Linux-stable-mirror

linux-stable-mirror@lists.linaro.org

351 participants
124263 discussions

[PATCH AUTOSEL 6.18-5.10] serial: sprd: Return -EPROBE_DEFER when uart clock is not ready

by Sasha Levin

From: Wenhua Lin <Wenhua.Lin(a)unisoc.com> [ Upstream commit 29e8a0c587e328ed458380a45d6028adf64d7487 ] In sprd_clk_init(), when devm_clk_get() returns -EPROBE_DEFER for either uart or source clock, we should propagate the error instead of just warning and continuing with NULL clocks. Currently the driver only emits a warning when clock acquisition fails and proceeds with NULL clock pointers. This can lead to issues later when the clocks are actually needed. More importantly, when the clock provider is not ready yet and returns -EPROBE_DEFER, we should return this error to allow deferred probing. This change adds explicit checks for -EPROBE_DEFER after both: 1. devm_clk_get(uport->dev, uart) 2. devm_clk_get(uport->dev, source) When -EPROBE_DEFER is encountered, the function now returns -EPROBE_DEFER to let the driver framework retry probing later when the clock dependencies are resolved. Signed-off-by: Wenhua Lin <Wenhua.Lin(a)unisoc.com> Link: https://patch.msgid.link/20251022030840.956589-1-Wenhua.Lin@unisoc.com Reviewed-by: Cixi Geng <cixi.geng(a)linux.dev> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- LLM Generated explanations, may be completely bogus: ## Commit Analysis: serial: sprd: Return -EPROBE_DEFER when uart clock is not ready ### 1. COMMIT MESSAGE ANALYSIS **Subject**: Fixes incorrect handling of `-EPROBE_DEFER` error from `devm_clk_get()` **Key observations**: - No `Cc: stable(a)vger.kernel.org` tag present - No `Fixes:` tag pointing to when the bug was introduced - Has `Reviewed-by:` from Cixi Geng - Accepted by Greg Kroah-Hartman (serial subsystem maintainer) ### 2. CODE CHANGE ANALYSIS The change adds two checks in `sprd_clk_init()`: ```c clk_uart = devm_clk_get(uport->dev, "uart"); if (IS_ERR(clk_uart)) { + if (PTR_ERR(clk_uart) == -EPROBE_DEFER) + return -EPROBE_DEFER; dev_warn(...); clk_uart = NULL; } clk_parent = devm_clk_get(uport->dev, "source"); if (IS_ERR(clk_parent)) { + if (PTR_ERR(clk_parent) == -EPROBE_DEFER) + return -EPROBE_DEFER; dev_warn(...); clk_parent = NULL; } ``` **Technical bug mechanism**: When clock providers aren't ready yet, `devm_clk_get()` returns `-EPROBE_DEFER`. The existing code ignores this error, sets the clock pointer to NULL, and continues. This bypasses the kernel's deferred probing mechanism which exists precisely to handle this dependency ordering scenario. **Existing pattern**: The function already has identical handling for the "enable" clock (visible in the context lines). This fix makes the handling consistent for all three clocks. ### 3. CLASSIFICATION **Type**: Bug fix (not a feature addition) This fixes incorrect error handling. The `-EPROBE_DEFER` mechanism is a fundamental kernel feature for handling driver load order dependencies. Not propagating this error is a bug. ### 4. SCOPE AND RISK ASSESSMENT - **Lines changed**: 6 lines added (two 3-line checks) - **Files touched**: 1 file (`drivers/tty/serial/sprd_serial.c`) - **Complexity**: Very low - follows identical pattern already in the same function - **Subsystem**: Hardware-specific serial driver for Spreadtrum/Unisoc UARTs **Risk**: Very low - Pattern is already established and proven in the same function - Only affects error handling during probe - No changes to normal operation when clocks are available - Worst case: probe failure happens earlier/more explicitly ### 5. USER IMPACT **Affected users**: Users of Spreadtrum/Unisoc UART hardware (embedded devices, some Android phones) **Severity**: Medium-High for affected users - without this fix, the serial port may not work correctly on systems where clock providers load after the serial driver. This is common in embedded systems with device tree-based configurations. **Real bug**: This is a practical issue in probe ordering scenarios. The driver would proceed with NULL clocks instead of waiting for dependencies to be ready. ### 6. STABILITY INDICATORS - Reviewed-by tag indicates code review - Maintainer accepted the change - Simple, straightforward change following existing code patterns - No complex logic introduced ### 7. DEPENDENCY CHECK - **Dependencies**: None - self-contained fix - **Code existence in stable**: The sprd_serial driver has existed since ~2015 (commit 3e1f2029a4b40), so it's present in all active stable trees ### SUMMARY **What it fixes**: Incorrect handling of `-EPROBE_DEFER` from `devm_clk_get()` for two clocks ("uart" and "source"), causing the driver to proceed with NULL clocks instead of deferring probe when clock providers aren't ready. **Stable kernel criteria**: - ✅ Obviously correct (follows identical pattern already in function) - ✅ Fixes a real bug (broken deferred probing) - ✅ Small and contained (6 lines, 1 file) - ✅ No new features - ✅ Low risk of regression **Concerns**: - No explicit `Cc: stable` tag (author/maintainer didn't flag for stable) - Relatively niche driver (Spreadtrum hardware) **Risk vs Benefit**: The fix is minimal risk (identical pattern to existing code) and addresses a real bug that could leave serial hardware non-functional on embedded systems. The benefit outweighs the minimal risk. The lack of `Cc: stable` tag is notable but not determinative - this is a straightforward bug fix that meets all stable criteria. The fix is small, obviously correct, and addresses a real probe ordering issue that embedded system users could encounter. **YES** drivers/tty/serial/sprd_serial.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/tty/serial/sprd_serial.c b/drivers/tty/serial/sprd_serial.c index 8c9366321f8e7..092755f356836 100644 --- a/drivers/tty/serial/sprd_serial.c +++ b/drivers/tty/serial/sprd_serial.c @@ -1133,6 +1133,9 @@ static int sprd_clk_init(struct uart_port *uport) clk_uart = devm_clk_get(uport->dev, "uart"); if (IS_ERR(clk_uart)) { + if (PTR_ERR(clk_uart) == -EPROBE_DEFER) + return -EPROBE_DEFER; + dev_warn(uport->dev, "uart%d can't get uart clock\n", uport->line); clk_uart = NULL; @@ -1140,6 +1143,9 @@ static int sprd_clk_init(struct uart_port *uport) clk_parent = devm_clk_get(uport->dev, "source"); if (IS_ERR(clk_parent)) { + if (PTR_ERR(clk_parent) == -EPROBE_DEFER) + return -EPROBE_DEFER; + dev_warn(uport->dev, "uart%d can't get source clock\n", uport->line); clk_parent = NULL; -- 2.51.0

1 month

[PATCH v2] media: dw2102: validate I2C messages in su3000_i2c_transfer()

by Dharanitharan R

syzbot reports a general protection fault caused by su3000_i2c_transfer() dereferencing msg->buf without validating the message length or buffer pointer. Although i2c-dev blocks zero-length messages, malformed I²C messages can still reach the driver through the DVB USB subsystem. Add strict validation of each message to prevent NULL-pointer dereferences. Reported-by: syzbot+d99f3a288cc7d8ef60fb(a)syzkaller.appspotmail.com Fixes: 0e148a522b84 ("media: dw2102: Don't translate i2c read into write") Closes: https://syzkaller.appspot.com/bug?extid=d99f3a288cc7d8ef60fb Cc: stable(a)vger.kernel.org Signed-off-by: Dharanitharan R <dharanitharan725(a)gmail.com> --- drivers/media/usb/dvb-usb/dw2102.c | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) diff --git a/drivers/media/usb/dvb-usb/dw2102.c b/drivers/media/usb/dvb-usb/dw2102.c index 4fecf2f965e9..0dd210ea16f3 100644 --- a/drivers/media/usb/dvb-usb/dw2102.c +++ b/drivers/media/usb/dvb-usb/dw2102.c @@ -733,6 +733,36 @@ static int su3000_i2c_transfer(struct i2c_adapter *adap, struct i2c_msg msg[], return -EAGAIN; } + /* Validate incoming I²C messages */ + if (!msg || num <= 0) { + mutex_unlock(&d->data_mutex); + mutex_unlock(&d->i2c_mutex); + return -EINVAL; + } + + for (j = 0; j < num; j++) { + /* msg buffer must exist */ + if (!msg[j].buf) { + mutex_unlock(&d->data_mutex); + mutex_unlock(&d->i2c_mutex); + return -EINVAL; + } + + /* zero or negative length is invalid */ + if (msg[j].len <= 0) { + mutex_unlock(&d->data_mutex); + mutex_unlock(&d->i2c_mutex); + return -EINVAL; + } + + /* protect against unreasonable sizes */ + if (msg[j].len > 256) { + mutex_unlock(&d->data_mutex); + mutex_unlock(&d->i2c_mutex); + return -EOPNOTSUPP; + } + } + j = 0; while (j < num) { switch (msg[j].addr) { -- 2.43.0

1 month

[PATCH AUTOSEL 6.18-6.17] functionfs: fix the open/removal races

by Sasha Levin

From: Al Viro <viro(a)zeniv.linux.org.uk> [ Upstream commit e5bf5ee266633cb18fff6f98f0b7d59a62819eee ] ffs_epfile_open() can race with removal, ending up with file->private_data pointing to freed object. There is a total count of opened files on functionfs (both ep0 and dynamic ones) and when it hits zero, dynamic files get removed. Unfortunately, that removal can happen while another thread is in ffs_epfile_open(), but has not incremented the count yet. In that case open will succeed, leaving us with UAF on any subsequent read() or write(). The root cause is that ffs->opened is misused; atomic_dec_and_test() vs. atomic_add_return() is not a good idea, when object remains visible all along. To untangle that * serialize openers on ffs->mutex (both for ep0 and for dynamic files) * have dynamic ones use atomic_inc_not_zero() and fail if we had zero ->opened; in that case the file we are opening is doomed. * have the inodes of dynamic files marked on removal (from the callback of simple_recursive_removal()) - clear ->i_private there. * have open of dynamic ones verify they hadn't been already removed, along with checking that state is FFS_ACTIVE. Reviewed-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Al Viro <viro(a)zeniv.linux.org.uk> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- LLM Generated explanations, may be completely bogus: ## Summary ### What the Commit Fixes This commit fixes a **Use-After-Free (UAF)** vulnerability in USB gadget functionfs. The race condition occurs between `ffs_epfile_open()` and file removal: 1. Thread A is in `ffs_epfile_open()` but hasn't incremented `ffs->opened` yet 2. Thread B closes the last file handle, triggering removal (since `opened` is 0) 3. Thread A's open succeeds with `file->private_data` pointing to freed memory 4. Subsequent `read()`/`write()` operations cause UAF ### Fix Mechanism The fix implements proper synchronization: - Serializes openers using `ffs->mutex` - Uses `atomic_inc_not_zero()` to fail if counter already zero - Uses `smp_load_acquire()`/`smp_store_release()` for memory ordering - Clears `i_private` during removal via `simple_recursive_removal()` callback - Uses `file->private_data` instead of `inode->i_private` in release path ### Stable Kernel Criteria Assessment | Criterion | Assessment | |-----------|------------| | Obviously correct | ✅ Uses standard kernel primitives, proper locking patterns | | Fixes real bug | ✅ UAF vulnerability, security-relevant | | Important issue | ✅ Security bug, potential for crashes/exploitation | | Small and contained | ✅ Single file, +43/-10 lines, localized changes | | No new features | ✅ Pure bug fix, no new APIs | ### Risk vs Benefit **Benefits:** - Fixes serious UAF vulnerability - USB gadget functionfs used in Android, embedded systems - Reviewed by Greg Kroah-Hartman (USB maintainer, stable maintainer) - Written by Al Viro (highly respected kernel developer) **Risks:** - Moderate complexity (changes locking behavior) - Recent commit (November 2025), limited mainline soak time - No explicit `Cc: stable(a)vger.kernel.org` tag ### Dependencies - `ffs_mutex_lock()` - exists in functionfs since early versions - `simple_recursive_removal()` with callback - available since ~5.x kernels - Standard kernel APIs (`atomic_inc_not_zero`, memory barriers) - universally available ### Concerns 1. **No Fixes: tag** - Makes it harder to determine which stable trees need this fix 2. **No Cc: stable tag** - May indicate maintainers wanted soak time, or an oversight given Greg KH reviewed it 3. **Backport effort** - May need adjustment for older stable trees depending on functionfs evolution ### Conclusion This is a legitimate UAF security fix that affects real-world users (Android, embedded USB gadgets). Despite moderate complexity, the fix: - Addresses a serious vulnerability class (UAF) - Uses correct synchronization patterns - Has been reviewed by the appropriate maintainer who also maintains stable trees - Is self-contained with no feature additions The lack of explicit stable tags appears to be an oversight given the security nature of the bug and Greg KH's review. UAF vulnerabilities typically warrant expedited backporting. **YES** drivers/usb/gadget/function/f_fs.c | 53 ++++++++++++++++++++++++------ 1 file changed, 43 insertions(+), 10 deletions(-) diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c index 47cfbe41fdff8..69f6e3c0f7e00 100644 --- a/drivers/usb/gadget/function/f_fs.c +++ b/drivers/usb/gadget/function/f_fs.c @@ -640,13 +640,22 @@ static ssize_t ffs_ep0_read(struct file *file, char __user *buf, static int ffs_ep0_open(struct inode *inode, struct file *file) { - struct ffs_data *ffs = inode->i_private; + struct ffs_data *ffs = inode->i_sb->s_fs_info; + int ret; - if (ffs->state == FFS_CLOSING) - return -EBUSY; + /* Acquire mutex */ + ret = ffs_mutex_lock(&ffs->mutex, file->f_flags & O_NONBLOCK); + if (ret < 0) + return ret; - file->private_data = ffs; ffs_data_opened(ffs); + if (ffs->state == FFS_CLOSING) { + ffs_data_closed(ffs); + mutex_unlock(&ffs->mutex); + return -EBUSY; + } + mutex_unlock(&ffs->mutex); + file->private_data = ffs; return stream_open(inode, file); } @@ -1193,14 +1202,33 @@ static ssize_t ffs_epfile_io(struct file *file, struct ffs_io_data *io_data) static int ffs_epfile_open(struct inode *inode, struct file *file) { - struct ffs_epfile *epfile = inode->i_private; + struct ffs_data *ffs = inode->i_sb->s_fs_info; + struct ffs_epfile *epfile; + int ret; - if (WARN_ON(epfile->ffs->state != FFS_ACTIVE)) + /* Acquire mutex */ + ret = ffs_mutex_lock(&ffs->mutex, file->f_flags & O_NONBLOCK); + if (ret < 0) + return ret; + + if (!atomic_inc_not_zero(&ffs->opened)) { + mutex_unlock(&ffs->mutex); + return -ENODEV; + } + /* + * we want the state to be FFS_ACTIVE; FFS_ACTIVE alone is + * not enough, though - we might have been through FFS_CLOSING + * and back to FFS_ACTIVE, with our file already removed. + */ + epfile = smp_load_acquire(&inode->i_private); + if (unlikely(ffs->state != FFS_ACTIVE || !epfile)) { + mutex_unlock(&ffs->mutex); + ffs_data_closed(ffs); return -ENODEV; + } + mutex_unlock(&ffs->mutex); file->private_data = epfile; - ffs_data_opened(epfile->ffs); - return stream_open(inode, file); } @@ -1332,7 +1360,7 @@ static void ffs_dmabuf_put(struct dma_buf_attachment *attach) static int ffs_epfile_release(struct inode *inode, struct file *file) { - struct ffs_epfile *epfile = inode->i_private; + struct ffs_epfile *epfile = file->private_data; struct ffs_dmabuf_priv *priv, *tmp; struct ffs_data *ffs = epfile->ffs; @@ -2352,6 +2380,11 @@ static int ffs_epfiles_create(struct ffs_data *ffs) return 0; } +static void clear_one(struct dentry *dentry) +{ + smp_store_release(&dentry->d_inode->i_private, NULL); +} + static void ffs_epfiles_destroy(struct ffs_epfile *epfiles, unsigned count) { struct ffs_epfile *epfile = epfiles; @@ -2359,7 +2392,7 @@ static void ffs_epfiles_destroy(struct ffs_epfile *epfiles, unsigned count) for (; count; --count, ++epfile) { BUG_ON(mutex_is_locked(&epfile->mutex)); if (epfile->dentry) { - simple_recursive_removal(epfile->dentry, NULL); + simple_recursive_removal(epfile->dentry, clear_one); epfile->dentry = NULL; } } -- 2.51.0

1 month

[PATCH v2] pmdomain: imx: Fix reference count leak in imx_gpc_probe()

by Wentao Liang

of_get_child_by_name() returns a node pointer with refcount incremented, we should use the __free() attribute to manage the pgc_node reference. This ensures automatic of_node_put() cleanup when pgc_node goes out of scope, eliminating the need for explicit error handling paths and avoiding reference count leaks. Fixes: 721cabf6c660 ("soc: imx: move PGC handling to a new GPC driver") Cc: stable(a)vger.kernel.org Signed-off-by: Wentao Liang <vulab(a)iscas.ac.cn> --- Change in V2: - Use __free() attribute instead of explicit of_node_put() calls --- drivers/pmdomain/imx/gpc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pmdomain/imx/gpc.c b/drivers/pmdomain/imx/gpc.c index f18c7e6e75dd..89d5d68c055d 100644 --- a/drivers/pmdomain/imx/gpc.c +++ b/drivers/pmdomain/imx/gpc.c @@ -403,7 +403,7 @@ static int imx_gpc_old_dt_init(struct device *dev, struct regmap *regmap, static int imx_gpc_probe(struct platform_device *pdev) { const struct imx_gpc_dt_data *of_id_data = device_get_match_data(&pdev->dev); - struct device_node *pgc_node; + struct device_node *pgc_node __free(pgc_node); struct regmap *regmap; void __iomem *base; int ret; -- 2.34.1

1 month

[PATCH] media: cx23885: Add missing unmap in snd_cx23885_hw_params()

by Haoxiang Li

In error path, add cx23885_alsa_dma_unmap() to release the resource acquired by cx23885_alsa_dma_map(). Fixes: 9529a4b0cf49 ("[media] cx23885: drop videobuf abuse in cx23885-alsa") Cc: stable(a)vger.kernel.org Signed-off-by: Haoxiang Li <lihaoxiang(a)isrc.iscas.ac.cn> --- drivers/media/pci/cx23885/cx23885-alsa.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/media/pci/cx23885/cx23885-alsa.c b/drivers/media/pci/cx23885/cx23885-alsa.c index 25dc8d4dc5b7..717fc6c9ef21 100644 --- a/drivers/media/pci/cx23885/cx23885-alsa.c +++ b/drivers/media/pci/cx23885/cx23885-alsa.c @@ -392,8 +392,10 @@ static int snd_cx23885_hw_params(struct snd_pcm_substream *substream, ret = cx23885_risc_databuffer(chip->pci, &buf->risc, buf->sglist, chip->period_size, chip->num_periods, 1); - if (ret < 0) + if (ret < 0) { + cx23885_alsa_dma_unmap(chip); goto error; + } /* Loop back to start of program */ buf->risc.jmp[0] = cpu_to_le32(RISC_JUMP|RISC_IRQ1|RISC_CNT_INC); -- 2.25.1

1 month

[PATCH] media: cx88: Add missing unmap in snd_cx88_hw_params()

by Haoxiang Li

In error path, add cx88_alsa_dma_unmap() to release resource acquired by cx88_alsa_dma_map(). Fixes: b2c75abde0de ("[media] cx88: drop videobuf abuse in cx88-alsa") Cc: stable(a)vger.kernel.org Signed-off-by: Haoxiang Li <lihaoxiang(a)isrc.iscas.ac.cn> --- drivers/media/pci/cx88/cx88-alsa.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/media/pci/cx88/cx88-alsa.c b/drivers/media/pci/cx88/cx88-alsa.c index 29fb1311e443..4e574d8390b4 100644 --- a/drivers/media/pci/cx88/cx88-alsa.c +++ b/drivers/media/pci/cx88/cx88-alsa.c @@ -483,8 +483,10 @@ static int snd_cx88_hw_params(struct snd_pcm_substream *substream, ret = cx88_risc_databuffer(chip->pci, &buf->risc, buf->sglist, chip->period_size, chip->num_periods, 1); - if (ret < 0) + if (ret < 0) { + cx88_alsa_dma_unmap(chip); goto error; + } /* Loop back to start of program */ buf->risc.jmp[0] = cpu_to_le32(RISC_JUMP | RISC_IRQ1 | RISC_CNT_INC); -- 2.25.1

1 month

[PATCH] media: cx25821: Add missing unmap in snd_cx25821_hw_params()

by Haoxiang Li

In error path, add cx25821_alsa_dma_unmap() to release the resource acquired by cx25821_alsa_dma_map() Fixes: b2c75abde0de ("[media] cx88: drop videobuf abuse in cx88-alsa") Cc: stable(a)vger.kernel.org Signed-off-by: Haoxiang Li <lihaoxiang(a)isrc.iscas.ac.cn> --- drivers/media/pci/cx25821/cx25821-alsa.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/media/pci/cx25821/cx25821-alsa.c b/drivers/media/pci/cx25821/cx25821-alsa.c index a42f0c03a7ca..f463365163b7 100644 --- a/drivers/media/pci/cx25821/cx25821-alsa.c +++ b/drivers/media/pci/cx25821/cx25821-alsa.c @@ -535,6 +535,7 @@ static int snd_cx25821_hw_params(struct snd_pcm_substream *substream, chip->period_size, chip->num_periods, 1); if (ret < 0) { pr_info("DEBUG: ERROR after cx25821_risc_databuffer_audio()\n"); + cx25821_alsa_dma_unmap(chip); goto error; } -- 2.25.1

1 month

[PATCH v2] x86/elf: Fix core dump truncation on CPUs with no extended xfeatures

by yongxin.liu＠windriver.com

From: Yongxin Liu <yongxin.liu(a)windriver.com> Zero can be a valid value of num_records. For example, on Intel Atom x6425RE, only x87 and SSE are supported (features 0, 1), and fpu_user_cfg.max_features is 3. The for_each_extended_xfeature() loop only iterates feature 2, which is not enabled, so num_records = 0. This is valid and should not cause core dump failure. The issue is that dump_xsave_layout_desc() returns 0 for both genuine errors (dump_emit() failure) and valid cases (no extended features). Use negative return values for errors and only abort on genuine failures. Cc: stable(a)vger.kernel.org Fixes: ba386777a30b ("x86/elf: Add a new FPU buffer layout info to x86 core files") Signed-off-by: Yongxin Liu <yongxin.liu(a)windriver.com> --- V2: Keep error checking but use negative value for genuine error V1: Remove error checking entirely --- arch/x86/kernel/fpu/xstate.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c index 48113c5193aa..76153dfb58c9 100644 --- a/arch/x86/kernel/fpu/xstate.c +++ b/arch/x86/kernel/fpu/xstate.c @@ -1946,7 +1946,7 @@ static int dump_xsave_layout_desc(struct coredump_params *cprm) }; if (!dump_emit(cprm, &xc, sizeof(xc))) - return 0; + return -1; num_records++; } @@ -1984,7 +1984,7 @@ int elf_coredump_extra_notes_write(struct coredump_params *cprm) return 1; num_records = dump_xsave_layout_desc(cprm); - if (!num_records) + if (num_records < 0) return 1; /* Total size should be equal to the number of records */ -- 2.46.2

1 month

[PATCH] x86/elf: Fix core dump truncation on CPUs with no extended xfeatures

by yongxin.liu＠windriver.com

From: Yongxin Liu <yongxin.liu(a)windriver.com> Zero can be a valid value of num_records. For example, on Intel Atom x6425RE, only x87 and SSE are supported (features 0, 1), and fpu_user_cfg.max_features is 3. The for_each_extended_xfeature() loop only iterates feature 2, which is not enabled, so num_records = 0. This is valid and should not cause core dump failure. The size check already validates consistency: if num_records = 0, then en.n_descsz = 0, so the check passes. Cc: stable(a)vger.kernel.org Fixes: ba386777a30b ("x86/elf: Add a new FPU buffer layout info to x86 core files") Signed-off-by: Yongxin Liu <yongxin.liu(a)windriver.com> --- arch/x86/kernel/fpu/xstate.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c index 48113c5193aa..b1dd30eb21a8 100644 --- a/arch/x86/kernel/fpu/xstate.c +++ b/arch/x86/kernel/fpu/xstate.c @@ -1984,8 +1984,6 @@ int elf_coredump_extra_notes_write(struct coredump_params *cprm) return 1; num_records = dump_xsave_layout_desc(cprm); - if (!num_records) - return 1; /* Total size should be equal to the number of records */ if ((sizeof(struct x86_xfeat_component) * num_records) != en.n_descsz) -- 2.46.2

1 month

[PATCH v5 1/6] drm/amdgpu: Fix gfx9 update PTE mtype flag

by Philip Yang

Fix copy&paste error, that should have been an assignment instead of an or, otherwise MTYPE_UC 0x3 can not be updated to MTYPE_RW 0x1. CC stables. cc: stable(a)vger.kernel.org Signed-off-by: Philip Yang <Philip.Yang(a)amd.com> Reviewed-by: Christian König <christian.koenig(a)amd.com> --- drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 97a04e3171f2..205c34eb8d11 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -1204,16 +1204,16 @@ static void gmc_v9_0_get_vm_pte(struct amdgpu_device *adev, *flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_NC); break; case AMDGPU_VM_MTYPE_WC: - *flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_WC); + *flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_WC); break; case AMDGPU_VM_MTYPE_RW: - *flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_RW); + *flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_RW); break; case AMDGPU_VM_MTYPE_CC: - *flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_CC); + *flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_CC); break; case AMDGPU_VM_MTYPE_UC: - *flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_UC); + *flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_UC); break; } -- 2.50.1

1 month

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror