6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christian Heusel christian@heusel.eu

commit 765f253e28909f161b0211f85cf0431cfee7d6df upstream.

This reverts commit 67d1a8956d2d62fe6b4c13ebabb57806098511d8. Since this commit has been proven to be problematic for the setup of USB-tethered ethernet connections and the related breakage is very noticeable for users it should be reverted until a fixed version of the change can be rolled out.

Closes: https://lore.kernel.org/all/e0df2d85-1296-4317-b717-bd757e3ab928@heusel.eu/ Link: https://chaos.social/@gromit/114377862699921553 Link: https://bugzilla.kernel.org/show_bug.cgi?id=220002 Link: https://bugs.gentoo.org/953555 Link: https://bbs.archlinux.org/viewtopic.php?id=304892 Cc: stable@vger.kernel.org Acked-by: Lubomir Rintel lkundrak@v3.sk Signed-off-by: Christian Heusel christian@heusel.eu Link: https://patch.msgid.link/20250424-usb-tethering-fix-v1-1-b65cf97c740e@heusel... Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/net/usb/rndis_host.c | 16 ++-------------- 1 file changed, 2 insertions(+), 14 deletions(-)

--- a/drivers/net/usb/rndis_host.c +++ b/drivers/net/usb/rndis_host.c @@ -630,16 +630,6 @@ static const struct driver_info zte_rndi .tx_fixup = rndis_tx_fixup, };

-static const struct driver_info wwan_rndis_info = { - .description = "Mobile Broadband RNDIS device", - .flags = FLAG_WWAN | FLAG_POINTTOPOINT | FLAG_FRAMING_RN | FLAG_NO_SETINT, - .bind = rndis_bind, - .unbind = rndis_unbind, - .status = rndis_status, - .rx_fixup = rndis_rx_fixup, - .tx_fixup = rndis_tx_fixup, -}; - /*-------------------------------------------------------------------------*/

static const struct usb_device_id products [] = { @@ -676,11 +666,9 @@ static const struct usb_device_id produc USB_INTERFACE_INFO(USB_CLASS_WIRELESS_CONTROLLER, 1, 3), .driver_info = (unsigned long) &rndis_info, }, { - /* Mobile Broadband Modem, seen in Novatel Verizon USB730L and - * Telit FN990A (RNDIS) - */ + /* Novatel Verizon USB730L */ USB_INTERFACE_INFO(USB_CLASS_MISC, 4, 1), - .driver_info = (unsigned long)&wwan_rndis_info, + .driver_info = (unsigned long) &rndis_info, }, { }, // END };

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Joachim Priesner joachim.priesner@web.de

commit 1149719442d28c96dc63cad432b5a6db7c300e1a upstream.

There seem to be multiple USB device IDs used for these; the one I have reports as 0b0e:030c when powered on. (When powered off, it reports as 0b0e:0311.)

Signed-off-by: Joachim Priesner joachim.priesner@web.de Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20250428053606.9237-1-joachim.priesner@web.de Signed-off-by: Takashi Iwai tiwai@suse.de Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- sound/usb/format.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)

--- a/sound/usb/format.c +++ b/sound/usb/format.c @@ -263,7 +263,8 @@ static int parse_audio_format_rates_v1(s }

/* Jabra Evolve 65 headset */ - if (chip->usb_id == USB_ID(0x0b0e, 0x030b)) { + if (chip->usb_id == USB_ID(0x0b0e, 0x030b) || + chip->usb_id == USB_ID(0x0b0e, 0x030c)) { /* only 48kHz for playback while keeping 16kHz for capture */ if (fp->nr_rates != 1) return set_fixed_rate(fp, 48000, SNDRV_PCM_RATE_48000);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tvrtko Ursulin tvrtko.ursulin@igalia.com

commit 5b1834d6202f86180e451ad1a2a8a193a1da18fc upstream.

If we unbind a driver from the PCI device with an active DRM client, subsequent read of the fdinfo data associated with the file descriptor in question will not end well.

Protect the path with a drm_dev_enter/exit() pair.

Signed-off-by: Tvrtko Ursulin tvrtko.ursulin@igalia.com Cc: Christian König christian.koenig@amd.com Cc: Lucas De Marchi lucas.demarchi@intel.com Cc: Rodrigo Vivi rodrigo.vivi@intel.com Cc: Umesh Nerlige Ramappa umesh.nerlige.ramappa@intel.com Reviewed-by: Christian König christian.koenig@amd.com Fixes: 3f09a0cd4ea3 ("drm: Add common fdinfo helper") Cc: stable@vger.kernel.org # v6.5+ Signed-off-by: Christian König christian.koenig@amd.com Link: https://lore.kernel.org/r/20250418162512.72324-1-tvrtko.ursulin@igalia.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/gpu/drm/drm_file.c | 6 ++++++ 1 file changed, 6 insertions(+)

--- a/drivers/gpu/drm/drm_file.c +++ b/drivers/gpu/drm/drm_file.c @@ -1015,6 +1015,10 @@ void drm_show_fdinfo(struct seq_file *m, struct drm_file *file = f->private_data; struct drm_device *dev = file->minor->dev; struct drm_printer p = drm_seq_file_printer(m); + int idx; + + if (!drm_dev_enter(dev, &idx)) + return;

drm_printf(&p, "drm-driver:\t%s\n", dev->driver->name); drm_printf(&p, "drm-client-id:\t%llu\n", file->client_id); @@ -1029,6 +1033,8 @@ void drm_show_fdinfo(struct seq_file *m,

if (dev->driver->show_fdinfo) dev->driver->show_fdinfo(&p, file); + + drm_dev_exit(idx); } EXPORT_SYMBOL(drm_show_fdinfo);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Niravkumar L Rabara niravkumar.l.rabara@altera.com

commit 4fb7b8fceb0beebbe00712c3daf49ade0386076a upstream.

Test correct structure member, ecc_cecnt_offset, before using it.

[ bp: Massage commit message. ]

Fixes: 73bcc942f427 ("EDAC, altera: Add Arria10 EDAC support") Signed-off-by: Niravkumar L Rabara niravkumar.l.rabara@altera.com Signed-off-by: Matthew Gerlach matthew.gerlach@altera.com Signed-off-by: Borislav Petkov (AMD) bp@alien8.de Acked-by: Dinh Nguyen dinguyen@kernel.org Cc: stable@kernel.org Link: https://lore.kernel.org/20250425142640.33125-2-matthew.gerlach@altera.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/edac/altera_edac.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/drivers/edac/altera_edac.c +++ b/drivers/edac/altera_edac.c @@ -98,7 +98,7 @@ static irqreturn_t altr_sdram_mc_err_han if (status & priv->ecc_stat_ce_mask) { regmap_read(drvdata->mc_vbase, priv->ecc_saddr_offset, &err_addr); - if (priv->ecc_uecnt_offset) + if (priv->ecc_cecnt_offset) regmap_read(drvdata->mc_vbase, priv->ecc_cecnt_offset, &err_count); edac_mc_handle_error(HW_EVENT_ERR_CORRECTED, mci, err_count,

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Clark Wang xiaoning.wang@nxp.com

commit b1852c5de2f2a37dd4462f7837c9e3e678f9e546 upstream.

Deferred probe with pm_runtime_put() may delay clock disable, causing incorrect clock usage count. Use pm_runtime_put_sync() to ensure the clock is disabled immediately.

Fixes: 13d6eb20fc79 ("i2c: imx-lpi2c: add runtime pm support") Signed-off-by: Clark Wang xiaoning.wang@nxp.com Signed-off-by: Carlos Song carlos.song@nxp.com Cc: stable@vger.kernel.org # v4.16+ Link: https://lore.kernel.org/r/20250421062341.2471922-1-carlos.song@nxp.com Signed-off-by: Andi Shyti andi.shyti@kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/i2c/busses/i2c-imx-lpi2c.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/drivers/i2c/busses/i2c-imx-lpi2c.c +++ b/drivers/i2c/busses/i2c-imx-lpi2c.c @@ -616,9 +616,9 @@ static int lpi2c_imx_probe(struct platfo return 0;

rpm_disable: - pm_runtime_put(&pdev->dev); - pm_runtime_disable(&pdev->dev); pm_runtime_dont_use_autosuspend(&pdev->dev); + pm_runtime_put_sync(&pdev->dev); + pm_runtime_disable(&pdev->dev);

return ret; }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Helge Deller deller@gmx.de

commit de3629baf5a33af1919dec7136d643b0662e85ef upstream.

Camm noticed that on parisc a SIGFPE exception will crash an application with a second SIGFPE in the signal handler. Dave analyzed it, and it happens because glibc uses a double-word floating-point store to atomically update function descriptors. As a result of lazy binding, we hit a floating-point store in fpe_func almost immediately.

When the T bit is set, an assist exception trap occurs when when the co-processor encounters *any* floating-point instruction except for a double store of register %fr0. The latter cancels all pending traps. Let's fix this by clearing the Trap (T) bit in the FP status register before returning to the signal handler in userspace.

The issue can be reproduced with this test program:

root@parisc:~# cat fpe.c

static void fpe_func(int sig, siginfo_t *i, void *v) { sigset_t set; sigemptyset(&set); sigaddset(&set, SIGFPE); sigprocmask(SIG_UNBLOCK, &set, NULL); printf("GOT signal %d with si_code %ld\n", sig, i->si_code); }

int main() { struct sigaction action = { .sa_sigaction = fpe_func, .sa_flags = SA_RESTART|SA_SIGINFO }; sigaction(SIGFPE, &action, 0); feenableexcept(FE_OVERFLOW); return printf("%lf\n",1.7976931348623158E308*1.7976931348623158E308); }

root@parisc:~# gcc fpe.c -lm root@parisc:~# ./a.out Floating point exception

root@parisc:~# strace -f ./a.out execve("./a.out", ["./a.out"], 0xf9ac7034 /* 20 vars */) = 0 getrlimit(RLIMIT_STACK, {rlim_cur=8192*1024, rlim_max=RLIM_INFINITY}) = 0 ... rt_sigaction(SIGFPE, {sa_handler=0x1110a, sa_mask=[], sa_flags=SA_RESTART|SA_SIGINFO}, NULL, 8) = 0 --- SIGFPE {si_signo=SIGFPE, si_code=FPE_FLTOVF, si_addr=0x1078f} --- --- SIGFPE {si_signo=SIGFPE, si_code=FPE_FLTOVF, si_addr=0xf8f21237} --- +++ killed by SIGFPE +++ Floating point exception

Signed-off-by: Helge Deller deller@gmx.de Suggested-by: John David Anglin dave.anglin@bell.net Reported-by: Camm Maguire camm@maguirefamily.org Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- arch/parisc/math-emu/driver.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-)

--- a/arch/parisc/math-emu/driver.c +++ b/arch/parisc/math-emu/driver.c @@ -103,9 +103,19 @@ handle_fpe(struct pt_regs *regs)

memcpy(regs->fr, frcopy, sizeof regs->fr); if (signalcode != 0) { - force_sig_fault(signalcode >> 24, signalcode & 0xffffff, - (void __user *) regs->iaoq[0]); - return -1; + int sig = signalcode >> 24; + + if (sig == SIGFPE) { + /* + * Clear floating point trap bit to avoid trapping + * again on the first floating-point instruction in + * the userspace signal handler. + */ + regs->fr[0] &= ~(1ULL << 38); + } + force_sig_fault(sig, signalcode & 0xffffff, + (void __user *) regs->iaoq[0]); + return -1; }

return signalcode ? -1 : 0;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vishal Badole Vishal.Badole@amd.com

commit f04dd30f1bef1ed2e74a4050af6e5e5e3869bac3 upstream.

According to the XGMAC specification, enabling features such as Layer 3 and Layer 4 Packet Filtering, Split Header and Virtualized Network support automatically selects the IPC Full Checksum Offload Engine on the receive side.

When RX checksum offload is disabled, these dependent features must also be disabled to prevent abnormal behavior caused by mismatched feature dependencies.

Ensure that toggling RX checksum offload (disabling or enabling) properly disables or enables all dependent features, maintaining consistent and expected behavior in the network device.

Cc: stable@vger.kernel.org Fixes: 1a510ccf5869 ("amd-xgbe: Add support for VXLAN offload capabilities") Signed-off-by: Vishal Badole Vishal.Badole@amd.com Reviewed-by: Simon Horman horms@kernel.org Link: https://patch.msgid.link/20250424130248.428865-1-Vishal.Badole@amd.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/net/ethernet/amd/xgbe/xgbe-desc.c | 9 +++++++-- drivers/net/ethernet/amd/xgbe/xgbe-dev.c | 24 ++++++++++++++++++++++-- drivers/net/ethernet/amd/xgbe/xgbe-drv.c | 11 +++++++++-- drivers/net/ethernet/amd/xgbe/xgbe.h | 4 ++++ 4 files changed, 42 insertions(+), 6 deletions(-)

--- a/drivers/net/ethernet/amd/xgbe/xgbe-desc.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-desc.c @@ -373,8 +373,13 @@ static int xgbe_map_rx_buffer(struct xgb }

/* Set up the header page info */ - xgbe_set_buffer_data(&rdata->rx.hdr, &ring->rx_hdr_pa, - XGBE_SKB_ALLOC_SIZE); + if (pdata->netdev->features & NETIF_F_RXCSUM) { + xgbe_set_buffer_data(&rdata->rx.hdr, &ring->rx_hdr_pa, + XGBE_SKB_ALLOC_SIZE); + } else { + xgbe_set_buffer_data(&rdata->rx.hdr, &ring->rx_hdr_pa, + pdata->rx_buf_size); + }

/* Set up the buffer page info */ xgbe_set_buffer_data(&rdata->rx.buf, &ring->rx_buf_pa, --- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c @@ -320,6 +320,18 @@ static void xgbe_config_sph_mode(struct XGMAC_IOWRITE_BITS(pdata, MAC_RCR, HDSMS, XGBE_SPH_HDSMS_SIZE); }

+static void xgbe_disable_sph_mode(struct xgbe_prv_data *pdata) +{ + unsigned int i; + + for (i = 0; i < pdata->channel_count; i++) { + if (!pdata->channel[i]->rx_ring) + break; + + XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_CR, SPH, 0); + } +} + static int xgbe_write_rss_reg(struct xgbe_prv_data *pdata, unsigned int type, unsigned int index, unsigned int val) { @@ -3545,8 +3557,12 @@ static int xgbe_init(struct xgbe_prv_dat xgbe_config_tx_coalesce(pdata); xgbe_config_rx_buffer_size(pdata); xgbe_config_tso_mode(pdata); - xgbe_config_sph_mode(pdata); - xgbe_config_rss(pdata); + + if (pdata->netdev->features & NETIF_F_RXCSUM) { + xgbe_config_sph_mode(pdata); + xgbe_config_rss(pdata); + } + desc_if->wrapper_tx_desc_init(pdata); desc_if->wrapper_rx_desc_init(pdata); xgbe_enable_dma_interrupts(pdata); @@ -3702,5 +3718,9 @@ void xgbe_init_function_ptrs_dev(struct hw_if->disable_vxlan = xgbe_disable_vxlan; hw_if->set_vxlan_id = xgbe_set_vxlan_id;

+ /* For Split Header*/ + hw_if->enable_sph = xgbe_config_sph_mode; + hw_if->disable_sph = xgbe_disable_sph_mode; + DBGPR("<--xgbe_init_function_ptrs\n"); } --- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c @@ -2257,10 +2257,17 @@ static int xgbe_set_features(struct net_ if (ret) return ret;

- if ((features & NETIF_F_RXCSUM) && !rxcsum) + if ((features & NETIF_F_RXCSUM) && !rxcsum) { + hw_if->enable_sph(pdata); + hw_if->enable_vxlan(pdata); hw_if->enable_rx_csum(pdata); - else if (!(features & NETIF_F_RXCSUM) && rxcsum) + schedule_work(&pdata->restart_work); + } else if (!(features & NETIF_F_RXCSUM) && rxcsum) { + hw_if->disable_sph(pdata); + hw_if->disable_vxlan(pdata); hw_if->disable_rx_csum(pdata); + schedule_work(&pdata->restart_work); + }

if ((features & NETIF_F_HW_VLAN_CTAG_RX) && !rxvlan) hw_if->enable_rx_vlan_stripping(pdata); --- a/drivers/net/ethernet/amd/xgbe/xgbe.h +++ b/drivers/net/ethernet/amd/xgbe/xgbe.h @@ -865,6 +865,10 @@ struct xgbe_hw_if { void (*enable_vxlan)(struct xgbe_prv_data *); void (*disable_vxlan)(struct xgbe_prv_data *); void (*set_vxlan_id)(struct xgbe_prv_data *); + + /* For Split Header */ + void (*enable_sph)(struct xgbe_prv_data *pdata); + void (*disable_sph)(struct xgbe_prv_data *pdata); };

/* This structure represents implementation specific routines for an

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wei Yang richard.weiyang@gmail.com

commit 06eaa824fd239edd1eab2754f29b2d03da313003 upstream.

The second parameter of memblock_set_node() is size instead of end.

Since it iterates from lower address to higher address, finally the node id is correct. But during the process, some of them are wrong.

Pass size instead of end.

Fixes: 61167ad5fecd ("mm: pass nid to reserve_bootmem_region()") Signed-off-by: Wei Yang richard.weiyang@gmail.com CC: Mike Rapoport rppt@kernel.org CC: Yajun Deng yajun.deng@linux.dev CC: stable@vger.kernel.org Reviewed-by: Anshuman Khandual anshuman.khandual@arm.com Link: https://lore.kernel.org/r/20250318071948.23854-2-richard.weiyang@gmail.com Signed-off-by: Mike Rapoport (Microsoft) rppt@kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- mm/memblock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/mm/memblock.c +++ b/mm/memblock.c @@ -2135,7 +2135,7 @@ static void __init memmap_init_reserved_ if (memblock_is_nomap(region)) reserve_bootmem_region(start, end, nid);

- memblock_set_node(start, end, &memblock.reserved, nid); + memblock_set_node(start, region->size, &memblock.reserved, nid); }

/* initialize struct pages for the reserved regions */

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ruslan Piasetskyi ruslan.piasetskyi@gmail.com

commit 649b50a82f09fa44c2f7a65618e4584072145ab7 upstream.

After moving tmio_mmc_host_probe down, error handling has to be adjusted.

Fixes: 74f45de394d9 ("mmc: renesas_sdhi: register irqs before registering controller") Reviewed-by: Ihar Salauyou salauyou.ihar@gmail.com Signed-off-by: Ruslan Piasetskyi ruslan.piasetskyi@gmail.com Reviewed-by: Geert Uytterhoeven geert+renesas@glider.be Reviewed-by: Wolfram Sang wsa+renesas@sang-engineering.com Tested-by: Wolfram Sang wsa+renesas@sang-engineering.com Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250326220638.460083-1-ruslan.piasetskyi@gmail.co... Signed-off-by: Ulf Hansson ulf.hansson@linaro.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/mmc/host/renesas_sdhi_core.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-)

--- a/drivers/mmc/host/renesas_sdhi_core.c +++ b/drivers/mmc/host/renesas_sdhi_core.c @@ -1107,26 +1107,26 @@ int renesas_sdhi_probe(struct platform_d num_irqs = platform_irq_count(pdev); if (num_irqs < 0) { ret = num_irqs; - goto eirq; + goto edisclk; }

/* There must be at least one IRQ source */ if (!num_irqs) { ret = -ENXIO; - goto eirq; + goto edisclk; }

for (i = 0; i < num_irqs; i++) { irq = platform_get_irq(pdev, i); if (irq < 0) { ret = irq; - goto eirq; + goto edisclk; }

ret = devm_request_irq(&pdev->dev, irq, tmio_mmc_irq, 0, dev_name(&pdev->dev), host); if (ret) - goto eirq; + goto edisclk; }

ret = tmio_mmc_host_probe(host); @@ -1138,8 +1138,6 @@ int renesas_sdhi_probe(struct platform_d

return ret;

-eirq: - tmio_mmc_host_remove(host); edisclk: renesas_sdhi_clk_disable(host); efree:

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Steven Rostedt rostedt@goodmis.org

commit 0a8f11f8569e7ed16cbcedeb28c4350f6378fea6 upstream.

On some paths in print_event_fields() it takes the trace_event_sem for read, even though it should always be held when the function is called.

Remove the taking of that mutex and add a lockdep_assert_held_read() to make sure the trace_event_sem is held when print_event_fields() is called.

Cc: stable@vger.kernel.org Cc: Masami Hiramatsu mhiramat@kernel.org Cc: Mathieu Desnoyers mathieu.desnoyers@efficios.com Link: https://lore.kernel.org/20250501224128.0b1f0571@batman.local.home Fixes: 80a76994b2d88 ("tracing: Add "fields" option to show raw trace event fields") Reported-by: syzbot+441582c1592938fccf09@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/6813ff5e.050a0220.14dd7d.001b.GAE@google.com/ Signed-off-by: Steven Rostedt (Google) rostedt@goodmis.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- kernel/trace/trace_output.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -950,11 +950,12 @@ enum print_line_t print_event_fields(str struct trace_event_call *call; struct list_head *head;

+ lockdep_assert_held_read(&trace_event_sem); + /* ftrace defined events have separate call structures */ if (event->type <= __TRACE_LAST_TYPE) { bool found = false;

- down_read(&trace_event_sem); list_for_each_entry(call, &ftrace_events, list) { if (call->event.type == event->type) { found = true; @@ -964,7 +965,6 @@ enum print_line_t print_event_fields(str if (call->event.type > __TRACE_LAST_TYPE) break; } - up_read(&trace_event_sem); if (!found) { trace_seq_printf(&iter->seq, "UNKNOWN TYPE %d\n", event->type); goto out;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: LongPing Wei weilongping@oppo.com

commit a3d8f0a7f5e8b193db509c7191fefeed3533fc44 upstream.

A BUG was reported as below when CONFIG_DEBUG_ATOMIC_SLEEP and try_verify_in_tasklet are enabled. [ 129.444685][ T934] BUG: sleeping function called from invalid context at drivers/md/dm-bufio.c:2421 [ 129.444723][ T934] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 934, name: kworker/1:4 [ 129.444740][ T934] preempt_count: 201, expected: 0 [ 129.444756][ T934] RCU nest depth: 0, expected: 0 [ 129.444781][ T934] Preemption disabled at: [ 129.444789][ T934] [<ffffffd816231900>] shrink_work+0x21c/0x248 [ 129.445167][ T934] kernel BUG at kernel/sched/walt/walt_debug.c:16! [ 129.445183][ T934] Internal error: Oops - BUG: 00000000f2000800 [#1] PREEMPT SMP [ 129.445204][ T934] Skip md ftrace buffer dump for: 0x1609e0 [ 129.447348][ T934] CPU: 1 PID: 934 Comm: kworker/1:4 Tainted: G W OE 6.6.56-android15-8-o-g6f82312b30b9-debug #1 1400000003000000474e5500b3187743670464e8 [ 129.447362][ T934] Hardware name: Qualcomm Technologies, Inc. Parrot QRD, Alpha-M (DT) [ 129.447373][ T934] Workqueue: dm_bufio_cache shrink_work [ 129.447394][ T934] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 129.447406][ T934] pc : android_rvh_schedule_bug+0x0/0x8 [sched_walt_debug] [ 129.447435][ T934] lr : __traceiter_android_rvh_schedule_bug+0x44/0x6c [ 129.447451][ T934] sp : ffffffc0843dbc90 [ 129.447459][ T934] x29: ffffffc0843dbc90 x28: ffffffffffffffff x27: 0000000000000c8b [ 129.447479][ T934] x26: 0000000000000040 x25: ffffff804b3d6260 x24: ffffffd816232b68 [ 129.447497][ T934] x23: ffffff805171c5b4 x22: 0000000000000000 x21: ffffffd816231900 [ 129.447517][ T934] x20: ffffff80306ba898 x19: 0000000000000000 x18: ffffffc084159030 [ 129.447535][ T934] x17: 00000000d2b5dd1f x16: 00000000d2b5dd1f x15: ffffffd816720358 [ 129.447554][ T934] x14: 0000000000000004 x13: ffffff89ef978000 x12: 0000000000000003 [ 129.447572][ T934] x11: ffffffd817a823c4 x10: 0000000000000202 x9 : 7e779c5735de9400 [ 129.447591][ T934] x8 : ffffffd81560d004 x7 : 205b5d3938373434 x6 : ffffffd8167397c8 [ 129.447610][ T934] x5 : 0000000000000000 x4 : 0000000000000001 x3 : ffffffc0843db9e0 [ 129.447629][ T934] x2 : 0000000000002f15 x1 : 0000000000000000 x0 : 0000000000000000 [ 129.447647][ T934] Call trace: [ 129.447655][ T934] android_rvh_schedule_bug+0x0/0x8 [sched_walt_debug 1400000003000000474e550080cce8a8a78606b6] [ 129.447681][ T934] __might_resched+0x190/0x1a8 [ 129.447694][ T934] shrink_work+0x180/0x248 [ 129.447706][ T934] process_one_work+0x260/0x624 [ 129.447718][ T934] worker_thread+0x28c/0x454 [ 129.447729][ T934] kthread+0x118/0x158 [ 129.447742][ T934] ret_from_fork+0x10/0x20 [ 129.447761][ T934] Code: ???????? ???????? ???????? d2b5dd1f (d4210000) [ 129.447772][ T934] ---[ end trace 0000000000000000 ]---

dm_bufio_lock will call spin_lock_bh when try_verify_in_tasklet is enabled, and __scan will be called in atomic context.

Fixes: 7cd326747f46 ("dm bufio: remove dm_bufio_cond_resched()") Signed-off-by: LongPing Wei weilongping@oppo.com Cc: stable@vger.kernel.org Signed-off-by: Mikulas Patocka mpatocka@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/md/dm-bufio.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)

--- a/drivers/md/dm-bufio.c +++ b/drivers/md/dm-bufio.c @@ -68,6 +68,8 @@ #define LIST_DIRTY 1 #define LIST_SIZE 2

+#define SCAN_RESCHED_CYCLE 16 + /*--------------------------------------------------------------*/

/* @@ -2387,7 +2389,12 @@ static void __scan(struct dm_bufio_clien

atomic_long_dec(&c->need_shrink); freed++; - cond_resched(); + + if (unlikely(freed % SCAN_RESCHED_CYCLE == 0)) { + dm_bufio_unlock(c); + cond_resched(); + dm_bufio_lock(c); + } } } }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Benjamin Marzinski bmarzins@redhat.com

commit 5a2a6c428190f945c5cbf5791f72dbea83e97f66 upstream.

realloc_argv() was only updating the array size if it was called with old_argv already allocated. The first time it was called to create an argv array, it would allocate the array but return the array size as zero. dm_split_args() would think that it couldn't store any arguments in the array and would call realloc_argv() again, causing it to reallocate the initial slots (this time using GPF_KERNEL) and finally return a size. Aside from being wasteful, this could cause deadlocks on targets that need to process messages without starting new IO. Instead, realloc_argv should always update the allocated array size on success.

Fixes: a0651926553c ("dm table: don't copy from a NULL pointer in realloc_argv()") Cc: stable@vger.kernel.org Signed-off-by: Benjamin Marzinski bmarzins@redhat.com Signed-off-by: Mikulas Patocka mpatocka@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/md/dm-table.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)

--- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -500,9 +500,10 @@ static char **realloc_argv(unsigned int gfp = GFP_NOIO; } argv = kmalloc_array(new_size, sizeof(*argv), gfp); - if (argv && old_argv) { - memcpy(argv, old_argv, *size * sizeof(*argv)); + if (argv) { *size = new_size; + if (old_argv) + memcpy(argv, old_argv, *size * sizeof(*argv)); }

kfree(old_argv);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mingcong Bai jeffbai@aosc.io

commit 2c8a7c66c90832432496616a9a3c07293f1364f3 upstream.

On the Lenovo ThinkPad X201, when Intel VT-d is enabled in the BIOS, the kernel boots with errors related to DMAR, the graphical interface appeared quite choppy, and the system resets erratically within a minute after it booted:

DMAR: DRHD: handling fault status reg 3 DMAR: [DMA Write NO_PASID] Request device [00:02.0] fault addr 0xb97ff000 [fault reason 0x05] PTE Write access is not set

Upon comparing boot logs with VT-d on/off, I found that the Intel Calpella quirk (`quirk_calpella_no_shadow_gtt()') correctly applied the igfx IOMMU disable/quirk correctly:

pci 0000:00:00.0: DMAR: BIOS has allocated no shadow GTT; disabling IOMMU for graphics

Whereas with VT-d on, it went into the "else" branch, which then triggered the DMAR handling fault above:

... else if (!disable_igfx_iommu) { /* we have to ensure the gfx device is idle before we flush */ pci_info(dev, "Disabling batched IOTLB flush on Ironlake\n"); iommu_set_dma_strict(); }

Now, this is not exactly scientific, but moving 0x0044 to quirk_iommu_igfx seems to have fixed the aforementioned issue. Running a few `git blame' runs on the function, I have found that the quirk was originally introduced as a fix specific to ThinkPad X201:

commit 9eecabcb9a92 ("intel-iommu: Abort IOMMU setup for igfx if BIOS gave no shadow GTT space")

Which was later revised twice to the "else" branch we saw above:

- 2011: commit 6fbcfb3e467a ("intel-iommu: Workaround IOTLB hang on Ironlake GPU") - 2024: commit ba00196ca41c ("iommu/vt-d: Decouple igfx_off from graphic identity mapping")

I'm uncertain whether further testings on this particular laptops were done in 2011 and (honestly I'm not sure) 2024, but I would be happy to do some distro-specific testing if that's what would be required to verify this patch.

P.S., I also see IDs 0x0040, 0x0062, and 0x006a listed under the same `quirk_calpella_no_shadow_gtt()' quirk, but I'm not sure how similar these chipsets are (if they share the same issue with VT-d or even, indeed, if this issue is specific to a bug in the Lenovo BIOS). With regards to 0x0062, it seems to be a Centrino wireless card, but not a chipset?

I have also listed a couple (distro and kernel) bug reports below as references (some of them are from 7-8 years ago!), as they seem to be similar issue found on different Westmere/Ironlake, Haswell, and Broadwell hardware setups.

Cc: stable@vger.kernel.org Fixes: 6fbcfb3e467a ("intel-iommu: Workaround IOTLB hang on Ironlake GPU") Fixes: ba00196ca41c ("iommu/vt-d: Decouple igfx_off from graphic identity mapping") Link: https://groups.google.com/g/qubes-users/c/4NP4goUds2c?pli=1 Link: https://bugs.archlinux.org/task/65362 Link: https://bbs.archlinux.org/viewtopic.php?id=230323 Reported-by: Wenhao Sun weiguangtwk@outlook.com Closes: https://bugzilla.kernel.org/show_bug.cgi?id=197029 Signed-off-by: Mingcong Bai jeffbai@aosc.io Link: https://lore.kernel.org/r/20250415133330.12528-1-jeffbai@aosc.io Signed-off-by: Lu Baolu baolu.lu@linux.intel.com Signed-off-by: Joerg Roedel jroedel@suse.de Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/iommu/intel/iommu.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)

--- a/drivers/iommu/intel/iommu.c +++ b/drivers/iommu/intel/iommu.c @@ -4855,6 +4855,9 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_I DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e40, quirk_iommu_igfx); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e90, quirk_iommu_igfx);

+/* QM57/QS57 integrated gfx malfunctions with dmar */ +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0044, quirk_iommu_igfx); + /* Broadwell igfx malfunctions with dmar */ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x1606, quirk_iommu_igfx); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x160B, quirk_iommu_igfx); @@ -4932,7 +4935,6 @@ static void quirk_calpella_no_shadow_gtt } } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0040, quirk_calpella_no_shadow_gtt); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0044, quirk_calpella_no_shadow_gtt); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0062, quirk_calpella_no_shadow_gtt); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x006a, quirk_calpella_no_shadow_gtt);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shouye Liu shouyeliu@tencent.com

commit 8d6955ed76e8a47115f2ea1d9c263ee6f505d737 upstream.

In certain situations, the sysfs for uncore may not be present when all CPUs in a package are offlined and then brought back online after boot.

This issue can occur if there is an error in adding the sysfs entry due to a memory allocation failure. Retrying to bring the CPUs online will not resolve the issue, as the uncore_cpu_mask is already set for the package before the failure condition occurs.

This issue does not occur if the failure happens during module initialization, as the module will fail to load in the event of any error.

To address this, ensure that the uncore_cpu_mask is not set until the successful return of uncore_freq_add_entry().

Fixes: dbce412a7733 ("platform/x86/intel-uncore-freq: Split common and enumeration part") Signed-off-by: Shouye Liu shouyeliu@tencent.com Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250417032321.75580-1-shouyeliu@gmail.com Reviewed-by: Ilpo Järvinen ilpo.jarvinen@linux.intel.com Signed-off-by: Ilpo Järvinen ilpo.jarvinen@linux.intel.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/platform/x86/intel/uncore-frequency/uncore-frequency.c | 13 ++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)

--- a/drivers/platform/x86/intel/uncore-frequency/uncore-frequency.c +++ b/drivers/platform/x86/intel/uncore-frequency/uncore-frequency.c @@ -121,15 +121,13 @@ static int uncore_event_cpu_online(unsig { struct uncore_data *data; int target; + int ret;

/* Check if there is an online cpu in the package for uncore MSR */ target = cpumask_any_and(&uncore_cpu_mask, topology_die_cpumask(cpu)); if (target < nr_cpu_ids) return 0;

- /* Use this CPU on this die as a control CPU */ - cpumask_set_cpu(cpu, &uncore_cpu_mask); - data = uncore_get_instance(cpu); if (!data) return 0; @@ -138,7 +136,14 @@ static int uncore_event_cpu_online(unsig data->die_id = topology_die_id(cpu); data->domain_id = UNCORE_DOMAIN_ID_INVALID;

- return uncore_freq_add_entry(data, cpu); + ret = uncore_freq_add_entry(data, cpu); + if (ret) + return ret; + + /* Use this CPU on this die as a control CPU */ + cpumask_set_cpu(cpu, &uncore_cpu_mask); + + return 0; }

static int uncore_event_cpu_offline(unsigned int cpu)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jethro Donaldson devel@jro.nz

commit 74c72419ec8da5cbc9c49410d3c44bb954538bdd upstream.

SMB create requests issued via smb311_posix_mkdir() have an incorrect length of zero bytes for the POSIX create context data. ksmbd server rejects such requests and logs "cli req too short" causing mkdir to fail with "invalid argument" on the client side. It also causes subsequent rmmod to crash in cifs_destroy_request_bufs()

Inspection of packets sent by cifs.ko using wireshark show valid data for the SMB2_POSIX_CREATE_CONTEXT is appended with the correct offset, but with an incorrect length of zero bytes. Fails with ksmbd+cifs.ko only as Windows server/client does not use POSIX extensions.

Fix smb311_posix_mkdir() to set req->CreateContextsLength as part of appending the POSIX creation context to the request.

Signed-off-by: Jethro Donaldson devel@jro.nz Acked-by: Paulo Alcantara (Red Hat) pc@manguebit.com Reviewed-by: Namjae Jeon linkinjeon@kernel.org Cc: stable@vger.kernel.org Signed-off-by: Steve French stfrench@microsoft.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- fs/smb/client/smb2pdu.c | 1 + 1 file changed, 1 insertion(+)

--- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -2932,6 +2932,7 @@ replay_again: req->CreateContextsOffset = cpu_to_le32( sizeof(struct smb2_create_req) + iov[1].iov_len); + le32_add_cpu(&req->CreateContextsLength, iov[n_iov-1].iov_len); pc_buf = iov[n_iov-1].iov_base; }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafael J. Wysocki rafael.j.wysocki@intel.com

commit b79028039f440e7d2c4df6ab243060c4e3803e84 upstream.

Commit 7491cdf46b5c ("cpufreq: Avoid using inconsistent policy->min and policy->max") overlooked the fact that policy->min and policy->max were accessed directly in cpufreq_frequency_table_target() and in the functions called by it. Consequently, the changes made by that commit led to problems with setting policy limits.

Address this by passing the target frequency limits to __resolve_freq() and cpufreq_frequency_table_target() and propagating them to the functions called by the latter.

Fixes: 7491cdf46b5c ("cpufreq: Avoid using inconsistent policy->min and policy->max") Cc: 5.16+ stable@vger.kernel.org # 5.16+ Closes: https://lore.kernel.org/linux-pm/aAplED3IA_J0eZN0@linaro.org/ Reported-by: Stephan Gerhold stephan.gerhold@linaro.org Signed-off-by: Rafael J. Wysocki rafael.j.wysocki@intel.com Tested-by: Stephan Gerhold stephan.gerhold@linaro.org Reviewed-by: Lifeng Zheng zhenglifeng1@huawei.com Link: https://patch.msgid.link/5896780.DvuYhMxLoT@rjwysocki.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/cpufreq/cpufreq.c | 22 ++++++--- drivers/cpufreq/cpufreq_ondemand.c | 3 - drivers/cpufreq/freq_table.c | 6 +- include/linux/cpufreq.h | 83 ++++++++++++++++++++++++------------- 4 files changed, 73 insertions(+), 41 deletions(-)

--- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -534,14 +534,18 @@ void cpufreq_disable_fast_switch(struct EXPORT_SYMBOL_GPL(cpufreq_disable_fast_switch);

static unsigned int __resolve_freq(struct cpufreq_policy *policy, - unsigned int target_freq, unsigned int relation) + unsigned int target_freq, + unsigned int min, unsigned int max, + unsigned int relation) { unsigned int idx;

+ target_freq = clamp_val(target_freq, min, max); + if (!policy->freq_table) return target_freq;

- idx = cpufreq_frequency_table_target(policy, target_freq, relation); + idx = cpufreq_frequency_table_target(policy, target_freq, min, max, relation); policy->cached_resolved_idx = idx; policy->cached_target_freq = target_freq; return policy->freq_table[idx].frequency; @@ -575,8 +579,7 @@ unsigned int cpufreq_driver_resolve_freq if (unlikely(min > max)) min = max;

- return __resolve_freq(policy, clamp_val(target_freq, min, max), - CPUFREQ_RELATION_LE); + return __resolve_freq(policy, target_freq, min, max, CPUFREQ_RELATION_LE); } EXPORT_SYMBOL_GPL(cpufreq_driver_resolve_freq);

@@ -2348,8 +2351,8 @@ int __cpufreq_driver_target(struct cpufr if (cpufreq_disabled()) return -ENODEV;

- target_freq = clamp_val(target_freq, policy->min, policy->max); - target_freq = __resolve_freq(policy, target_freq, relation); + target_freq = __resolve_freq(policy, target_freq, policy->min, + policy->max, relation);

pr_debug("target for CPU %u: %u kHz, relation %u, requested %u kHz\n", policy->cpu, target_freq, relation, old_target_freq); @@ -2644,8 +2647,11 @@ static int cpufreq_set_policy(struct cpu * compiler optimizations around them because they may be accessed * concurrently by cpufreq_driver_resolve_freq() during the update. */ - WRITE_ONCE(policy->max, __resolve_freq(policy, new_data.max, CPUFREQ_RELATION_H)); - new_data.min = __resolve_freq(policy, new_data.min, CPUFREQ_RELATION_L); + WRITE_ONCE(policy->max, __resolve_freq(policy, new_data.max, + new_data.min, new_data.max, + CPUFREQ_RELATION_H)); + new_data.min = __resolve_freq(policy, new_data.min, new_data.min, + new_data.max, CPUFREQ_RELATION_L); WRITE_ONCE(policy->min, new_data.min > policy->max ? policy->max : new_data.min);

trace_cpu_frequency_limits(policy); --- a/drivers/cpufreq/cpufreq_ondemand.c +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -77,7 +77,8 @@ static unsigned int generic_powersave_bi return freq_next; }

- index = cpufreq_frequency_table_target(policy, freq_next, relation); + index = cpufreq_frequency_table_target(policy, freq_next, policy->min, + policy->max, relation); freq_req = freq_table[index].frequency; freq_reduc = freq_req * od_tuners->powersave_bias / 1000; freq_avg = freq_req - freq_reduc; --- a/drivers/cpufreq/freq_table.c +++ b/drivers/cpufreq/freq_table.c @@ -116,8 +116,8 @@ int cpufreq_generic_frequency_table_veri EXPORT_SYMBOL_GPL(cpufreq_generic_frequency_table_verify);

int cpufreq_table_index_unsorted(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) + unsigned int target_freq, unsigned int min, + unsigned int max, unsigned int relation) { struct cpufreq_frequency_table optimal = { .driver_data = ~0, @@ -148,7 +148,7 @@ int cpufreq_table_index_unsorted(struct cpufreq_for_each_valid_entry_idx(pos, table, i) { freq = pos->frequency;

- if ((freq < policy->min) || (freq > policy->max)) + if (freq < min || freq > max) continue; if (freq == target_freq) { optimal.driver_data = i; --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -787,8 +787,8 @@ int cpufreq_frequency_table_verify(struc int cpufreq_generic_frequency_table_verify(struct cpufreq_policy_data *policy);

int cpufreq_table_index_unsorted(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation); + unsigned int target_freq, unsigned int min, + unsigned int max, unsigned int relation); int cpufreq_frequency_table_get_index(struct cpufreq_policy *policy, unsigned int freq);

@@ -853,12 +853,12 @@ static inline int cpufreq_table_find_ind return best; }

-/* Works only on sorted freq-tables */ -static inline int cpufreq_table_find_index_l(struct cpufreq_policy *policy, - unsigned int target_freq, - bool efficiencies) +static inline int find_index_l(struct cpufreq_policy *policy, + unsigned int target_freq, + unsigned int min, unsigned int max, + bool efficiencies) { - target_freq = clamp_val(target_freq, policy->min, policy->max); + target_freq = clamp_val(target_freq, min, max);

if (policy->freq_table_sorted == CPUFREQ_TABLE_SORTED_ASCENDING) return cpufreq_table_find_index_al(policy, target_freq, @@ -868,6 +868,14 @@ static inline int cpufreq_table_find_ind efficiencies); }

+/* Works only on sorted freq-tables */ +static inline int cpufreq_table_find_index_l(struct cpufreq_policy *policy, + unsigned int target_freq, + bool efficiencies) +{ + return find_index_l(policy, target_freq, policy->min, policy->max, efficiencies); +} + /* Find highest freq at or below target in a table in ascending order */ static inline int cpufreq_table_find_index_ah(struct cpufreq_policy *policy, unsigned int target_freq, @@ -921,12 +929,12 @@ static inline int cpufreq_table_find_ind return best; }

-/* Works only on sorted freq-tables */ -static inline int cpufreq_table_find_index_h(struct cpufreq_policy *policy, - unsigned int target_freq, - bool efficiencies) +static inline int find_index_h(struct cpufreq_policy *policy, + unsigned int target_freq, + unsigned int min, unsigned int max, + bool efficiencies) { - target_freq = clamp_val(target_freq, policy->min, policy->max); + target_freq = clamp_val(target_freq, min, max);

if (policy->freq_table_sorted == CPUFREQ_TABLE_SORTED_ASCENDING) return cpufreq_table_find_index_ah(policy, target_freq, @@ -936,6 +944,14 @@ static inline int cpufreq_table_find_ind efficiencies); }

+/* Works only on sorted freq-tables */ +static inline int cpufreq_table_find_index_h(struct cpufreq_policy *policy, + unsigned int target_freq, + bool efficiencies) +{ + return find_index_h(policy, target_freq, policy->min, policy->max, efficiencies); +} + /* Find closest freq to target in a table in ascending order */ static inline int cpufreq_table_find_index_ac(struct cpufreq_policy *policy, unsigned int target_freq, @@ -1006,12 +1022,12 @@ static inline int cpufreq_table_find_ind return best; }

-/* Works only on sorted freq-tables */ -static inline int cpufreq_table_find_index_c(struct cpufreq_policy *policy, - unsigned int target_freq, - bool efficiencies) +static inline int find_index_c(struct cpufreq_policy *policy, + unsigned int target_freq, + unsigned int min, unsigned int max, + bool efficiencies) { - target_freq = clamp_val(target_freq, policy->min, policy->max); + target_freq = clamp_val(target_freq, min, max);

if (policy->freq_table_sorted == CPUFREQ_TABLE_SORTED_ASCENDING) return cpufreq_table_find_index_ac(policy, target_freq, @@ -1021,7 +1037,17 @@ static inline int cpufreq_table_find_ind efficiencies); }

-static inline bool cpufreq_is_in_limits(struct cpufreq_policy *policy, int idx) +/* Works only on sorted freq-tables */ +static inline int cpufreq_table_find_index_c(struct cpufreq_policy *policy, + unsigned int target_freq, + bool efficiencies) +{ + return find_index_c(policy, target_freq, policy->min, policy->max, efficiencies); +} + +static inline bool cpufreq_is_in_limits(struct cpufreq_policy *policy, + unsigned int min, unsigned int max, + int idx) { unsigned int freq;

@@ -1030,11 +1056,13 @@ static inline bool cpufreq_is_in_limits(

freq = policy->freq_table[idx].frequency;

- return freq == clamp_val(freq, policy->min, policy->max); + return freq == clamp_val(freq, min, max); }

static inline int cpufreq_frequency_table_target(struct cpufreq_policy *policy, unsigned int target_freq, + unsigned int min, + unsigned int max, unsigned int relation) { bool efficiencies = policy->efficiencies_available && @@ -1045,29 +1073,26 @@ static inline int cpufreq_frequency_tabl relation &= ~CPUFREQ_RELATION_E;

if (unlikely(policy->freq_table_sorted == CPUFREQ_TABLE_UNSORTED)) - return cpufreq_table_index_unsorted(policy, target_freq, - relation); + return cpufreq_table_index_unsorted(policy, target_freq, min, + max, relation); retry: switch (relation) { case CPUFREQ_RELATION_L: - idx = cpufreq_table_find_index_l(policy, target_freq, - efficiencies); + idx = find_index_l(policy, target_freq, min, max, efficiencies); break; case CPUFREQ_RELATION_H: - idx = cpufreq_table_find_index_h(policy, target_freq, - efficiencies); + idx = find_index_h(policy, target_freq, min, max, efficiencies); break; case CPUFREQ_RELATION_C: - idx = cpufreq_table_find_index_c(policy, target_freq, - efficiencies); + idx = find_index_c(policy, target_freq, min, max, efficiencies); break; default: WARN_ON_ONCE(1); return 0; }

- /* Limit frequency index to honor policy->min/max */ - if (!cpufreq_is_in_limits(policy, idx) && efficiencies) { + /* Limit frequency index to honor min and max */ + if (!cpufreq_is_in_limits(policy, min, max, idx) && efficiencies) { efficiencies = false; goto retry; }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eduard Zingerman eddyz87@gmail.com

commit 27e88bc4df1d80888fe1aaca786a7cc6e69587e2 upstream.

Add a utility function, looking for a subprogram containing a given instruction index, rewrite find_subprog() to use this function.

Signed-off-by: Eduard Zingerman eddyz87@gmail.com Link: https://lore.kernel.org/r/20241210041100.1898468-2-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov ast@kernel.org Signed-off-by: Shung-Hsi Yu shung-hsi.yu@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- kernel/bpf/verifier.c | 28 ++++++++++++++++++++++++---- 1 file changed, 24 insertions(+), 4 deletions(-)

--- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -2636,16 +2636,36 @@ static int cmp_subprogs(const void *a, c ((struct bpf_subprog_info *)b)->start; }

+/* Find subprogram that contains instruction at 'off' */ +static struct bpf_subprog_info *find_containing_subprog(struct bpf_verifier_env *env, int off) +{ + struct bpf_subprog_info *vals = env->subprog_info; + int l, r, m; + + if (off >= env->prog->len || off < 0 || env->subprog_cnt == 0) + return NULL; + + l = 0; + r = env->subprog_cnt - 1; + while (l < r) { + m = l + (r - l + 1) / 2; + if (vals[m].start <= off) + l = m; + else + r = m - 1; + } + return &vals[l]; +} + +/* Find subprogram that starts exactly at 'off' */ static int find_subprog(struct bpf_verifier_env *env, int off) { struct bpf_subprog_info *p;

- p = bsearch(&off, env->subprog_info, env->subprog_cnt, - sizeof(env->subprog_info[0]), cmp_subprogs); - if (!p) + p = find_containing_subprog(env, off); + if (!p || p->start != off) return -ENOENT; return p - env->subprog_info; - }

static int add_subprog(struct bpf_verifier_env *env, int off)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eduard Zingerman eddyz87@gmail.com

commit 51081a3f25c742da5a659d7fc6fd77ebfdd555be upstream.

When processing calls to certain helpers, verifier invalidates all packet pointers in a current state. For example, consider the following program:

__attribute__((__noinline__)) long skb_pull_data(struct __sk_buff *sk, __u32 len) { return bpf_skb_pull_data(sk, len); }

SEC("tc") int test_invalidate_checks(struct __sk_buff *sk) { int *p = (void *)(long)sk->data; if ((void *)(p + 1) > (void *)(long)sk->data_end) return TCX_DROP; skb_pull_data(sk, 0); *p = 42; return TCX_PASS; }

After a call to bpf_skb_pull_data() the pointer 'p' can't be used safely. See function filter.c:bpf_helper_changes_pkt_data() for a list of such helpers.

At the moment verifier invalidates packet pointers when processing helper function calls, and does not traverse global sub-programs when processing calls to global sub-programs. This means that calls to helpers done from global sub-programs do not invalidate pointers in the caller state. E.g. the program above is unsafe, but is not rejected by verifier.

This commit fixes the omission by computing field bpf_subprog_info->changes_pkt_data for each sub-program before main verification pass. changes_pkt_data should be set if: - subprogram calls helper for which bpf_helper_changes_pkt_data returns true; - subprogram calls a global function, for which bpf_subprog_info->changes_pkt_data should be set.

The verifier.c:check_cfg() pass is modified to compute this information. The commit relies on depth first instruction traversal done by check_cfg() and absence of recursive function calls: - check_cfg() would eventually visit every call to subprogram S in a state when S is fully explored; - when S is fully explored: - every direct helper call within S is explored (and thus changes_pkt_data is set if needed); - every call to subprogram S1 called by S was visited with S1 fully explored (and thus S inherits changes_pkt_data from S1).

The downside of such approach is that dead code elimination is not taken into account: if a helper call inside global function is dead because of current configuration, verifier would conservatively assume that the call occurs for the purpose of the changes_pkt_data computation.

Reported-by: Nick Zavaritsky mejedi@gmail.com Closes: https://lore.kernel.org/bpf/0498CA22-5779-4767-9C0C-A9515CEA711F@gmail.com/ Signed-off-by: Eduard Zingerman eddyz87@gmail.com Link: https://lore.kernel.org/r/20241210041100.1898468-4-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov ast@kernel.org [shung-hsi.yu: do not use bitfield in "struct bpf_subprog_info" because commit 406a6fa44bfb ("bpf: use bitfields for simple per-subprog bool flags") is not present and minor context difference in check_func_call() because commit 491dd8edecbc ("bpf: Emit global subprog name in verifier logs") is not present. ] Signed-off-by: Shung-Hsi Yu shung-hsi.yu@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- include/linux/bpf_verifier.h | 1 + kernel/bpf/verifier.c | 32 +++++++++++++++++++++++++++++++- 2 files changed, 32 insertions(+), 1 deletion(-)

--- a/include/linux/bpf_verifier.h +++ b/include/linux/bpf_verifier.h @@ -573,6 +573,7 @@ struct bpf_subprog_info { bool tail_call_reachable; bool has_ld_abs; bool is_async_cb; + bool changes_pkt_data; };

struct bpf_verifier_env; --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -9364,6 +9364,8 @@ static int check_func_call(struct bpf_ve

if (env->log.level & BPF_LOG_LEVEL) verbose(env, "Func#%d is global and valid. Skipping.\n", subprog); + if (env->subprog_info[subprog].changes_pkt_data) + clear_all_pkt_pointers(env); clear_caller_saved_regs(env, caller->regs);

/* All global functions return a 64-bit SCALAR_VALUE */ @@ -15114,6 +15116,29 @@ static int check_return_code(struct bpf_ return 0; }

+static void mark_subprog_changes_pkt_data(struct bpf_verifier_env *env, int off) +{ + struct bpf_subprog_info *subprog; + + subprog = find_containing_subprog(env, off); + subprog->changes_pkt_data = true; +} + +/* 't' is an index of a call-site. + * 'w' is a callee entry point. + * Eventually this function would be called when env->cfg.insn_state[w] == EXPLORED. + * Rely on DFS traversal order and absence of recursive calls to guarantee that + * callee's change_pkt_data marks would be correct at that moment. + */ +static void merge_callee_effects(struct bpf_verifier_env *env, int t, int w) +{ + struct bpf_subprog_info *caller, *callee; + + caller = find_containing_subprog(env, t); + callee = find_containing_subprog(env, w); + caller->changes_pkt_data |= callee->changes_pkt_data; +} + /* non-recursive DFS pseudo code * 1 procedure DFS-iterative(G,v): * 2 label v as discovered @@ -15247,6 +15272,7 @@ static int visit_func_call_insn(int t, s bool visit_callee) { int ret, insn_sz; + int w;

insn_sz = bpf_is_ldimm64(&insns[t]) ? 2 : 1; ret = push_insn(t, t + insn_sz, FALLTHROUGH, env); @@ -15258,8 +15284,10 @@ static int visit_func_call_insn(int t, s mark_jmp_point(env, t + insn_sz);

if (visit_callee) { + w = t + insns[t].imm + 1; mark_prune_point(env, t); - ret = push_insn(t, t + insns[t].imm + 1, BRANCH, env); + merge_callee_effects(env, t, w); + ret = push_insn(t, w, BRANCH, env); } return ret; } @@ -15311,6 +15339,8 @@ static int visit_insn(int t, struct bpf_ mark_prune_point(env, t); mark_jmp_point(env, t); } + if (bpf_helper_call(insn) && bpf_helper_changes_pkt_data(insn->imm)) + mark_subprog_changes_pkt_data(env, t); if (insn->src_reg == BPF_PSEUDO_KFUNC_CALL) { struct bpf_kfunc_call_arg_meta meta;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eduard Zingerman eddyz87@gmail.com

commit 81f6d0530ba031b5f038a091619bf2ff29568852 upstream.

When processing calls to global sub-programs, verifier decides whether to invalidate all packet pointers in current state depending on the changes_pkt_data property of the global sub-program.

Because of this, an extension program replacing a global sub-program must be compatible with changes_pkt_data property of the sub-program being replaced.

This commit: - adds changes_pkt_data flag to struct bpf_prog_aux: - this flag is set in check_cfg() for main sub-program; - in jit_subprogs() for other sub-programs; - modifies bpf_check_attach_btf_id() to check changes_pkt_data flag; - moves call to check_attach_btf_id() after the call to check_cfg(), because it needs changes_pkt_data flag to be set:

bpf_check: ... ... - check_attach_btf_id resolve_pseudo_ldimm64 resolve_pseudo_ldimm64 --> bpf_prog_is_offloaded bpf_prog_is_offloaded check_cfg check_cfg + check_attach_btf_id ... ...

The following fields are set by check_attach_btf_id(): - env->ops - prog->aux->attach_btf_trace - prog->aux->attach_func_name - prog->aux->attach_func_proto - prog->aux->dst_trampoline - prog->aux->mod - prog->aux->saved_dst_attach_type - prog->aux->saved_dst_prog_type - prog->expected_attach_type

Neither of these fields are used by resolve_pseudo_ldimm64() or bpf_prog_offload_verifier_prep() (for netronome and netdevsim drivers), so the reordering is safe.

Suggested-by: Alexei Starovoitov alexei.starovoitov@gmail.com Signed-off-by: Eduard Zingerman eddyz87@gmail.com Link: https://lore.kernel.org/r/20241210041100.1898468-6-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov ast@kernel.org [ shung-hsi.yu: adapt to missing fields in "struct bpf_prog_aux". Context difference in jit_subprogs() because BPF Exception is not supported. Context difference in bpf_check() because commit 5b5f51bff1b6 "bpf: no_caller_saved_registers attribute for helper calls" is not present. ] Signed-off-by: Shung-Hsi Yu shung-hsi.yu@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- include/linux/bpf.h | 1 + kernel/bpf/verifier.c | 16 ++++++++++++---- 2 files changed, 13 insertions(+), 4 deletions(-)

--- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -1430,6 +1430,7 @@ struct bpf_prog_aux { bool sleepable; bool tail_call_reachable; bool xdp_has_frags; + bool changes_pkt_data; /* BTF_KIND_FUNC_PROTO for valid attach_btf_id */ const struct btf_type *attach_func_proto; /* function name for valid attach_btf_id */ --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -15462,6 +15462,7 @@ static int check_cfg(struct bpf_verifier } } ret = 0; /* cfg looks good */ + env->prog->aux->changes_pkt_data = env->subprog_info[0].changes_pkt_data;

err_free: kvfree(insn_state); @@ -18622,6 +18623,7 @@ static int jit_subprogs(struct bpf_verif } func[i]->aux->num_exentries = num_exentries; func[i]->aux->tail_call_reachable = env->subprog_info[i].tail_call_reachable; + func[i]->aux->changes_pkt_data = env->subprog_info[i].changes_pkt_data; func[i] = bpf_int_jit_compile(func[i]); if (!func[i]->jited) { err = -ENOTSUPP; @@ -19934,6 +19936,12 @@ int bpf_check_attach_target(struct bpf_v "Extension programs should be JITed\n"); return -EINVAL; } + if (prog->aux->changes_pkt_data && + !aux->func[subprog]->aux->changes_pkt_data) { + bpf_log(log, + "Extension program changes packet data, while original does not\n"); + return -EINVAL; + } } if (!tgt_prog->jited) { bpf_log(log, "Can attach to only JITed progs\n"); @@ -20393,10 +20401,6 @@ int bpf_check(struct bpf_prog **prog, un if (ret < 0) goto skip_full_check;

- ret = check_attach_btf_id(env); - if (ret) - goto skip_full_check; - ret = resolve_pseudo_ldimm64(env); if (ret < 0) goto skip_full_check; @@ -20411,6 +20415,10 @@ int bpf_check(struct bpf_prog **prog, un if (ret < 0) goto skip_full_check;

+ ret = check_attach_btf_id(env); + if (ret) + goto skip_full_check; + ret = do_check_subprogs(env); ret = ret ?: do_check_main(env);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eduard Zingerman eddyz87@gmail.com

commit 1a4607ffba35bf2a630aab299e34dd3f6e658d70 upstream.

Tail-called programs could execute any of the helpers that invalidate packet pointers. Hence, conservatively assume that each tail call invalidates packet pointers.

Making the change in bpf_helper_changes_pkt_data() automatically makes use of check_cfg() logic that computes 'changes_pkt_data' effect for global sub-programs, such that the following program could be rejected:

int tail_call(struct __sk_buff *sk) { bpf_tail_call_static(sk, &jmp_table, 0); return 0; }

SEC("tc") int not_safe(struct __sk_buff *sk) { int *p = (void *)(long)sk->data; ... make p valid ... tail_call(sk); *p = 42; /* this is unsafe */ ... }

The tc_bpf2bpf.c:subprog_tc() needs change: mark it as a function that can invalidate packet pointers. Otherwise, it can't be freplaced with tailcall_freplace.c:entry_freplace() that does a tail call.

Signed-off-by: Eduard Zingerman eddyz87@gmail.com Link: https://lore.kernel.org/r/20241210041100.1898468-8-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov ast@kernel.org [ shung-hsi.yu: drop changes to tools/testing/selftests/bpf/progs/tc_bpf2bpf.c because it is not present. ] Signed-off-by: Shung-Hsi Yu shung-hsi.yu@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/core/filter.c | 2 ++ 1 file changed, 2 insertions(+)

--- a/net/core/filter.c +++ b/net/core/filter.c @@ -7893,6 +7893,8 @@ bool bpf_helper_changes_pkt_data(enum bp case BPF_FUNC_xdp_adjust_head: case BPF_FUNC_xdp_adjust_meta: case BPF_FUNC_xdp_adjust_tail: + /* tail-called program could call any of the above */ + case BPF_FUNC_tail_call: return true; default: return false;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eduard Zingerman eddyz87@gmail.com

commit ac6542ad92759cda383ad62b4e4cbfc28136abc1 upstream.

bpf_prog_aux->func field might be NULL if program does not have subprograms except for main sub-program. The fixed commit does bpf_prog_aux->func access unconditionally, which might lead to null pointer dereference.

The bug could be triggered by replacing the following BPF program:

SEC("tc") int main_changes(struct __sk_buff *sk) { bpf_skb_pull_data(sk, 0); return 0; }

With the following BPF program:

SEC("freplace") long changes_pkt_data(struct __sk_buff *sk) { return bpf_skb_pull_data(sk, 0); }

bpf_prog_aux instance itself represents the main sub-program, use this property to fix the bug.

Fixes: 81f6d0530ba0 ("bpf: check changes_pkt_data property for extension programs") Reported-by: kernel test robot lkp@intel.com Reported-by: Dan Carpenter dan.carpenter@linaro.org Closes: https://lore.kernel.org/r/202412111822.qGw6tOyB-lkp@intel.com/ Signed-off-by: Eduard Zingerman eddyz87@gmail.com Link: https://lore.kernel.org/r/20241212070711.427443-1-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov ast@kernel.org Signed-off-by: Shung-Hsi Yu shung-hsi.yu@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- kernel/bpf/verifier.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)

--- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -19908,6 +19908,7 @@ int bpf_check_attach_target(struct bpf_v } if (tgt_prog) { struct bpf_prog_aux *aux = tgt_prog->aux; + bool tgt_changes_pkt_data;

if (bpf_prog_is_dev_bound(prog->aux) && !bpf_prog_dev_bound_match(prog, tgt_prog)) { @@ -19936,8 +19937,10 @@ int bpf_check_attach_target(struct bpf_v "Extension programs should be JITed\n"); return -EINVAL; } - if (prog->aux->changes_pkt_data && - !aux->func[subprog]->aux->changes_pkt_data) { + tgt_changes_pkt_data = aux->func + ? aux->func[subprog]->aux->changes_pkt_data + : aux->changes_pkt_data; + if (prog->aux->changes_pkt_data && !tgt_changes_pkt_data) { bpf_log(log, "Extension program changes packet data, while original does not\n"); return -EINVAL;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ryan Matthews ryanmatthews@fastmail.com

This reverts commit 2a12efc567a270a155e3b886258297abd79cdea0 which is commit f068ffdd034c93f0c768acdc87d4d2d7023c1379 upstream.

This is a backport mistake.

Deleting "IMX7D" here skips more than just controller_id logic. It skips reset assignments too, which causes:

imx6q-pcie 33800000.pcie: PCIe PLL lock timeout

In my case, in addition to broken PCIe, kernel boot hangs entirely.

This isn't a problem upstream because before this, they moved the rest of the code out of the switch case.

Signed-off-by: Ryan Matthews ryanmatthews@fastmail.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/pci/controller/dwc/pci-imx6.c | 1 + 1 file changed, 1 insertion(+)

--- a/drivers/pci/controller/dwc/pci-imx6.c +++ b/drivers/pci/controller/dwc/pci-imx6.c @@ -1281,6 +1281,7 @@ static int imx6_pcie_probe(struct platfo switch (imx6_pcie->drvdata->variant) { case IMX8MQ: case IMX8MQ_EP: + case IMX7D: if (dbi_base->start == IMX8MQ_PCIE2_BASE_ADDR) imx6_pcie->controller_id = 1;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson seanjc@google.com

commit c2fee09fc167c74a64adb08656cb993ea475197e upstream.

Move the conditional loading of hardware DR6 with the guest's DR6 value out of the core .vcpu_run() loop to fix a bug where KVM can load hardware with a stale vcpu->arch.dr6.

When the guest accesses a DR and host userspace isn't debugging the guest, KVM disables DR interception and loads the guest's values into hardware on VM-Enter and saves them on VM-Exit. This allows the guest to access DRs at will, e.g. so that a sequence of DR accesses to configure a breakpoint only generates one VM-Exit.

For DR0-DR3, the logic/behavior is identical between VMX and SVM, and also identical between KVM_DEBUGREG_BP_ENABLED (userspace debugging the guest) and KVM_DEBUGREG_WONT_EXIT (guest using DRs), and so KVM handles loading DR0-DR3 in common code, _outside_ of the core kvm_x86_ops.vcpu_run() loop.

But for DR6, the guest's value doesn't need to be loaded into hardware for KVM_DEBUGREG_BP_ENABLED, and SVM provides a dedicated VMCB field whereas VMX requires software to manually load the guest value, and so loading the guest's value into DR6 is handled by {svm,vmx}_vcpu_run(), i.e. is done _inside_ the core run loop.

Unfortunately, saving the guest values on VM-Exit is initiated by common x86, again outside of the core run loop. If the guest modifies DR6 (in hardware, when DR interception is disabled), and then the next VM-Exit is a fastpath VM-Exit, KVM will reload hardware DR6 with vcpu->arch.dr6 and clobber the guest's actual value.

The bug shows up primarily with nested VMX because KVM handles the VMX preemption timer in the fastpath, and the window between hardware DR6 being modified (in guest context) and DR6 being read by guest software is orders of magnitude larger in a nested setup. E.g. in non-nested, the VMX preemption timer would need to fire precisely between #DB injection and the #DB handler's read of DR6, whereas with a KVM-on-KVM setup, the window where hardware DR6 is "dirty" extends all the way from L1 writing DR6 to VMRESUME (in L1).

L1's view: ========== <L1 disables DR interception> CPU 0/KVM-7289 [023] d.... 2925.640961: kvm_entry: vcpu 0 A: L1 Writes DR6 CPU 0/KVM-7289 [023] d.... 2925.640963: <hack>: Set DRs, DR6 = 0xffff0ff1

B: CPU 0/KVM-7289 [023] d.... 2925.640967: kvm_exit: vcpu 0 reason EXTERNAL_INTERRUPT intr_info 0x800000ec

D: L1 reads DR6, arch.dr6 = 0 CPU 0/KVM-7289 [023] d.... 2925.640969: <hack>: Sync DRs, DR6 = 0xffff0ff0

CPU 0/KVM-7289 [023] d.... 2925.640976: kvm_entry: vcpu 0 L2 reads DR6, L1 disables DR interception CPU 0/KVM-7289 [023] d.... 2925.640980: kvm_exit: vcpu 0 reason DR_ACCESS info1 0x0000000000000216 CPU 0/KVM-7289 [023] d.... 2925.640983: kvm_entry: vcpu 0

CPU 0/KVM-7289 [023] d.... 2925.640983: <hack>: Set DRs, DR6 = 0xffff0ff0

L2 detects failure CPU 0/KVM-7289 [023] d.... 2925.640987: kvm_exit: vcpu 0 reason HLT L1 reads DR6 (confirms failure) CPU 0/KVM-7289 [023] d.... 2925.640990: <hack>: Sync DRs, DR6 = 0xffff0ff0

L0's view: ========== L2 reads DR6, arch.dr6 = 0 CPU 23/KVM-5046 [001] d.... 3410.005610: kvm_exit: vcpu 23 reason DR_ACCESS info1 0x0000000000000216 CPU 23/KVM-5046 [001] ..... 3410.005610: kvm_nested_vmexit: vcpu 23 reason DR_ACCESS info1 0x0000000000000216

L2 => L1 nested VM-Exit CPU 23/KVM-5046 [001] ..... 3410.005610: kvm_nested_vmexit_inject: reason: DR_ACCESS ext_inf1: 0x0000000000000216

CPU 23/KVM-5046 [001] d.... 3410.005610: kvm_entry: vcpu 23 CPU 23/KVM-5046 [001] d.... 3410.005611: kvm_exit: vcpu 23 reason VMREAD CPU 23/KVM-5046 [001] d.... 3410.005611: kvm_entry: vcpu 23 CPU 23/KVM-5046 [001] d.... 3410.005612: kvm_exit: vcpu 23 reason VMREAD CPU 23/KVM-5046 [001] d.... 3410.005612: kvm_entry: vcpu 23

L1 writes DR7, L0 disables DR interception CPU 23/KVM-5046 [001] d.... 3410.005612: kvm_exit: vcpu 23 reason DR_ACCESS info1 0x0000000000000007 CPU 23/KVM-5046 [001] d.... 3410.005613: kvm_entry: vcpu 23

L0 writes DR6 = 0 (arch.dr6) CPU 23/KVM-5046 [001] d.... 3410.005613: <hack>: Set DRs, DR6 = 0xffff0ff0

A: <L1 writes DR6 = 1, no interception, arch.dr6 is still '0'>

B: CPU 23/KVM-5046 [001] d.... 3410.005614: kvm_exit: vcpu 23 reason PREEMPTION_TIMER CPU 23/KVM-5046 [001] d.... 3410.005614: kvm_entry: vcpu 23

C: L0 writes DR6 = 0 (arch.dr6) CPU 23/KVM-5046 [001] d.... 3410.005614: <hack>: Set DRs, DR6 = 0xffff0ff0

L1 => L2 nested VM-Enter CPU 23/KVM-5046 [001] d.... 3410.005616: kvm_exit: vcpu 23 reason VMRESUME

L0 reads DR6, arch.dr6 = 0

Reported-by: John Stultz jstultz@google.com Closes: https://lkml.kernel.org/r/CANDhNCq5_F3HfFYABqFGCA1bPd_%2BxgNj-iDQhH4tDk%2Bwi... Fixes: 375e28ffc0cf ("KVM: X86: Set host DR6 only on VMX and for KVM_DEBUGREG_WONT_EXIT") Fixes: d67668e9dd76 ("KVM: x86, SVM: isolate vcpu->arch.dr6 from vmcb->save.dr6") Cc: stable@vger.kernel.org Cc: Jim Mattson jmattson@google.com Tested-by: John Stultz jstultz@google.com Link: https://lore.kernel.org/r/20250125011833.3644371-1-seanjc@google.com Signed-off-by: Sean Christopherson seanjc@google.com [jth: Handled conflicts with kvm_x86_ops reshuffle] Signed-off-by: James Houghton jthoughton@google.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- arch/x86/include/asm/kvm-x86-ops.h | 1 + arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/svm/svm.c | 13 ++++++------- arch/x86/kvm/vmx/vmx.c | 11 +++++++---- arch/x86/kvm/x86.c | 3 +++ 5 files changed, 18 insertions(+), 11 deletions(-)

--- a/arch/x86/include/asm/kvm-x86-ops.h +++ b/arch/x86/include/asm/kvm-x86-ops.h @@ -48,6 +48,7 @@ KVM_X86_OP(set_idt) KVM_X86_OP(get_gdt) KVM_X86_OP(set_gdt) KVM_X86_OP(sync_dirty_debug_regs) +KVM_X86_OP(set_dr6) KVM_X86_OP(set_dr7) KVM_X86_OP(cache_reg) KVM_X86_OP(get_rflags) --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -1595,6 +1595,7 @@ struct kvm_x86_ops { void (*get_gdt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt); void (*set_gdt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt); void (*sync_dirty_debug_regs)(struct kvm_vcpu *vcpu); + void (*set_dr6)(struct kvm_vcpu *vcpu, unsigned long value); void (*set_dr7)(struct kvm_vcpu *vcpu, unsigned long value); void (*cache_reg)(struct kvm_vcpu *vcpu, enum kvm_reg reg); unsigned long (*get_rflags)(struct kvm_vcpu *vcpu); --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -2014,11 +2014,11 @@ static void new_asid(struct vcpu_svm *sv svm->asid = sd->next_asid++; }

-static void svm_set_dr6(struct vcpu_svm *svm, unsigned long value) +static void svm_set_dr6(struct kvm_vcpu *vcpu, unsigned long value) { - struct vmcb *vmcb = svm->vmcb; + struct vmcb *vmcb = to_svm(vcpu)->vmcb;

- if (svm->vcpu.arch.guest_state_protected) + if (vcpu->arch.guest_state_protected) return;

if (unlikely(value != vmcb->save.dr6)) { @@ -4220,10 +4220,8 @@ static __no_kcsan fastpath_t svm_vcpu_ru * Run with all-zero DR6 unless needed, so that we can get the exact cause * of a #DB. */ - if (unlikely(vcpu->arch.switch_db_regs & KVM_DEBUGREG_WONT_EXIT)) - svm_set_dr6(svm, vcpu->arch.dr6); - else - svm_set_dr6(svm, DR6_ACTIVE_LOW); + if (likely(!(vcpu->arch.switch_db_regs & KVM_DEBUGREG_WONT_EXIT))) + svm_set_dr6(vcpu, DR6_ACTIVE_LOW);

clgi(); kvm_load_guest_xsave_state(vcpu); @@ -5002,6 +5000,7 @@ static struct kvm_x86_ops svm_x86_ops __ .set_idt = svm_set_idt, .get_gdt = svm_get_gdt, .set_gdt = svm_set_gdt, + .set_dr6 = svm_set_dr6, .set_dr7 = svm_set_dr7, .sync_dirty_debug_regs = svm_sync_dirty_debug_regs, .cache_reg = svm_cache_reg, --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -5617,6 +5617,12 @@ static void vmx_sync_dirty_debug_regs(st set_debugreg(DR6_RESERVED, 6); }

+static void vmx_set_dr6(struct kvm_vcpu *vcpu, unsigned long val) +{ + lockdep_assert_irqs_disabled(); + set_debugreg(vcpu->arch.dr6, 6); +} + static void vmx_set_dr7(struct kvm_vcpu *vcpu, unsigned long val) { vmcs_writel(GUEST_DR7, val); @@ -7356,10 +7362,6 @@ static fastpath_t vmx_vcpu_run(struct kv vmx->loaded_vmcs->host_state.cr4 = cr4; }

- /* When KVM_DEBUGREG_WONT_EXIT, dr6 is accessible in guest. */ - if (unlikely(vcpu->arch.switch_db_regs & KVM_DEBUGREG_WONT_EXIT)) - set_debugreg(vcpu->arch.dr6, 6); - /* When single-stepping over STI and MOV SS, we must clear the * corresponding interruptibility bits in the guest state. Otherwise * vmentry fails as it then expects bit 14 (BS) in pending debug @@ -8292,6 +8294,7 @@ static struct kvm_x86_ops vmx_x86_ops __ .set_idt = vmx_set_idt, .get_gdt = vmx_get_gdt, .set_gdt = vmx_set_gdt, + .set_dr6 = vmx_set_dr6, .set_dr7 = vmx_set_dr7, .sync_dirty_debug_regs = vmx_sync_dirty_debug_regs, .cache_reg = vmx_cache_reg, --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -10772,6 +10772,9 @@ static int vcpu_enter_guest(struct kvm_v set_debugreg(vcpu->arch.eff_db[1], 1); set_debugreg(vcpu->arch.eff_db[2], 2); set_debugreg(vcpu->arch.eff_db[3], 3); + /* When KVM_DEBUGREG_WONT_EXIT, dr6 is accessible in guest. */ + if (unlikely(vcpu->arch.switch_db_regs & KVM_DEBUGREG_WONT_EXIT)) + static_call(kvm_x86_set_dr6)(vcpu, vcpu->arch.dr6); } else if (unlikely(hw_breakpoint_active())) { set_debugreg(0, 7); }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sheetal sheetal@nvidia.com

[ Upstream commit 9aff2e8df240e84a36f2607f98a0a9924a24e65d ]

Issue: When multiple audio streams share a common BE DAI, the BE DAI widget can be powered up before its hardware parameters are configured. This incorrect sequence leads to intermittent pcm_write errors.

For example, the below Tegra use-case throws an error: aplay(2 streams) -> AMX(mux) -> ADX(demux) -> arecord(2 streams), here, 'AMX TX' and 'ADX RX' are common BE DAIs.

For above usecase when failure happens below sequence is observed: aplay(1) FE open() - BE DAI callbacks added to the list - BE DAI state = SND_SOC_DPCM_STATE_OPEN aplay(2) FE open() - BE DAI callbacks are not added to the list as the state is already SND_SOC_DPCM_STATE_OPEN during aplay(1) FE open(). aplay(2) FE hw_params() - BE DAI hw_params() callback ignored aplay(2) FE prepare() - Widget is powered ON without BE DAI hw_params() call aplay(1) FE hw_params() - BE DAI hw_params() is now called

Fix: Add BE DAIs in the list if its state is either SND_SOC_DPCM_STATE_OPEN or SND_SOC_DPCM_STATE_HW_PARAMS as well.

It ensures the widget is powered ON after BE DAI hw_params() callback.

Fixes: 0c25db3f7621 ("ASoC: soc-pcm: Don't reconnect an already active BE") Signed-off-by: Sheetal sheetal@nvidia.com Link: https://patch.msgid.link/20250404105953.2784819-1-sheetal@nvidia.com Signed-off-by: Mark Brown broonie@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- sound/soc/soc-pcm.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/sound/soc/soc-pcm.c b/sound/soc/soc-pcm.c index 60248a6820aac..30e93f9aad762 100644 --- a/sound/soc/soc-pcm.c +++ b/sound/soc/soc-pcm.c @@ -1534,10 +1534,13 @@ static int dpcm_add_paths(struct snd_soc_pcm_runtime *fe, int stream, /* * Filter for systems with 'component_chaining' enabled. * This helps to avoid unnecessary re-configuration of an - * already active BE on such systems. + * already active BE on such systems and ensures the BE DAI + * widget is powered ON after hw_params() BE DAI callback. */ if (fe->card->component_chaining && (be->dpcm[stream].state != SND_SOC_DPCM_STATE_NEW) && + (be->dpcm[stream].state != SND_SOC_DPCM_STATE_OPEN) && + (be->dpcm[stream].state != SND_SOC_DPCM_STATE_HW_PARAMS) && (be->dpcm[stream].state != SND_SOC_DPCM_STATE_CLOSE)) continue;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Madhavan Srinivasan maddy@linux.ibm.com

[ Upstream commit b2accfe7ca5bc9f9af28e603b79bdd5ad8df5c0b ]

Commit 579aee9fc594 ("powerpc: suppress some linker warnings in recent linker versions") enabled support to add linker option "--no-warn-rwx-segments", if the version is greater than 2.39. Similar build warning were reported recently from linker version 2.35.2.

ld: warning: arch/powerpc/boot/zImage.epapr has a LOAD segment with RWX permissions ld: warning: arch/powerpc/boot/zImage.pseries has a LOAD segment with RWX permissions

Fix the warning by checking for "--no-warn-rwx-segments" option support in linker to enable it, instead of checking for the version range.

Fixes: 579aee9fc594 ("powerpc: suppress some linker warnings in recent linker versions") Reported-by: Venkat Rao Bagalkote venkat88@linux.ibm.com Suggested-by: Christophe Leroy christophe.leroy@csgroup.eu Tested-by: Venkat Rao Bagalkote venkat88@linux.ibm.com Closes: https://lore.kernel.org/linuxppc-dev/61cf556c-4947-4bd6-af63-892fc0966dad@li... Signed-off-by: Madhavan Srinivasan maddy@linux.ibm.com Link: https://patch.msgid.link/20250401004218.24869-1-maddy@linux.ibm.com Signed-off-by: Sasha Levin sashal@kernel.org --- arch/powerpc/boot/wrapper | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/arch/powerpc/boot/wrapper b/arch/powerpc/boot/wrapper index 352d7de24018f..fea9694f1047e 100755 --- a/arch/powerpc/boot/wrapper +++ b/arch/powerpc/boot/wrapper @@ -234,10 +234,8 @@ fi

# suppress some warnings in recent ld versions nowarn="-z noexecstack" -if ! ld_is_lld; then - if [ "$LD_VERSION" -ge "$(echo 2.39 | ld_version)" ]; then - nowarn="$nowarn --no-warn-rwx-segments" - fi +if [ $(${CROSS}ld -v --no-warn-rwx-segments &>/dev/null; echo $?) -eq 0 ]; then + nowarn="$nowarn --no-warn-rwx-segments" fi

platformo=$object/"$platform".o

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Murad Masimov m.masimov@mt-integration.ru

[ Upstream commit 0fb15ae3b0a9221be01715dac0335647c79f3362 ]

plfxlc_mac_release() asserts that mac->lock is held. This assertion is incorrect, because even if it was possible, it would not be the valid behaviour. The function is used when probe fails or after the device is disconnected. In both cases mac->lock can not be held as the driver is not working with the device at the moment. All functions that use mac->lock unlock it just after it was held. There is also no need to hold mac->lock for plfxlc_mac_release() itself, as mac data is not affected, except for mac->flags, which is modified atomically.

This bug leads to the following warning: ================================================================ WARNING: CPU: 0 PID: 127 at drivers/net/wireless/purelifi/plfxlc/mac.c:106 plfxlc_mac_release+0x7d/0xa0 Modules linked in: CPU: 0 PID: 127 Comm: kworker/0:2 Not tainted 6.1.124-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/13/2024 Workqueue: usb_hub_wq hub_event RIP: 0010:plfxlc_mac_release+0x7d/0xa0 drivers/net/wireless/purelifi/plfxlc/mac.c:106 Call Trace: <TASK> probe+0x941/0xbd0 drivers/net/wireless/purelifi/plfxlc/usb.c:694 usb_probe_interface+0x5c0/0xaf0 drivers/usb/core/driver.c:396 really_probe+0x2ab/0xcb0 drivers/base/dd.c:639 __driver_probe_device+0x1a2/0x3d0 drivers/base/dd.c:785 driver_probe_device+0x50/0x420 drivers/base/dd.c:815 __device_attach_driver+0x2cf/0x510 drivers/base/dd.c:943 bus_for_each_drv+0x183/0x200 drivers/base/bus.c:429 __device_attach+0x359/0x570 drivers/base/dd.c:1015 bus_probe_device+0xba/0x1e0 drivers/base/bus.c:489 device_add+0xb48/0xfd0 drivers/base/core.c:3696 usb_set_configuration+0x19dd/0x2020 drivers/usb/core/message.c:2165 usb_generic_driver_probe+0x84/0x140 drivers/usb/core/generic.c:238 usb_probe_device+0x130/0x260 drivers/usb/core/driver.c:293 really_probe+0x2ab/0xcb0 drivers/base/dd.c:639 __driver_probe_device+0x1a2/0x3d0 drivers/base/dd.c:785 driver_probe_device+0x50/0x420 drivers/base/dd.c:815 __device_attach_driver+0x2cf/0x510 drivers/base/dd.c:943 bus_for_each_drv+0x183/0x200 drivers/base/bus.c:429 __device_attach+0x359/0x570 drivers/base/dd.c:1015 bus_probe_device+0xba/0x1e0 drivers/base/bus.c:489 device_add+0xb48/0xfd0 drivers/base/core.c:3696 usb_new_device+0xbdd/0x18f0 drivers/usb/core/hub.c:2620 hub_port_connect drivers/usb/core/hub.c:5477 [inline] hub_port_connect_change drivers/usb/core/hub.c:5617 [inline] port_event drivers/usb/core/hub.c:5773 [inline] hub_event+0x2efe/0x5730 drivers/usb/core/hub.c:5855 process_one_work+0x8a9/0x11d0 kernel/workqueue.c:2292 worker_thread+0xa47/0x1200 kernel/workqueue.c:2439 kthread+0x28d/0x320 kernel/kthread.c:376 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295 </TASK> ================================================================

Found by Linux Verification Center (linuxtesting.org) with Syzkaller.

Fixes: 68d57a07bfe5 ("wireless: add plfxlc driver for pureLiFi X, XL, XC devices") Reported-by: syzbot+7d4f142f6c288de8abfe@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=7d4f142f6c288de8abfe Signed-off-by: Murad Masimov m.masimov@mt-integration.ru Link: https://patch.msgid.link/20250321185226.71-2-m.masimov@mt-integration.ru Signed-off-by: Johannes Berg johannes.berg@intel.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/wireless/purelifi/plfxlc/mac.c | 1 - 1 file changed, 1 deletion(-)

diff --git a/drivers/net/wireless/purelifi/plfxlc/mac.c b/drivers/net/wireless/purelifi/plfxlc/mac.c index 506d2f31efb5a..7ebc0df0944cb 100644 --- a/drivers/net/wireless/purelifi/plfxlc/mac.c +++ b/drivers/net/wireless/purelifi/plfxlc/mac.c @@ -103,7 +103,6 @@ int plfxlc_mac_init_hw(struct ieee80211_hw *hw) void plfxlc_mac_release(struct plfxlc_mac *mac) { plfxlc_chip_release(&mac->chip); - lockdep_assert_held(&mac->lock); }

int plfxlc_op_start(struct ieee80211_hw *hw)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel idosch@nvidia.com

[ Upstream commit 087a9eb9e5978e3ba362e1163691e41097e8ca20 ]

When a VNI is deleted from a VXLAN device in 'vnifilter' mode, the FDB entry associated with the default remote (assuming one was configured) is deleted without holding the hash lock. This is wrong and will result in a warning [1] being generated by the lockdep annotation that was added by commit ebe642067455 ("vxlan: Create wrappers for FDB lookup").

Reproducer:

# ip link add vx0 up type vxlan dstport 4789 external vnifilter local 192.0.2.1 # bridge vni add vni 10010 remote 198.51.100.1 dev vx0 # bridge vni del vni 10010 dev vx0

Fix by acquiring the hash lock before the deletion and releasing it afterwards. Blame the original commit that introduced the issue rather than the one that exposed it.

[1] WARNING: CPU: 3 PID: 392 at drivers/net/vxlan/vxlan_core.c:417 vxlan_find_mac+0x17f/0x1a0 [...] RIP: 0010:vxlan_find_mac+0x17f/0x1a0 [...] Call Trace: <TASK> __vxlan_fdb_delete+0xbe/0x560 vxlan_vni_delete_group+0x2ba/0x940 vxlan_vni_del.isra.0+0x15f/0x580 vxlan_process_vni_filter+0x38b/0x7b0 vxlan_vnifilter_process+0x3bb/0x510 rtnetlink_rcv_msg+0x2f7/0xb70 netlink_rcv_skb+0x131/0x360 netlink_unicast+0x426/0x710 netlink_sendmsg+0x75a/0xc20 __sock_sendmsg+0xc1/0x150 ____sys_sendmsg+0x5aa/0x7b0 ___sys_sendmsg+0xfc/0x180 __sys_sendmsg+0x121/0x1b0 do_syscall_64+0xbb/0x1d0 entry_SYSCALL_64_after_hwframe+0x4b/0x53

Fixes: f9c4bb0b245c ("vxlan: vni filtering support on collect metadata device") Signed-off-by: Ido Schimmel idosch@nvidia.com Reviewed-by: Nikolay Aleksandrov razor@blackwall.org Link: https://patch.msgid.link/20250423145131.513029-1-idosch@nvidia.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/vxlan/vxlan_vnifilter.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/vxlan/vxlan_vnifilter.c b/drivers/net/vxlan/vxlan_vnifilter.c index 6e6e9f05509ab..06d19e90eadb5 100644 --- a/drivers/net/vxlan/vxlan_vnifilter.c +++ b/drivers/net/vxlan/vxlan_vnifilter.c @@ -627,7 +627,11 @@ static void vxlan_vni_delete_group(struct vxlan_dev *vxlan, * default dst remote_ip previously added for this vni */ if (!vxlan_addr_any(&vninode->remote_ip) || - !vxlan_addr_any(&dst->remote_ip)) + !vxlan_addr_any(&dst->remote_ip)) { + u32 hash_index = fdb_head_index(vxlan, all_zeros_mac, + vninode->vni); + + spin_lock_bh(&vxlan->hash_lock[hash_index]); __vxlan_fdb_delete(vxlan, all_zeros_mac, (vxlan_addr_any(&vninode->remote_ip) ? dst->remote_ip : vninode->remote_ip), @@ -635,6 +639,8 @@ static void vxlan_vni_delete_group(struct vxlan_dev *vxlan, vninode->vni, vninode->vni, dst->remote_ifindex, true); + spin_unlock_bh(&vxlan->hash_lock[hash_index]); + }

if (vxlan->dev->flags & IFF_UP) { if (vxlan_addr_multicast(&vninode->remote_ip) &&

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chris Mi cmi@nvidia.com

[ Upstream commit 90538d23278a981e344d364e923162fce752afeb ]

The cited commit assumes enabling roce always succeeds. But it is not true. Add error handling for it.

Fixes: 80f09dfc237f ("net/mlx5: Eswitch, enable RoCE loopback traffic") Signed-off-by: Chris Mi cmi@nvidia.com Reviewed-by: Roi Dayan roid@nvidia.com Reviewed-by: Maor Gottlieb maorg@nvidia.com Signed-off-by: Mark Bloch mbloch@nvidia.com Reviewed-by: Michal Swiatkowski michal.swiatkowski@linux.intel.com Link: https://patch.msgid.link/20250423083611.324567-6-mbloch@nvidia.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- .../net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 5 ++++- drivers/net/ethernet/mellanox/mlx5/core/rdma.c | 9 +++++---- drivers/net/ethernet/mellanox/mlx5/core/rdma.h | 4 ++-- 3 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 7eba3a5bb97ca..326c72b3df867 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -3499,7 +3499,9 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) int err;

mutex_init(&esw->offloads.termtbl_mutex); - mlx5_rdma_enable_roce(esw->dev); + err = mlx5_rdma_enable_roce(esw->dev); + if (err) + goto err_roce;

err = mlx5_esw_host_number_init(esw); if (err) @@ -3560,6 +3562,7 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) esw_offloads_metadata_uninit(esw); err_metadata: mlx5_rdma_disable_roce(esw->dev); +err_roce: mutex_destroy(&esw->offloads.termtbl_mutex); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/rdma.c b/drivers/net/ethernet/mellanox/mlx5/core/rdma.c index f585ef5a34243..5c552b71e371c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/rdma.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/rdma.c @@ -140,17 +140,17 @@ void mlx5_rdma_disable_roce(struct mlx5_core_dev *dev) mlx5_nic_vport_disable_roce(dev); }

-void mlx5_rdma_enable_roce(struct mlx5_core_dev *dev) +int mlx5_rdma_enable_roce(struct mlx5_core_dev *dev) { int err;

if (!MLX5_CAP_GEN(dev, roce)) - return; + return 0;

err = mlx5_nic_vport_enable_roce(dev); if (err) { mlx5_core_err(dev, "Failed to enable RoCE: %d\n", err); - return; + return err; }

err = mlx5_rdma_add_roce_addr(dev); @@ -165,10 +165,11 @@ void mlx5_rdma_enable_roce(struct mlx5_core_dev *dev) goto del_roce_addr; }

- return; + return err;

del_roce_addr: mlx5_rdma_del_roce_addr(dev); disable_roce: mlx5_nic_vport_disable_roce(dev); + return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/rdma.h b/drivers/net/ethernet/mellanox/mlx5/core/rdma.h index 750cff2a71a4b..3d9e76c3d42fb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/rdma.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/rdma.h @@ -8,12 +8,12 @@

#ifdef CONFIG_MLX5_ESWITCH

-void mlx5_rdma_enable_roce(struct mlx5_core_dev *dev); +int mlx5_rdma_enable_roce(struct mlx5_core_dev *dev); void mlx5_rdma_disable_roce(struct mlx5_core_dev *dev);

#else /* CONFIG_MLX5_ESWITCH */

-static inline void mlx5_rdma_enable_roce(struct mlx5_core_dev *dev) {} +static inline int mlx5_rdma_enable_roce(struct mlx5_core_dev *dev) { return 0; } static inline void mlx5_rdma_disable_roce(struct mlx5_core_dev *dev) {}

#endif /* CONFIG_MLX5_ESWITCH */

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Abhishek Chauhan quic_abchauha@quicinc.com

[ Upstream commit 4d25ca2d6801cfcf26f7f39c561611ba5be99bf8 ]

mono_delivery_time was added to check if skb->tstamp has delivery time in mono clock base (i.e. EDT) otherwise skb->tstamp has timestamp in ingress and delivery_time at egress.

Renaming the bitfield from mono_delivery_time to tstamp_type is for extensibilty for other timestamps such as userspace timestamp (i.e. SO_TXTIME) set via sock opts.

As we are renaming the mono_delivery_time to tstamp_type, it makes sense to start assigning tstamp_type based on enum defined in this commit.

Earlier we used bool arg flag to check if the tstamp is mono in function skb_set_delivery_time, Now the signature of the functions accepts tstamp_type to distinguish between mono and real time.

Also skb_set_delivery_type_by_clockid is a new function which accepts clockid to determine the tstamp_type.

In future tstamp_type:1 can be extended to support userspace timestamp by increasing the bitfield.

Signed-off-by: Abhishek Chauhan quic_abchauha@quicinc.com Reviewed-by: Willem de Bruijn willemb@google.com Reviewed-by: Martin KaFai Lau martin.lau@kernel.org Link: https://lore.kernel.org/r/20240509211834.3235191-2-quic_abchauha@quicinc.com Signed-off-by: Martin KaFai Lau martin.lau@kernel.org Stable-dep-of: 3908feb1bd7f ("Bluetooth: L2CAP: copy RX timestamp to new fragments") Signed-off-by: Sasha Levin sashal@kernel.org --- include/linux/skbuff.h | 52 ++++++++++++++++------ include/net/inet_frag.h | 4 +- net/bridge/netfilter/nf_conntrack_bridge.c | 6 +-- net/core/dev.c | 2 +- net/core/filter.c | 10 ++--- net/ieee802154/6lowpan/reassembly.c | 2 +- net/ipv4/inet_fragment.c | 2 +- net/ipv4/ip_fragment.c | 2 +- net/ipv4/ip_output.c | 9 ++-- net/ipv4/tcp_output.c | 14 +++--- net/ipv6/ip6_output.c | 6 +-- net/ipv6/netfilter.c | 6 +-- net/ipv6/netfilter/nf_conntrack_reasm.c | 2 +- net/ipv6/reassembly.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- net/sched/act_bpf.c | 4 +- net/sched/cls_bpf.c | 4 +- 17 files changed, 78 insertions(+), 51 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 5f11f98733419..f7d392d849be5 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -685,6 +685,11 @@ typedef unsigned int sk_buff_data_t; typedef unsigned char *sk_buff_data_t; #endif

+enum skb_tstamp_type { + SKB_CLOCK_REALTIME, + SKB_CLOCK_MONOTONIC, +}; + /** * DOC: Basic sk_buff geometry * @@ -804,10 +809,8 @@ typedef unsigned char *sk_buff_data_t; * @dst_pending_confirm: need to confirm neighbour * @decrypted: Decrypted SKB * @slow_gro: state present at GRO time, slower prepare step required - * @mono_delivery_time: When set, skb->tstamp has the - * delivery_time in mono clock base (i.e. EDT). Otherwise, the - * skb->tstamp has the (rcv) timestamp at ingress and - * delivery_time at egress. + * @tstamp_type: When set, skb->tstamp has the + * delivery_time clock base of skb->tstamp. * @napi_id: id of the NAPI struct this skb came from * @sender_cpu: (aka @napi_id) source CPU in XPS * @alloc_cpu: CPU which did the skb allocation. @@ -935,7 +938,7 @@ struct sk_buff { /* private: */ __u8 __mono_tc_offset[0]; /* public: */ - __u8 mono_delivery_time:1; /* See SKB_MONO_DELIVERY_TIME_MASK */ + __u8 tstamp_type:1; /* See skb_tstamp_type */ #ifdef CONFIG_NET_XGRESS __u8 tc_at_ingress:1; /* See TC_AT_INGRESS_MASK */ __u8 tc_skip_classify:1; @@ -4189,7 +4192,7 @@ static inline void skb_get_new_timestampns(const struct sk_buff *skb, static inline void __net_timestamp(struct sk_buff *skb) { skb->tstamp = ktime_get_real(); - skb->mono_delivery_time = 0; + skb->tstamp_type = SKB_CLOCK_REALTIME; }

static inline ktime_t net_timedelta(ktime_t t) @@ -4198,10 +4201,33 @@ static inline ktime_t net_timedelta(ktime_t t) }

static inline void skb_set_delivery_time(struct sk_buff *skb, ktime_t kt, - bool mono) + u8 tstamp_type) { skb->tstamp = kt; - skb->mono_delivery_time = kt && mono; + + if (kt) + skb->tstamp_type = tstamp_type; + else + skb->tstamp_type = SKB_CLOCK_REALTIME; +} + +static inline void skb_set_delivery_type_by_clockid(struct sk_buff *skb, + ktime_t kt, clockid_t clockid) +{ + u8 tstamp_type = SKB_CLOCK_REALTIME; + + switch (clockid) { + case CLOCK_REALTIME: + break; + case CLOCK_MONOTONIC: + tstamp_type = SKB_CLOCK_MONOTONIC; + break; + default: + WARN_ON_ONCE(1); + kt = 0; + } + + skb_set_delivery_time(skb, kt, tstamp_type); }

DECLARE_STATIC_KEY_FALSE(netstamp_needed_key); @@ -4211,8 +4237,8 @@ DECLARE_STATIC_KEY_FALSE(netstamp_needed_key); */ static inline void skb_clear_delivery_time(struct sk_buff *skb) { - if (skb->mono_delivery_time) { - skb->mono_delivery_time = 0; + if (skb->tstamp_type) { + skb->tstamp_type = SKB_CLOCK_REALTIME; if (static_branch_unlikely(&netstamp_needed_key)) skb->tstamp = ktime_get_real(); else @@ -4222,7 +4248,7 @@ static inline void skb_clear_delivery_time(struct sk_buff *skb)

static inline void skb_clear_tstamp(struct sk_buff *skb) { - if (skb->mono_delivery_time) + if (skb->tstamp_type) return;

skb->tstamp = 0; @@ -4230,7 +4256,7 @@ static inline void skb_clear_tstamp(struct sk_buff *skb)

static inline ktime_t skb_tstamp(const struct sk_buff *skb) { - if (skb->mono_delivery_time) + if (skb->tstamp_type) return 0;

return skb->tstamp; @@ -4238,7 +4264,7 @@ static inline ktime_t skb_tstamp(const struct sk_buff *skb)

static inline ktime_t skb_tstamp_cond(const struct sk_buff *skb, bool cond) { - if (!skb->mono_delivery_time && skb->tstamp) + if (skb->tstamp_type != SKB_CLOCK_MONOTONIC && skb->tstamp) return skb->tstamp;

if (static_branch_unlikely(&netstamp_needed_key) || cond) diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h index 153960663ce4c..5af6eb14c5db1 100644 --- a/include/net/inet_frag.h +++ b/include/net/inet_frag.h @@ -76,7 +76,7 @@ struct frag_v6_compare_key { * @stamp: timestamp of the last received fragment * @len: total length of the original datagram * @meat: length of received fragments so far - * @mono_delivery_time: stamp has a mono delivery time (EDT) + * @tstamp_type: stamp has a mono delivery time (EDT) * @flags: fragment queue flags * @max_size: maximum received fragment size * @fqdir: pointer to struct fqdir @@ -97,7 +97,7 @@ struct inet_frag_queue { ktime_t stamp; int len; int meat; - u8 mono_delivery_time; + u8 tstamp_type; __u8 flags; u16 max_size; struct fqdir *fqdir; diff --git a/net/bridge/netfilter/nf_conntrack_bridge.c b/net/bridge/netfilter/nf_conntrack_bridge.c index 6ef04f9fe481b..4fbfbafdfa027 100644 --- a/net/bridge/netfilter/nf_conntrack_bridge.c +++ b/net/bridge/netfilter/nf_conntrack_bridge.c @@ -32,7 +32,7 @@ static int nf_br_ip_fragment(struct net *net, struct sock *sk, struct sk_buff *)) { int frag_max_size = BR_INPUT_SKB_CB(skb)->frag_max_size; - bool mono_delivery_time = skb->mono_delivery_time; + u8 tstamp_type = skb->tstamp_type; unsigned int hlen, ll_rs, mtu; ktime_t tstamp = skb->tstamp; struct ip_frag_state state; @@ -82,7 +82,7 @@ static int nf_br_ip_fragment(struct net *net, struct sock *sk, if (iter.frag) ip_fraglist_prepare(skb, &iter);

- skb_set_delivery_time(skb, tstamp, mono_delivery_time); + skb_set_delivery_time(skb, tstamp, tstamp_type); err = output(net, sk, data, skb); if (err || !iter.frag) break; @@ -113,7 +113,7 @@ static int nf_br_ip_fragment(struct net *net, struct sock *sk, goto blackhole; }

- skb_set_delivery_time(skb2, tstamp, mono_delivery_time); + skb_set_delivery_time(skb2, tstamp, tstamp_type); err = output(net, sk, data, skb2); if (err) goto blackhole; diff --git a/net/core/dev.c b/net/core/dev.c index c31a7f7bedf3d..4006fd164b7bc 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2189,7 +2189,7 @@ EXPORT_SYMBOL(net_disable_timestamp); static inline void net_timestamp_set(struct sk_buff *skb) { skb->tstamp = 0; - skb->mono_delivery_time = 0; + skb->tstamp_type = SKB_CLOCK_REALTIME; if (static_branch_unlikely(&netstamp_needed_key)) skb->tstamp = ktime_get_real(); } diff --git a/net/core/filter.c b/net/core/filter.c index c4c27e03279c9..066277b91a1be 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -7734,13 +7734,13 @@ BPF_CALL_3(bpf_skb_set_tstamp, struct sk_buff *, skb, if (!tstamp) return -EINVAL; skb->tstamp = tstamp; - skb->mono_delivery_time = 1; + skb->tstamp_type = SKB_CLOCK_MONOTONIC; break; case BPF_SKB_TSTAMP_UNSPEC: if (tstamp) return -EINVAL; skb->tstamp = 0; - skb->mono_delivery_time = 0; + skb->tstamp_type = SKB_CLOCK_REALTIME; break; default: return -EINVAL; @@ -9438,7 +9438,7 @@ static struct bpf_insn *bpf_convert_tstamp_read(const struct bpf_prog *prog, TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK); *insn++ = BPF_JMP32_IMM(BPF_JNE, tmp_reg, TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK, 2); - /* skb->tc_at_ingress && skb->mono_delivery_time, + /* skb->tc_at_ingress && skb->tstamp_type, * read 0 as the (rcv) timestamp. */ *insn++ = BPF_MOV64_IMM(value_reg, 0); @@ -9463,7 +9463,7 @@ static struct bpf_insn *bpf_convert_tstamp_write(const struct bpf_prog *prog, * the bpf prog is aware the tstamp could have delivery time. * Thus, write skb->tstamp as is if tstamp_type_access is true. * Otherwise, writing at ingress will have to clear the - * mono_delivery_time bit also. + * skb->tstamp_type bit also. */ if (!prog->tstamp_type_access) { __u8 tmp_reg = BPF_REG_AX; @@ -9473,7 +9473,7 @@ static struct bpf_insn *bpf_convert_tstamp_write(const struct bpf_prog *prog, *insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg, TC_AT_INGRESS_MASK, 1); /* goto <store> */ *insn++ = BPF_JMP_A(2); - /* <clear>: mono_delivery_time */ + /* <clear>: skb->tstamp_type */ *insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg, ~SKB_MONO_DELIVERY_TIME_MASK); *insn++ = BPF_STX_MEM(BPF_B, skb_reg, tmp_reg, SKB_BF_MONO_TC_OFFSET); } diff --git a/net/ieee802154/6lowpan/reassembly.c b/net/ieee802154/6lowpan/reassembly.c index 6dd960ec558cf..ba0455ad77019 100644 --- a/net/ieee802154/6lowpan/reassembly.c +++ b/net/ieee802154/6lowpan/reassembly.c @@ -130,7 +130,7 @@ static int lowpan_frag_queue(struct lowpan_frag_queue *fq, goto err;

fq->q.stamp = skb->tstamp; - fq->q.mono_delivery_time = skb->mono_delivery_time; + fq->q.tstamp_type = skb->tstamp_type; if (frag_type == LOWPAN_DISPATCH_FRAG1) fq->q.flags |= INET_FRAG_FIRST_IN;

diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index c88c9034d6300..496308c023848 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -619,7 +619,7 @@ void inet_frag_reasm_finish(struct inet_frag_queue *q, struct sk_buff *head, skb_mark_not_on_list(head); head->prev = NULL; head->tstamp = q->stamp; - head->mono_delivery_time = q->mono_delivery_time; + head->tstamp_type = q->tstamp_type;

if (sk) refcount_add(sum_truesize - head_truesize, &sk->sk_wmem_alloc); diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index 877d1e03150c7..484edc8513e4b 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -360,7 +360,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb) qp->iif = dev->ifindex;

qp->q.stamp = skb->tstamp; - qp->q.mono_delivery_time = skb->mono_delivery_time; + qp->q.tstamp_type = skb->tstamp_type; qp->q.meat += skb->len; qp->ecn |= ecn; add_frag_mem_limit(qp->q.fqdir, skb->truesize); diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 765bd3f2a8408..b8cfe6afc84b8 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -764,7 +764,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, { struct iphdr *iph; struct sk_buff *skb2; - bool mono_delivery_time = skb->mono_delivery_time; + u8 tstamp_type = skb->tstamp_type; struct rtable *rt = skb_rtable(skb); unsigned int mtu, hlen, ll_rs; struct ip_fraglist_iter iter; @@ -856,7 +856,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, } }

- skb_set_delivery_time(skb, tstamp, mono_delivery_time); + skb_set_delivery_time(skb, tstamp, tstamp_type); err = output(net, sk, skb);

if (!err) @@ -912,7 +912,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, /* * Put this fragment into the sending queue. */ - skb_set_delivery_time(skb2, tstamp, mono_delivery_time); + skb_set_delivery_time(skb2, tstamp, tstamp_type); err = output(net, sk, skb2); if (err) goto fail; @@ -1648,7 +1648,8 @@ void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb, arg->csumoffset) = csum_fold(csum_add(nskb->csum, arg->csum)); nskb->ip_summed = CHECKSUM_NONE; - nskb->mono_delivery_time = !!transmit_time; + if (transmit_time) + nskb->tstamp_type = SKB_CLOCK_MONOTONIC; if (txhash) skb_set_hash(nskb, txhash, PKT_HASH_TYPE_L4); ip_push_pending_frames(sk, &fl4); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 3771ed22c2f56..560273e7f7736 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1266,7 +1266,7 @@ static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, tp = tcp_sk(sk); prior_wstamp = tp->tcp_wstamp_ns; tp->tcp_wstamp_ns = max(tp->tcp_wstamp_ns, tp->tcp_clock_cache); - skb_set_delivery_time(skb, tp->tcp_wstamp_ns, true); + skb_set_delivery_time(skb, tp->tcp_wstamp_ns, SKB_CLOCK_MONOTONIC); if (clone_it) { oskb = skb;

@@ -1607,7 +1607,7 @@ int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue,

skb_split(skb, buff, len);

- skb_set_delivery_time(buff, skb->tstamp, true); + skb_set_delivery_time(buff, skb->tstamp, SKB_CLOCK_MONOTONIC); tcp_fragment_tstamp(skb, buff);

old_factor = tcp_skb_pcount(skb); @@ -2703,7 +2703,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, if (unlikely(tp->repair) && tp->repair_queue == TCP_SEND_QUEUE) { /* "skb_mstamp_ns" is used as a start point for the retransmit timer */ tp->tcp_wstamp_ns = tp->tcp_clock_cache; - skb_set_delivery_time(skb, tp->tcp_wstamp_ns, true); + skb_set_delivery_time(skb, tp->tcp_wstamp_ns, SKB_CLOCK_MONOTONIC); list_move_tail(&skb->tcp_tsorted_anchor, &tp->tsorted_sent_queue); tcp_init_tso_segs(skb, mss_now); goto repair; /* Skip network transmission */ @@ -3688,11 +3688,11 @@ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst, #ifdef CONFIG_SYN_COOKIES if (unlikely(synack_type == TCP_SYNACK_COOKIE && ireq->tstamp_ok)) skb_set_delivery_time(skb, cookie_init_timestamp(req, now), - true); + SKB_CLOCK_MONOTONIC); else #endif { - skb_set_delivery_time(skb, now, true); + skb_set_delivery_time(skb, now, SKB_CLOCK_MONOTONIC); if (!tcp_rsk(req)->snt_synack) /* Timestamp first SYNACK */ tcp_rsk(req)->snt_synack = tcp_skb_timestamp_us(skb); } @@ -3741,7 +3741,7 @@ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst, bpf_skops_write_hdr_opt((struct sock *)sk, skb, req, syn_skb, synack_type, &opts);

- skb_set_delivery_time(skb, now, true); + skb_set_delivery_time(skb, now, SKB_CLOCK_MONOTONIC); tcp_add_tx_delay(skb, tp);

return skb; @@ -3923,7 +3923,7 @@ static int tcp_send_syn_data(struct sock *sk, struct sk_buff *syn)

err = tcp_transmit_skb(sk, syn_data, 1, sk->sk_allocation);

- skb_set_delivery_time(syn, syn_data->skb_mstamp_ns, true); + skb_set_delivery_time(syn, syn_data->skb_mstamp_ns, SKB_CLOCK_MONOTONIC);

/* Now full SYN+DATA was cloned and sent (or not), * remove the SYN from the original skb (syn_data) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index cd89a2b35dfb5..c86d5dca29df0 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -864,7 +864,7 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, struct rt6_info *rt = dst_rt6_info(skb_dst(skb)); struct ipv6_pinfo *np = skb->sk && !dev_recursion_level() ? inet6_sk(skb->sk) : NULL; - bool mono_delivery_time = skb->mono_delivery_time; + u8 tstamp_type = skb->tstamp_type; struct ip6_frag_state state; unsigned int mtu, hlen, nexthdr_offset; ktime_t tstamp = skb->tstamp; @@ -958,7 +958,7 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, if (iter.frag) ip6_fraglist_prepare(skb, &iter);

- skb_set_delivery_time(skb, tstamp, mono_delivery_time); + skb_set_delivery_time(skb, tstamp, tstamp_type); err = output(net, sk, skb); if (!err) IP6_INC_STATS(net, ip6_dst_idev(&rt->dst), @@ -1019,7 +1019,7 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, /* * Put this fragment into the sending queue. */ - skb_set_delivery_time(frag, tstamp, mono_delivery_time); + skb_set_delivery_time(frag, tstamp, tstamp_type); err = output(net, sk, frag); if (err) goto fail; diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 857713d7a38a5..7c4af48d529e1 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -126,7 +126,7 @@ int br_ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, struct sk_buff *)) { int frag_max_size = BR_INPUT_SKB_CB(skb)->frag_max_size; - bool mono_delivery_time = skb->mono_delivery_time; + u8 tstamp_type = skb->tstamp_type; ktime_t tstamp = skb->tstamp; struct ip6_frag_state state; u8 *prevhdr, nexthdr = 0; @@ -192,7 +192,7 @@ int br_ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, if (iter.frag) ip6_fraglist_prepare(skb, &iter);

- skb_set_delivery_time(skb, tstamp, mono_delivery_time); + skb_set_delivery_time(skb, tstamp, tstamp_type); err = output(net, sk, data, skb); if (err || !iter.frag) break; @@ -225,7 +225,7 @@ int br_ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, goto blackhole; }

- skb_set_delivery_time(skb2, tstamp, mono_delivery_time); + skb_set_delivery_time(skb2, tstamp, tstamp_type); err = output(net, sk, data, skb2); if (err) goto blackhole; diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index c78b13ea5b196..82e51b2ec4f51 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -268,7 +268,7 @@ static int nf_ct_frag6_queue(struct frag_queue *fq, struct sk_buff *skb, fq->iif = dev->ifindex;

fq->q.stamp = skb->tstamp; - fq->q.mono_delivery_time = skb->mono_delivery_time; + fq->q.tstamp_type = skb->tstamp_type; fq->q.meat += skb->len; fq->ecn |= ecn; if (payload_len > fq->q.max_size) diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 2af98edef87ee..cb219d4bdf25e 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -198,7 +198,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, fq->iif = dev->ifindex;

fq->q.stamp = skb->tstamp; - fq->q.mono_delivery_time = skb->mono_delivery_time; + fq->q.tstamp_type = skb->tstamp_type; fq->q.meat += skb->len; fq->ecn |= ecn; add_frag_mem_limit(fq->q.fqdir, skb->truesize); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index f285e52b8b857..624ab1424eba7 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -934,7 +934,7 @@ static void tcp_v6_send_response(const struct sock *sk, struct sk_buff *skb, u32 mark = inet_twsk(sk)->tw_mark; else mark = READ_ONCE(sk->sk_mark); - skb_set_delivery_time(buff, tcp_transmit_time(sk), true); + skb_set_delivery_time(buff, tcp_transmit_time(sk), SKB_CLOCK_MONOTONIC); } if (txhash) { /* autoflowlabel/skb_get_hash_flowi6 rely on buff->hash */ diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index b0455fda7d0ba..ac87fcff4795e 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -54,8 +54,8 @@ TC_INDIRECT_SCOPE int tcf_bpf_act(struct sk_buff *skb, bpf_compute_data_pointers(skb); filter_res = bpf_prog_run(filter, skb); } - if (unlikely(!skb->tstamp && skb->mono_delivery_time)) - skb->mono_delivery_time = 0; + if (unlikely(!skb->tstamp && skb->tstamp_type)) + skb->tstamp_type = SKB_CLOCK_REALTIME; if (skb_sk_is_prefetched(skb) && filter_res != TC_ACT_OK) skb_orphan(skb);

diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c index 382c7a71f81f2..db7151c6b70b7 100644 --- a/net/sched/cls_bpf.c +++ b/net/sched/cls_bpf.c @@ -104,8 +104,8 @@ TC_INDIRECT_SCOPE int cls_bpf_classify(struct sk_buff *skb, bpf_compute_data_pointers(skb); filter_res = bpf_prog_run(prog->filter, skb); } - if (unlikely(!skb->tstamp && skb->mono_delivery_time)) - skb->mono_delivery_time = 0; + if (unlikely(!skb->tstamp && skb->tstamp_type)) + skb->tstamp_type = SKB_CLOCK_REALTIME;

if (prog->exts_integrated) { res->class = 0;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vladimir Oltean vladimir.oltean@nxp.com

[ Upstream commit 36dd1141be70b5966906919714dc504a24c65ddf ]

I was revisiting the topic of 802.1ad treatment in the Ocelot switch [0] and realized that not only is its basic VLAN classification pipeline improper for offloading vlan_protocol 802.1ad bridges, but also improper for offloading regular 802.1Q bridges already.

Namely, 802.1ad-tagged traffic should be treated as VLAN-untagged by bridged ports, but this switch treats it as if it was 802.1Q-tagged with the same VID as in the 802.1ad header. This is markedly different to what the Linux bridge expects; see the "other_tpid()" function in tools/testing/selftests/net/forwarding/bridge_vlan_aware.sh.

An idea came to me that the VCAP IS1 TCAM is more powerful than I'm giving it credit for, and that it actually overwrites the classified VID before the VLAN Table lookup takes place. In other words, it can be used even to save a packet from being dropped on ingress due to VLAN membership.

Add a sophisticated TCAM rule hardcoded into the driver to force the switch to behave like a Linux bridge with vlan_filtering 1 vlan_protocol 802.1Q.

Regarding the lifetime of the filter: eventually the bridge will disappear, and vlan_filtering on the port will be restored to 0 for standalone mode. Then the filter will be deleted.

[0]: https://lore.kernel.org/netdev/20201009122947.nvhye4hvcha3tljh@skbuf/

Fixes: 7142529f1688 ("net: mscc: ocelot: add VLAN filtering") Signed-off-by: Vladimir Oltean vladimir.oltean@nxp.com Signed-off-by: David S. Miller davem@davemloft.net Stable-dep-of: 5ec6d7d737a4 ("net: mscc: ocelot: delete PVID VLAN when readding it as non-PVID") Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/mscc/ocelot.c | 188 ++++++++++++++++++++++-- drivers/net/ethernet/mscc/ocelot_vcap.c | 1 + include/soc/mscc/ocelot_vcap.h | 2 + 3 files changed, 180 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c index f6aa5d6b6597e..d3e7aff5a794a 100644 --- a/drivers/net/ethernet/mscc/ocelot.c +++ b/drivers/net/ethernet/mscc/ocelot.c @@ -453,9 +453,158 @@ static u16 ocelot_vlan_unaware_pvid(struct ocelot *ocelot, return VLAN_N_VID - bridge_num - 1; }

+/** + * ocelot_update_vlan_reclassify_rule() - Make switch aware only to bridge VLAN TPID + * + * @ocelot: Switch private data structure + * @port: Index of ingress port + * + * IEEE 802.1Q-2018 clauses "5.5 C-VLAN component conformance" and "5.6 S-VLAN + * component conformance" suggest that a C-VLAN component should only recognize + * and filter on C-Tags, and an S-VLAN component should only recognize and + * process based on C-Tags. + * + * In Linux, as per commit 1a0b20b25732 ("Merge branch 'bridge-next'"), C-VLAN + * components are largely represented by a bridge with vlan_protocol 802.1Q, + * and S-VLAN components by a bridge with vlan_protocol 802.1ad. + * + * Currently the driver only offloads vlan_protocol 802.1Q, but the hardware + * design is non-conformant, because the switch assigns each frame to a VLAN + * based on an entirely different question, as detailed in figure "Basic VLAN + * Classification Flow" from its manual and reproduced below. + * + * Set TAG_TYPE, PCP, DEI, VID to port-default values in VLAN_CFG register + * if VLAN_AWARE_ENA[port] and frame has outer tag then: + * if VLAN_INNER_TAG_ENA[port] and frame has inner tag then: + * TAG_TYPE = (Frame.InnerTPID <> 0x8100) + * Set PCP, DEI, VID to values from inner VLAN header + * else: + * TAG_TYPE = (Frame.OuterTPID <> 0x8100) + * Set PCP, DEI, VID to values from outer VLAN header + * if VID == 0 then: + * VID = VLAN_CFG.VLAN_VID + * + * Summarized, the switch will recognize both 802.1Q and 802.1ad TPIDs as VLAN + * "with equal rights", and just set the TAG_TYPE bit to 0 (if 802.1Q) or to 1 + * (if 802.1ad). It will classify based on whichever of the tags is "outer", no + * matter what TPID that may have (or "inner", if VLAN_INNER_TAG_ENA[port]). + * + * In the VLAN Table, the TAG_TYPE information is not accessible - just the + * classified VID is - so it is as if each VLAN Table entry is for 2 VLANs: + * C-VLAN X, and S-VLAN X. + * + * Whereas the Linux bridge behavior is to only filter on frames with a TPID + * equal to the vlan_protocol, and treat everything else as VLAN-untagged. + * + * Consider an ingress packet tagged with 802.1ad VID=3 and 802.1Q VID=5, + * received on a bridge vlan_filtering=1 vlan_protocol=802.1Q port. This frame + * should be treated as 802.1Q-untagged, and classified to the PVID of that + * bridge port. Not to VID=3, and not to VID=5. + * + * The VCAP IS1 TCAM has everything we need to overwrite the choices made in + * the basic VLAN classification pipeline: it can match on TAG_TYPE in the key, + * and it can modify the classified VID in the action. Thus, for each port + * under a vlan_filtering bridge, we can insert a rule in VCAP IS1 lookup 0 to + * match on 802.1ad tagged frames and modify their classified VID to the 802.1Q + * PVID of the port. This effectively makes it appear to the outside world as + * if those packets were processed as VLAN-untagged. + * + * The rule needs to be updated each time the bridge PVID changes, and needs + * to be deleted if the bridge PVID is deleted, or if the port becomes + * VLAN-unaware. + */ +static int ocelot_update_vlan_reclassify_rule(struct ocelot *ocelot, int port) +{ + unsigned long cookie = OCELOT_VCAP_IS1_VLAN_RECLASSIFY(ocelot, port); + struct ocelot_vcap_block *block_vcap_is1 = &ocelot->block[VCAP_IS1]; + struct ocelot_port *ocelot_port = ocelot->ports[port]; + const struct ocelot_bridge_vlan *pvid_vlan; + struct ocelot_vcap_filter *filter; + int err, val, pcp, dei; + bool vid_replace_ena; + u16 vid; + + pvid_vlan = ocelot_port->pvid_vlan; + vid_replace_ena = ocelot_port->vlan_aware && pvid_vlan; + + filter = ocelot_vcap_block_find_filter_by_id(block_vcap_is1, cookie, + false); + if (!vid_replace_ena) { + /* If the reclassification filter doesn't need to exist, delete + * it if it was previously installed, and exit doing nothing + * otherwise. + */ + if (filter) + return ocelot_vcap_filter_del(ocelot, filter); + + return 0; + } + + /* The reclassification rule must apply. See if it already exists + * or if it must be created. + */ + + /* Treating as VLAN-untagged means using as classified VID equal to + * the bridge PVID, and PCP/DEI set to the port default QoS values. + */ + vid = pvid_vlan->vid; + val = ocelot_read_gix(ocelot, ANA_PORT_QOS_CFG, port); + pcp = ANA_PORT_QOS_CFG_QOS_DEFAULT_VAL_X(val); + dei = !!(val & ANA_PORT_QOS_CFG_DP_DEFAULT_VAL); + + if (filter) { + bool changed = false; + + /* Filter exists, just update it */ + if (filter->action.vid != vid) { + filter->action.vid = vid; + changed = true; + } + if (filter->action.pcp != pcp) { + filter->action.pcp = pcp; + changed = true; + } + if (filter->action.dei != dei) { + filter->action.dei = dei; + changed = true; + } + + if (!changed) + return 0; + + return ocelot_vcap_filter_replace(ocelot, filter); + } + + /* Filter doesn't exist, create it */ + filter = kzalloc(sizeof(*filter), GFP_KERNEL); + if (!filter) + return -ENOMEM; + + filter->key_type = OCELOT_VCAP_KEY_ANY; + filter->ingress_port_mask = BIT(port); + filter->vlan.tpid = OCELOT_VCAP_BIT_1; + filter->prio = 1; + filter->id.cookie = cookie; + filter->id.tc_offload = false; + filter->block_id = VCAP_IS1; + filter->type = OCELOT_VCAP_FILTER_OFFLOAD; + filter->lookup = 0; + filter->action.vid_replace_ena = true; + filter->action.pcp_dei_ena = true; + filter->action.vid = vid; + filter->action.pcp = pcp; + filter->action.dei = dei; + + err = ocelot_vcap_filter_add(ocelot, filter, NULL); + if (err) + kfree(filter); + + return err; +} + /* Default vlan to clasify for untagged frames (may be zero) */ -static void ocelot_port_set_pvid(struct ocelot *ocelot, int port, - const struct ocelot_bridge_vlan *pvid_vlan) +static int ocelot_port_set_pvid(struct ocelot *ocelot, int port, + const struct ocelot_bridge_vlan *pvid_vlan) { struct ocelot_port *ocelot_port = ocelot->ports[port]; u16 pvid = ocelot_vlan_unaware_pvid(ocelot, ocelot_port->bridge); @@ -475,15 +624,23 @@ static void ocelot_port_set_pvid(struct ocelot *ocelot, int port, * happens automatically), but also 802.1p traffic which gets * classified to VLAN 0, but that is always in our RX filter, so it * would get accepted were it not for this setting. + * + * Also, we only support the bridge 802.1Q VLAN protocol, so + * 802.1ad-tagged frames (carrying S-Tags) should be considered + * 802.1Q-untagged, and also dropped. */ if (!pvid_vlan && ocelot_port->vlan_aware) val = ANA_PORT_DROP_CFG_DROP_PRIO_S_TAGGED_ENA | - ANA_PORT_DROP_CFG_DROP_PRIO_C_TAGGED_ENA; + ANA_PORT_DROP_CFG_DROP_PRIO_C_TAGGED_ENA | + ANA_PORT_DROP_CFG_DROP_S_TAGGED_ENA;

ocelot_rmw_gix(ocelot, val, ANA_PORT_DROP_CFG_DROP_PRIO_S_TAGGED_ENA | - ANA_PORT_DROP_CFG_DROP_PRIO_C_TAGGED_ENA, + ANA_PORT_DROP_CFG_DROP_PRIO_C_TAGGED_ENA | + ANA_PORT_DROP_CFG_DROP_S_TAGGED_ENA, ANA_PORT_DROP_CFG, port); + + return ocelot_update_vlan_reclassify_rule(ocelot, port); }

static struct ocelot_bridge_vlan *ocelot_bridge_vlan_find(struct ocelot *ocelot, @@ -631,7 +788,10 @@ int ocelot_port_vlan_filtering(struct ocelot *ocelot, int port, ANA_PORT_VLAN_CFG_VLAN_POP_CNT_M, ANA_PORT_VLAN_CFG, port);

- ocelot_port_set_pvid(ocelot, port, ocelot_port->pvid_vlan); + err = ocelot_port_set_pvid(ocelot, port, ocelot_port->pvid_vlan); + if (err) + return err; + ocelot_port_manage_port_tag(ocelot, port);

return 0; @@ -684,9 +844,12 @@ int ocelot_vlan_add(struct ocelot *ocelot, int port, u16 vid, bool pvid, return err;

/* Default ingress vlan classification */ - if (pvid) - ocelot_port_set_pvid(ocelot, port, - ocelot_bridge_vlan_find(ocelot, vid)); + if (pvid) { + err = ocelot_port_set_pvid(ocelot, port, + ocelot_bridge_vlan_find(ocelot, vid)); + if (err) + return err; + }

/* Untagged egress vlan clasification */ ocelot_port_manage_port_tag(ocelot, port); @@ -712,8 +875,11 @@ int ocelot_vlan_del(struct ocelot *ocelot, int port, u16 vid) return err;

/* Ingress */ - if (del_pvid) - ocelot_port_set_pvid(ocelot, port, NULL); + if (del_pvid) { + err = ocelot_port_set_pvid(ocelot, port, NULL); + if (err) + return err; + }

/* Egress */ ocelot_port_manage_port_tag(ocelot, port); @@ -2607,7 +2773,7 @@ int ocelot_port_set_default_prio(struct ocelot *ocelot, int port, u8 prio) ANA_PORT_QOS_CFG, port);

- return 0; + return ocelot_update_vlan_reclassify_rule(ocelot, port); } EXPORT_SYMBOL_GPL(ocelot_port_set_default_prio);

diff --git a/drivers/net/ethernet/mscc/ocelot_vcap.c b/drivers/net/ethernet/mscc/ocelot_vcap.c index 73cdec5ca6a34..5734b86aed5b5 100644 --- a/drivers/net/ethernet/mscc/ocelot_vcap.c +++ b/drivers/net/ethernet/mscc/ocelot_vcap.c @@ -695,6 +695,7 @@ static void is1_entry_set(struct ocelot *ocelot, int ix, vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_L2_MC, filter->dmac_mc); vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_L2_BC, filter->dmac_bc); vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_VLAN_TAGGED, tag->tagged); + vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_TPID, tag->tpid); vcap_key_set(vcap, &data, VCAP_IS1_HK_VID, tag->vid.value, tag->vid.mask); vcap_key_set(vcap, &data, VCAP_IS1_HK_PCP, diff --git a/include/soc/mscc/ocelot_vcap.h b/include/soc/mscc/ocelot_vcap.h index c601a4598b0da..eb19668a06db1 100644 --- a/include/soc/mscc/ocelot_vcap.h +++ b/include/soc/mscc/ocelot_vcap.h @@ -13,6 +13,7 @@ */ #define OCELOT_VCAP_ES0_TAG_8021Q_RXVLAN(ocelot, port, upstream) ((upstream) << 16 | (port)) #define OCELOT_VCAP_IS1_TAG_8021Q_TXVLAN(ocelot, port) (port) +#define OCELOT_VCAP_IS1_VLAN_RECLASSIFY(ocelot, port) ((ocelot)->num_phys_ports + (port)) #define OCELOT_VCAP_IS2_TAG_8021Q_TXVLAN(ocelot, port) (port) #define OCELOT_VCAP_IS2_MRP_REDIRECT(ocelot, port) ((ocelot)->num_phys_ports + (port)) #define OCELOT_VCAP_IS2_MRP_TRAP(ocelot) ((ocelot)->num_phys_ports * 2) @@ -499,6 +500,7 @@ struct ocelot_vcap_key_vlan { struct ocelot_vcap_u8 pcp; /* PCP (3 bit) */ enum ocelot_vcap_bit dei; /* DEI */ enum ocelot_vcap_bit tagged; /* Tagged/untagged frame */ + enum ocelot_vcap_bit tpid; };

struct ocelot_vcap_key_etype {

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Louis-Alexis Eyraud louisalexis.eyraud@collabora.com

[ Upstream commit 6fe0866014486736cc3ba1c6fd4606d3dbe55c9c ]

Use spin_lock_irqsave and spin_unlock_irqrestore instead of spin_lock and spin_unlock in mtk_star_emac driver to avoid spinlock recursion occurrence that can happen when enabling the DMA interrupts again in rx/tx poll.

``` BUG: spinlock recursion on CPU#0, swapper/0/0 lock: 0xffff00000db9cf20, .magic: dead4ead, .owner: swapper/0/0, .owner_cpu: 0 CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Not tainted 6.15.0-rc2-next-20250417-00001-gf6a27738686c-dirty #28 PREEMPT Hardware name: MediaTek MT8365 Open Platform EVK (DT) Call trace: show_stack+0x18/0x24 (C) dump_stack_lvl+0x60/0x80 dump_stack+0x18/0x24 spin_dump+0x78/0x88 do_raw_spin_lock+0x11c/0x120 _raw_spin_lock+0x20/0x2c mtk_star_handle_irq+0xc0/0x22c [mtk_star_emac] __handle_irq_event_percpu+0x48/0x140 handle_irq_event+0x4c/0xb0 handle_fasteoi_irq+0xa0/0x1bc handle_irq_desc+0x34/0x58 generic_handle_domain_irq+0x1c/0x28 gic_handle_irq+0x4c/0x120 do_interrupt_handler+0x50/0x84 el1_interrupt+0x34/0x68 el1h_64_irq_handler+0x18/0x24 el1h_64_irq+0x6c/0x70 regmap_mmio_read32le+0xc/0x20 (P) _regmap_bus_reg_read+0x6c/0xac _regmap_read+0x60/0xdc regmap_read+0x4c/0x80 mtk_star_rx_poll+0x2f4/0x39c [mtk_star_emac] __napi_poll+0x38/0x188 net_rx_action+0x164/0x2c0 handle_softirqs+0x100/0x244 __do_softirq+0x14/0x20 ____do_softirq+0x10/0x20 call_on_irq_stack+0x24/0x64 do_softirq_own_stack+0x1c/0x40 __irq_exit_rcu+0xd4/0x10c irq_exit_rcu+0x10/0x1c el1_interrupt+0x38/0x68 el1h_64_irq_handler+0x18/0x24 el1h_64_irq+0x6c/0x70 cpuidle_enter_state+0xac/0x320 (P) cpuidle_enter+0x38/0x50 do_idle+0x1e4/0x260 cpu_startup_entry+0x34/0x3c rest_init+0xdc/0xe0 console_on_rootfs+0x0/0x6c __primary_switched+0x88/0x90 ```

Fixes: 0a8bd81fd6aa ("net: ethernet: mtk-star-emac: separate tx/rx handling with two NAPIs") Signed-off-by: Louis-Alexis Eyraud louisalexis.eyraud@collabora.com Reviewed-by: Maxime Chevallier maxime.chevallier@bootlin.com Acked-by: Bartosz Golaszewski bartosz.golaszewski@linaro.org Link: https://patch.msgid.link/20250424-mtk_star_emac-fix-spinlock-recursion-issue... Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/mediatek/mtk_star_emac.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_star_emac.c b/drivers/net/ethernet/mediatek/mtk_star_emac.c index 25989c79c92e6..47a00e02365a2 100644 --- a/drivers/net/ethernet/mediatek/mtk_star_emac.c +++ b/drivers/net/ethernet/mediatek/mtk_star_emac.c @@ -1163,6 +1163,7 @@ static int mtk_star_tx_poll(struct napi_struct *napi, int budget) struct net_device *ndev = priv->ndev; unsigned int head = ring->head; unsigned int entry = ring->tail; + unsigned long flags;

while (entry != head && count < (MTK_STAR_RING_NUM_DESCS - 1)) { ret = mtk_star_tx_complete_one(priv); @@ -1182,9 +1183,9 @@ static int mtk_star_tx_poll(struct napi_struct *napi, int budget) netif_wake_queue(ndev);

if (napi_complete(napi)) { - spin_lock(&priv->lock); + spin_lock_irqsave(&priv->lock, flags); mtk_star_enable_dma_irq(priv, false, true); - spin_unlock(&priv->lock); + spin_unlock_irqrestore(&priv->lock, flags); }

return 0; @@ -1341,6 +1342,7 @@ static int mtk_star_rx(struct mtk_star_priv *priv, int budget) static int mtk_star_rx_poll(struct napi_struct *napi, int budget) { struct mtk_star_priv *priv; + unsigned long flags; int work_done = 0;

priv = container_of(napi, struct mtk_star_priv, rx_napi); @@ -1348,9 +1350,9 @@ static int mtk_star_rx_poll(struct napi_struct *napi, int budget) work_done = mtk_star_rx(priv, budget); if (work_done < budget) { napi_complete_done(napi, work_done); - spin_lock(&priv->lock); + spin_lock_irqsave(&priv->lock, flags); mtk_star_enable_dma_irq(priv, true, false); - spin_unlock(&priv->lock); + spin_unlock_irqrestore(&priv->lock, flags); }

return work_done;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shannon Nelson shannon.nelson@amd.com

[ Upstream commit f7b5bd725b737de3f2c4a836e07c82ba156d75df ]

Similar to what we do in the AdminQ, check for devcmd health while waiting for an answer.

Signed-off-by: Shannon Nelson shannon.nelson@amd.com Reviewed-by: Brett Creeley brett.creeley@amd.com Signed-off-by: David S. Miller davem@davemloft.net Stable-dep-of: dfd76010f8e8 ("pds_core: remove write-after-free of client_id") Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/amd/pds_core/dev.c | 11 +++++++++-- include/linux/pds/pds_core_if.h | 1 + 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/amd/pds_core/dev.c b/drivers/net/ethernet/amd/pds_core/dev.c index f0e39ab400450..e65a1632df505 100644 --- a/drivers/net/ethernet/amd/pds_core/dev.c +++ b/drivers/net/ethernet/amd/pds_core/dev.c @@ -42,6 +42,8 @@ int pdsc_err_to_errno(enum pds_core_status_code code) return -ERANGE; case PDS_RC_BAD_ADDR: return -EFAULT; + case PDS_RC_BAD_PCI: + return -ENXIO; case PDS_RC_EOPCODE: case PDS_RC_EINTR: case PDS_RC_DEV_CMD: @@ -65,7 +67,7 @@ bool pdsc_is_fw_running(struct pdsc *pdsc) /* Firmware is useful only if the running bit is set and * fw_status != 0xff (bad PCI read) */ - return (pdsc->fw_status != 0xff) && + return (pdsc->fw_status != PDS_RC_BAD_PCI) && (pdsc->fw_status & PDS_CORE_FW_STS_F_RUNNING); }

@@ -131,6 +133,7 @@ static int pdsc_devcmd_wait(struct pdsc *pdsc, u8 opcode, int max_seconds) unsigned long max_wait; unsigned long duration; int timeout = 0; + bool running; int done = 0; int err = 0; int status; @@ -139,6 +142,10 @@ static int pdsc_devcmd_wait(struct pdsc *pdsc, u8 opcode, int max_seconds) max_wait = start_time + (max_seconds * HZ);

while (!done && !timeout) { + running = pdsc_is_fw_running(pdsc); + if (!running) + break; + done = pdsc_devcmd_done(pdsc); if (done) break; @@ -155,7 +162,7 @@ static int pdsc_devcmd_wait(struct pdsc *pdsc, u8 opcode, int max_seconds) dev_dbg(dev, "DEVCMD %d %s after %ld secs\n", opcode, pdsc_devcmd_str(opcode), duration / HZ);

- if (!done || timeout) { + if ((!done || timeout) && running) { dev_err(dev, "DEVCMD %d %s timeout, done %d timeout %d max_seconds=%d\n", opcode, pdsc_devcmd_str(opcode), done, timeout, max_seconds); diff --git a/include/linux/pds/pds_core_if.h b/include/linux/pds/pds_core_if.h index e838a2b90440c..17a87c1a55d7c 100644 --- a/include/linux/pds/pds_core_if.h +++ b/include/linux/pds/pds_core_if.h @@ -79,6 +79,7 @@ enum pds_core_status_code { PDS_RC_EVFID = 31, /* VF ID does not exist */ PDS_RC_BAD_FW = 32, /* FW file is invalid or corrupted */ PDS_RC_ECLIENT = 33, /* No such client id */ + PDS_RC_BAD_PCI = 255, /* Broken PCI when reading status */ };

/**

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shannon Nelson shannon.nelson@amd.com

[ Upstream commit e8562da829432d04a0de1830146984c89844f35e ]

Since there really is no useful return, advertising a return value is rather misleading. Make pdsc_auxbus_dev_del() a void function.

Link: https://patch.msgid.link/r/20250320194412.67983-2-shannon.nelson@amd.com Reviewed-by: Leon Romanovsky leonro@nvidia.com Reviewed-by: Jonathan Cameron Jonathan.Cameron@huawei.com Reviewed-by: Kalesh AP kalesh-anakkur.purayil@broadcom.com Reviewed-by: Dave Jiang dave.jiang@intel.com Signed-off-by: Shannon Nelson shannon.nelson@amd.com Signed-off-by: Jason Gunthorpe jgg@nvidia.com Stable-dep-of: dfd76010f8e8 ("pds_core: remove write-after-free of client_id") Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/amd/pds_core/auxbus.c | 7 +------ drivers/net/ethernet/amd/pds_core/core.h | 2 +- drivers/net/ethernet/amd/pds_core/devlink.c | 6 ++++-- 3 files changed, 6 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/amd/pds_core/auxbus.c b/drivers/net/ethernet/amd/pds_core/auxbus.c index b76a9b7e0aed6..d53b2124b1498 100644 --- a/drivers/net/ethernet/amd/pds_core/auxbus.c +++ b/drivers/net/ethernet/amd/pds_core/auxbus.c @@ -172,13 +172,9 @@ static struct pds_auxiliary_dev *pdsc_auxbus_dev_register(struct pdsc *cf, return padev; }

-int pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf) +void pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf) { struct pds_auxiliary_dev *padev; - int err = 0; - - if (!cf) - return -ENODEV;

mutex_lock(&pf->config_lock);

@@ -192,7 +188,6 @@ int pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf) pf->vfs[cf->vf_id].padev = NULL;

mutex_unlock(&pf->config_lock); - return err; }

int pdsc_auxbus_dev_add(struct pdsc *cf, struct pdsc *pf) diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h index 858bebf797762..480f9e8cbc4d5 100644 --- a/drivers/net/ethernet/amd/pds_core/core.h +++ b/drivers/net/ethernet/amd/pds_core/core.h @@ -301,7 +301,7 @@ int pdsc_register_notify(struct notifier_block *nb); void pdsc_unregister_notify(struct notifier_block *nb); void pdsc_notify(unsigned long event, void *data); int pdsc_auxbus_dev_add(struct pdsc *cf, struct pdsc *pf); -int pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf); +void pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf);

void pdsc_process_adminq(struct pdsc_qcq *qcq); void pdsc_work_thread(struct work_struct *work); diff --git a/drivers/net/ethernet/amd/pds_core/devlink.c b/drivers/net/ethernet/amd/pds_core/devlink.c index 0032e8e351811..9b77bb73d25db 100644 --- a/drivers/net/ethernet/amd/pds_core/devlink.c +++ b/drivers/net/ethernet/amd/pds_core/devlink.c @@ -55,8 +55,10 @@ int pdsc_dl_enable_set(struct devlink *dl, u32 id, for (vf_id = 0; vf_id < pdsc->num_vfs; vf_id++) { struct pdsc *vf = pdsc->vfs[vf_id].vf;

- err = ctx->val.vbool ? pdsc_auxbus_dev_add(vf, pdsc) : - pdsc_auxbus_dev_del(vf, pdsc); + if (ctx->val.vbool) + err = pdsc_auxbus_dev_add(vf, pdsc); + else + pdsc_auxbus_dev_del(vf, pdsc); }

return err;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shannon Nelson shannon.nelson@amd.com

[ Upstream commit dfd76010f8e821b66116dec3c7d90dd2403d1396 ]

A use-after-free error popped up in stress testing:

[Mon Apr 21 21:21:33 2025] BUG: KFENCE: use-after-free write in pdsc_auxbus_dev_del+0xef/0x160 [pds_core] [Mon Apr 21 21:21:33 2025] Use-after-free write at 0x000000007013ecd1 (in kfence-#47): [Mon Apr 21 21:21:33 2025] pdsc_auxbus_dev_del+0xef/0x160 [pds_core] [Mon Apr 21 21:21:33 2025] pdsc_remove+0xc0/0x1b0 [pds_core] [Mon Apr 21 21:21:33 2025] pci_device_remove+0x24/0x70 [Mon Apr 21 21:21:33 2025] device_release_driver_internal+0x11f/0x180 [Mon Apr 21 21:21:33 2025] driver_detach+0x45/0x80 [Mon Apr 21 21:21:33 2025] bus_remove_driver+0x83/0xe0 [Mon Apr 21 21:21:33 2025] pci_unregister_driver+0x1a/0x80

The actual device uninit usually happens on a separate thread scheduled after this code runs, but there is no guarantee of order of thread execution, so this could be a problem. There's no actual need to clear the client_id at this point, so simply remove the offending code.

Fixes: 10659034c622 ("pds_core: add the aux client API") Signed-off-by: Shannon Nelson shannon.nelson@amd.com Reviewed-by: Simon Horman horms@kernel.org Link: https://patch.msgid.link/20250425203857.71547-1-shannon.nelson@amd.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/amd/pds_core/auxbus.c | 1 - 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/amd/pds_core/auxbus.c b/drivers/net/ethernet/amd/pds_core/auxbus.c index 4d3387bebe6a4..889a18962270a 100644 --- a/drivers/net/ethernet/amd/pds_core/auxbus.c +++ b/drivers/net/ethernet/amd/pds_core/auxbus.c @@ -186,7 +186,6 @@ void pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf, pds_client_unregister(pf, padev->client_id); auxiliary_device_delete(&padev->aux_dev); auxiliary_device_uninit(&padev->aux_dev); - padev->client_id = 0; *pd_ptr = NULL;

mutex_unlock(&pf->config_lock);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Victor Nogueira victor@mojatatu.com

[ Upstream commit 141d34391abbb315d68556b7c67ad97885407547 ]

As described in Gerrard's report [1], we have a UAF case when an hfsc class has a netem child qdisc. The crux of the issue is that hfsc is assuming that checking for cl->qdisc->q.qlen == 0 guarantees that it hasn't inserted the class in the vttree or eltree (which is not true for the netem duplicate case).

This patch checks the n_active class variable to make sure that the code won't insert the class in the vttree or eltree twice, catering for the reentrant case.

[1] https://lore.kernel.org/netdev/CAHcdcOm+03OD2j6R0=YHKqmy=VgJ8xEOKuP6c7mSgnp-...

Fixes: 37d9cf1a3ce3 ("sched: Fix detection of empty queues in child qdiscs") Reported-by: Gerrard Tai gerrard.tai@starlabs.sg Acked-by: Jamal Hadi Salim jhs@mojatatu.com Signed-off-by: Victor Nogueira victor@mojatatu.com Link: https://patch.msgid.link/20250425220710.3964791-3-victor@mojatatu.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- net/sched/sch_hfsc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index 371255e624332..4be8760b8bc3c 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1566,7 +1566,7 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) return err; }

- if (first) { + if (first && !cl->cl_nactive) { if (cl->cl_flags & HFSC_RSC) init_ed(cl, len); if (cl->cl_flags & HFSC_FSC)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Victor Nogueira victor@mojatatu.com

[ Upstream commit f139f37dcdf34b67f5bf92bc8e0f7f6b3ac63aa4 ]

As described in Gerrard's report [1], there are use cases where a netem child qdisc will make the parent qdisc's enqueue callback reentrant. In the case of qfq, there won't be a UAF, but the code will add the same classifier to the list twice, which will cause memory corruption.

This patch checks whether the class was already added to the agg->active list (cl_is_active) before doing the addition to cater for the reentrant case.

[1] https://lore.kernel.org/netdev/CAHcdcOm+03OD2j6R0=YHKqmy=VgJ8xEOKuP6c7mSgnp-...

Fixes: 37d9cf1a3ce3 ("sched: Fix detection of empty queues in child qdiscs") Acked-by: Jamal Hadi Salim jhs@mojatatu.com Signed-off-by: Victor Nogueira victor@mojatatu.com Link: https://patch.msgid.link/20250425220710.3964791-5-victor@mojatatu.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- net/sched/sch_qfq.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index 546c10adcacde..c7d4ee8cedff0 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -202,6 +202,11 @@ struct qfq_sched { */ enum update_reason {enqueue, requeue};

+static bool cl_is_active(struct qfq_class *cl) +{ + return !list_empty(&cl->alist); +} + static struct qfq_class *qfq_find_class(struct Qdisc *sch, u32 classid) { struct qfq_sched *q = qdisc_priv(sch); @@ -1217,7 +1222,6 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct qfq_class *cl; struct qfq_aggregate *agg; int err = 0; - bool first;

cl = qfq_classify(skb, sch, &err); if (cl == NULL) { @@ -1239,7 +1243,6 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch, }

gso_segs = skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1; - first = !cl->qdisc->q.qlen; err = qdisc_enqueue(skb, cl->qdisc, to_free); if (unlikely(err != NET_XMIT_SUCCESS)) { pr_debug("qfq_enqueue: enqueue failed %d\n", err); @@ -1255,8 +1258,8 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch, ++sch->q.qlen;

agg = cl->agg; - /* if the queue was not empty, then done here */ - if (!first) { + /* if the class is active, then done here */ + if (cl_is_active(cl)) { if (unlikely(skb == cl->qdisc->ops->peek(cl->qdisc)) && list_first_entry(&agg->active, struct qfq_class, alist) == cl && cl->deficit < len)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Takashi Iwai tiwai@suse.de

[ Upstream commit 56f1f30e6795b890463d9b20b11e576adf5a2f77 ]

The conversion function from MIDI 1.0 to UMP packet contains an internal buffer to keep the incoming MIDI bytes, and its size is 4, as it was supposed to be the max size for a MIDI1 UMP packet data. However, the implementation overlooked that SysEx is handled in a different format, and it can be up to 6 bytes, as found in do_convert_to_ump(). It leads eventually to a buffer overflow, and may corrupt the memory when a longer SysEx message is received.

The fix is simply to extend the buffer size to 6 to fit with the SysEx UMP message.

Fixes: 0b5288f5fe63 ("ALSA: ump: Add legacy raw MIDI support") Reported-by: Argusee vr@darknavy.com Link: https://patch.msgid.link/20250429124845.25128-1-tiwai@suse.de Signed-off-by: Takashi Iwai tiwai@suse.de Signed-off-by: Sasha Levin sashal@kernel.org --- include/sound/ump_convert.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/sound/ump_convert.h b/include/sound/ump_convert.h index d099ae27f8491..682499b871eac 100644 --- a/include/sound/ump_convert.h +++ b/include/sound/ump_convert.h @@ -19,7 +19,7 @@ struct ump_cvt_to_ump_bank { /* context for converting from MIDI1 byte stream to UMP packet */ struct ump_cvt_to_ump { /* MIDI1 intermediate buffer */ - unsigned char buf[4]; + unsigned char buf[6]; /* up to 6 bytes for SysEx */ int len; int cmd_bytes;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Simon Horman horms@kernel.org

[ Upstream commit e7e5ae71831c44d58627a991e603845a2fed2cab ]

As it's name suggests, parse_eeprom() parses EEPROM data.

This is done by reading data, 16 bits at a time as follows:

for (i = 0; i < 128; i++) ((__le16 *) sromdata)[i] = cpu_to_le16(read_eeprom(np, i));

sromdata is at the same memory location as psrom. And the type of psrom is a pointer to struct t_SROM.

As can be seen in the loop above, data is stored in sromdata, and thus psrom, as 16-bit little-endian values.

However, the integer fields of t_SROM are host byte order integers. And in the case of led_mode this leads to a little endian value being incorrectly treated as host byte order.

Looking at rio_set_led_mode, this does appear to be a bug as that code masks led_mode with 0x1, 0x2 and 0x8. Logic that would be effected by a reversed byte order.

This problem would only manifest on big endian hosts.

Found by inspection while investigating a sparse warning regarding the crc field of t_SROM.

I believe that warning is a false positive. And although I plan to send a follow-up to use little-endian types for other the integer fields of PSROM_t I do not believe that will involve any bug fixes.

Compile tested only.

Fixes: c3f45d322cbd ("dl2k: Add support for IP1000A-based cards") Signed-off-by: Simon Horman horms@kernel.org Link: https://patch.msgid.link/20250425-dlink-led-mode-v1-1-6bae3c36e736@kernel.or... Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/dlink/dl2k.c | 2 +- drivers/net/ethernet/dlink/dl2k.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c index db6615aa921b1..ce46f3ac3b5a1 100644 --- a/drivers/net/ethernet/dlink/dl2k.c +++ b/drivers/net/ethernet/dlink/dl2k.c @@ -352,7 +352,7 @@ parse_eeprom (struct net_device *dev) eth_hw_addr_set(dev, psrom->mac_addr);

if (np->chip_id == CHIP_IP1000A) { - np->led_mode = psrom->led_mode; + np->led_mode = le16_to_cpu(psrom->led_mode); return 0; }

diff --git a/drivers/net/ethernet/dlink/dl2k.h b/drivers/net/ethernet/dlink/dl2k.h index 195dc6cfd8955..0e33e2eaae960 100644 --- a/drivers/net/ethernet/dlink/dl2k.h +++ b/drivers/net/ethernet/dlink/dl2k.h @@ -335,7 +335,7 @@ typedef struct t_SROM { u16 sub_system_id; /* 0x06 */ u16 pci_base_1; /* 0x08 (IP1000A only) */ u16 pci_base_2; /* 0x0a (IP1000A only) */ - u16 led_mode; /* 0x0c (IP1000A only) */ + __le16 led_mode; /* 0x0c (IP1000A only) */ u16 reserved1[9]; /* 0x0e-0x1f */ u8 mac_addr[6]; /* 0x20-0x25 */ u8 reserved2[10]; /* 0x26-0x2f */

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jacob Keller jacob.e.keller@intel.com

[ Upstream commit c7d6cb96d5c33b5148f3dc76fcd30a9b8cd9e973 ]

Commit 1a931c4f5e68 ("igc: add lock preventing multiple simultaneous PTM transactions") added a new mutex to protect concurrent PTM transactions. This lock is acquired in igc_ptp_reset() in order to ensure the PTM registers are properly disabled after a device reset.

The flow where the lock is acquired already holds a spinlock, so acquiring a mutex leads to a sleep-while-locking bug, reported both by smatch, and the kernel test robot.

The critical section in igc_ptp_reset() does correctly use the readx_poll_timeout_atomic variants, but the standard PTM flow uses regular sleeping variants. This makes converting the mutex to a spinlock a bit tricky.

Instead, re-order the locking in igc_ptp_reset. Acquire the mutex first, and then the tmreg_lock spinlock. This is safe because there is no other ordering dependency on these locks, as this is the only place where both locks were acquired simultaneously. Indeed, any other flow acquiring locks in that order would be wrong regardless.

Signed-off-by: Jacob Keller jacob.e.keller@intel.com Fixes: 1a931c4f5e68 ("igc: add lock preventing multiple simultaneous PTM transactions") Link: https://lore.kernel.org/intel-wired-lan/Z_-P-Hc1yxcw0lTB@stanley.mountain/ Link: https://lore.kernel.org/intel-wired-lan/202504211511.f7738f5d-lkp@intel.com/... Reviewed-by: Przemek Kitszel przemyslaw.kitszel@intel.com Reviewed-by: Vitaly Lifshits vitaly.lifshits@intel.com Tested-by: Mor Bar-Gabay morx.bar.gabay@intel.com Signed-off-by: Tony Nguyen anthony.l.nguyen@intel.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/intel/igc/igc_ptp.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c index b6bb01a486d9d..a82af96e6bd12 100644 --- a/drivers/net/ethernet/intel/igc/igc_ptp.c +++ b/drivers/net/ethernet/intel/igc/igc_ptp.c @@ -1237,6 +1237,8 @@ void igc_ptp_reset(struct igc_adapter *adapter) /* reset the tstamp_config */ igc_ptp_set_timestamp_mode(adapter, &adapter->tstamp_config);

+ mutex_lock(&adapter->ptm_lock); + spin_lock_irqsave(&adapter->tmreg_lock, flags);

switch (adapter->hw.mac.type) { @@ -1255,7 +1257,6 @@ void igc_ptp_reset(struct igc_adapter *adapter) if (!igc_is_crosststamp_supported(adapter)) break;

- mutex_lock(&adapter->ptm_lock); wr32(IGC_PCIE_DIG_DELAY, IGC_PCIE_DIG_DELAY_DEFAULT); wr32(IGC_PCIE_PHY_DELAY, IGC_PCIE_PHY_DELAY_DEFAULT);

@@ -1279,7 +1280,6 @@ void igc_ptp_reset(struct igc_adapter *adapter) netdev_err(adapter->netdev, "Timeout reading IGC_PTM_STAT register\n");

igc_ptm_reset(hw); - mutex_unlock(&adapter->ptm_lock); break; default: /* No work to do. */ @@ -1296,5 +1296,7 @@ void igc_ptp_reset(struct igc_adapter *adapter) out: spin_unlock_irqrestore(&adapter->tmreg_lock, flags);

+ mutex_unlock(&adapter->ptm_lock); + wrfl(); }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vladimir Oltean vladimir.oltean@nxp.com

[ Upstream commit 426d487bca38b34f39c483edfc6313a036446b33 ]

Simplest setup to reproduce the issue: connect 2 ports of the LS1028A-RDB together (eno0 with swp0) and run:

$ ip link set eno0 up && ip link set swp0 up $ tc qdisc replace dev swp0 parent root handle 100 taprio num_tc 8 \ queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 map 0 1 2 3 4 5 6 7 \ base-time 0 sched-entry S 20 300000 sched-entry S 10 200000 \ sched-entry S 20 300000 sched-entry S 48 200000 \ sched-entry S 20 300000 sched-entry S 83 200000 \ sched-entry S 40 300000 sched-entry S 00 200000 flags 2 $ ptp4l -i eno0 -f /etc/linuxptp/configs/gPTP.cfg -m & $ ptp4l -i swp0 -f /etc/linuxptp/configs/gPTP.cfg -m

One will observe that the PTP state machine on swp0 starts synchronizing, then it attempts to do a clock step, and after that, it never fails to recover from the condition below.

ptp4l[82.427]: selected best master clock 00049f.fffe.05f627 ptp4l[82.428]: port 1 (swp0): MASTER to UNCALIBRATED on RS_SLAVE ptp4l[83.252]: port 1 (swp0): UNCALIBRATED to SLAVE on MASTER_CLOCK_SELECTED ptp4l[83.886]: rms 4537731277 max 9075462553 freq -18518 +/- 11467 delay 818 +/- 0 ptp4l[84.170]: timed out while polling for tx timestamp ptp4l[84.171]: increasing tx_timestamp_timeout or increasing kworker priority may correct this issue, but a driver bug likely causes it ptp4l[84.172]: port 1 (swp0): send peer delay request failed ptp4l[84.173]: port 1 (swp0): clearing fault immediately ptp4l[84.269]: port 1 (swp0): SLAVE to LISTENING on INIT_COMPLETE ptp4l[85.303]: timed out while polling for tx timestamp ptp4l[84.171]: increasing tx_timestamp_timeout or increasing kworker priority may correct this issue, but a driver bug likely causes it ptp4l[84.172]: port 1 (swp0): send peer delay request failed ptp4l[84.173]: port 1 (swp0): clearing fault immediately ptp4l[84.269]: port 1 (swp0): SLAVE to LISTENING on INIT_COMPLETE ptp4l[85.303]: timed out while polling for tx timestamp ptp4l[85.304]: increasing tx_timestamp_timeout or increasing kworker priority may correct this issue, but a driver bug likely causes it ptp4l[85.305]: port 1 (swp0): send peer delay response failed ptp4l[85.306]: port 1 (swp0): clearing fault immediately ptp4l[86.304]: timed out while polling for tx timestamp

A hint is given by the non-zero statistics for dropped packets which were expecting hardware TX timestamps:

$ ethtool --include-statistics -T swp0 (...) Statistics: tx_pkts: 30 tx_lost: 11 tx_err: 0

We know that when PTP clock stepping takes place (from ocelot_ptp_settime64() or from ocelot_ptp_adjtime()), vsc9959_tas_clock_adjust() is called.

Another interesting hint is that placing an early return in vsc9959_tas_clock_adjust(), so as to neutralize this function, fixes the issue and TX timestamps are no longer dropped.

The debugging function written by me and included below is intended to read the GCL RAM, after the admin schedule became operational, through the two status registers available for this purpose: QSYS_GCL_STATUS_REG_1 and QSYS_GCL_STATUS_REG_2.

static void vsc9959_print_tas_gcl(struct ocelot *ocelot) { u32 val, list_length, interval, gate_state; int i, err;

err = read_poll_timeout(ocelot_read, val, !(val & QSYS_PARAM_STATUS_REG_8_CONFIG_PENDING), 10, 100000, false, ocelot, QSYS_PARAM_STATUS_REG_8); if (err) { dev_err(ocelot->dev, "Failed to wait for TAS config pending bit to clear: %pe\n", ERR_PTR(err)); return; }

val = ocelot_read(ocelot, QSYS_PARAM_STATUS_REG_3); list_length = QSYS_PARAM_STATUS_REG_3_LIST_LENGTH_X(val);

dev_info(ocelot->dev, "GCL length: %u\n", list_length);

for (i = 0; i < list_length; i++) { ocelot_rmw(ocelot, QSYS_GCL_STATUS_REG_1_GCL_ENTRY_NUM(i), QSYS_GCL_STATUS_REG_1_GCL_ENTRY_NUM_M, QSYS_GCL_STATUS_REG_1); interval = ocelot_read(ocelot, QSYS_GCL_STATUS_REG_2); val = ocelot_read(ocelot, QSYS_GCL_STATUS_REG_1); gate_state = QSYS_GCL_STATUS_REG_1_GATE_STATE_X(val);

dev_info(ocelot->dev, "GCL entry %d: states 0x%x interval %u\n", i, gate_state, interval); } }

Calling it from two places: after the initial QSYS_TAS_PARAM_CFG_CTRL_CONFIG_CHANGE performed by vsc9959_qos_port_tas_set(), and after the one done by vsc9959_tas_clock_adjust(), I notice the following difference.

From the tc-taprio process context, where the schedule was initially

configured, the GCL looks like this:

mscc_felix 0000:00:00.5: GCL length: 8 mscc_felix 0000:00:00.5: GCL entry 0: states 0x20 interval 300000 mscc_felix 0000:00:00.5: GCL entry 1: states 0x10 interval 200000 mscc_felix 0000:00:00.5: GCL entry 2: states 0x20 interval 300000 mscc_felix 0000:00:00.5: GCL entry 3: states 0x48 interval 200000 mscc_felix 0000:00:00.5: GCL entry 4: states 0x20 interval 300000 mscc_felix 0000:00:00.5: GCL entry 5: states 0x83 interval 200000 mscc_felix 0000:00:00.5: GCL entry 6: states 0x40 interval 300000 mscc_felix 0000:00:00.5: GCL entry 7: states 0x0 interval 200000

But from the ptp4l clock stepping process context, when the vsc9959_tas_clock_adjust() hook is called, the GCL RAM of the operational schedule now looks like this:

mscc_felix 0000:00:00.5: GCL length: 8 mscc_felix 0000:00:00.5: GCL entry 0: states 0x0 interval 0 mscc_felix 0000:00:00.5: GCL entry 1: states 0x0 interval 0 mscc_felix 0000:00:00.5: GCL entry 2: states 0x0 interval 0 mscc_felix 0000:00:00.5: GCL entry 3: states 0x0 interval 0 mscc_felix 0000:00:00.5: GCL entry 4: states 0x0 interval 0 mscc_felix 0000:00:00.5: GCL entry 5: states 0x0 interval 0 mscc_felix 0000:00:00.5: GCL entry 6: states 0x0 interval 0 mscc_felix 0000:00:00.5: GCL entry 7: states 0x0 interval 0

I do not have a formal explanation, just experimental conclusions. It appears that after triggering QSYS_TAS_PARAM_CFG_CTRL_CONFIG_CHANGE for a port's TAS, the GCL entry RAM is updated anyway, despite what the documentation claims: "Specify the time interval in QSYS::GCL_CFG_REG_2.TIME_INTERVAL. This triggers the actual RAM write with the gate state and the time interval for the entry number specified". We don't touch that register (through vsc9959_tas_gcl_set()) from vsc9959_tas_clock_adjust(), yet the GCL RAM is updated anyway.

It seems to be updated with effectively stale memory, which in my testing can hold a variety of things, including even pieces of the previously applied schedule, for particular schedule lengths.

As such, in most circumstances it is very difficult to pinpoint this issue, because the newly updated schedule would "behave strangely", but ultimately might still pass traffic to some extent, due to some gate entries still being present in the stale GCL entry RAM. It is easy to miss.

With the particular schedule given at the beginning, the GCL RAM "happens" to be reproducibly rewritten with all zeroes, and this is consistent with what we see: when the time-aware shaper has gate entries with all gates closed, traffic is dropped on TX, no wonder we can't retrieve TX timestamps.

Rewriting the GCL entry RAM when reapplying the new base time fixes the observed issue.

Fixes: 8670dc33f48b ("net: dsa: felix: update base time of time-aware shaper when adjusting PTP time") Reported-by: Richie Pearn richard.pearn@nxp.com Signed-off-by: Vladimir Oltean vladimir.oltean@nxp.com Link: https://patch.msgid.link/20250426144859.3128352-2-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/dsa/ocelot/felix_vsc9959.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c index 8d27933c3733b..f91f25578f075 100644 --- a/drivers/net/dsa/ocelot/felix_vsc9959.c +++ b/drivers/net/dsa/ocelot/felix_vsc9959.c @@ -1543,7 +1543,7 @@ static void vsc9959_tas_clock_adjust(struct ocelot *ocelot) struct tc_taprio_qopt_offload *taprio; struct ocelot_port *ocelot_port; struct timespec64 base_ts; - int port; + int i, port; u32 val;

mutex_lock(&ocelot->fwd_domain_lock); @@ -1575,6 +1575,9 @@ static void vsc9959_tas_clock_adjust(struct ocelot *ocelot) QSYS_PARAM_CFG_REG_3_BASE_TIME_SEC_MSB_M, QSYS_PARAM_CFG_REG_3);

+ for (i = 0; i < taprio->num_entries; i++) + vsc9959_tas_gcl_set(ocelot, i, &taprio->entries[i]); + ocelot_rmw(ocelot, QSYS_TAS_PARAM_CFG_CTRL_CONFIG_CHANGE, QSYS_TAS_PARAM_CFG_CTRL_CONFIG_CHANGE, QSYS_TAS_PARAM_CFG_CTRL);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shruti Parab shruti.parab@broadcom.com

[ Upstream commit ea9376cf68230e05492f22ca45d329f16e262c7b ]

When handling HWRM_DBG_COREDUMP_LIST FW command in bnxt_hwrm_dbg_dma_data(), the allocated buffer info->dest_buf is not freed in the error path. In the normal path, info->dest_buf is assigned to coredump->data and it will eventually be freed after the coredump is collected.

Free info->dest_buf immediately inside bnxt_hwrm_dbg_dma_data() in the error path.

Fixes: c74751f4c392 ("bnxt_en: Return error if FW returns more data than dump length") Reported-by: Michael Chan michael.chan@broadcom.com Reviewed-by: Kalesh AP kalesh-anakkur.purayil@broadcom.com Signed-off-by: Shruti Parab shruti.parab@broadcom.com Signed-off-by: Michael Chan michael.chan@broadcom.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c | 5 +++++ 1 file changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c index c067898820360..b57d2a25ae276 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c @@ -72,6 +72,11 @@ static int bnxt_hwrm_dbg_dma_data(struct bnxt *bp, void *msg, memcpy(info->dest_buf + off, dma_buf, len); } else { rc = -ENOBUFS; + if (cmn_req->req_type == + cpu_to_le16(HWRM_DBG_COREDUMP_LIST)) { + kfree(info->dest_buf); + info->dest_buf = NULL; + } break; } }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Michael Chan michael.chan@broadcom.com

[ Upstream commit 02e8be5a032cae0f4ca33c6053c44d83cf4acc93 ]

For version 1 register dump that includes the PCIe stats, the existing code incorrectly assumes that all PCIe stats are 64-bit values. Fix it by using an array containing the starting and ending index of the 32-bit values. The loop in bnxt_get_regs() will use the array to do proper endian swap for the 32-bit values.

Fixes: b5d600b027eb ("bnxt_en: Add support for 'ethtool -d'") Reviewed-by: Shruti Parab shruti.parab@broadcom.com Reviewed-by: Kalesh AP kalesh-anakkur.purayil@broadcom.com Reviewed-by: Andy Gospodarek andrew.gospodarek@broadcom.com Signed-off-by: Michael Chan michael.chan@broadcom.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- .../net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 38 ++++++++++++++++--- 1 file changed, 32 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 2e7ddbca9d53b..dcedafa4d2e14 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -1393,6 +1393,17 @@ static int bnxt_get_regs_len(struct net_device *dev) return reg_len; }

+#define BNXT_PCIE_32B_ENTRY(start, end) \ + { offsetof(struct pcie_ctx_hw_stats, start), \ + offsetof(struct pcie_ctx_hw_stats, end) } + +static const struct { + u16 start; + u16 end; +} bnxt_pcie_32b_entries[] = { + BNXT_PCIE_32B_ENTRY(pcie_ltssm_histogram[0], pcie_ltssm_histogram[3]), +}; + static void bnxt_get_regs(struct net_device *dev, struct ethtool_regs *regs, void *_p) { @@ -1424,12 +1435,27 @@ static void bnxt_get_regs(struct net_device *dev, struct ethtool_regs *regs, req->pcie_stat_host_addr = cpu_to_le64(hw_pcie_stats_addr); rc = hwrm_req_send(bp, req); if (!rc) { - __le64 *src = (__le64 *)hw_pcie_stats; - u64 *dst = (u64 *)(_p + BNXT_PXP_REG_LEN); - int i; - - for (i = 0; i < sizeof(*hw_pcie_stats) / sizeof(__le64); i++) - dst[i] = le64_to_cpu(src[i]); + u8 *dst = (u8 *)(_p + BNXT_PXP_REG_LEN); + u8 *src = (u8 *)hw_pcie_stats; + int i, j; + + for (i = 0, j = 0; i < sizeof(*hw_pcie_stats); ) { + if (i >= bnxt_pcie_32b_entries[j].start && + i <= bnxt_pcie_32b_entries[j].end) { + u32 *dst32 = (u32 *)(dst + i); + + *dst32 = le32_to_cpu(*(__le32 *)(src + i)); + i += 4; + if (i > bnxt_pcie_32b_entries[j].end && + j < ARRAY_SIZE(bnxt_pcie_32b_entries) - 1) + j++; + } else { + u64 *dst64 = (u64 *)(dst + i); + + *dst64 = le64_to_cpu(*(__le64 *)(src + i)); + i += 8; + } + } } hwrm_req_drop(bp, req); }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thangaraj Samynathan thangaraj.s@microchip.com

[ Upstream commit 2d52e2e38b85c8b7bc00dca55c2499f46f8c8198 ]

Always map the `skb` to the LS descriptor. Previously skb was mapped to EXT descriptor when the number of fragments is zero with GSO enabled. Mapping the skb to EXT descriptor prevents it from being freed, leading to a memory leak

Fixes: 23f0703c125b ("lan743x: Add main source files for new lan743x driver") Signed-off-by: Thangaraj Samynathan thangaraj.s@microchip.com Reviewed-by: Jacob Keller jacob.e.keller@intel.com Link: https://patch.msgid.link/20250429052527.10031-1-thangaraj.s@microchip.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/microchip/lan743x_main.c | 8 ++++++-- drivers/net/ethernet/microchip/lan743x_main.h | 1 + 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c index 92010bfe5e413..5d2ceff72784f 100644 --- a/drivers/net/ethernet/microchip/lan743x_main.c +++ b/drivers/net/ethernet/microchip/lan743x_main.c @@ -1949,6 +1949,7 @@ static void lan743x_tx_frame_add_lso(struct lan743x_tx *tx, if (nr_frags <= 0) { tx->frame_data0 |= TX_DESC_DATA0_LS_; tx->frame_data0 |= TX_DESC_DATA0_IOC_; + tx->frame_last = tx->frame_first; } tx_descriptor = &tx->ring_cpu_ptr[tx->frame_tail]; tx_descriptor->data0 = cpu_to_le32(tx->frame_data0); @@ -2018,6 +2019,7 @@ static int lan743x_tx_frame_add_fragment(struct lan743x_tx *tx, tx->frame_first = 0; tx->frame_data0 = 0; tx->frame_tail = 0; + tx->frame_last = 0; return -ENOMEM; }

@@ -2058,16 +2060,18 @@ static void lan743x_tx_frame_end(struct lan743x_tx *tx, TX_DESC_DATA0_DTYPE_DATA_) { tx->frame_data0 |= TX_DESC_DATA0_LS_; tx->frame_data0 |= TX_DESC_DATA0_IOC_; + tx->frame_last = tx->frame_tail; }

- tx_descriptor = &tx->ring_cpu_ptr[tx->frame_tail]; - buffer_info = &tx->buffer_info[tx->frame_tail]; + tx_descriptor = &tx->ring_cpu_ptr[tx->frame_last]; + buffer_info = &tx->buffer_info[tx->frame_last]; buffer_info->skb = skb; if (time_stamp) buffer_info->flags |= TX_BUFFER_INFO_FLAG_TIMESTAMP_REQUESTED; if (ignore_sync) buffer_info->flags |= TX_BUFFER_INFO_FLAG_IGNORE_SYNC;

+ tx_descriptor = &tx->ring_cpu_ptr[tx->frame_tail]; tx_descriptor->data0 = cpu_to_le32(tx->frame_data0); tx->frame_tail = lan743x_tx_next_index(tx, tx->frame_tail); tx->last_tail = tx->frame_tail; diff --git a/drivers/net/ethernet/microchip/lan743x_main.h b/drivers/net/ethernet/microchip/lan743x_main.h index 3b2c6046eb3ad..b6c83c68241e6 100644 --- a/drivers/net/ethernet/microchip/lan743x_main.h +++ b/drivers/net/ethernet/microchip/lan743x_main.h @@ -974,6 +974,7 @@ struct lan743x_tx { u32 frame_first; u32 frame_data0; u32 frame_tail; + u32 frame_last;

struct lan743x_tx_buffer_info *buffer_info;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sathesh B Edara sedara@marvell.com

[ Upstream commit 34f42736b325287a7b2ce37e415838f539767bda ]

When the host loses heartbeat messages from the device, the driver calls the device-specific ndo_stop function, which frees the resources. If the driver is unloaded in this scenario, it calls ndo_stop again, attempting to free resources that have already been freed, leading to a host hang issue. To resolve this, dev_close should be called instead of the device-specific stop function.dev_close internally calls ndo_stop to stop the network interface and performs additional cleanup tasks. During the driver unload process, if the device is already down, ndo_stop is not called.

Fixes: 5cb96c29aa0e ("octeon_ep: add heartbeat monitor") Signed-off-by: Sathesh B Edara sedara@marvell.com Reviewed-by: Simon Horman horms@kernel.org Link: https://patch.msgid.link/20250429114624.19104-1-sedara@marvell.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/marvell/octeon_ep/octep_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c index 6f1fe7e283d4e..7a30095b3486f 100644 --- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c +++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c @@ -917,7 +917,7 @@ static void octep_hb_timeout_task(struct work_struct *work) miss_cnt); rtnl_lock(); if (netif_running(oct->netdev)) - octep_stop(oct->netdev); + dev_close(oct->netdev); rtnl_unlock(); }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yonglong Liu liuyonglong@huawei.com

[ Upstream commit 8e6b9c6ea5a55045eed6526d8ee49e93192d1a58 ]

When a VF is passthrough to a VM, and the VM is killed, the reported interrupt may not been handled, it will remain, and won't be clear by the nic engine even with a flr or tqp reset. When the VM restart, the interrupt of the first vector may be dropped by the second enable_irq in vfio, see the issue below: https://gitlab.com/qemu-project/qemu/-/issues/2884#note_2423361621

We notice that the vfio has always behaved this way, and the interrupt is a residue of the nic engine, so we fix the problem by moving the vector enable process out of the enable_irq loop.

Fixes: 08a100689d4b ("net: hns3: re-organize vector handle") Signed-off-by: Yonglong Liu liuyonglong@huawei.com Signed-off-by: Jijie Shao shaojijie@huawei.com Link: https://patch.msgid.link/20250430093052.2400464-3-shaojijie@huawei.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- .../net/ethernet/hisilicon/hns3/hns3_enet.c | 82 +++++++++---------- 1 file changed, 39 insertions(+), 43 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index 801801e8803e9..0ed01f4d68061 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -473,20 +473,14 @@ static void hns3_mask_vector_irq(struct hns3_enet_tqp_vector *tqp_vector, writel(mask_en, tqp_vector->mask_addr); }

-static void hns3_vector_enable(struct hns3_enet_tqp_vector *tqp_vector) +static void hns3_irq_enable(struct hns3_enet_tqp_vector *tqp_vector) { napi_enable(&tqp_vector->napi); enable_irq(tqp_vector->vector_irq); - - /* enable vector */ - hns3_mask_vector_irq(tqp_vector, 1); }

-static void hns3_vector_disable(struct hns3_enet_tqp_vector *tqp_vector) +static void hns3_irq_disable(struct hns3_enet_tqp_vector *tqp_vector) { - /* disable vector */ - hns3_mask_vector_irq(tqp_vector, 0); - disable_irq(tqp_vector->vector_irq); napi_disable(&tqp_vector->napi); cancel_work_sync(&tqp_vector->rx_group.dim.work); @@ -707,11 +701,42 @@ static int hns3_set_rx_cpu_rmap(struct net_device *netdev) return 0; }

+static void hns3_enable_irqs_and_tqps(struct net_device *netdev) +{ + struct hns3_nic_priv *priv = netdev_priv(netdev); + struct hnae3_handle *h = priv->ae_handle; + u16 i; + + for (i = 0; i < priv->vector_num; i++) + hns3_irq_enable(&priv->tqp_vector[i]); + + for (i = 0; i < priv->vector_num; i++) + hns3_mask_vector_irq(&priv->tqp_vector[i], 1); + + for (i = 0; i < h->kinfo.num_tqps; i++) + hns3_tqp_enable(h->kinfo.tqp[i]); +} + +static void hns3_disable_irqs_and_tqps(struct net_device *netdev) +{ + struct hns3_nic_priv *priv = netdev_priv(netdev); + struct hnae3_handle *h = priv->ae_handle; + u16 i; + + for (i = 0; i < h->kinfo.num_tqps; i++) + hns3_tqp_disable(h->kinfo.tqp[i]); + + for (i = 0; i < priv->vector_num; i++) + hns3_mask_vector_irq(&priv->tqp_vector[i], 0); + + for (i = 0; i < priv->vector_num; i++) + hns3_irq_disable(&priv->tqp_vector[i]); +} + static int hns3_nic_net_up(struct net_device *netdev) { struct hns3_nic_priv *priv = netdev_priv(netdev); struct hnae3_handle *h = priv->ae_handle; - int i, j; int ret;

ret = hns3_nic_reset_all_ring(h); @@ -720,23 +745,13 @@ static int hns3_nic_net_up(struct net_device *netdev)

clear_bit(HNS3_NIC_STATE_DOWN, &priv->state);

- /* enable the vectors */ - for (i = 0; i < priv->vector_num; i++) - hns3_vector_enable(&priv->tqp_vector[i]); - - /* enable rcb */ - for (j = 0; j < h->kinfo.num_tqps; j++) - hns3_tqp_enable(h->kinfo.tqp[j]); + hns3_enable_irqs_and_tqps(netdev);

/* start the ae_dev */ ret = h->ae_algo->ops->start ? h->ae_algo->ops->start(h) : 0; if (ret) { set_bit(HNS3_NIC_STATE_DOWN, &priv->state); - while (j--) - hns3_tqp_disable(h->kinfo.tqp[j]); - - for (j = i - 1; j >= 0; j--) - hns3_vector_disable(&priv->tqp_vector[j]); + hns3_disable_irqs_and_tqps(netdev); }

return ret; @@ -823,17 +838,9 @@ static void hns3_reset_tx_queue(struct hnae3_handle *h) static void hns3_nic_net_down(struct net_device *netdev) { struct hns3_nic_priv *priv = netdev_priv(netdev); - struct hnae3_handle *h = hns3_get_handle(netdev); const struct hnae3_ae_ops *ops; - int i;

- /* disable vectors */ - for (i = 0; i < priv->vector_num; i++) - hns3_vector_disable(&priv->tqp_vector[i]); - - /* disable rcb */ - for (i = 0; i < h->kinfo.num_tqps; i++) - hns3_tqp_disable(h->kinfo.tqp[i]); + hns3_disable_irqs_and_tqps(netdev);

/* stop ae_dev */ ops = priv->ae_handle->ae_algo->ops; @@ -5870,8 +5877,6 @@ int hns3_set_channels(struct net_device *netdev, void hns3_external_lb_prepare(struct net_device *ndev, bool if_running) { struct hns3_nic_priv *priv = netdev_priv(ndev); - struct hnae3_handle *h = priv->ae_handle; - int i;

if (!if_running) return; @@ -5882,11 +5887,7 @@ void hns3_external_lb_prepare(struct net_device *ndev, bool if_running) netif_carrier_off(ndev); netif_tx_disable(ndev);

- for (i = 0; i < priv->vector_num; i++) - hns3_vector_disable(&priv->tqp_vector[i]); - - for (i = 0; i < h->kinfo.num_tqps; i++) - hns3_tqp_disable(h->kinfo.tqp[i]); + hns3_disable_irqs_and_tqps(ndev);

/* delay ring buffer clearing to hns3_reset_notify_uninit_enet * during reset process, because driver may not be able @@ -5902,7 +5903,6 @@ void hns3_external_lb_restore(struct net_device *ndev, bool if_running) { struct hns3_nic_priv *priv = netdev_priv(ndev); struct hnae3_handle *h = priv->ae_handle; - int i;

if (!if_running) return; @@ -5918,11 +5918,7 @@ void hns3_external_lb_restore(struct net_device *ndev, bool if_running)

clear_bit(HNS3_NIC_STATE_DOWN, &priv->state);

- for (i = 0; i < priv->vector_num; i++) - hns3_vector_enable(&priv->tqp_vector[i]); - - for (i = 0; i < h->kinfo.num_tqps; i++) - hns3_tqp_enable(h->kinfo.tqp[i]); + hns3_enable_irqs_and_tqps(ndev);

netif_tx_wake_all_queues(ndev);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jian Shen shenjian15@huawei.com

[ Upstream commit 4971394d9d624f91689d766f31ce668d169d9959 ]

Currently the ptp_clock_register() is called before relative ptp resource ready. It may cause unexpected result when upper layer called the ptp API during the timewindow. Fix it by moving the ptp_clock_register() to the function end.

Fixes: 0bf5eb788512 ("net: hns3: add support for PTP") Signed-off-by: Jian Shen shenjian15@huawei.com Signed-off-by: Jijie Shao shaojijie@huawei.com Reviewed-by: Vadim Fedorenko vadim.fedorenko@linux.dev Link: https://patch.msgid.link/20250430093052.2400464-5-shaojijie@huawei.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- .../net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c index ddc691424c816..9a806ac727cf5 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c @@ -440,6 +440,13 @@ static int hclge_ptp_create_clock(struct hclge_dev *hdev) ptp->info.settime64 = hclge_ptp_settime;

ptp->info.n_alarm = 0; + + spin_lock_init(&ptp->lock); + ptp->io_base = hdev->hw.hw.io_base + HCLGE_PTP_REG_OFFSET; + ptp->ts_cfg.rx_filter = HWTSTAMP_FILTER_NONE; + ptp->ts_cfg.tx_type = HWTSTAMP_TX_OFF; + hdev->ptp = ptp; + ptp->clock = ptp_clock_register(&ptp->info, &hdev->pdev->dev); if (IS_ERR(ptp->clock)) { dev_err(&hdev->pdev->dev, @@ -451,12 +458,6 @@ static int hclge_ptp_create_clock(struct hclge_dev *hdev) return -ENODEV; }

- spin_lock_init(&ptp->lock); - ptp->io_base = hdev->hw.hw.io_base + HCLGE_PTP_REG_OFFSET; - ptp->ts_cfg.rx_filter = HWTSTAMP_FILTER_NONE; - ptp->ts_cfg.tx_type = HWTSTAMP_TX_OFF; - hdev->ptp = ptp; - return 0; }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Stefan Wahren wahrenst@gmx.net

[ Upstream commit 74987089ec678b4018dba0a609e9f4bf6ef7f4ad ]

The LEN_MASK for CMD_RTS doesn't cover the whole parameter mask. The Bit 11 is reserved, so adjust LEN_MASK accordingly.

Fixes: 2f207cbf0dd4 ("net: vertexcom: Add MSE102x SPI support") Signed-off-by: Stefan Wahren wahrenst@gmx.net Reviewed-by: Andrew Lunn andrew@lunn.ch Link: https://patch.msgid.link/20250430133043.7722-3-wahrenst@gmx.net Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/vertexcom/mse102x.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/vertexcom/mse102x.c b/drivers/net/ethernet/vertexcom/mse102x.c index 45f4d2cb5b31a..55f34d9d111c4 100644 --- a/drivers/net/ethernet/vertexcom/mse102x.c +++ b/drivers/net/ethernet/vertexcom/mse102x.c @@ -33,7 +33,7 @@ #define CMD_CTR (0x2 << CMD_SHIFT)

#define CMD_MASK GENMASK(15, CMD_SHIFT) -#define LEN_MASK GENMASK(CMD_SHIFT - 1, 0) +#define LEN_MASK GENMASK(CMD_SHIFT - 2, 0)

#define DET_CMD_LEN 4 #define DET_SOF_LEN 2

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Stefan Wahren wahrenst@gmx.net

[ Upstream commit ee512922ddd7d64afe2b28830a88f19063217649 ]

In case the CMD_RTS got corrupted by interferences, the MSE102x doesn't allow a retransmission of the command. Instead the Ethernet frame must be shifted out of the SPI FIFO. Since the actual length is unknown, assume the maximum possible value.

Fixes: 2f207cbf0dd4 ("net: vertexcom: Add MSE102x SPI support") Signed-off-by: Stefan Wahren wahrenst@gmx.net Reviewed-by: Andrew Lunn andrew@lunn.ch Link: https://patch.msgid.link/20250430133043.7722-5-wahrenst@gmx.net Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/vertexcom/mse102x.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/vertexcom/mse102x.c b/drivers/net/ethernet/vertexcom/mse102x.c index 2b1aac72601d0..060a566bc6aae 100644 --- a/drivers/net/ethernet/vertexcom/mse102x.c +++ b/drivers/net/ethernet/vertexcom/mse102x.c @@ -263,7 +263,7 @@ static int mse102x_tx_frame_spi(struct mse102x_net *mse, struct sk_buff *txp, }

static int mse102x_rx_frame_spi(struct mse102x_net *mse, u8 *buff, - unsigned int frame_len) + unsigned int frame_len, bool drop) { struct mse102x_net_spi *mses = to_mse102x_spi(mse); struct spi_transfer *xfer = &mses->spi_xfer; @@ -281,6 +281,9 @@ static int mse102x_rx_frame_spi(struct mse102x_net *mse, u8 *buff, netdev_err(mse->ndev, "%s: spi_sync() failed: %d\n", __func__, ret); mse->stats.xfer_err++; + } else if (drop) { + netdev_dbg(mse->ndev, "%s: Drop frame\n", __func__); + ret = -EINVAL; } else if (*sof != cpu_to_be16(DET_SOF)) { netdev_dbg(mse->ndev, "%s: SPI start of frame is invalid (0x%04x)\n", __func__, *sof); @@ -308,6 +311,7 @@ static void mse102x_rx_pkt_spi(struct mse102x_net *mse) struct sk_buff *skb; unsigned int rxalign; unsigned int rxlen; + bool drop = false; __be16 rx = 0; u16 cmd_resp; u8 *rxpkt; @@ -330,7 +334,8 @@ static void mse102x_rx_pkt_spi(struct mse102x_net *mse) net_dbg_ratelimited("%s: Unexpected response (0x%04x)\n", __func__, cmd_resp); mse->stats.invalid_rts++; - return; + drop = true; + goto drop; }

net_dbg_ratelimited("%s: Unexpected response to first CMD\n", @@ -342,9 +347,16 @@ static void mse102x_rx_pkt_spi(struct mse102x_net *mse) net_dbg_ratelimited("%s: Invalid frame length: %d\n", __func__, rxlen); mse->stats.invalid_len++; - return; + drop = true; }

+ /* In case of a invalid CMD_RTS, the frame must be consumed anyway. + * So assume the maximum possible frame length. + */ +drop: + if (drop) + rxlen = VLAN_ETH_FRAME_LEN; + rxalign = ALIGN(rxlen + DET_SOF_LEN + DET_DFT_LEN, 4); skb = netdev_alloc_skb_ip_align(mse->ndev, rxalign); if (!skb) @@ -355,7 +367,7 @@ static void mse102x_rx_pkt_spi(struct mse102x_net *mse) * They are copied, but ignored. */ rxpkt = skb_put(skb, rxlen) - DET_SOF_LEN; - if (mse102x_rx_frame_spi(mse, rxpkt, rxlen)) { + if (mse102x_rx_frame_spi(mse, rxpkt, rxlen, drop)) { mse->ndev->stats.rx_errors++; dev_kfree_skb(skb); return;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Geert Uytterhoeven geert+renesas@glider.be

commit 6eab7034579917f207ca6d8e3f4e11e85e0ab7d5 upstream.

On R-Car:

OF: /sound: Read of boolean property 'simple-audio-card,bitclock-master' with a value. OF: /sound: Read of boolean property 'simple-audio-card,frame-master' with a value.

or:

OF: /soc/sound@ec500000/ports/port@0/endpoint: Read of boolean property 'bitclock-master' with a value. OF: /soc/sound@ec500000/ports/port@0/endpoint: Read of boolean property 'frame-master' with a value.

The use of of_property_read_bool() for non-boolean properties is deprecated in favor of of_property_present() when testing for property presence.

Replace testing for presence before calling of_property_read_u32() by testing for an -EINVAL return value from the latter, to simplify the code.

Signed-off-by: Geert Uytterhoeven geert+renesas@glider.be Link: https://patch.msgid.link/db10e96fbda121e7456d70e97a013cbfc9755f4d.1737533954... Signed-off-by: Mark Brown broonie@kernel.org Cc: Christophe Leroy christophe.leroy@csgroup.eu Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- sound/soc/soc-core.c | 32 +++++++++++++------------------- 1 file changed, 13 insertions(+), 19 deletions(-)

--- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -2935,7 +2935,7 @@ int snd_soc_of_parse_pin_switches(struct unsigned int i, nb_controls; int ret;

- if (!of_property_read_bool(dev->of_node, prop)) + if (!of_property_present(dev->of_node, prop)) return 0;

strings = devm_kcalloc(dev, nb_controls_max, @@ -3009,23 +3009,17 @@ int snd_soc_of_parse_tdm_slot(struct dev if (rx_mask) snd_soc_of_get_slot_mask(np, "dai-tdm-slot-rx-mask", rx_mask);

- if (of_property_read_bool(np, "dai-tdm-slot-num")) { - ret = of_property_read_u32(np, "dai-tdm-slot-num", &val); - if (ret) - return ret; - - if (slots) - *slots = val; - } - - if (of_property_read_bool(np, "dai-tdm-slot-width")) { - ret = of_property_read_u32(np, "dai-tdm-slot-width", &val); - if (ret) - return ret; + ret = of_property_read_u32(np, "dai-tdm-slot-num", &val); + if (ret && ret != -EINVAL) + return ret; + if (!ret && slots) + *slots = val;

- if (slot_width) - *slot_width = val; - } + ret = of_property_read_u32(np, "dai-tdm-slot-width", &val); + if (ret && ret != -EINVAL) + return ret; + if (!ret && slot_width) + *slot_width = val;

return 0; } @@ -3289,12 +3283,12 @@ unsigned int snd_soc_daifmt_parse_clock_ * check "[prefix]frame-master" */ snprintf(prop, sizeof(prop), "%sbitclock-master", prefix); - bit = of_property_read_bool(np, prop); + bit = of_property_present(np, prop); if (bit && bitclkmaster) *bitclkmaster = of_parse_phandle(np, prop, 0);

snprintf(prop, sizeof(prop), "%sframe-master", prefix); - frame = of_property_read_bool(np, prop); + frame = of_property_present(np, prop); if (frame && framemaster) *framemaster = of_parse_phandle(np, prop, 0);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cong Wang xiyou.wangcong@gmail.com

commit 5ba8b837b522d7051ef81bacf3d95383ff8edce5 upstream.

htb_qlen_notify() always deactivates the HTB class and in fact could trigger a warning if it is already deactivated. Therefore, it is not idempotent and not friendly to its callers, like fq_codel_dequeue().

Let's make it idempotent to ease qdisc_tree_reduce_backlog() callers' life.

Reported-by: Gerrard Tai gerrard.tai@starlabs.sg Signed-off-by: Cong Wang xiyou.wangcong@gmail.com Reviewed-by: Simon Horman horms@kernel.org Link: https://patch.msgid.link/20250403211033.166059-2-xiyou.wangcong@gmail.com Acked-by: Jamal Hadi Salim jhs@mojatatu.com Signed-off-by: Paolo Abeni pabeni@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/sched/sch_htb.c | 2 ++ 1 file changed, 2 insertions(+)

--- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1485,6 +1485,8 @@ static void htb_qlen_notify(struct Qdisc { struct htb_class *cl = (struct htb_class *)arg;

+ if (!cl->prio_activity) + return; htb_deactivate(qdisc_priv(sch), cl); }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cong Wang xiyou.wangcong@gmail.com

commit 51eb3b65544c9efd6a1026889ee5fb5aa62da3bb upstream.

hfsc_qlen_notify() is not idempotent either and not friendly to its callers, like fq_codel_dequeue(). Let's make it idempotent to ease qdisc_tree_reduce_backlog() callers' life:

1. update_vf() decreases cl->cl_nactive, so we can check whether it is non-zero before calling it.

2. eltree_remove() always removes RB node cl->el_node, but we can use RB_EMPTY_NODE() + RB_CLEAR_NODE() to make it safe.

Reported-by: Gerrard Tai gerrard.tai@starlabs.sg Signed-off-by: Cong Wang xiyou.wangcong@gmail.com Reviewed-by: Simon Horman horms@kernel.org Link: https://patch.msgid.link/20250403211033.166059-4-xiyou.wangcong@gmail.com Acked-by: Jamal Hadi Salim jhs@mojatatu.com Signed-off-by: Paolo Abeni pabeni@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/sched/sch_hfsc.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)

--- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -203,7 +203,10 @@ eltree_insert(struct hfsc_class *cl) static inline void eltree_remove(struct hfsc_class *cl) { - rb_erase(&cl->el_node, &cl->sched->eligible); + if (!RB_EMPTY_NODE(&cl->el_node)) { + rb_erase(&cl->el_node, &cl->sched->eligible); + RB_CLEAR_NODE(&cl->el_node); + } }

static inline void @@ -1224,7 +1227,8 @@ hfsc_qlen_notify(struct Qdisc *sch, unsi /* vttree is now handled in update_vf() so that update_vf(cl, 0, 0) * needs to be called explicitly to remove a class from vttree. */ - update_vf(cl, 0, 0); + if (cl->cl_nactive) + update_vf(cl, 0, 0); if (cl->cl_flags & HFSC_RSC) eltree_remove(cl); }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cong Wang xiyou.wangcong@gmail.com

commit a7a15f39c682ac4268624da2abdb9114bdde96d5 upstream.

est_qlen_notify() deletes its class from its active list with list_del() when qlen is 0, therefore, it is not idempotent and not friendly to its callers, like fq_codel_dequeue().

Let's make it idempotent to ease qdisc_tree_reduce_backlog() callers' life. Also change other list_del()'s to list_del_init() just to be extra safe.

Reported-by: Gerrard Tai gerrard.tai@starlabs.sg Signed-off-by: Cong Wang xiyou.wangcong@gmail.com Link: https://patch.msgid.link/20250403211033.166059-6-xiyou.wangcong@gmail.com Acked-by: Jamal Hadi Salim jhs@mojatatu.com Signed-off-by: Paolo Abeni pabeni@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/sched/sch_ets.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)

--- a/net/sched/sch_ets.c +++ b/net/sched/sch_ets.c @@ -298,7 +298,7 @@ static void ets_class_qlen_notify(struct * to remove them. */ if (!ets_class_is_strict(q, cl) && sch->q.qlen) - list_del(&cl->alist); + list_del_init(&cl->alist); }

static int ets_class_dump(struct Qdisc *sch, unsigned long arg, @@ -491,7 +491,7 @@ static struct sk_buff *ets_qdisc_dequeue if (unlikely(!skb)) goto out; if (cl->qdisc->q.qlen == 0) - list_del(&cl->alist); + list_del_init(&cl->alist); return ets_qdisc_dequeue_skb(sch, skb); }

@@ -660,7 +660,7 @@ static int ets_qdisc_change(struct Qdisc } for (i = q->nbands; i < oldbands; i++) { if (i >= q->nstrict && q->classes[i].qdisc->q.qlen) - list_del(&q->classes[i].alist); + list_del_init(&q->classes[i].alist); qdisc_tree_flush_backlog(q->classes[i].qdisc); } q->nstrict = nstrict; @@ -716,7 +716,7 @@ static void ets_qdisc_reset(struct Qdisc

for (band = q->nstrict; band < q->nbands; band++) { if (q->classes[band].qdisc->q.qlen) - list_del(&q->classes[band].alist); + list_del_init(&q->classes[band].alist); } for (band = 0; band < q->nbands; band++) qdisc_reset(q->classes[band].qdisc);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sudeep Holla sudeep.holla@arm.com

[ Upstream commit 4567bdaaaaa1744da3d7da07d9aca2f941f5b4e5 ]

Completion of the FFA_PARTITION_INFO_GET ABI transfers the ownership of the caller’s Rx buffer from the producer(typically partition mnager) to the consumer(this driver/OS). FFA_RX_RELEASE transfers the ownership from the consumer back to the producer.

However, when we set the flag to just return the count of partitions deployed in the system corresponding to the specified UUID while invoking FFA_PARTITION_INFO_GET, the Rx buffer ownership shouldn't be transferred to this driver. We must be able to skip transferring back the ownership to the partition manager when we request just to get the count of the partitions as the buffers are not acquired in this case.

Firmware may return FFA_RET_DENIED or other error for the ffa_rx_release() in such cases.

Fixes: bb1be7498500 ("firmware: arm_ffa: Add v1.1 get_partition_info support") Message-Id: 20250321115700.3525197-1-sudeep.holla@arm.com Signed-off-by: Sudeep Holla sudeep.holla@arm.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/firmware/arm_ffa/driver.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/firmware/arm_ffa/driver.c b/drivers/firmware/arm_ffa/driver.c index 7cd6b1564e801..7c2db3f017651 100644 --- a/drivers/firmware/arm_ffa/driver.c +++ b/drivers/firmware/arm_ffa/driver.c @@ -225,7 +225,8 @@ __ffa_partition_info_get(u32 uuid0, u32 uuid1, u32 uuid2, u32 uuid3, memcpy(buffer + idx, drv_info->rx_buffer + idx * sz, buf_sz);

- ffa_rx_release(); + if (!(flags & PARTITION_INFO_GET_RETURN_COUNT_ONLY)) + ffa_rx_release();

mutex_unlock(&drv_info->rx_lock);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christian Bruel christian.bruel@foss.st.com

[ Upstream commit de2b2107d5a41a91ab603e135fb6e408abbee28e ]

Use gic-400 compatible and remove address-cells = <1> on aarch64

Fixes: 5d30d03aaf785 ("arm64: dts: st: introduce stm32mp25 SoCs family") Signed-off-by: Christian Bruel christian.bruel@foss.st.com Link: https://lore.kernel.org/r/20250415111654.2103767-2-christian.bruel@foss.st.c... Signed-off-by: Alexandre Torgue alexandre.torgue@foss.st.com Signed-off-by: Arnd Bergmann arnd@arndb.de Signed-off-by: Sasha Levin sashal@kernel.org --- arch/arm64/boot/dts/st/stm32mp251.dtsi | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/arch/arm64/boot/dts/st/stm32mp251.dtsi b/arch/arm64/boot/dts/st/stm32mp251.dtsi index 5268a43218415..3219a8ea1e6a7 100644 --- a/arch/arm64/boot/dts/st/stm32mp251.dtsi +++ b/arch/arm64/boot/dts/st/stm32mp251.dtsi @@ -73,9 +73,8 @@ };

intc: interrupt-controller@4ac00000 { - compatible = "arm,cortex-a7-gic"; + compatible = "arm,gic-400"; #interrupt-cells = <3>; - #address-cells = <1>; interrupt-controller; reg = <0x0 0x4ac10000 0x0 0x1000>, <0x0 0x4ac20000 0x0 0x2000>,

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christian Hewitt christianshewitt@gmail.com

[ Upstream commit f37bb5486ea536c1d61df89feeaeff3f84f0b560 ]

This reverts commit bfbc68e.

The patch does permit the offending YUV420 @ 59.94 phy_freq and vclk_freq mode to match in calculations. It also results in all fractional rates being unavailable for use. This was unintended and requires the patch to be reverted.

Fixes: bfbc68e4d869 ("drm/meson: vclk: fix calculation of 59.94 fractional rates") Cc: stable@vger.kernel.org Signed-off-by: Christian Hewitt christianshewitt@gmail.com Signed-off-by: Martin Blumenstingl martin.blumenstingl@googlemail.com Reviewed-by: Neil Armstrong neil.armstrong@linaro.org Link: https://lore.kernel.org/r/20250421201300.778955-2-martin.blumenstingl@google... Signed-off-by: Neil Armstrong neil.armstrong@linaro.org Link: https://lore.kernel.org/r/20250421201300.778955-2-martin.blumenstingl@google... Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/gpu/drm/meson/meson_vclk.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/meson/meson_vclk.c b/drivers/gpu/drm/meson/meson_vclk.c index 2a942dc6a6dc2..2a82119eb58ed 100644 --- a/drivers/gpu/drm/meson/meson_vclk.c +++ b/drivers/gpu/drm/meson/meson_vclk.c @@ -790,13 +790,13 @@ meson_vclk_vic_supported_freq(struct meson_drm *priv, unsigned int phy_freq, FREQ_1000_1001(params[i].pixel_freq)); DRM_DEBUG_DRIVER("i = %d phy_freq = %d alt = %d\n", i, params[i].phy_freq, - FREQ_1000_1001(params[i].phy_freq/1000)*1000); + FREQ_1000_1001(params[i].phy_freq/10)*10); /* Match strict frequency */ if (phy_freq == params[i].phy_freq && vclk_freq == params[i].vclk_freq) return MODE_OK; /* Match 1000/1001 variant */ - if (phy_freq == (FREQ_1000_1001(params[i].phy_freq/1000)*1000) && + if (phy_freq == (FREQ_1000_1001(params[i].phy_freq/10)*10) && vclk_freq == FREQ_1000_1001(params[i].vclk_freq)) return MODE_OK; } @@ -1070,7 +1070,7 @@ void meson_vclk_setup(struct meson_drm *priv, unsigned int target,

for (freq = 0 ; params[freq].pixel_freq ; ++freq) { if ((phy_freq == params[freq].phy_freq || - phy_freq == FREQ_1000_1001(params[freq].phy_freq/1000)*1000) && + phy_freq == FREQ_1000_1001(params[freq].phy_freq/10)*10) && (vclk_freq == params[freq].vclk_freq || vclk_freq == FREQ_1000_1001(params[freq].vclk_freq))) { if (vclk_freq != params[freq].vclk_freq)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jonathan Bell jonathan@raspberrypi.com

[ Upstream commit 28084d3fcc3c8445542917f32e382c45b5343cc2 ]

Users have reported log spam created by "Event Ring Full" xHC event TRBs. These are caused by interrupt latency in conjunction with a very busy set of devices on the bus. The errors are benign, but throughput will suffer as the xHC will pause processing of transfers until the Event Ring is drained by the kernel.

Commit dc0ffbea5729 ("usb: host: xhci: update event ring dequeue pointer on purpose") mitigated the issue by advancing the Event Ring Dequeue Pointer already after half a segment has been processed. Nevertheless, providing a larger Event Ring would be useful to cope with load peaks.

Expand the number of event TRB slots available by increasing the number of Event Ring segments in the ERST.

Controllers have a hardware-defined limit as to the number of ERST entries they can process, but with up to 32k it can be excessively high (sec 5.3.4). So cap the actual number at 2 (configurable through the ERST_MAX_SEGS macro), which seems like a reasonable quantity. It is supported by any xHC because the limit in the HCSPARAMS2 register is defined as a power of 2. Renesas uPD720201 and VIA VL805 controllers do not support more than 2 ERST entries.

An alternative to increasing the number of Event Ring segments would be an increase of the segment size. But that requires allocating multiple contiguous pages, which may be impossible if memory is fragmented.

Signed-off-by: Jonathan Bell jonathan@raspberrypi.com Signed-off-by: Lukas Wunner lukas@wunner.de Signed-off-by: Mathias Nyman mathias.nyman@linux.intel.com Link: https://lore.kernel.org/r/20231019102924.2797346-6-mathias.nyman@linux.intel... Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Stable-dep-of: bea5892d0ed2 ("xhci: Limit time spent with xHC interrupts disabled during bus resume") Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/usb/host/xhci-mem.c | 10 +++++++--- drivers/usb/host/xhci.h | 5 +++-- 2 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index f236fba5cd248..45240299fa171 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -2237,14 +2237,18 @@ xhci_alloc_interrupter(struct xhci_hcd *xhci, gfp_t flags) { struct device *dev = xhci_to_hcd(xhci)->self.sysdev; struct xhci_interrupter *ir; + unsigned int num_segs; int ret;

ir = kzalloc_node(sizeof(*ir), flags, dev_to_node(dev)); if (!ir) return NULL;

- ir->event_ring = xhci_ring_alloc(xhci, ERST_NUM_SEGS, 1, TYPE_EVENT, - 0, flags); + num_segs = min_t(unsigned int, 1 << HCS_ERST_MAX(xhci->hcs_params2), + ERST_MAX_SEGS); + + ir->event_ring = xhci_ring_alloc(xhci, num_segs, 1, TYPE_EVENT, 0, + flags); if (!ir->event_ring) { xhci_warn(xhci, "Failed to allocate interrupter event ring\n"); kfree(ir); @@ -2280,7 +2284,7 @@ xhci_add_interrupter(struct xhci_hcd *xhci, struct xhci_interrupter *ir, /* set ERST count with the number of entries in the segment table */ erst_size = readl(&ir->ir_set->erst_size); erst_size &= ERST_SIZE_MASK; - erst_size |= ERST_NUM_SEGS; + erst_size |= ir->event_ring->num_segs; writel(erst_size, &ir->ir_set->erst_size);

erst_base = xhci_read_64(xhci, &ir->ir_set->erst_base); diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index 0325fccfaa2a4..76a3010b8b74a 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -1427,8 +1427,9 @@ struct urb_priv { * Each segment table entry is 4*32bits long. 1K seems like an ok size: * (1K bytes * 8bytes/bit) / (4*32 bits) = 64 segment entries in the table, * meaning 64 ring segments. - * Initial allocated size of the ERST, in number of entries */ -#define ERST_NUM_SEGS 1 + * Reasonable limit for number of Event Ring segments (spec allows 32k) + */ +#define ERST_MAX_SEGS 2 /* Poll every 60 seconds */ #define POLL_TIMEOUT 60 /* Stop endpoint command timeout (secs) for URB cancellation watchdog timer */

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mathias Nyman mathias.nyman@linux.intel.com

[ Upstream commit 47f503cf5f799ec02e5f4b7c3b9afe145eca2aef ]

The current function that both removes and frees an interrupter isn't optimal when using several interrupters. The array of interrupters need to be protected with a lock while removing interrupters, but the default xhci spin lock can't be used while freeing the interrupters event ring segment table as dma_free_coherent() should be called with IRQs enabled.

There is no need to free the interrupter under the lock, so split this code into separate unlocked free part, and a lock protected remove part.

Signed-off-by: Mathias Nyman mathias.nyman@linux.intel.com Link: https://lore.kernel.org/r/20231019102924.2797346-17-mathias.nyman@linux.inte... Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Stable-dep-of: bea5892d0ed2 ("xhci: Limit time spent with xHC interrupts disabled during bus resume") Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/usb/host/xhci-mem.c | 32 +++++++++++++++++++++----------- 1 file changed, 21 insertions(+), 11 deletions(-)

diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index 45240299fa171..f00e96c9ca57a 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -1803,22 +1803,13 @@ int xhci_alloc_erst(struct xhci_hcd *xhci, }

static void -xhci_free_interrupter(struct xhci_hcd *xhci, struct xhci_interrupter *ir) +xhci_remove_interrupter(struct xhci_hcd *xhci, struct xhci_interrupter *ir) { - struct device *dev = xhci_to_hcd(xhci)->self.sysdev; - size_t erst_size; u32 tmp;

if (!ir) return;

- erst_size = sizeof(struct xhci_erst_entry) * ir->erst.num_entries; - if (ir->erst.entries) - dma_free_coherent(dev, erst_size, - ir->erst.entries, - ir->erst.erst_dma_addr); - ir->erst.entries = NULL; - /* * Clean out interrupter registers except ERSTBA. Clearing either the * low or high 32 bits of ERSTBA immediately causes the controller to @@ -1831,10 +1822,28 @@ xhci_free_interrupter(struct xhci_hcd *xhci, struct xhci_interrupter *ir)

xhci_write_64(xhci, ERST_EHB, &ir->ir_set->erst_dequeue); } +} + +static void +xhci_free_interrupter(struct xhci_hcd *xhci, struct xhci_interrupter *ir) +{ + struct device *dev = xhci_to_hcd(xhci)->self.sysdev; + size_t erst_size; + + if (!ir) + return; + + erst_size = sizeof(struct xhci_erst_entry) * ir->erst.num_entries; + if (ir->erst.entries) + dma_free_coherent(dev, erst_size, + ir->erst.entries, + ir->erst.erst_dma_addr); + ir->erst.entries = NULL;

- /* free interrrupter event ring */ + /* free interrupter event ring */ if (ir->event_ring) xhci_ring_free(xhci, ir->event_ring); + ir->event_ring = NULL;

kfree(ir); @@ -1847,6 +1856,7 @@ void xhci_mem_cleanup(struct xhci_hcd *xhci)

cancel_delayed_work_sync(&xhci->cmd_timer);

+ xhci_remove_interrupter(xhci, xhci->interrupter); xhci_free_interrupter(xhci, xhci->interrupter); xhci->interrupter = NULL; xhci_dbg_trace(xhci, trace_xhci_dbg_init, "Freed primary event ring");

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mathias Nyman mathias.nyman@linux.intel.com

[ Upstream commit ace21625878f78708b75b7a872ec7a0e2ed15ca4 ]

Add a helper to set the interrupt moderation interval for an interrupter. Each interrupter can have its own moderation value.

Hardware has a 16bit register for the moderation value, each step is 250ns.

Helper function imod_interval argument is in nanoseconds.

Values from 0 to 16383750 (250 x 0xffff) are accepted. 0 means no interrupt throttling.

Signed-off-by: Mathias Nyman mathias.nyman@linux.intel.com Signed-off-by: Wesley Cheng quic_wcheng@quicinc.com Link: https://lore.kernel.org/r/20240217001017.29969-3-quic_wcheng@quicinc.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Stable-dep-of: bea5892d0ed2 ("xhci: Limit time spent with xHC interrupts disabled during bus resume") Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/usb/host/xhci.c | 25 ++++++++++++++++++------- 1 file changed, 18 insertions(+), 7 deletions(-)

diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 5c3250989047e..d6a0c79e5fada 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -323,6 +323,23 @@ static int xhci_disable_interrupter(struct xhci_interrupter *ir) return 0; }

+/* interrupt moderation interval imod_interval in nanoseconds */ +static int xhci_set_interrupter_moderation(struct xhci_interrupter *ir, + u32 imod_interval) +{ + u32 imod; + + if (!ir || !ir->ir_set || imod_interval > U16_MAX * 250) + return -EINVAL; + + imod = readl(&ir->ir_set->irq_control); + imod &= ~ER_IRQ_INTERVAL_MASK; + imod |= (imod_interval / 250) & ER_IRQ_INTERVAL_MASK; + writel(imod, &ir->ir_set->irq_control); + + return 0; +} + static void compliance_mode_recovery(struct timer_list *t) { struct xhci_hcd *xhci; @@ -505,7 +522,6 @@ static int xhci_run_finished(struct xhci_hcd *xhci) */ int xhci_run(struct usb_hcd *hcd) { - u32 temp; u64 temp_64; int ret; struct xhci_hcd *xhci = hcd_to_xhci(hcd); @@ -525,12 +541,7 @@ int xhci_run(struct usb_hcd *hcd) xhci_dbg_trace(xhci, trace_xhci_dbg_init, "ERST deq = 64'h%0lx", (long unsigned int) temp_64);

- xhci_dbg_trace(xhci, trace_xhci_dbg_init, - "// Set the interrupt modulation register"); - temp = readl(&ir->ir_set->irq_control); - temp &= ~ER_IRQ_INTERVAL_MASK; - temp |= (xhci->imod_interval / 250) & ER_IRQ_INTERVAL_MASK; - writel(temp, &ir->ir_set->irq_control); + xhci_set_interrupter_moderation(ir, xhci->imod_interval);

if (xhci->quirks & XHCI_NEC_HOST) { struct xhci_command *command;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mathias Nyman mathias.nyman@linux.intel.com

[ Upstream commit 9c0c11bb87b09a8b7cdc21ca1090e7b36abe9d09 ]

Allow creators of seconday interrupters to specify the interrupt moderation interval value in nanoseconds when creating the interrupter.

If not sure what value to use then use the xhci driver default xhci->imod_interval

Suggested-by: Wesley Cheng quic_wcheng@quicinc.com Signed-off-by: Mathias Nyman mathias.nyman@linux.intel.com Link: https://lore.kernel.org/r/20240905143300.1959279-13-mathias.nyman@linux.inte... Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Stable-dep-of: bea5892d0ed2 ("xhci: Limit time spent with xHC interrupts disabled during bus resume") Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/usb/host/xhci-mem.c | 8 +++++++- drivers/usb/host/xhci.c | 4 ++-- drivers/usb/host/xhci.h | 5 ++++- 3 files changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index 489f54cf9a8a2..2c44855be75ed 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -2353,7 +2353,8 @@ xhci_add_interrupter(struct xhci_hcd *xhci, struct xhci_interrupter *ir, }

struct xhci_interrupter * -xhci_create_secondary_interrupter(struct usb_hcd *hcd, unsigned int segs) +xhci_create_secondary_interrupter(struct usb_hcd *hcd, unsigned int segs, + u32 imod_interval) { struct xhci_hcd *xhci = hcd_to_xhci(hcd); struct xhci_interrupter *ir; @@ -2386,6 +2387,11 @@ xhci_create_secondary_interrupter(struct usb_hcd *hcd, unsigned int segs) return NULL; }

+ err = xhci_set_interrupter_moderation(ir, imod_interval); + if (err) + xhci_warn(xhci, "Failed to set interrupter %d moderation to %uns\n", + i, imod_interval); + xhci_dbg(xhci, "Add secondary interrupter %d, max interrupters %d\n", i, xhci->max_interrupters);

diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index d6a0c79e5fada..0af298c5af65a 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -324,8 +324,8 @@ static int xhci_disable_interrupter(struct xhci_interrupter *ir) }

/* interrupt moderation interval imod_interval in nanoseconds */ -static int xhci_set_interrupter_moderation(struct xhci_interrupter *ir, - u32 imod_interval) +int xhci_set_interrupter_moderation(struct xhci_interrupter *ir, + u32 imod_interval) { u32 imod;

diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index 156e43977cdd4..9d05a21392bb8 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -1867,7 +1867,8 @@ struct xhci_container_ctx *xhci_alloc_container_ctx(struct xhci_hcd *xhci, void xhci_free_container_ctx(struct xhci_hcd *xhci, struct xhci_container_ctx *ctx); struct xhci_interrupter * -xhci_create_secondary_interrupter(struct usb_hcd *hcd, unsigned int segs); +xhci_create_secondary_interrupter(struct usb_hcd *hcd, unsigned int segs, + u32 imod_interval); void xhci_remove_secondary_interrupter(struct usb_hcd *hcd, struct xhci_interrupter *ir);

@@ -1905,6 +1906,8 @@ int xhci_alloc_tt_info(struct xhci_hcd *xhci, struct xhci_virt_device *virt_dev, struct usb_device *hdev, struct usb_tt *tt, gfp_t mem_flags); +int xhci_set_interrupter_moderation(struct xhci_interrupter *ir, + u32 imod_interval);

/* xHCI ring, segment, TRB, and TD functions */ dma_addr_t xhci_trb_virt_to_dma(struct xhci_segment *seg, union xhci_trb *trb);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shakeel Butt shakeel.butt@linux.dev

[ Upstream commit 9f01b4954490d4ccdbcc2b9be34a9921ceee9cbb ]

Currently on cpu hotplug teardown, only memcg stock is drained but we need to drain the obj stock as well otherwise we will miss the stats accumulated on the target cpu as well as the nr_bytes cached. The stats include MEMCG_KMEM, NR_SLAB_RECLAIMABLE_B & NR_SLAB_UNRECLAIMABLE_B. In addition we are leaking reference to struct obj_cgroup object.

Link: https://lkml.kernel.org/r/20250310230934.2913113-1-shakeel.butt@linux.dev Fixes: bf4f059954dc ("mm: memcg/slab: obj_cgroup API") Signed-off-by: Shakeel Butt shakeel.butt@linux.dev Reviewed-by: Roman Gushchin roman.gushchin@linux.dev Acked-by: Johannes Weiner hannes@cmpxchg.org Cc: Michal Hocko mhocko@kernel.org Cc: Muchun Song muchun.song@linux.dev Cc: stable@vger.kernel.org Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Sasha Levin sashal@kernel.org --- mm/memcontrol.c | 9 +++++++++ 1 file changed, 9 insertions(+)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 9bf5a69e20d87..ab9afcd372a93 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -2378,9 +2378,18 @@ static void drain_all_stock(struct mem_cgroup *root_memcg) static int memcg_hotplug_cpu_dead(unsigned int cpu) { struct memcg_stock_pcp *stock; + struct obj_cgroup *old; + unsigned long flags;

stock = &per_cpu(memcg_stock, cpu); + + /* drain_obj_stock requires stock_lock */ + local_lock_irqsave(&memcg_stock.stock_lock, flags); + old = drain_obj_stock(stock); + local_unlock_irqrestore(&memcg_stock.stock_lock, flags); + drain_stock(stock); + obj_cgroup_put(old);

return 0; }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shyam Saini shyamsaini@linux.microsoft.com

[ Upstream commit bbc9462f0cb0c8917a4908e856731708f0cee910 ]

The locate_module_kobject() function looks up an existing module_kobject for a given module name. If it cannot find the corresponding module_kobject, it creates one for the given name.

This commit renames locate_module_kobject() to lookup_or_create_module_kobject() to better describe its operations.

This doesn't change anything functionality wise.

Suggested-by: Rasmus Villemoes linux@rasmusvillemoes.dk Signed-off-by: Shyam Saini shyamsaini@linux.microsoft.com Link: https://lore.kernel.org/r/20250227184930.34163-2-shyamsaini@linux.microsoft.... Signed-off-by: Petr Pavlu petr.pavlu@suse.com Stable-dep-of: f95bbfe18512 ("drivers: base: handle module_kobject creation") Signed-off-by: Sasha Levin sashal@kernel.org --- kernel/params.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/params.c b/kernel/params.c index 2d4a0564697e8..8d48a6bfe68da 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -759,7 +759,7 @@ void destroy_params(const struct kernel_param *params, unsigned num) params[i].ops->free(params[i].arg); }

-static struct module_kobject * __init locate_module_kobject(const char *name) +static struct module_kobject * __init lookup_or_create_module_kobject(const char *name) { struct module_kobject *mk; struct kobject *kobj; @@ -801,7 +801,7 @@ static void __init kernel_add_sysfs_param(const char *name, struct module_kobject *mk; int err;

- mk = locate_module_kobject(name); + mk = lookup_or_create_module_kobject(name); if (!mk) return;

@@ -872,7 +872,7 @@ static void __init version_sysfs_builtin(void) int err;

for (vattr = __start___modver; vattr < __stop___modver; vattr++) { - mk = locate_module_kobject(vattr->module_name); + mk = lookup_or_create_module_kobject(vattr->module_name); if (mk) { err = sysfs_create_file(&mk->kobj, &vattr->mattr.attr); WARN_ON_ONCE(err);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shyam Saini shyamsaini@linux.microsoft.com

[ Upstream commit f95bbfe18512c5c018720468959edac056a17196 ]

module_add_driver() relies on module_kset list for /sys/module/<built-in-module>/drivers directory creation.

Since, commit 96a1a2412acba ("kernel/params.c: defer most of param_sysfs_init() to late_initcall time") drivers which are initialized from subsys_initcall() or any other higher precedence initcall couldn't find the related kobject entry in the module_kset list because module_kset is not fully populated by the time module_add_driver() refers it. As a consequence, module_add_driver() returns early without calling make_driver_name(). Therefore, /sys/module/<built-in-module>/drivers is never created.

Fix this issue by letting module_add_driver() handle module_kobject creation itself.

Fixes: 96a1a2412acb ("kernel/params.c: defer most of param_sysfs_init() to late_initcall time") Cc: stable@vger.kernel.org # requires all other patches from the series Suggested-by: Rasmus Villemoes linux@rasmusvillemoes.dk Signed-off-by: Shyam Saini shyamsaini@linux.microsoft.com Acked-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Link: https://lore.kernel.org/r/20250227184930.34163-5-shyamsaini@linux.microsoft.... Signed-off-by: Petr Pavlu petr.pavlu@suse.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/base/module.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/drivers/base/module.c b/drivers/base/module.c index a33663d92256d..955582b34e54a 100644 --- a/drivers/base/module.c +++ b/drivers/base/module.c @@ -42,16 +42,13 @@ int module_add_driver(struct module *mod, struct device_driver *drv) if (mod) mk = &mod->mkobj; else if (drv->mod_name) { - struct kobject *mkobj; - - /* Lookup built-in module entry in /sys/modules */ - mkobj = kset_find_obj(module_kset, drv->mod_name); - if (mkobj) { - mk = container_of(mkobj, struct module_kobject, kobj); + /* Lookup or create built-in module entry in /sys/modules */ + mk = lookup_or_create_module_kobject(drv->mod_name); + if (mk) { /* remember our module structure */ drv->p->mkobj = mk; - /* kset_find_obj took a reference */ - kobject_put(mkobj); + /* lookup_or_create_module_kobject took a reference */ + kobject_put(&mk->kobj); } }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nicolin Chen nicolinc@nvidia.com

[ Upstream commit b00d24997a11c10d3e420614f0873b83ce358a34 ]

ASPEED VGA card has two built-in devices: 0008:06:00.0 PCI bridge: ASPEED Technology, Inc. AST1150 PCI-to-PCI Bridge (rev 06) 0008:07:00.0 VGA compatible controller: ASPEED Technology, Inc. ASPEED Graphics Family (rev 52)

Its toplogy looks like this: +-[0008:00]---00.0-[01-09]--+-00.0-[02-09]--+-00.0-[03]----00.0 Sandisk Corp Device 5017 | +-01.0-[04]-- | +-02.0-[05]----00.0 NVIDIA Corporation Device | +-03.0-[06-07]----00.0-[07]----00.0 ASPEED Technology, Inc. ASPEED Graphics Family | +-04.0-[08]----00.0 Renesas Technology Corp. uPD720201 USB 3.0 Host Controller | -05.0-[09]----00.0 Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller -00.1 PMC-Sierra Inc. Device 4028

The IORT logic populaties two identical IDs into the fwspec->ids array via DMA aliasing in iort_pci_iommu_init() called by pci_for_each_dma_alias().

Though the SMMU driver had been able to handle this situation since commit 563b5cbe334e ("iommu/arm-smmu-v3: Cope with duplicated Stream IDs"), that got broken by the later commit cdf315f907d4 ("iommu/arm-smmu-v3: Maintain a SID->device structure"), which ended up with allocating separate streams with the same stuffing.

On a kernel prior to v6.15-rc1, there has been an overlooked warning: pci 0008:07:00.0: vgaarb: setting as boot VGA device pci 0008:07:00.0: vgaarb: bridge control possible pci 0008:07:00.0: vgaarb: VGA device added: decodes=io+mem,owns=none,locks=none pcieport 0008:06:00.0: Adding to iommu group 14 ast 0008:07:00.0: stream 67328 already in tree <===== WARNING ast 0008:07:00.0: enabling device (0002 -> 0003) ast 0008:07:00.0: Using default configuration ast 0008:07:00.0: AST 2600 detected ast 0008:07:00.0: [drm] Using analog VGA ast 0008:07:00.0: [drm] dram MCLK=396 Mhz type=1 bus_width=16 [drm] Initialized ast 0.1.0 for 0008:07:00.0 on minor 0 ast 0008:07:00.0: [drm] fb0: astdrmfb frame buffer device

With v6.15-rc, since the commit bcb81ac6ae3c ("iommu: Get DT/ACPI parsing into the proper probe path"), the error returned with the warning is moved to the SMMU device probe flow: arm_smmu_probe_device+0x15c/0x4c0 __iommu_probe_device+0x150/0x4f8 probe_iommu_group+0x44/0x80 bus_for_each_dev+0x7c/0x100 bus_iommu_probe+0x48/0x1a8 iommu_device_register+0xb8/0x178 arm_smmu_device_probe+0x1350/0x1db0 which then fails the entire SMMU driver probe: pci 0008:06:00.0: Adding to iommu group 21 pci 0008:07:00.0: stream 67328 already in tree arm-smmu-v3 arm-smmu-v3.9.auto: Failed to register iommu arm-smmu-v3 arm-smmu-v3.9.auto: probe with driver arm-smmu-v3 failed with error -22

Since SMMU driver had been already expecting a potential duplicated Stream ID in arm_smmu_install_ste_for_dev(), change the arm_smmu_insert_master() routine to ignore a duplicated ID from the fwspec->sids array as well.

Note: this has been failing the iommu_device_probe() since 2021, although a recent iommu commit in v6.15-rc1 that moves iommu_device_probe() started to fail the SMMU driver probe. Since nobody has cared about DMA Alias support, leave that as it was but fix the fundamental iommu_device_probe() breakage.

Fixes: cdf315f907d4 ("iommu/arm-smmu-v3: Maintain a SID->device structure") Cc: stable@vger.kernel.org Suggested-by: Jason Gunthorpe jgg@nvidia.com Reviewed-by: Jason Gunthorpe jgg@nvidia.com Signed-off-by: Nicolin Chen nicolinc@nvidia.com Link: https://lore.kernel.org/r/20250415185620.504299-1-nicolinc@nvidia.com Signed-off-by: Will Deacon will@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c index 2cab4798e7a0d..f2260f45728e7 100644 --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c @@ -2597,6 +2597,7 @@ static int arm_smmu_insert_master(struct arm_smmu_device *smmu, mutex_lock(&smmu->streams_mutex); for (i = 0; i < fwspec->num_ids; i++) { struct arm_smmu_stream *new_stream = &master->streams[i]; + struct rb_node *existing; u32 sid = fwspec->ids[i];

new_stream->id = sid; @@ -2607,10 +2608,20 @@ static int arm_smmu_insert_master(struct arm_smmu_device *smmu, break;

/* Insert into SID tree */ - if (rb_find_add(&new_stream->node, &smmu->streams, - arm_smmu_streams_cmp_node)) { - dev_warn(master->dev, "stream %u already in tree\n", - sid); + existing = rb_find_add(&new_stream->node, &smmu->streams, + arm_smmu_streams_cmp_node); + if (existing) { + struct arm_smmu_master *existing_master = + rb_entry(existing, struct arm_smmu_stream, node) + ->master; + + /* Bridged PCI devices may end up with duplicated IDs */ + if (existing_master == master) + continue; + + dev_warn(master->dev, + "stream %u already in tree from dev %s\n", sid, + dev_name(existing_master->dev)); ret = -EINVAL; break; }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chris Bainbridge chris.bainbridge@gmail.com

[ Upstream commit be593d9d91c5a3a363d456b9aceb71029aeb3f1d ]

The HDCP code in amdgpu_dm_hdcp.c copies pointers to amdgpu_dm_connector objects without incrementing the kref reference counts. When using a USB-C dock, and the dock is unplugged, the corresponding amdgpu_dm_connector objects are freed, creating dangling pointers in the HDCP code. When the dock is plugged back, the dangling pointers are dereferenced, resulting in a slab-use-after-free:

[ 66.775837] BUG: KASAN: slab-use-after-free in event_property_validate+0x42f/0x6c0 [amdgpu] [ 66.776171] Read of size 4 at addr ffff888127804120 by task kworker/0:1/10

[ 66.776179] CPU: 0 UID: 0 PID: 10 Comm: kworker/0:1 Not tainted 6.14.0-rc7-00180-g54505f727a38-dirty #233 [ 66.776183] Hardware name: HP HP Pavilion Aero Laptop 13-be0xxx/8916, BIOS F.17 12/18/2024 [ 66.776186] Workqueue: events event_property_validate [amdgpu] [ 66.776494] Call Trace: [ 66.776496] <TASK> [ 66.776497] dump_stack_lvl+0x70/0xa0 [ 66.776504] print_report+0x175/0x555 [ 66.776507] ? __virt_addr_valid+0x243/0x450 [ 66.776510] ? kasan_complete_mode_report_info+0x66/0x1c0 [ 66.776515] kasan_report+0xeb/0x1c0 [ 66.776518] ? event_property_validate+0x42f/0x6c0 [amdgpu] [ 66.776819] ? event_property_validate+0x42f/0x6c0 [amdgpu] [ 66.777121] __asan_report_load4_noabort+0x14/0x20 [ 66.777124] event_property_validate+0x42f/0x6c0 [amdgpu] [ 66.777342] ? __lock_acquire+0x6b40/0x6b40 [ 66.777347] ? enable_assr+0x250/0x250 [amdgpu] [ 66.777571] process_one_work+0x86b/0x1510 [ 66.777575] ? pwq_dec_nr_in_flight+0xcf0/0xcf0 [ 66.777578] ? assign_work+0x16b/0x280 [ 66.777580] ? lock_is_held_type+0xa3/0x130 [ 66.777583] worker_thread+0x5c0/0xfa0 [ 66.777587] ? process_one_work+0x1510/0x1510 [ 66.777588] kthread+0x3a2/0x840 [ 66.777591] ? kthread_is_per_cpu+0xd0/0xd0 [ 66.777594] ? trace_hardirqs_on+0x4f/0x60 [ 66.777597] ? _raw_spin_unlock_irq+0x27/0x60 [ 66.777599] ? calculate_sigpending+0x77/0xa0 [ 66.777602] ? kthread_is_per_cpu+0xd0/0xd0 [ 66.777605] ret_from_fork+0x40/0x90 [ 66.777607] ? kthread_is_per_cpu+0xd0/0xd0 [ 66.777609] ret_from_fork_asm+0x11/0x20 [ 66.777614] </TASK>

[ 66.777643] Allocated by task 10: [ 66.777646] kasan_save_stack+0x39/0x60 [ 66.777649] kasan_save_track+0x14/0x40 [ 66.777652] kasan_save_alloc_info+0x37/0x50 [ 66.777655] __kasan_kmalloc+0xbb/0xc0 [ 66.777658] __kmalloc_cache_noprof+0x1c8/0x4b0 [ 66.777661] dm_dp_add_mst_connector+0xdd/0x5c0 [amdgpu] [ 66.777880] drm_dp_mst_port_add_connector+0x47e/0x770 [drm_display_helper] [ 66.777892] drm_dp_send_link_address+0x1554/0x2bf0 [drm_display_helper] [ 66.777901] drm_dp_check_and_send_link_address+0x187/0x1f0 [drm_display_helper] [ 66.777909] drm_dp_mst_link_probe_work+0x2b8/0x410 [drm_display_helper] [ 66.777917] process_one_work+0x86b/0x1510 [ 66.777919] worker_thread+0x5c0/0xfa0 [ 66.777922] kthread+0x3a2/0x840 [ 66.777925] ret_from_fork+0x40/0x90 [ 66.777927] ret_from_fork_asm+0x11/0x20

[ 66.777932] Freed by task 1713: [ 66.777935] kasan_save_stack+0x39/0x60 [ 66.777938] kasan_save_track+0x14/0x40 [ 66.777940] kasan_save_free_info+0x3b/0x60 [ 66.777944] __kasan_slab_free+0x52/0x70 [ 66.777946] kfree+0x13f/0x4b0 [ 66.777949] dm_dp_mst_connector_destroy+0xfa/0x150 [amdgpu] [ 66.778179] drm_connector_free+0x7d/0xb0 [ 66.778184] drm_mode_object_put.part.0+0xee/0x160 [ 66.778188] drm_mode_object_put+0x37/0x50 [ 66.778191] drm_atomic_state_default_clear+0x220/0xd60 [ 66.778194] __drm_atomic_state_free+0x16e/0x2a0 [ 66.778197] drm_mode_atomic_ioctl+0x15ed/0x2ba0 [ 66.778200] drm_ioctl_kernel+0x17a/0x310 [ 66.778203] drm_ioctl+0x584/0xd10 [ 66.778206] amdgpu_drm_ioctl+0xd2/0x1c0 [amdgpu] [ 66.778375] __x64_sys_ioctl+0x139/0x1a0 [ 66.778378] x64_sys_call+0xee7/0xfb0 [ 66.778381] do_syscall_64+0x87/0x140 [ 66.778385] entry_SYSCALL_64_after_hwframe+0x4b/0x53

Fix this by properly incrementing and decrementing the reference counts when making and deleting copies of the amdgpu_dm_connector pointers.

(Mario: rebase on current code and update fixes tag)

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4006 Signed-off-by: Chris Bainbridge chris.bainbridge@gmail.com Fixes: da3fd7ac0bcf3 ("drm/amd/display: Update CP property based on HW query") Reviewed-by: Alex Hung alex.hung@amd.com Link: https://lore.kernel.org/r/20250417215005.37964-1-mario.limonciello@amd.com Signed-off-by: Mario Limonciello mario.limonciello@amd.com Signed-off-by: Alex Deucher alexander.deucher@amd.com (cherry picked from commit d4673f3c3b3dcb74e36e53cdfc880baa7a87b330) Cc: stable@vger.kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- .../amd/display/amdgpu_dm/amdgpu_dm_hdcp.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c index 4330d37022fa3..a048022d9865a 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c @@ -173,6 +173,9 @@ void hdcp_update_display(struct hdcp_workqueue *hdcp_work, unsigned int conn_index = aconnector->base.index;

guard(mutex)(&hdcp_w->mutex); + drm_connector_get(&aconnector->base); + if (hdcp_w->aconnector[conn_index]) + drm_connector_put(&hdcp_w->aconnector[conn_index]->base); hdcp_w->aconnector[conn_index] = aconnector;

memset(&link_adjust, 0, sizeof(link_adjust)); @@ -220,7 +223,6 @@ static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work, unsigned int conn_index = aconnector->base.index;

guard(mutex)(&hdcp_w->mutex); - hdcp_w->aconnector[conn_index] = aconnector;

/* the removal of display will invoke auth reset -> hdcp destroy and * we'd expect the Content Protection (CP) property changed back to @@ -236,7 +238,10 @@ static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work, }

mod_hdcp_remove_display(&hdcp_w->hdcp, aconnector->base.index, &hdcp_w->output); - + if (hdcp_w->aconnector[conn_index]) { + drm_connector_put(&hdcp_w->aconnector[conn_index]->base); + hdcp_w->aconnector[conn_index] = NULL; + } process_output(hdcp_w); }

@@ -254,6 +259,10 @@ void hdcp_reset_display(struct hdcp_workqueue *hdcp_work, unsigned int link_inde for (conn_index = 0; conn_index < AMDGPU_DM_MAX_DISPLAY_INDEX; conn_index++) { hdcp_w->encryption_status[conn_index] = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; + if (hdcp_w->aconnector[conn_index]) { + drm_connector_put(&hdcp_w->aconnector[conn_index]->base); + hdcp_w->aconnector[conn_index] = NULL; + } }

process_output(hdcp_w); @@ -489,6 +498,7 @@ static void update_config(void *handle, struct cp_psp_stream_config *config) struct hdcp_workqueue *hdcp_work = handle; struct amdgpu_dm_connector *aconnector = config->dm_stream_ctx; int link_index = aconnector->dc_link->link_index; + unsigned int conn_index = aconnector->base.index; struct mod_hdcp_display *display = &hdcp_work[link_index].display; struct mod_hdcp_link *link = &hdcp_work[link_index].link; struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index]; @@ -545,7 +555,10 @@ static void update_config(void *handle, struct cp_psp_stream_config *config) guard(mutex)(&hdcp_w->mutex);

mod_hdcp_add_display(&hdcp_w->hdcp, link, display, &hdcp_w->output); - + drm_connector_get(&aconnector->base); + if (hdcp_w->aconnector[conn_index]) + drm_connector_put(&hdcp_w->aconnector[conn_index]->base); + hdcp_w->aconnector[conn_index] = aconnector; process_output(hdcp_w); }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mathias Nyman mathias.nyman@linux.intel.com

commit a54a594d72f25b08f39d743880a76721fba9ae77 upstream.

Don't try to remove a secondary interrupter that is known to be invalid. Also check if the interrupter is valid inside the spinlock that protects the array of interrupters.

Found by smatch static checker

Reported-by: Dan Carpenter dan.carpenter@linaro.org Closes: https://lore.kernel.org/linux-usb/ffaa0a1b-5984-4a1f-bfd3-9184630a97b9@morot... Fixes: c99b38c41234 ("xhci: add support to allocate several interrupters") Signed-off-by: Mathias Nyman mathias.nyman@linux.intel.com Link: https://lore.kernel.org/r/20240125152737.2983959-2-mathias.nyman@linux.intel... Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/usb/host/xhci-mem.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)

--- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -1855,14 +1855,14 @@ void xhci_remove_secondary_interrupter(s struct xhci_hcd *xhci = hcd_to_xhci(hcd); unsigned int intr_num;

+ spin_lock_irq(&xhci->lock); + /* interrupter 0 is primary interrupter, don't touch it */ - if (!ir || !ir->intr_num || ir->intr_num >= xhci->max_interrupters) + if (!ir || !ir->intr_num || ir->intr_num >= xhci->max_interrupters) { xhci_dbg(xhci, "Invalid secondary interrupter, can't remove\n"); - - /* fixme, should we check xhci->interrupter[intr_num] == ir */ - /* fixme locking */ - - spin_lock_irq(&xhci->lock); + spin_unlock_irq(&xhci->lock); + return; + }

intr_num = ir->intr_num;