April 2025 - Linux-stable-mirror

[PATCH AUTOSEL 5.15 1/8] MIPS: cm: Detect CM quirks from device tree

by Sasha Levin

From: Gregory CLEMENT <gregory.clement(a)bootlin.com> [ Upstream commit e27fbe16af5cfc40639de4ced67d1a866a1953e9 ] Some information that should be retrieved at runtime for the Coherence Manager can be either absent or wrong. This patch allows checking if some of this information is available from the device tree and updates the internal variable accordingly. For now, only the compatible string associated with the broken HCI is being retrieved. Signed-off-by: Gregory CLEMENT <gregory.clement(a)bootlin.com> Signed-off-by: Thomas Bogendoerfer <tsbogend(a)alpha.franken.de> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- arch/mips/include/asm/mips-cm.h | 22 ++++++++++++++++++++++ arch/mips/kernel/mips-cm.c | 14 ++++++++++++++ 2 files changed, 36 insertions(+) diff --git a/arch/mips/include/asm/mips-cm.h b/arch/mips/include/asm/mips-cm.h index 696b40beb774f..0f31324998c0a 100644 --- a/arch/mips/include/asm/mips-cm.h +++ b/arch/mips/include/asm/mips-cm.h @@ -47,6 +47,16 @@ extern phys_addr_t __mips_cm_phys_base(void); */ extern int mips_cm_is64; +/* + * mips_cm_is_l2_hci_broken - determine if HCI is broken + * + * Some CM reports show that Hardware Cache Initialization is + * complete, but in reality it's not the case. They also incorrectly + * indicate that Hardware Cache Initialization is supported. This + * flags allows warning about this broken feature. + */ +extern bool mips_cm_is_l2_hci_broken; + /** * mips_cm_error_report - Report CM cache errors */ @@ -85,6 +95,18 @@ static inline bool mips_cm_present(void) #endif } +/** + * mips_cm_update_property - update property from the device tree + * + * Retrieve the properties from the device tree if a CM node exist and + * update the internal variable based on this. + */ +#ifdef CONFIG_MIPS_CM +extern void mips_cm_update_property(void); +#else +static void mips_cm_update_property(void) {} +#endif + /** * mips_cm_has_l2sync - determine whether an L2-only sync region is present * diff --git a/arch/mips/kernel/mips-cm.c b/arch/mips/kernel/mips-cm.c index b4f7d950c8468..e21c2fd761674 100644 --- a/arch/mips/kernel/mips-cm.c +++ b/arch/mips/kernel/mips-cm.c @@ -5,6 +5,7 @@ */ #include <linux/errno.h> +#include <linux/of.h> #include <linux/percpu.h> #include <linux/spinlock.h> @@ -14,6 +15,7 @@ void __iomem *mips_gcr_base; void __iomem *mips_cm_l2sync_base; int mips_cm_is64; +bool mips_cm_is_l2_hci_broken; static char *cm2_tr[8] = { "mem", "gcr", "gic", "mmio", @@ -238,6 +240,18 @@ static void mips_cm_probe_l2sync(void) mips_cm_l2sync_base = ioremap(addr, MIPS_CM_L2SYNC_SIZE); } +void mips_cm_update_property(void) +{ + struct device_node *cm_node; + + cm_node = of_find_compatible_node(of_root, NULL, "mobileye,eyeq6-cm"); + if (!cm_node) + return; + pr_info("HCI (Hardware Cache Init for the L2 cache) in GCR_L2_RAM_CONFIG from the CM3 is broken"); + mips_cm_is_l2_hci_broken = true; + of_node_put(cm_node); +} + int mips_cm_probe(void) { phys_addr_t addr; -- 2.39.5

3 weeks

1
7
0 0

[PATCH AUTOSEL 6.1 01/10] pinctrl: renesas: rza2: Fix potential NULL pointer dereference

by Sasha Levin

From: Chenyuan Yang <chenyuan0y(a)gmail.com> [ Upstream commit f752ee5b5b86b5f88a5687c9eb0ef9b39859b908 ] `chip.label` in rza2_gpio_register() could be NULL. Add the missing check. Signed-off-by: Chenyuan Yang <chenyuan0y(a)gmail.com> Reviewed-by: Geert Uytterhoeven <geert+renesas(a)glider.be> Reviewed-by: Matthias Brugger <matthias.bgg(a)gmail.com> Link: https://lore.kernel.org/20250210232552.1545887-1-chenyuan0y@gmail.com Signed-off-by: Geert Uytterhoeven <geert+renesas(a)glider.be> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- drivers/pinctrl/renesas/pinctrl-rza2.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/pinctrl/renesas/pinctrl-rza2.c b/drivers/pinctrl/renesas/pinctrl-rza2.c index 12126e30dc20f..e94a6120e1e81 100644 --- a/drivers/pinctrl/renesas/pinctrl-rza2.c +++ b/drivers/pinctrl/renesas/pinctrl-rza2.c @@ -242,6 +242,9 @@ static int rza2_gpio_register(struct rza2_pinctrl_priv *priv) int ret; chip.label = devm_kasprintf(priv->dev, GFP_KERNEL, "%pOFn", np); + if (!chip.label) + return -ENOMEM; + chip.parent = priv->dev; chip.ngpio = priv->npins; -- 2.39.5

3 weeks

1
9
0 0

[PATCH AUTOSEL 6.6 01/16] pinctrl: renesas: rza2: Fix potential NULL pointer dereference

by Sasha Levin

From: Chenyuan Yang <chenyuan0y(a)gmail.com> [ Upstream commit f752ee5b5b86b5f88a5687c9eb0ef9b39859b908 ] `chip.label` in rza2_gpio_register() could be NULL. Add the missing check. Signed-off-by: Chenyuan Yang <chenyuan0y(a)gmail.com> Reviewed-by: Geert Uytterhoeven <geert+renesas(a)glider.be> Reviewed-by: Matthias Brugger <matthias.bgg(a)gmail.com> Link: https://lore.kernel.org/20250210232552.1545887-1-chenyuan0y@gmail.com Signed-off-by: Geert Uytterhoeven <geert+renesas(a)glider.be> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- drivers/pinctrl/renesas/pinctrl-rza2.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/pinctrl/renesas/pinctrl-rza2.c b/drivers/pinctrl/renesas/pinctrl-rza2.c index 990b96d459671..c5dc63d9bd3fc 100644 --- a/drivers/pinctrl/renesas/pinctrl-rza2.c +++ b/drivers/pinctrl/renesas/pinctrl-rza2.c @@ -243,6 +243,9 @@ static int rza2_gpio_register(struct rza2_pinctrl_priv *priv) int ret; chip.label = devm_kasprintf(priv->dev, GFP_KERNEL, "%pOFn", np); + if (!chip.label) + return -ENOMEM; + chip.parent = priv->dev; chip.ngpio = priv->npins; -- 2.39.5

3 weeks

1
15
0 0

[PATCH AUTOSEL 6.12 01/20] selftests/bpf: Fix stdout race condition in traffic monitor

by Sasha Levin

From: Amery Hung <ameryhung(a)gmail.com> [ Upstream commit b99f27e90268b1a814c13f8bd72ea1db448ea257 ] Fix a race condition between the main test_progs thread and the traffic monitoring thread. The traffic monitor thread tries to print a line using multiple printf and use flockfile() to prevent the line from being torn apart. Meanwhile, the main thread doing io redirection can reassign or close stdout when going through tests. A deadlock as shown below can happen. main traffic_monitor_thread ==== ====================== show_transport() -> flockfile(stdout) stdio_hijack_init() -> stdout = open_memstream(log_buf, log_cnt); ... env.subtest_state->stdout_saved = stdout; ... funlockfile(stdout) stdio_restore_cleanup() -> fclose(env.subtest_state->stdout_saved); After the traffic monitor thread lock stdout, A new memstream can be assigned to stdout by the main thread. Therefore, the traffic monitor thread later will not be able to unlock the original stdout. As the main thread tries to access the old stdout, it will hang indefinitely as it is still locked by the traffic monitor thread. The deadlock can be reproduced by running test_progs repeatedly with traffic monitor enabled: for ((i=1;i<=100;i++)); do ./test_progs -a flow_dissector_skb* -m '*' done Fix this by only calling printf once and remove flockfile()/funlockfile(). Signed-off-by: Amery Hung <ameryhung(a)gmail.com> Signed-off-by: Martin KaFai Lau <martin.lau(a)kernel.org> Link: https://patch.msgid.link/20250213233217.553258-1-ameryhung@gmail.com Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- tools/testing/selftests/bpf/network_helpers.c | 33 ++++++++----------- 1 file changed, 13 insertions(+), 20 deletions(-) diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c index 27784946b01b8..af0ee70a53f9f 100644 --- a/tools/testing/selftests/bpf/network_helpers.c +++ b/tools/testing/selftests/bpf/network_helpers.c @@ -771,12 +771,13 @@ static const char *pkt_type_str(u16 pkt_type) return "Unknown"; } +#define MAX_FLAGS_STRLEN 21 /* Show the information of the transport layer in the packet */ static void show_transport(const u_char *packet, u16 len, u32 ifindex, const char *src_addr, const char *dst_addr, u16 proto, bool ipv6, u8 pkt_type) { - char *ifname, _ifname[IF_NAMESIZE]; + char *ifname, _ifname[IF_NAMESIZE], flags[MAX_FLAGS_STRLEN] = ""; const char *transport_str; u16 src_port, dst_port; struct udphdr *udp; @@ -817,29 +818,21 @@ static void show_transport(const u_char *packet, u16 len, u32 ifindex, /* TCP or UDP*/ - flockfile(stdout); + if (proto == IPPROTO_TCP) + snprintf(flags, MAX_FLAGS_STRLEN, "%s%s%s%s", + tcp->fin ? ", FIN" : "", + tcp->syn ? ", SYN" : "", + tcp->rst ? ", RST" : "", + tcp->ack ? ", ACK" : ""); + if (ipv6) - printf("%-7s %-3s IPv6 %s.%d > %s.%d: %s, length %d", + printf("%-7s %-3s IPv6 %s.%d > %s.%d: %s, length %d%s\n", ifname, pkt_type_str(pkt_type), src_addr, src_port, - dst_addr, dst_port, transport_str, len); + dst_addr, dst_port, transport_str, len, flags); else - printf("%-7s %-3s IPv4 %s:%d > %s:%d: %s, length %d", + printf("%-7s %-3s IPv4 %s:%d > %s:%d: %s, length %d%s\n", ifname, pkt_type_str(pkt_type), src_addr, src_port, - dst_addr, dst_port, transport_str, len); - - if (proto == IPPROTO_TCP) { - if (tcp->fin) - printf(", FIN"); - if (tcp->syn) - printf(", SYN"); - if (tcp->rst) - printf(", RST"); - if (tcp->ack) - printf(", ACK"); - } - - printf("\n"); - funlockfile(stdout); + dst_addr, dst_port, transport_str, len, flags); } static void show_ipv6_packet(const u_char *packet, u32 ifindex, u8 pkt_type) -- 2.39.5

3 weeks

1
19
0 0

[PATCH AUTOSEL 6.13 01/22] crypto: ecdsa - Harden against integer overflows in DIV_ROUND_UP()

by Sasha Levin

From: Lukas Wunner <lukas(a)wunner.de> [ Upstream commit b16510a530d1e6ab9683f04f8fb34f2e0f538275 ] Herbert notes that DIV_ROUND_UP() may overflow unnecessarily if an ecdsa implementation's ->key_size() callback returns an unusually large value. Herbert instead suggests (for a division by 8): X / 8 + !!(X & 7) Based on this formula, introduce a generic DIV_ROUND_UP_POW2() macro and use it in lieu of DIV_ROUND_UP() for ->key_size() return values. Additionally, use the macro in ecc_digits_from_bytes(), whose "nbytes" parameter is a ->key_size() return value in some instances, or a user-specified ASN.1 length in the case of ecdsa_get_signature_rs(). Link: https://lore.kernel.org/r/Z3iElsILmoSu6FuC@gondor.apana.org.au/ Signed-off-by: Lukas Wunner <lukas(a)wunner.de> Signed-off-by: Lukas Wunner <lukas(a)wunner.de> Signed-off-by: Herbert Xu <herbert(a)gondor.apana.org.au> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- crypto/ecc.c | 2 +- crypto/ecdsa-p1363.c | 2 +- crypto/ecdsa-x962.c | 4 ++-- include/linux/math.h | 12 ++++++++++++ 4 files changed, 16 insertions(+), 4 deletions(-) diff --git a/crypto/ecc.c b/crypto/ecc.c index 50ad2d4ed672c..6cf9a945fc6c2 100644 --- a/crypto/ecc.c +++ b/crypto/ecc.c @@ -71,7 +71,7 @@ EXPORT_SYMBOL(ecc_get_curve); void ecc_digits_from_bytes(const u8 *in, unsigned int nbytes, u64 *out, unsigned int ndigits) { - int diff = ndigits - DIV_ROUND_UP(nbytes, sizeof(u64)); + int diff = ndigits - DIV_ROUND_UP_POW2(nbytes, sizeof(u64)); unsigned int o = nbytes & 7; __be64 msd = 0; diff --git a/crypto/ecdsa-p1363.c b/crypto/ecdsa-p1363.c index eaae7214d69bc..4454f1f8f33f5 100644 --- a/crypto/ecdsa-p1363.c +++ b/crypto/ecdsa-p1363.c @@ -22,7 +22,7 @@ static int ecdsa_p1363_verify(struct crypto_sig *tfm, { struct ecdsa_p1363_ctx *ctx = crypto_sig_ctx(tfm); unsigned int keylen = crypto_sig_keysize(ctx->child); - unsigned int ndigits = DIV_ROUND_UP(keylen, sizeof(u64)); + unsigned int ndigits = DIV_ROUND_UP_POW2(keylen, sizeof(u64)); struct ecdsa_raw_sig sig; if (slen != 2 * keylen) diff --git a/crypto/ecdsa-x962.c b/crypto/ecdsa-x962.c index 6a77c13e192b1..90a04f4b9a2f5 100644 --- a/crypto/ecdsa-x962.c +++ b/crypto/ecdsa-x962.c @@ -81,8 +81,8 @@ static int ecdsa_x962_verify(struct crypto_sig *tfm, struct ecdsa_x962_signature_ctx sig_ctx; int err; - sig_ctx.ndigits = DIV_ROUND_UP(crypto_sig_keysize(ctx->child), - sizeof(u64)); + sig_ctx.ndigits = DIV_ROUND_UP_POW2(crypto_sig_keysize(ctx->child), + sizeof(u64)); err = asn1_ber_decoder(&ecdsasignature_decoder, &sig_ctx, src, slen); if (err < 0) diff --git a/include/linux/math.h b/include/linux/math.h index f5f18dc3616b0..0198c92cbe3ef 100644 --- a/include/linux/math.h +++ b/include/linux/math.h @@ -34,6 +34,18 @@ */ #define round_down(x, y) ((x) & ~__round_mask(x, y)) +/** + * DIV_ROUND_UP_POW2 - divide and round up + * @n: numerator + * @d: denominator (must be a power of 2) + * + * Divides @n by @d and rounds up to next multiple of @d (which must be a power + * of 2). Avoids integer overflows that may occur with __KERNEL_DIV_ROUND_UP(). + * Performance is roughly equivalent to __KERNEL_DIV_ROUND_UP(). + */ +#define DIV_ROUND_UP_POW2(n, d) \ + ((n) / (d) + !!((n) & ((d) - 1))) + #define DIV_ROUND_UP __KERNEL_DIV_ROUND_UP #define DIV_ROUND_DOWN_ULL(ll, d) \ -- 2.39.5

3 weeks

1
21
0 0

[PATCH AUTOSEL 6.14 01/23] crypto: ecdsa - Harden against integer overflows in DIV_ROUND_UP()

by Sasha Levin

From: Lukas Wunner <lukas(a)wunner.de> [ Upstream commit b16510a530d1e6ab9683f04f8fb34f2e0f538275 ] Herbert notes that DIV_ROUND_UP() may overflow unnecessarily if an ecdsa implementation's ->key_size() callback returns an unusually large value. Herbert instead suggests (for a division by 8): X / 8 + !!(X & 7) Based on this formula, introduce a generic DIV_ROUND_UP_POW2() macro and use it in lieu of DIV_ROUND_UP() for ->key_size() return values. Additionally, use the macro in ecc_digits_from_bytes(), whose "nbytes" parameter is a ->key_size() return value in some instances, or a user-specified ASN.1 length in the case of ecdsa_get_signature_rs(). Link: https://lore.kernel.org/r/Z3iElsILmoSu6FuC@gondor.apana.org.au/ Signed-off-by: Lukas Wunner <lukas(a)wunner.de> Signed-off-by: Lukas Wunner <lukas(a)wunner.de> Signed-off-by: Herbert Xu <herbert(a)gondor.apana.org.au> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- crypto/ecc.c | 2 +- crypto/ecdsa-p1363.c | 2 +- crypto/ecdsa-x962.c | 4 ++-- include/linux/math.h | 12 ++++++++++++ 4 files changed, 16 insertions(+), 4 deletions(-) diff --git a/crypto/ecc.c b/crypto/ecc.c index 50ad2d4ed672c..6cf9a945fc6c2 100644 --- a/crypto/ecc.c +++ b/crypto/ecc.c @@ -71,7 +71,7 @@ EXPORT_SYMBOL(ecc_get_curve); void ecc_digits_from_bytes(const u8 *in, unsigned int nbytes, u64 *out, unsigned int ndigits) { - int diff = ndigits - DIV_ROUND_UP(nbytes, sizeof(u64)); + int diff = ndigits - DIV_ROUND_UP_POW2(nbytes, sizeof(u64)); unsigned int o = nbytes & 7; __be64 msd = 0; diff --git a/crypto/ecdsa-p1363.c b/crypto/ecdsa-p1363.c index eaae7214d69bc..4454f1f8f33f5 100644 --- a/crypto/ecdsa-p1363.c +++ b/crypto/ecdsa-p1363.c @@ -22,7 +22,7 @@ static int ecdsa_p1363_verify(struct crypto_sig *tfm, { struct ecdsa_p1363_ctx *ctx = crypto_sig_ctx(tfm); unsigned int keylen = crypto_sig_keysize(ctx->child); - unsigned int ndigits = DIV_ROUND_UP(keylen, sizeof(u64)); + unsigned int ndigits = DIV_ROUND_UP_POW2(keylen, sizeof(u64)); struct ecdsa_raw_sig sig; if (slen != 2 * keylen) diff --git a/crypto/ecdsa-x962.c b/crypto/ecdsa-x962.c index 6a77c13e192b1..90a04f4b9a2f5 100644 --- a/crypto/ecdsa-x962.c +++ b/crypto/ecdsa-x962.c @@ -81,8 +81,8 @@ static int ecdsa_x962_verify(struct crypto_sig *tfm, struct ecdsa_x962_signature_ctx sig_ctx; int err; - sig_ctx.ndigits = DIV_ROUND_UP(crypto_sig_keysize(ctx->child), - sizeof(u64)); + sig_ctx.ndigits = DIV_ROUND_UP_POW2(crypto_sig_keysize(ctx->child), + sizeof(u64)); err = asn1_ber_decoder(&ecdsasignature_decoder, &sig_ctx, src, slen); if (err < 0) diff --git a/include/linux/math.h b/include/linux/math.h index f5f18dc3616b0..0198c92cbe3ef 100644 --- a/include/linux/math.h +++ b/include/linux/math.h @@ -34,6 +34,18 @@ */ #define round_down(x, y) ((x) & ~__round_mask(x, y)) +/** + * DIV_ROUND_UP_POW2 - divide and round up + * @n: numerator + * @d: denominator (must be a power of 2) + * + * Divides @n by @d and rounds up to next multiple of @d (which must be a power + * of 2). Avoids integer overflows that may occur with __KERNEL_DIV_ROUND_UP(). + * Performance is roughly equivalent to __KERNEL_DIV_ROUND_UP(). + */ +#define DIV_ROUND_UP_POW2(n, d) \ + ((n) / (d) + !!((n) & ((d) - 1))) + #define DIV_ROUND_UP __KERNEL_DIV_ROUND_UP #define DIV_ROUND_DOWN_ULL(ll, d) \ -- 2.39.5

3 weeks

1
22
0 0

[PATCH net] gve: handle overflow when reporting TX consumed descriptors

by Harshitha Ramamurthy

From: Joshua Washington <joshwash(a)google.com> When the tx tail is less than the head (in cases of wraparound), the TX consumed descriptor statistic in DQ will be reported as UINT32_MAX - head + tail, which is incorrect. Mask the difference of head and tail according to the ring size when reporting the statistic. Cc: stable(a)vger.kernel.org Fixes: 2c9198356d56 ("gve: Add consumed counts to ethtool stats") Signed-off-by: Joshua Washington <joshwash(a)google.com> Signed-off-by: Harshitha Ramamurthy <hramamurthy(a)google.com> --- drivers/net/ethernet/google/gve/gve_ethtool.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c index 31a21ccf4863..4dea1fdce748 100644 --- a/drivers/net/ethernet/google/gve/gve_ethtool.c +++ b/drivers/net/ethernet/google/gve/gve_ethtool.c @@ -392,7 +392,9 @@ gve_get_ethtool_stats(struct net_device *netdev, */ data[i++] = 0; data[i++] = 0; - data[i++] = tx->dqo_tx.tail - tx->dqo_tx.head; + data[i++] = + (tx->dqo_tx.tail - tx->dqo_tx.head) & + tx->mask; } do { start = -- 2.49.0.472.ge94155a9ec-goog

3 weeks

4
3
0 0

[PATCH v4 0/4] media: uvcvideo: Introduce V4L2_META_FMT_UVC_MSXU_1_5 + other meta fixes

by Ricardo Ribalda

This series introduces a new metadata format for UVC cameras and adds a couple of improvements to the UVC metadata handling. Signed-off-by: Ricardo Ribalda <ribalda(a)chromium.org> --- Changes in v4: - Rename format to V4L2_META_FMT_UVC_MSXU_1_5 (Thanks Mauro) - Flag the new format with a quirk. - Autodetect MSXU devices. - Link to v3: https://lore.kernel.org/linux-media/20250313-uvc-metadata-v3-0-c467af869c60… Changes in v3: - Fix doc syntax errors. - Link to v2: https://lore.kernel.org/r/20250306-uvc-metadata-v2-0-7e939857cad5@chromium.… Changes in v2: - Add metadata invalid fix - Move doc note to a separate patch - Introuce V4L2_META_FMT_UVC_CUSTOM (thanks HdG!). - Link to v1: https://lore.kernel.org/r/20250226-uvc-metadata-v1-1-6cd6fe5ec2cb@chromium.… --- Ricardo Ribalda (4): media: uvcvideo: Do not mark valid metadata as invalid media: Documentation: Add note about UVCH length field media: uvcvideo: Introduce V4L2_META_FMT_UVC_MSXU_1_5 media: uvcvideo: Auto-set UVC_QUIRK_MSXU_META .../userspace-api/media/v4l/meta-formats.rst | 1 + .../media/v4l/metafmt-uvc-msxu-1-5.rst | 23 ++++++ .../userspace-api/media/v4l/metafmt-uvc.rst | 4 +- MAINTAINERS | 1 + drivers/media/usb/uvc/uvc_metadata.c | 96 ++++++++++++++++++++-- drivers/media/usb/uvc/uvc_video.c | 12 +-- drivers/media/usb/uvc/uvcvideo.h | 1 + drivers/media/v4l2-core/v4l2-ioctl.c | 1 + include/linux/usb/uvc.h | 3 + include/uapi/linux/videodev2.h | 1 + 10 files changed, 130 insertions(+), 13 deletions(-) --- base-commit: 4e82c87058f45e79eeaa4d5bcc3b38dd3dce7209 change-id: 20250403-uvc-meta-e556773d12ae Best regards, -- Ricardo Ribalda <ribalda(a)chromium.org>

3 weeks

1
1
0 0

[PATCH v4 1/2] string: Add load_unaligned_zeropad() code path to sized_strscpy()

by Peter Collingbourne

The call to read_word_at_a_time() in sized_strscpy() is problematic with MTE because it may trigger a tag check fault when reading across a tag granule (16 bytes) boundary. To make this code MTE compatible, let's start using load_unaligned_zeropad() on architectures where it is available (i.e. architectures that define CONFIG_DCACHE_WORD_ACCESS). Because load_unaligned_zeropad() takes care of page boundaries as well as tag granule boundaries, also disable the code preventing crossing page boundaries when using load_unaligned_zeropad(). Signed-off-by: Peter Collingbourne <pcc(a)google.com> Link: https://linux-review.googlesource.com/id/If4b22e43b5a4ca49726b4bf98ada827fd… Fixes: 94ab5b61ee16 ("kasan, arm64: enable CONFIG_KASAN_HW_TAGS") Cc: stable(a)vger.kernel.org --- v2: - new approach lib/string.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/lib/string.c b/lib/string.c index eb4486ed40d25..b632c71df1a50 100644 --- a/lib/string.c +++ b/lib/string.c @@ -119,6 +119,7 @@ ssize_t sized_strscpy(char *dest, const char *src, size_t count) if (count == 0 || WARN_ON_ONCE(count > INT_MAX)) return -E2BIG; +#ifndef CONFIG_DCACHE_WORD_ACCESS #ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS /* * If src is unaligned, don't cross a page boundary, @@ -133,12 +134,14 @@ ssize_t sized_strscpy(char *dest, const char *src, size_t count) /* If src or dest is unaligned, don't do word-at-a-time. */ if (((long) dest | (long) src) & (sizeof(long) - 1)) max = 0; +#endif #endif /* - * read_word_at_a_time() below may read uninitialized bytes after the - * trailing zero and use them in comparisons. Disable this optimization - * under KMSAN to prevent false positive reports. + * load_unaligned_zeropad() or read_word_at_a_time() below may read + * uninitialized bytes after the trailing zero and use them in + * comparisons. Disable this optimization under KMSAN to prevent + * false positive reports. */ if (IS_ENABLED(CONFIG_KMSAN)) max = 0; @@ -146,7 +149,11 @@ ssize_t sized_strscpy(char *dest, const char *src, size_t count) while (max >= sizeof(unsigned long)) { unsigned long c, data; +#ifdef CONFIG_DCACHE_WORD_ACCESS + c = load_unaligned_zeropad(src+res); +#else c = read_word_at_a_time(src+res); +#endif if (has_zero(c, &data, &constants)) { data = prep_zero_mask(c, data, &constants); data = create_zero_mask(data); -- 2.49.0.472.ge94155a9ec-goog

3 weeks

3
4
0 0

[PATCH] drm/xe/dma_buf: stop relying on placement in unmap

by Matthew Auld

The is_vram() is checking the current placement, however if we consider exported VRAM with dynamic dma-buf, it looks possible for the xe driver to async evict the memory, notifying the importer, however importer does not have to call unmap_attachment() immediately, but rather just as "soon as possible", like when the dma-resv idles. Following from this we would then pipeline the move, attaching the fence to the manager, and then update the current placement. But when the unmap_attachment() runs at some later point we might see that is_vram() is now false, and take the complete wrong path when dma-unmapping the sg, leading to explosions. To fix this rather make a note in the attachment if the sg was originally mapping vram or tt pages. Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4563 Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs") Signed-off-by: Matthew Auld <matthew.auld(a)intel.com> Cc: Thomas Hellström <thomas.hellstrom(a)linux.intel.com> Cc: Matthew Brost <matthew.brost(a)intel.com> Cc: <stable(a)vger.kernel.org> # v6.8+ --- drivers/gpu/drm/xe/xe_dma_buf.c | 23 ++++++++++++++++++++--- 1 file changed, 20 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_dma_buf.c b/drivers/gpu/drm/xe/xe_dma_buf.c index f67803e15a0e..b71058e26820 100644 --- a/drivers/gpu/drm/xe/xe_dma_buf.c +++ b/drivers/gpu/drm/xe/xe_dma_buf.c @@ -22,6 +22,22 @@ MODULE_IMPORT_NS("DMA_BUF"); +/** + * struct xe_sg_info - Track the exported sg info + */ +struct xe_sg_info { + /** @is_vram: True if this sg is mapping VRAM. */ + bool is_vram; +}; + +static struct xe_sg_info tt_sg_info = { + .is_vram = false, +}; + +static struct xe_sg_info vram_sg_info = { + .is_vram = true, +}; + static int xe_dma_buf_attach(struct dma_buf *dmabuf, struct dma_buf_attachment *attach) { @@ -118,6 +134,7 @@ static struct sg_table *xe_dma_buf_map(struct dma_buf_attachment *attach, if (dma_map_sgtable(attach->dev, sgt, dir, DMA_ATTR_SKIP_CPU_SYNC)) goto error_free; + attach->priv = &tt_sg_info; break; case XE_PL_VRAM0: @@ -128,6 +145,7 @@ static struct sg_table *xe_dma_buf_map(struct dma_buf_attachment *attach, dir, &sgt); if (r) return ERR_PTR(r); + attach->priv = &vram_sg_info; break; default: return ERR_PTR(-EINVAL); @@ -145,10 +163,9 @@ static void xe_dma_buf_unmap(struct dma_buf_attachment *attach, struct sg_table *sgt, enum dma_data_direction dir) { - struct dma_buf *dma_buf = attach->dmabuf; - struct xe_bo *bo = gem_to_xe_bo(dma_buf->priv); + struct xe_sg_info *sg_info = attach->priv; - if (!xe_bo_is_vram(bo)) { + if (!sg_info->is_vram) { dma_unmap_sgtable(attach->dev, sgt, dir, 0); sg_free_table(sgt); kfree(sgt); -- 2.49.0

3 weeks

2
2
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror April 2025