes-stable] mm-close-theoretical-race-where-stale-tlb-entries-could-linger.patch removed from -mm tree

by Andrew Morton

The quilt patch titled Subject: mm: close theoretical race where stale TLB entries could linger has been removed from the -mm tree. Its filename was mm-close-theoretical-race-where-stale-tlb-entries-could-linger.patch This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm ------------------------------------------------------ From: Ryan Roberts <ryan.roberts(a)arm.com> Subject: mm: close theoretical race where stale TLB entries could linger Date: Fri, 6 Jun 2025 10:28:07 +0100 Commit 3ea277194daa ("mm, mprotect: flush TLB if potentially racing with a parallel reclaim leaving stale TLB entries") described a theoretical race as such: """ Nadav Amit identified a theoretical race between page reclaim and mprotect due to TLB flushes being batched outside of the PTL being held. He described the race as follows: CPU0 CPU1 ---- ---- user accesses memory using RW PTE [PTE now cached in TLB] try_to_unmap_one() ==> ptep_get_and_clear() ==> set_tlb_ubc_flush_pending() mprotect(addr, PROT_READ) ==> change_pte_range() ==> [ PTE non-present - no flush ] user writes using cached RW PTE ... try_to_unmap_flush() The same type of race exists for reads when protecting for PROT_NONE and also exists for operations that can leave an old TLB entry behind such as munmap, mremap and madvise. """ The solution was to introduce flush_tlb_batched_pending() and call it under the PTL from mprotect/madvise/munmap/mremap to complete any pending tlb flushes. However, while madvise_free_pte_range() and madvise_cold_or_pageout_pte_range() were both retro-fitted to call flush_tlb_batched_pending() immediately after initially acquiring the PTL, they both temporarily release the PTL to split a large folio if they stumble upon one. In this case, where re-acquiring the PTL flush_tlb_batched_pending() must be called again, but it previously was not. Let's fix that. There are 2 Fixes: tags here: the first is the commit that fixed madvise_free_pte_range(). The second is the commit that added madvise_cold_or_pageout_pte_range(), which looks like it copy/pasted the faulty pattern from madvise_free_pte_range(). This is a theoretical bug discovered during code review. Link: https://lkml.kernel.org/r/20250606092809.4194056-1-ryan.roberts@arm.com Fixes: 3ea277194daa ("mm, mprotect: flush TLB if potentially racing with a parallel reclaim leaving stale TLB entries") Fixes: 9c276cc65a58 ("mm: introduce MADV_COLD") Signed-off-by: Ryan Roberts <ryan.roberts(a)arm.com> Reviewed-by: Jann Horn <jannh(a)google.com> Acked-by: David Hildenbrand <david(a)redhat.com> Cc: Liam Howlett <liam.howlett(a)oracle.com> Cc: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Cc: Mel Gorman <mgorman <mgorman(a)suse.de> Cc: Vlastimil Babka <vbabka(a)suse.cz> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- mm/madvise.c | 2 ++ 1 file changed, 2 insertions(+) --- a/mm/madvise.c~mm-close-theoretical-race-where-stale-tlb-entries-could-linger +++ a/mm/madvise.c @@ -508,6 +508,7 @@ restart: pte_offset_map_lock(mm, pmd, addr, &ptl); if (!start_pte) break; + flush_tlb_batched_pending(mm); arch_enter_lazy_mmu_mode(); if (!err) nr = 0; @@ -741,6 +742,7 @@ static int madvise_free_pte_range(pmd_t start_pte = pte; if (!start_pte) break; + flush_tlb_batched_pending(mm); arch_enter_lazy_mmu_mode(); if (!err) nr = 0; _ Patches currently in -mm which might be from ryan.roberts(a)arm.com are mm-readahead-honour-new_order-in-page_cache_ra_order.patch mm-readahead-terminate-async-readahead-on-natural-boundary.patch mm-readahead-make-space-in-struct-file_ra_state.patch mm-readahead-store-folio-order-in-struct-file_ra_state.patch mm-filemap-allow-arch-to-request-folio-size-for-exec-memory.patch mm-remove-arch_flush_tlb_batched_pending-arch-helper.patch

3 months

1
0
0 0

[merged mm-hotfixes-stable] mm-vma-reset-vma-iterator-on-commit_merge-oom-failure.patch removed from -mm tree

by Andrew Morton

The quilt patch titled Subject: mm/vma: reset VMA iterator on commit_merge() OOM failure has been removed from the -mm tree. Its filename was mm-vma-reset-vma-iterator-on-commit_merge-oom-failure.patch This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm ------------------------------------------------------ From: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Subject: mm/vma: reset VMA iterator on commit_merge() OOM failure Date: Fri, 6 Jun 2025 13:50:32 +0100 While an OOM failure in commit_merge() isn't really feasible due to the allocation which might fail (a maple tree pre-allocation) being 'too small to fail', we do need to handle this case correctly regardless. In vma_merge_existing_range(), we can theoretically encounter failures which result in an OOM error in two ways - firstly dup_anon_vma() might fail with an OOM error, and secondly commit_merge() failing, ultimately, to pre-allocate a maple tree node. The abort logic for dup_anon_vma() resets the VMA iterator to the initial range, ensuring that any logic looping on this iterator will correctly proceed to the next VMA. However the commit_merge() abort logic does not do the same thing. This resulted in a syzbot report occurring because mlockall() iterates through VMAs, is tolerant of errors, but ended up with an incorrect previous VMA being specified due to incorrect iterator state. While making this change, it became apparent we are duplicating logic - the logic introduced in commit 41e6ddcaa0f1 ("mm/vma: add give_up_on_oom option on modify/merge, use in uffd release") duplicates the vmg->give_up_on_oom check in both abort branches. Additionally, we observe that we can perform the anon_dup check safely on dup_anon_vma() failure, as this will not be modified should this call fail. Finally, we need to reset the iterator in both cases, so now we can simply use the exact same code to abort for both. We remove the VM_WARN_ON(err != -ENOMEM) as it would be silly for this to be otherwise and it allows us to implement the abort check more neatly. Link: https://lkml.kernel.org/r/20250606125032.164249-1-lorenzo.stoakes@oracle.com Fixes: 47b16d0462a4 ("mm: abort vma_modify() on merge out of memory failure") Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Reported-by: syzbot+d16409ea9ecc16ed261a(a)syzkaller.appspotmail.com Closes: https://lore.kernel.org/linux-mm/6842cc67.a00a0220.29ac89.003b.GAE@google.c… Reviewed-by: Pedro Falcato <pfalcato(a)suse.de> Reviewed-by: Vlastimil Babka <vbabka(a)suse.cz> Reviewed-by: Liam R. Howlett <Liam.Howlett(a)oracle.com> Cc: Jann Horn <jannh(a)google.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- mm/vma.c | 22 ++++------------------ 1 file changed, 4 insertions(+), 18 deletions(-) --- a/mm/vma.c~mm-vma-reset-vma-iterator-on-commit_merge-oom-failure +++ a/mm/vma.c @@ -967,26 +967,9 @@ static __must_check struct vm_area_struc err = dup_anon_vma(next, middle, &anon_dup); } - if (err) + if (err || commit_merge(vmg)) goto abort; - err = commit_merge(vmg); - if (err) { - VM_WARN_ON(err != -ENOMEM); - - if (anon_dup) - unlink_anon_vmas(anon_dup); - - /* - * We've cleaned up any cloned anon_vma's, no VMAs have been - * modified, no harm no foul if the user requests that we not - * report this and just give up, leaving the VMAs unmerged. - */ - if (!vmg->give_up_on_oom) - vmg->state = VMA_MERGE_ERROR_NOMEM; - return NULL; - } - khugepaged_enter_vma(vmg->target, vmg->flags); vmg->state = VMA_MERGE_SUCCESS; return vmg->target; @@ -995,6 +978,9 @@ abort: vma_iter_set(vmg->vmi, start); vma_iter_load(vmg->vmi); + if (anon_dup) + unlink_anon_vmas(anon_dup); + /* * This means we have failed to clone anon_vma's correctly, but no * actual changes to VMAs have occurred, so no harm no foul - if the _ Patches currently in -mm which might be from lorenzo.stoakes(a)oracle.com are mm-add-mmap_prepare-compatibility-layer-for-nested-file-systems.patch mm-add-mmap_prepare-compatibility-layer-for-nested-file-systems-fix-2.patch docs-mm-expand-vma-doc-to-highlight-pte-freeing-non-vma-traversal.patch mm-ksm-have-ksm-vma-checks-not-require-a-vma-pointer.patch mm-ksm-refer-to-special-vmas-via-vm_special-in-ksm_compatible.patch mm-prevent-ksm-from-breaking-vma-merging-for-new-vmas.patch tools-testing-selftests-add-vma-merge-tests-for-ksm-merge.patch mm-pagewalk-split-walk_page_range_novma-into-kernel-user-parts.patch mm-mremap-introduce-more-mergeable-mremap-via-mremap_relocate_anon.patch mm-mremap-add-mremap_must_relocate_anon.patch mm-mremap-add-mremap_relocate_anon-support-for-large-folios.patch tools-uapi-update-copy-of-linux-mmanh-from-the-kernel-sources.patch tools-testing-selftests-add-sys_mremap-helper-to-vm_utilh.patch tools-testing-selftests-add-mremap-cases-that-merge-normally.patch tools-testing-selftests-add-mremap_relocate_anon-merge-test-cases.patch tools-testing-selftests-expand-mremap-tests-for-mremap_relocate_anon.patch tools-testing-selftests-have-cow-self-test-use-mremap_relocate_anon.patch tools-testing-selftests-test-relocate-anon-in-split-huge-page-test.patch tools-testing-selftests-add-mremap_relocate_anon-fork-tests.patch

3 months

1
0
0 0

[PATCH v6] clk: qcom: dispcc-sm8750: Fix setting rate byte and pixel clocks

by Krzysztof Kozlowski

On SM8750 the setting rate of pixel and byte clocks, while the parent DSI PHY PLL, fails with: disp_cc_mdss_byte0_clk_src: rcg didn't update its configuration. DSI PHY PLL has to be unprepared and its "PLL Power Down" bits in CMN_CTRL_0 asserted. Mark these clocks with CLK_OPS_PARENT_ENABLE to ensure the parent is enabled during rate changes. Cc: <stable(a)vger.kernel.org> Fixes: f1080d8dab0f ("clk: qcom: dispcc-sm8750: Add SM8750 Display clock controller") Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski(a)linaro.org> --- Changes in v6: 1. Add CLK_OPS_PARENT_ENABLE also to pclk1, pclk2 and byte1. 2. Add Fixes tag and cc-stable Previously part of v5 (thus b4 diff might not work nice here): https://lore.kernel.org/r/20250430-b4-sm8750-display-v5-6-8cab30c3e4df@lina… Changes in v5: 1. New patch in above patchset. Cc: Abhinav Kumar <quic_abhinavk(a)quicinc.com> Cc: Dmitry Baryshkov <lumag(a)kernel.org> --- drivers/clk/qcom/dispcc-sm8750.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/clk/qcom/dispcc-sm8750.c b/drivers/clk/qcom/dispcc-sm8750.c index 877b40d50e6f..ca09da111a50 100644 --- a/drivers/clk/qcom/dispcc-sm8750.c +++ b/drivers/clk/qcom/dispcc-sm8750.c @@ -393,7 +393,7 @@ static struct clk_rcg2 disp_cc_mdss_byte0_clk_src = { .name = "disp_cc_mdss_byte0_clk_src", .parent_data = disp_cc_parent_data_1, .num_parents = ARRAY_SIZE(disp_cc_parent_data_1), - .flags = CLK_SET_RATE_PARENT, + .flags = CLK_SET_RATE_PARENT | CLK_OPS_PARENT_ENABLE, .ops = &clk_byte2_ops, }, }; @@ -408,7 +408,7 @@ static struct clk_rcg2 disp_cc_mdss_byte1_clk_src = { .name = "disp_cc_mdss_byte1_clk_src", .parent_data = disp_cc_parent_data_1, .num_parents = ARRAY_SIZE(disp_cc_parent_data_1), - .flags = CLK_SET_RATE_PARENT, + .flags = CLK_SET_RATE_PARENT | CLK_OPS_PARENT_ENABLE, .ops = &clk_byte2_ops, }, }; @@ -712,7 +712,7 @@ static struct clk_rcg2 disp_cc_mdss_pclk0_clk_src = { .name = "disp_cc_mdss_pclk0_clk_src", .parent_data = disp_cc_parent_data_1, .num_parents = ARRAY_SIZE(disp_cc_parent_data_1), - .flags = CLK_SET_RATE_PARENT, + .flags = CLK_SET_RATE_PARENT | CLK_OPS_PARENT_ENABLE, .ops = &clk_pixel_ops, }, }; @@ -727,7 +727,7 @@ static struct clk_rcg2 disp_cc_mdss_pclk1_clk_src = { .name = "disp_cc_mdss_pclk1_clk_src", .parent_data = disp_cc_parent_data_1, .num_parents = ARRAY_SIZE(disp_cc_parent_data_1), - .flags = CLK_SET_RATE_PARENT, + .flags = CLK_SET_RATE_PARENT | CLK_OPS_PARENT_ENABLE, .ops = &clk_pixel_ops, }, }; @@ -742,7 +742,7 @@ static struct clk_rcg2 disp_cc_mdss_pclk2_clk_src = { .name = "disp_cc_mdss_pclk2_clk_src", .parent_data = disp_cc_parent_data_1, .num_parents = ARRAY_SIZE(disp_cc_parent_data_1), - .flags = CLK_SET_RATE_PARENT, + .flags = CLK_SET_RATE_PARENT | CLK_OPS_PARENT_ENABLE, .ops = &clk_pixel_ops, }, }; -- 2.45.2

3 months

4
3
0 0

[PATCH] clk: qcom: gcc-ipq8074: fix broken freq table for nss_port6_tx_clk_src

by Christian Marangi

With the conversion done by commit e88f03230dc0 ("clk: qcom: gcc-ipq8074: rework nss_port5/6 clock to multiple conf") a Copy-Paste error was made for the nss_port6_tx_clk_src frequency table. This was caused by the wrong setting of the parent in ftbl_nss_port6_tx_clk_src that was wrongly set to P_UNIPHY1_RX instead of P_UNIPHY2_TX. This cause the UNIPHY2 port to malfunction when it needs to be scaled to higher clock. The malfunction was observed with the example scenario with an Aquantia 10G PHY connected and a speed higher than 1G (example 2.5G) Fix the broken frequency table to restore original functionality. Cc: stable(a)vger.kernel.org Fixes: e88f03230dc0 ("clk: qcom: gcc-ipq8074: rework nss_port5/6 clock to multiple conf") Signed-off-by: Christian Marangi <ansuelsmth(a)gmail.com> --- drivers/clk/qcom/gcc-ipq8074.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/clk/qcom/gcc-ipq8074.c b/drivers/clk/qcom/gcc-ipq8074.c index 7258ba5c0900..1329ea28d703 100644 --- a/drivers/clk/qcom/gcc-ipq8074.c +++ b/drivers/clk/qcom/gcc-ipq8074.c @@ -1895,10 +1895,10 @@ static const struct freq_conf ftbl_nss_port6_tx_clk_src_125[] = { static const struct freq_multi_tbl ftbl_nss_port6_tx_clk_src[] = { FMS(19200000, P_XO, 1, 0, 0), FM(25000000, ftbl_nss_port6_tx_clk_src_25), - FMS(78125000, P_UNIPHY1_RX, 4, 0, 0), + FMS(78125000, P_UNIPHY2_TX, 4, 0, 0), FM(125000000, ftbl_nss_port6_tx_clk_src_125), - FMS(156250000, P_UNIPHY1_RX, 2, 0, 0), - FMS(312500000, P_UNIPHY1_RX, 1, 0, 0), + FMS(156250000, P_UNIPHY2_TX, 2, 0, 0), + FMS(312500000, P_UNIPHY2_TX, 1, 0, 0), { } }; -- 2.48.1

3 months

3
2
0 0

[PATCH RESEND 5.10.y 0/4] serial: sh-sci: Backport fixes

by Claudiu

From: Claudiu Beznea <claudiu.beznea.uj(a)bp.renesas.com> Hi, Commit 653143ed73ec ("serial: sh-sci: Check if TX data was written to device in .tx_empty()") doesn't apply cleanly on top of v5.10.y stable tree. This series adjust it. Along with it, propose for backporting other sh-sci fixes. Please provide your feedback. Thank you, Claudiu Beznea Claudiu Beznea (4): serial: sh-sci: Check if TX data was written to device in .tx_empty() serial: sh-sci: Move runtime PM enable to sci_probe_single() serial: sh-sci: Clean sci_ports[0] after at earlycon exit serial: sh-sci: Increment the runtime usage counter for the earlycon device drivers/tty/serial/sh-sci.c | 97 ++++++++++++++++++++++++++++++------- 1 file changed, 79 insertions(+), 18 deletions(-) -- 2.43.0

3 months

2
8
0 0

[PATCH 6.1.y 0/4] serial: sh-sci: Backport fixes

by Claudiu

From: Claudiu Beznea <claudiu.beznea.uj(a)bp.renesas.com> Hi, Commit 653143ed73ec ("serial: sh-sci: Check if TX data was written to device in .tx_empty()") doesn't apply cleanly on top of v6.1.y stable tree. This series adjust it. Along with it, propose for backporting other sh-sci fixes. Please provide your feedback. Thank you, Claudiu Beznea Claudiu Beznea (4): serial: sh-sci: Check if TX data was written to device in .tx_empty() serial: sh-sci: Move runtime PM enable to sci_probe_single() serial: sh-sci: Clean sci_ports[0] after at earlycon exit serial: sh-sci: Increment the runtime usage counter for the earlycon device drivers/tty/serial/sh-sci.c | 97 ++++++++++++++++++++++++++++++------- 1 file changed, 79 insertions(+), 18 deletions(-) -- 2.43.0

3 months

2
8
0 0

[PATCH 5.15.y 0/4] serial: sh-sci: Backport fixes

by Claudiu

From: Claudiu Beznea <claudiu.beznea.uj(a)bp.renesas.com> Hi, Commit 653143ed73ec ("serial: sh-sci: Check if TX data was written to device in .tx_empty()") doesn't apply cleanly on top of v5.15.y stable tree. This series adjust it. Along with it, propose for backporting other sh-sci fixes. Please provide your feedback. Thank you, Claudiu Beznea Claudiu Beznea (4): serial: sh-sci: Check if TX data was written to device in .tx_empty() serial: sh-sci: Move runtime PM enable to sci_probe_single() serial: sh-sci: Clean sci_ports[0] after at earlycon exit serial: sh-sci: Increment the runtime usage counter for the earlycon device drivers/tty/serial/sh-sci.c | 97 ++++++++++++++++++++++++++++++------- 1 file changed, 79 insertions(+), 18 deletions(-) -- 2.43.0

3 months

2
8
0 0

[PATCH 5.4.y] NFSD: Fix NFSv3 SETATTR/CREATE's handling of large file sizes

by Larry Bassel

From: Chuck Lever <chuck.lever(a)oracle.com> [ Upstream commit a648fdeb7c0e17177a2280344d015dba3fbe3314 ] iattr::ia_size is a loff_t, so these NFSv3 procedures must be careful to deal with incoming client size values that are larger than s64_max without corrupting the value. Silently capping the value results in storing a different value than the client passed in which is unexpected behavior, so remove the min_t() check in decode_sattr3(). Note that RFC 1813 permits only the WRITE procedure to return NFS3ERR_FBIG. We believe that NFSv3 reference implementations also return NFS3ERR_FBIG when ia_size is too large. Cc: stable(a)vger.kernel.org Signed-off-by: Chuck Lever <chuck.lever(a)oracle.com> (cherry picked from commit a648fdeb7c0e17177a2280344d015dba3fbe3314) [Larry: backport to 5.4.y. Minor conflict resolved due to missing commit 9cde9360d18d NFSD: Update the SETATTR3args decoder to use struct xdr_stream] Signed-off-by: Larry Bassel <larry.bassel(a)oracle.com> --- fs/nfsd/nfs3xdr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index 03e8c45a52f3..25b6b4db0af2 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -122,7 +122,7 @@ decode_sattr3(__be32 *p, struct iattr *iap, struct user_namespace *userns) iap->ia_valid |= ATTR_SIZE; p = xdr_decode_hyper(p, &newsize); - iap->ia_size = min_t(u64, newsize, NFS_OFFSET_MAX); + iap->ia_size = newsize; } if ((tmp = ntohl(*p++)) == 1) { /* set to server time */ iap->ia_valid |= ATTR_ATIME; -- 2.46.0

3 months

2
1
0 0

[PATCH 6.6.y 0/4] serial: sh-sci: Backport fixes

by Claudiu

From: Claudiu Beznea <claudiu.beznea.uj(a)bp.renesas.com> Hi, Commit 653143ed73ec ("serial: sh-sci: Check if TX data was written to device in .tx_empty()") doesn't apply cleanly on top of v6.6.y stable tree. This series adjust it. Along with it, propose for backporting other sh-sci fixes. Please provide your feedback. Thank you, Claudiu Beznea Claudiu Beznea (4): serial: sh-sci: Check if TX data was written to device in .tx_empty() serial: sh-sci: Move runtime PM enable to sci_probe_single() serial: sh-sci: Clean sci_ports[0] after at earlycon exit serial: sh-sci: Increment the runtime usage counter for the earlycon device drivers/tty/serial/sh-sci.c | 97 ++++++++++++++++++++++++++++++------- 1 file changed, 79 insertions(+), 18 deletions(-) -- 2.43.0

3 months

2
8
0 0

[PATCH 5.4.y] NFSD: Fix ia_size underflow

by Larry Bassel

From: Chuck Lever <chuck.lever(a)oracle.com> [ Upstream commit e6faac3f58c7c4176b66f63def17a34232a17b0e ] iattr::ia_size is a loff_t, which is a signed 64-bit type. NFSv3 and NFSv4 both define file size as an unsigned 64-bit type. Thus there is a range of valid file size values an NFS client can send that is already larger than Linux can handle. Currently decode_fattr4() dumps a full u64 value into ia_size. If that value happens to be larger than S64_MAX, then ia_size underflows. I'm about to fix up the NFSv3 behavior as well, so let's catch the underflow in the common code path: nfsd_setattr(). Cc: stable(a)vger.kernel.org Signed-off-by: Chuck Lever <chuck.lever(a)oracle.com> (cherry picked from commit e6faac3f58c7c4176b66f63def17a34232a17b0e) [Larry: backport to 5.4.y. Minor conflict resolved due to missing commit 2f221d6f7b88 attr: handle idmapped mounts] Signed-off-by: Larry Bassel <larry.bassel(a)oracle.com> --- fs/nfsd/vfs.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 6aa968bee0ce..bee4fdf6e239 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -448,6 +448,10 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, .ia_size = iap->ia_size, }; + host_err = -EFBIG; + if (iap->ia_size < 0) + goto out_unlock; + host_err = notify_change(dentry, &size_attr, NULL); if (host_err) goto out_unlock; -- 2.46.0

3 months

2
1
0 0

[PATCH v3 4/5] scsi: fnic: Turn off FDMI ACTIVE flags on link down

by Karan Tilak Kumar

When the link goes down and comes up, FDMI requests are not sent out anymore. Fix bug by turning off FNIC_FDMI_ACTIVE when the link goes down. Fixes: 09c1e6ab4ab2 ("scsi: fnic: Add and integrate support for FDMI") Reviewed-by: Sesidhar Baddela <sebaddel(a)cisco.com> Reviewed-by: Arulprabhu Ponnusamy <arulponn(a)cisco.com> Reviewed-by: Gian Carlo Boffa <gcboffa(a)cisco.com> Reviewed-by: Arun Easi <aeasi(a)cisco.com> Tested-by: Karan Tilak Kumar <kartilak(a)cisco.com> Cc: <stable(a)vger.kernel.org> # 6.14.x Please see patch description Signed-off-by: Karan Tilak Kumar <kartilak(a)cisco.com> --- drivers/scsi/fnic/fdls_disc.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/fnic/fdls_disc.c b/drivers/scsi/fnic/fdls_disc.c index 9e9939d41fa8..14691db4d5f9 100644 --- a/drivers/scsi/fnic/fdls_disc.c +++ b/drivers/scsi/fnic/fdls_disc.c @@ -5078,9 +5078,12 @@ void fnic_fdls_link_down(struct fnic_iport_s *iport) fdls_delete_tport(iport, tport); } - if ((fnic_fdmi_support == 1) && (iport->fabric.fdmi_pending > 0)) { - timer_delete_sync(&iport->fabric.fdmi_timer); - iport->fabric.fdmi_pending = 0; + if (fnic_fdmi_support == 1) { + if (iport->fabric.fdmi_pending > 0) { + timer_delete_sync(&iport->fabric.fdmi_timer); + iport->fabric.fdmi_pending = 0; + } + iport->flags &= ~FNIC_FDMI_ACTIVE; } FNIC_FCS_DBG(KERN_INFO, fnic->host, fnic->fnic_num, -- 2.47.1

3 months

1
0
0 0

[PATCH net v3 1/2] net: clear the dst when changing skb protocol

by Jakub Kicinski

A not-so-careful NAT46 BPF program can crash the kernel if it indiscriminately flips ingress packets from v4 to v6: BUG: kernel NULL pointer dereference, address: 0000000000000000 ip6_rcv_core (net/ipv6/ip6_input.c:190:20) ipv6_rcv (net/ipv6/ip6_input.c:306:8) process_backlog (net/core/dev.c:6186:4) napi_poll (net/core/dev.c:6906:9) net_rx_action (net/core/dev.c:7028:13) do_softirq (kernel/softirq.c:462:3) netif_rx (net/core/dev.c:5326:3) dev_loopback_xmit (net/core/dev.c:4015:2) ip_mc_finish_output (net/ipv4/ip_output.c:363:8) NF_HOOK (./include/linux/netfilter.h:314:9) ip_mc_output (net/ipv4/ip_output.c:400:5) dst_output (./include/net/dst.h:459:9) ip_local_out (net/ipv4/ip_output.c:130:9) ip_send_skb (net/ipv4/ip_output.c:1496:8) udp_send_skb (net/ipv4/udp.c:1040:8) udp_sendmsg (net/ipv4/udp.c:1328:10) The output interface has a 4->6 program attached at ingress. We try to loop the multicast skb back to the sending socket. Ingress BPF runs as part of netif_rx(), pushes a valid v6 hdr and changes skb->protocol to v6. We enter ip6_rcv_core which tries to use skb_dst(). But the dst is still an IPv4 one left after IPv4 mcast output. Clear the dst in all BPF helpers which change the protocol. Try to preserve metadata dsts, those may carry non-routing metadata. Cc: stable(a)vger.kernel.org Reviewed-by: Maciej Żenczykowski <maze(a)google.com> Acked-by: Daniel Borkmann <daniel(a)iogearbox.net> Fixes: d219df60a70e ("bpf: Add ipip6 and ip6ip decap support for bpf_skb_adjust_room()") Fixes: 1b00e0dfe7d0 ("bpf: update skb->protocol in bpf_skb_net_grow") Fixes: 6578171a7ff0 ("bpf: add bpf_skb_change_proto helper") Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> --- v3: - go back to v1, the encap / decap which don't change proto will be added in -next - split out the test v2: https://lore.kernel.org/20250607204734.1588964-1-kuba@kernel.org - drop on encap/decap - fix typo (protcol) - add the test to the Makefile v1: https://lore.kernel.org/20250604210604.257036-1-kuba@kernel.org I wonder if we should not skip ingress (tc_skip_classify?) for looped back packets in the first place. But that doesn't seem robust enough vs multiple redirections to solve the crash. Ignoring LOOPBACK packets (like the NAT46 prog should) doesn't work either, since BPF can change pkt_type arbitrarily. CC: martin.lau(a)linux.dev CC: daniel(a)iogearbox.net CC: john.fastabend(a)gmail.com CC: eddyz87(a)gmail.com CC: sdf(a)fomichev.me CC: haoluo(a)google.com CC: willemb(a)google.com CC: william.xuanziyang(a)huawei.com CC: alan.maguire(a)oracle.com CC: bpf(a)vger.kernel.org CC: edumazet(a)google.com CC: maze(a)google.com CC: shuah(a)kernel.org CC: linux-kselftest(a)vger.kernel.org CC: yonghong.song(a)linux.dev --- net/core/filter.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) diff --git a/net/core/filter.c b/net/core/filter.c index 327ca73f9cd7..7a72f766aacf 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -3233,6 +3233,13 @@ static const struct bpf_func_proto bpf_skb_vlan_pop_proto = { .arg1_type = ARG_PTR_TO_CTX, }; +static void bpf_skb_change_protocol(struct sk_buff *skb, u16 proto) +{ + skb->protocol = htons(proto); + if (skb_valid_dst(skb)) + skb_dst_drop(skb); +} + static int bpf_skb_generic_push(struct sk_buff *skb, u32 off, u32 len) { /* Caller already did skb_cow() with len as headroom, @@ -3329,7 +3336,7 @@ static int bpf_skb_proto_4_to_6(struct sk_buff *skb) } } - skb->protocol = htons(ETH_P_IPV6); + bpf_skb_change_protocol(skb, ETH_P_IPV6); skb_clear_hash(skb); return 0; @@ -3359,7 +3366,7 @@ static int bpf_skb_proto_6_to_4(struct sk_buff *skb) } } - skb->protocol = htons(ETH_P_IP); + bpf_skb_change_protocol(skb, ETH_P_IP); skb_clear_hash(skb); return 0; @@ -3550,10 +3557,10 @@ static int bpf_skb_net_grow(struct sk_buff *skb, u32 off, u32 len_diff, /* Match skb->protocol to new outer l3 protocol */ if (skb->protocol == htons(ETH_P_IP) && flags & BPF_F_ADJ_ROOM_ENCAP_L3_IPV6) - skb->protocol = htons(ETH_P_IPV6); + bpf_skb_change_protocol(skb, ETH_P_IPV6); else if (skb->protocol == htons(ETH_P_IPV6) && flags & BPF_F_ADJ_ROOM_ENCAP_L3_IPV4) - skb->protocol = htons(ETH_P_IP); + bpf_skb_change_protocol(skb, ETH_P_IP); } if (skb_is_gso(skb)) { @@ -3606,10 +3613,10 @@ static int bpf_skb_net_shrink(struct sk_buff *skb, u32 off, u32 len_diff, /* Match skb->protocol to new outer l3 protocol */ if (skb->protocol == htons(ETH_P_IP) && flags & BPF_F_ADJ_ROOM_DECAP_L3_IPV6) - skb->protocol = htons(ETH_P_IPV6); + bpf_skb_change_protocol(skb, ETH_P_IPV6); else if (skb->protocol == htons(ETH_P_IPV6) && flags & BPF_F_ADJ_ROOM_DECAP_L3_IPV4) - skb->protocol = htons(ETH_P_IP); + bpf_skb_change_protocol(skb, ETH_P_IP); if (skb_is_gso(skb)) { struct skb_shared_info *shinfo = skb_shinfo(skb); -- 2.49.0

3 months

3
2
0 0

[PATCH v4] mm: userfaultfd: fix race of userfaultfd_move and swap cache

by Kairui Song

From: Kairui Song <kasong(a)tencent.com> On seeing a swap entry PTE, userfaultfd_move does a lockless swap cache lookup, and tries to move the found folio to the faulting vma. Currently, it relies on checking the PTE value to ensure that the moved folio still belongs to the src swap entry and that no new folio has been added to the swap cache, which turns out to be unreliable. While working and reviewing the swap table series with Barry, following existing races are observed and reproduced [1]: In the example below, move_pages_pte is moving src_pte to dst_pte, where src_pte is a swap entry PTE holding swap entry S1, and S1 is not in the swap cache: CPU1 CPU2 userfaultfd_move move_pages_pte() entry = pte_to_swp_entry(orig_src_pte); // Here it got entry = S1 ... < interrupted> ... <swapin src_pte, alloc and use folio A> // folio A is a new allocated folio // and get installed into src_pte <frees swap entry S1> // src_pte now points to folio A, S1 // has swap count == 0, it can be freed // by folio_swap_swap or swap // allocator's reclaim. <try to swap out another folio B> // folio B is a folio in another VMA. <put folio B to swap cache using S1 > // S1 is freed, folio B can use it // for swap out with no problem. ... folio = filemap_get_folio(S1) // Got folio B here !!! ... < interrupted again> ... <swapin folio B and free S1> // Now S1 is free to be used again. <swapout src_pte & folio A using S1> // Now src_pte is a swap entry PTE // holding S1 again. folio_trylock(folio) move_swap_pte double_pt_lock is_pte_pages_stable // Check passed because src_pte == S1 folio_move_anon_rmap(...) // Moved invalid folio B here !!! The race window is very short and requires multiple collisions of multiple rare events, so it's very unlikely to happen, but with a deliberately constructed reproducer and increased time window, it can be reproduced easily. This can be fixed by checking if the folio returned by filemap is the valid swap cache folio after acquiring the folio lock. Another similar race is possible: filemap_get_folio may return NULL, but folio (A) could be swapped in and then swapped out again using the same swap entry after the lookup. In such a case, folio (A) may remain in the swap cache, so it must be moved too: CPU1 CPU2 userfaultfd_move move_pages_pte() entry = pte_to_swp_entry(orig_src_pte); // Here it got entry = S1, and S1 is not in swap cache folio = filemap_get_folio(S1) // Got NULL ... < interrupted again> ... <swapin folio A and free S1> <swapout folio A re-using S1> move_swap_pte double_pt_lock is_pte_pages_stable // Check passed because src_pte == S1 folio_move_anon_rmap(...) // folio A is ignored !!! Fix this by checking the swap cache again after acquiring the src_pte lock. And to avoid the filemap overhead, we check swap_map directly [2]. The SWP_SYNCHRONOUS_IO path does make the problem more complex, but so far we don't need to worry about that, since folios can only be exposed to the swap cache in the swap out path, and this is covered in this patch by checking the swap cache again after acquiring the src_pte lock. Testing with a simple C program that allocates and moves several GB of memory did not show any observable performance change. Cc: <stable(a)vger.kernel.org> Fixes: adef440691ba ("userfaultfd: UFFDIO_MOVE uABI") Closes: https://lore.kernel.org/linux-mm/CAMgjq7B1K=6OOrK2OUZ0-tqCzi+EJt+2_K97TPGoS… [1] Link: https://lore.kernel.org/all/CAGsJ_4yJhJBo16XhiC-nUzSheyX-V3-nFE+tAi=8Y560K8… [2] Signed-off-by: Kairui Song <kasong(a)tencent.com> Reviewed-by: Lokesh Gidra <lokeshgidra(a)google.com> --- V1: https://lore.kernel.org/linux-mm/20250530201710.81365-1-ryncsn@gmail.com/ Changes: - Check swap_map instead of doing a filemap lookup after acquiring the PTE lock to minimize critical section overhead [ Barry Song, Lokesh Gidra ] V2: https://lore.kernel.org/linux-mm/20250601200108.23186-1-ryncsn@gmail.com/ Changes: - Move the folio and swap check inside move_swap_pte to avoid skipping the check and potential overhead [ Lokesh Gidra ] - Add a READ_ONCE for the swap_map read to ensure it reads a up to dated value. V3: https://lore.kernel.org/all/20250602181419.20478-1-ryncsn@gmail.com/ Changes: - Add more comments and more context in commit message. mm/userfaultfd.c | 33 +++++++++++++++++++++++++++++++-- 1 file changed, 31 insertions(+), 2 deletions(-) diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c index bc473ad21202..8253978ee0fb 100644 --- a/mm/userfaultfd.c +++ b/mm/userfaultfd.c @@ -1084,8 +1084,18 @@ static int move_swap_pte(struct mm_struct *mm, struct vm_area_struct *dst_vma, pte_t orig_dst_pte, pte_t orig_src_pte, pmd_t *dst_pmd, pmd_t dst_pmdval, spinlock_t *dst_ptl, spinlock_t *src_ptl, - struct folio *src_folio) + struct folio *src_folio, + struct swap_info_struct *si, swp_entry_t entry) { + /* + * Check if the folio still belongs to the target swap entry after + * acquiring the lock. Folio can be freed in the swap cache while + * not locked. + */ + if (src_folio && unlikely(!folio_test_swapcache(src_folio) || + entry.val != src_folio->swap.val)) + return -EAGAIN; + double_pt_lock(dst_ptl, src_ptl); if (!is_pte_pages_stable(dst_pte, src_pte, orig_dst_pte, orig_src_pte, @@ -1102,6 +1112,25 @@ static int move_swap_pte(struct mm_struct *mm, struct vm_area_struct *dst_vma, if (src_folio) { folio_move_anon_rmap(src_folio, dst_vma); src_folio->index = linear_page_index(dst_vma, dst_addr); + } else { + /* + * Check if the swap entry is cached after acquiring the src_pte + * lock. Otherwise, we might miss a newly loaded swap cache folio. + * + * Check swap_map directly to minimize overhead, READ_ONCE is sufficient. + * We are trying to catch newly added swap cache, the only possible case is + * when a folio is swapped in and out again staying in swap cache, using the + * same entry before the PTE check above. The PTL is acquired and released + * twice, each time after updating the swap_map's flag. So holding + * the PTL here ensures we see the updated value. False positive is possible, + * e.g. SWP_SYNCHRONOUS_IO swapin may set the flag without touching the + * cache, or during the tiny synchronization window between swap cache and + * swap_map, but it will be gone very quickly, worst result is retry jitters. + */ + if (READ_ONCE(si->swap_map[swp_offset(entry)]) & SWAP_HAS_CACHE) { + double_pt_unlock(dst_ptl, src_ptl); + return -EAGAIN; + } } orig_src_pte = ptep_get_and_clear(mm, src_addr, src_pte); @@ -1412,7 +1441,7 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd, } err = move_swap_pte(mm, dst_vma, dst_addr, src_addr, dst_pte, src_pte, orig_dst_pte, orig_src_pte, dst_pmd, dst_pmdval, - dst_ptl, src_ptl, src_folio); + dst_ptl, src_ptl, src_folio, si, entry); } out: -- 2.49.0

3 months

6
6
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror