iio: adc: xilinx: fix potential use-after-free on remove
Subsystems (19)
Sasha Levin
maintainerPATCH 4/81 acpi/nfit: Always dump _DSM output payload
In reply to Sasha LevinFrom: Dan Williams <[email protected]>
[ Upstream commit 351f339faa308c1c1461314a18c832239a841ca0 ]
The dynamic-debug statements for command payload output only get emitted when the command is not ND_CMD_CALL. Move the output payload dumping ahead of the early return path for ND_CMD_CALL.
--- drivers/acpi/nfit/core.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/acpi/nfit/core.c +6 -6
@@ -542,6 +542,12 @@ int acpi_nfit_ctl(struct nvdimm_bus_descriptor *nd_desc, struct nvdimm *nvdimm, goto out; } +dev_dbg (dev,"%s cmd: %s output length: %d\n" , dimm_name, + cmd_name, out_obj->buffer.length); +print_hex_dump_debug (cmd_name,DUMP_PREFIX_OFFSET ,4 ,4 , + out_obj->buffer.pointer, +min_t (u32,128 , out_obj->buffer.length), true); +if (call_pkg) { call_pkg->nd_fw_size = out_obj->buffer.length; memcpy(call_pkg->nd_payload + call_pkg->nd_size_in,
@@ -560,12 +566,6 @@ int acpi_nfit_ctl(struct nvdimm_bus_descriptor *nd_desc, struct nvdimm *nvdimm,return 0 ; } -dev_dbg (dev,"%s cmd: %s output length: %d\n" , dimm_name, - cmd_name, out_obj->buffer.length); -print_hex_dump_debug (cmd_name,DUMP_PREFIX_OFFSET ,4 ,4 , - out_obj->buffer.pointer, -min_t (u32,128 , out_obj->buffer.length), true); -for (i =0 , offset =0 ; i < desc->out_num; i++) { u32 out_size = nd_cmd_out_size(nvdimm, cmd, desc, i, buf, (u32 *) out_obj->buffer.pointer,
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 6/81 HID: input: add mapping for Expose/Overview key
In reply to Sasha LevinFrom: Dmitry Torokhov <[email protected]>
[ Upstream commit 96dd86871e1fffbc39e4fa61c9c75ec54ee9af0f ]
According to HUTRR77 usage 0x29f from the consumer page is reserved for the Desktop application to present all running user’s application windows. Linux defines KEY_SCALE to request Compiz Scale (Expose) mode, so let's add the mapping.
--- drivers/hid/hid-input.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/hid/hid-input.c +2 -0
@@ -1025,6 +1025,8 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fielcase 0x2cb :map_key_clear (KEY_KBDINPUTASSIST_ACCEPT );break ;case 0x2cc :map_key_clear (KEY_KBDINPUTASSIST_CANCEL );break ; +case 0x29f :map_key_clear (KEY_SCALE );break ; +default :map_key_clear (KEY_UNKNOWN ); }break ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 7/81 HID: input: add mapping for keyboard Brightness Up/Down/Toggle keys
In reply to Sasha LevinFrom: Dmitry Torokhov <[email protected]>
[ Upstream commit 7975a1d6a7afeb3eb61c971a153d24dd8fa032f3 ]
According to HUTRR73 usages 0x79, 0x7a and 0x7c from the consumer page correspond to Brightness Up/Down/Toggle keys, so let's add the mappings.
--- drivers/hid/hid-input.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/hid/hid-input.c +4 -0
@@ -895,6 +895,10 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fielcase 0x074 :map_key_clear (KEY_BRIGHTNESS_MAX );break ;case 0x075 :map_key_clear (KEY_BRIGHTNESS_AUTO );break ; +case 0x079 :map_key_clear (KEY_KBDILLUMUP );break ; +case 0x07a :map_key_clear (KEY_KBDILLUMDOWN );break ; +case 0x07c :map_key_clear (KEY_KBDILLUMTOGGLE );break ; +case 0x082 :map_key_clear (KEY_VIDEO_NEXT );break ;case 0x083 :map_key_clear (KEY_LAST );break ;case 0x084 :map_key_clear (KEY_ENTER );break ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 8/81 HID: input: add mapping for "Toggle Display" key
In reply to Sasha LevinFrom: Dmitry Torokhov <[email protected]>
[ Upstream commit c01908a14bf735b871170092807c618bb9dae654 ]
According to HUT 1.12 usage 0xb5 from the generic desktop page is reserved for switching between external and internal display, so let's add the mapping.
--- drivers/hid/hid-input.c | 8 ++++++++ 1 file changed, 8 insertions(+)
drivers/hid/hid-input.c +8 -0
@@ -677,6 +677,14 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fielbreak ; } +if ((usage->hid &0xf0 ) ==0xb0 ) {/* SC - Display */ +switch (usage->hid &0xf ) { +case 0x05 :map_key_clear (KEY_SWITCHVIDEOMODE );break ; +default : goto ignore; + } +break ; + } +/* * Some lazy vendors declare 255 usages for System Control, * leading to the creation of ABS_X|Y axis and too many others.
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 10/81 s390/dasd: Fix capacity calculation for large volumes
In reply to Sasha LevinFrom: Peter Oberparleiter <[email protected]>
[ Upstream commit 2cc9637ce825f3a9f51f8f78af7474e9e85bfa5f ]
The DASD driver incorrectly limits the maximum number of blocks of ECKD DASD volumes to 32 bit numbers. Volumes with a capacity greater than 2^32-1 blocks are incorrectly recognized as smaller volumes.
This results in the following volume capacity limits depending on the formatted block size:
BLKSIZE MAX_GB MAX_CYL
512 2047 5843492
1024 4095 8676701
2048 8191 13634816
4096 16383 23860929The same problem occurs when a volume with more than 17895697 cylinders is accessed in raw-track-access mode.
Fix this problem by adding an explicit type cast when calculating the maximum number of blocks.
--- drivers/s390/block/dasd_eckd.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/s390/block/dasd_eckd.c +3 -3
@@ -2004,14 +2004,14 @@ static int dasd_eckd_end_analysis(struct dasd_block *block) blk_per_trk =recs_per_track (&private->rdc_data,0 , block->bp_block); raw: - block->blocks = (private->real_cyl * + block->blocks = ((unsigned long ) private->real_cyl * private->rdc_data.trk_per_cyl * blk_per_trk); dev_info(&device->cdev->dev, - "DASD with %d KB/block, %d KB total size, %d KB/track, " + "DASD with %u KB/block, %lu KB total size, %u KB/track, " "%s\n", (block->bp_block >> 10), - ((private->real_cyl * + (((unsigned long) private->real_cyl * private->rdc_data.trk_per_cyl * blk_per_trk * (block->bp_block >> 9)) >> 1), ((blk_per_trk * block->bp_block) >> 10),
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 16/81 Input: snvs_pwrkey - initialize necessary driver data before enabling IRQ
In reply to Sasha LevinFrom: Anson Huang <[email protected]>
[ Upstream commit bf2a7ca39fd3ab47ef71c621a7ee69d1813b1f97 ]
SNVS IRQ is requested before necessary driver data initialized, if there is a pending IRQ during driver probe phase, kernel NULL pointer panic will occur in IRQ handler. To avoid such scenario, just initialize necessary driver data before enabling IRQ. This patch is inspired by NXP's internal kernel tree.
--- drivers/input/keyboard/snvs_pwrkey.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/input/keyboard/snvs_pwrkey.c +3 -3
@@ -148,6 +148,9 @@ static int imx_snvs_pwrkey_probe(struct platform_device *pdev)return error; } + pdata->input = input; +platform_set_drvdata (pdev, pdata); + error =devm_request_irq (&pdev->dev, pdata->irq, imx_snvs_pwrkey_interrupt,0 , pdev->name, pdev);
@@ -163,9 +166,6 @@ static int imx_snvs_pwrkey_probe(struct platform_device *pdev)return error; } - pdata->input = input; -platform_set_drvdata (pdev, pdata); -device_init_wakeup (&pdev->dev, pdata->wakeup);return 0 ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 19/81 s390/3270: fix lockdep false positive on view->lock
In reply to Sasha LevinFrom: Martin Schwidefsky <[email protected]>
[ Upstream commit 5712f3301a12c0c3de9cc423484496b0464f2faf ]
The spinlock in the raw3270_view structure is used by con3270, tty3270 and fs3270 in different ways. For con3270 the lock can be acquired in irq context, for tty3270 and fs3270 the highest context is bh.
Lockdep sees the view->lock as a single class and if the 3270 driver is used for the console the following message is generated:
WARNING: inconsistent lock state
5.1.0-rc3-05157-g5c168033979d #12 Not tainted
--------------------------------
inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage.
swapper/0/1 [HC0[0]:SC1[1]:HE1:SE0] takes:
(____ptrval____) (&(&view->lock)->rlock){?.-.}, at: tty3270_update+0x7c/0x330
Introduce a lockdep subclass for the view lock to distinguish bh from irq locks.
--- drivers/s390/char/con3270.c | 2 +- drivers/s390/char/fs3270.c | 3 ++- drivers/s390/char/raw3270.c | 3 ++- drivers/s390/char/raw3270.h | 4 +++- drivers/s390/char/tty3270.c | 3 ++- 5 files changed, 10 insertions(+), 5 deletions(-)
drivers/s390/char/con3270.c +1 -1
@@ -629,7 +629,7 @@ con3270_init(void) (void (*)(unsigned long )) con3270_read_tasklet, (unsigned long) condev->read); -raw3270_add_view (&condev->view, &con3270_fn,1 ); +raw3270_add_view (&condev->view, &con3270_fn,1 ,RAW3270_VIEW_LOCK_IRQ );INIT_LIST_HEAD (&condev->freemem);for (i =0 ; i <CON3270_STRING_PAGES ; i++) {
drivers/s390/char/fs3270.c +2 -1
@@ -463,7 +463,8 @@ fs3270_open(struct inode *inode, struct file *filp)init_waitqueue_head (&fp->wait); fp->fs_pid =get_pid (task_pid (current)); - rc =raw3270_add_view (&fp->view, &fs3270_fn, minor); + rc =raw3270_add_view (&fp->view, &fs3270_fn, minor, +RAW3270_VIEW_LOCK_BH );if (rc) {fs3270_free_view (&fp->view); goto out;
drivers/s390/char/raw3270.c +2 -1
@@ -920,7 +920,7 @@ raw3270_deactivate_view(struct raw3270_view *view)* Add view to device with minor "minor". */ int -raw3270_add_view (struct raw3270_view *view,struct raw3270_fn *fn,int minor) +raw3270_add_view (struct raw3270_view *view,struct raw3270_fn *fn,int minor,int subclass) {unsigned long flags;struct raw3270 *rp;
@@ -942,6 +942,7 @@ raw3270_add_view(struct raw3270_view *view, struct raw3270_fn *fn, int minor) view->cols = rp->cols; view->ascebc = rp->ascebc;spin_lock_init (&view->lock); +lockdep_set_subclass (&view->lock, subclass);list_add (&view->list, &rp->view_list); rc =0 ;spin_unlock_irqrestore (get_ccwdev_lock (rp->cdev), flags);
drivers/s390/char/raw3270.h +3 -1
@@ -150,6 +150,8 @@ struct raw3270_fn {struct raw3270_view {struct list_head list;spinlock_t lock; +#define RAW3270_VIEW_LOCK_IRQ 0 +#define RAW3270_VIEW_LOCK_BH 1atomic_t ref_count;struct raw3270 *dev;struct raw3270_fn *fn;
@@ -158,7 +160,7 @@ struct raw3270_view {unsigned char *ascebc;/* ascii -> ebcdic table */ }; -int raw3270_add_view (struct raw3270_view *,struct raw3270_fn *,int ); +int raw3270_add_view (struct raw3270_view *,struct raw3270_fn *,int ,int );int raw3270_activate_view (struct raw3270_view *);void raw3270_del_view (struct raw3270_view *);void raw3270_deactivate_view (struct raw3270_view *);
drivers/s390/char/tty3270.c +2 -1
@@ -980,7 +980,8 @@ static int tty3270_install(struct tty_driver *driver, struct tty_struct *tty)return PTR_ERR (tp); rc =raw3270_add_view (&tp->view, &tty3270_fn, - tty->index +RAW3270_FIRSTMINOR ); + tty->index +RAW3270_FIRSTMINOR , +RAW3270_VIEW_LOCK_BH );if (rc) {tty3270_free_view (tp);return rc;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 20/81 drm/amd/display: extending AUX SW Timeout
In reply to Sasha LevinFrom: Martin Leung <[email protected]>
[ Upstream commit f4bbebf8e7eb4d294b040ab2d2ba71e70e69b930 ]
[Why]
AUX takes longer to reply when using active DP-DVI dongle on some asics resulting in up to 2000+ us edid read (timeout).
[How]
1. Adjust AUX poll to match spec
2. Extend the SW timeout. This does not affect normal operation since we exit the loop as soon as AUX acks.
--- drivers/gpu/drm/amd/display/dc/dce/dce_aux.c | 9 ++++++--- drivers/gpu/drm/amd/display/dc/dce/dce_aux.h | 6 +++--- 2 files changed, 9 insertions(+), 6 deletions(-)
drivers/gpu/drm/amd/display/dc/dce/dce_aux.c +6 -3
@@ -189,6 +189,12 @@ static void submit_channel_request(1 , 0); } + +REG_UPDATE (AUX_INTERRUPT_CONTROL ,AUX_SW_DONE_ACK ,1 ); + +REG_WAIT (AUX_SW_STATUS ,AUX_SW_DONE ,0 , +10 , aux110->timeout_period/10 ); +/* set the delay and the number of bytes to write */ /* The length include
@@ -241,9 +247,6 @@ static void submit_channel_request( } } -REG_UPDATE (AUX_INTERRUPT_CONTROL ,AUX_SW_DONE_ACK ,1 ); -REG_WAIT (AUX_SW_STATUS ,AUX_SW_DONE ,0 , -10 , aux110->timeout_period/10 );REG_UPDATE (AUX_SW_CONTROL ,AUX_SW_GO ,1 ); }
drivers/gpu/drm/amd/display/dc/dce/dce_aux.h +3 -3
@@ -69,11 +69,11 @@ enum { /* This is the timeout as defined in DP 1.2a,* at most within ~240usec. That means, * increasing this timeout will not affect normal operation, * and we'll timeout after -* SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD = 1600usec. +* SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD = 2400usec. * This timeout is especially important for -* resume from S3 and CTS. +* converters, resume from S3, and CTS. */ -SW_AUX_TIMEOUT_PERIOD_MULTIPLIER =4 +SW_AUX_TIMEOUT_PERIOD_MULTIPLIER =6 };struct aux_engine_dce110 {struct aux_engine base;
Signature · 6 lines
2.20.1 _______________________________________________ amd-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/amd-gfx
Sasha Levin
maintainerPATCH 23/81 selftests: fib_tests: Fix 'Command line is not complete' errors
In reply to Sasha LevinFrom: David Ahern <dsahern at gmail.com>
[ Upstream commit a5f622984a623df9a84cf43f6b098d8dd76fbe05 ]
A couple of tests are verifying a route has been removed. The helper expects the prefix as the first part of the expected output. When checking that a route has been deleted the prefix is empty leading to an invalid ip command:
$ ip ro ls match Command line is not complete. Try option "help"
Fix by moving the comparison of expected output and output to a new function that is used by both check_route and check_route6. Use the new helper for the 2 checks on route removal.
Also, remove the reset of 'set -x' in route_setup which overrides the user managed setting.
--- tools/testing/selftests/net/fib_tests.sh | 94 ++++++++++-------------- 1 file changed, 40 insertions(+), 54 deletions(-)
tools/testing/selftests/net/fib_tests.sh +40 -54
@@ -602,6 +602,39 @@ run_cmd() return $rc } +check_expected() +{ + local out="$1" + local expected="$2" + local rc=0 + + [ "${out}" = "${expected}" ] && return 0 + + if [ -z "${out}" ]; then + if [ "$VERBOSE" = "1" ]; then + printf "\nNo route entry found\n" + printf "Expected:\n" + printf " ${expected}\n" + fi + return 1 + fi + + # tricky way to convert output to 1-line without ip's + # messy '\'; this drops all extra white space + out=$(echo ${out}) + if [ "${out}" != "${expected}" ]; then + rc=1 + if [ "${VERBOSE}" = "1" ]; then + printf " Unexpected route entry. Have:\n" + printf " ${out}\n" + printf " Expected:\n" + printf " ${expected}\n\n" + fi + fi + + return $rc +} + # add route for a prefix, flushing any existing routes first # expected to be the first step of a test add_route6()
@@ -646,31 +679,7 @@ check_route6() local rc=0 out=$($IP -6 ro ls match ${pfx} | sed -e 's/ pref medium//') - [ "${out}" = "${expected}" ] && return 0 - - if [ -z "${out}" ]; then - if [ "$VERBOSE" = "1" ]; then - printf "\nNo route entry found\n" - printf "Expected:\n" - printf " ${expected}\n" - fi - return 1 - fi - - # tricky way to convert output to 1-line without ip's - # messy '\'; this drops all extra white space - out=$(echo ${out}) - if [ "${out}" != "${expected}" ]; then - rc=1 - if [ "${VERBOSE}" = "1" ]; then - printf " Unexpected route entry. Have:\n" - printf " ${out}\n" - printf " Expected:\n" - printf " ${expected}\n\n" - fi - fi - - return $rc + check_expected "${out}" "${expected}" } route_cleanup()
@@ -714,7 +723,7 @@ route_setup() $IP addr add 172.16.103.2/24 dev veth4 $IP addr add 172.16.104.1/24 dev dummy1 - set +ex + set +e } # assumption is that basic add of a single path route works
@@ -949,7 +958,8 @@ ipv6_addr_metric_test() run_cmd "$IP li set dev dummy2 down" rc=$? if [ $rc -eq 0 ]; then - check_route6 "" + out=$($IP -6 ro ls match 2001:db8:104::/64) + check_expected "${out}" "" rc=$? fi log_test $rc 0 "Prefix route removed on link down"
@@ -1009,34 +1019,9 @@ check_route() local pfx="172.16.104.0/24" local expected="$1" local out - local rc=0 out=$($IP ro ls match ${pfx}) - [ "${out}" = "${expected}" ] && return 0 - - if [ -z "${out}" ]; then - if [ "$VERBOSE" = "1" ]; then - printf "\nNo route entry found\n" - printf "Expected:\n" - printf " ${expected}\n" - fi - return 1 - fi - - # tricky way to convert output to 1-line without ip's - # messy '\'; this drops all extra white space - out=$(echo ${out}) - if [ "${out}" != "${expected}" ]; then - rc=1 - if [ "${VERBOSE}" = "1" ]; then - printf " Unexpected route entry. Have:\n" - printf " ${out}\n" - printf " Expected:\n" - printf " ${expected}\n\n" - fi - fi - - return $rc + check_expected "${out}" "${expected}" } # assumption is that basic add of a single path route works
@@ -1301,7 +1286,8 @@ ipv4_addr_metric_test() run_cmd "$IP li set dev dummy2 down" rc=$? if [ $rc -eq 0 ]; then - check_route "" + out=$($IP ro ls match 172.16.104.0/24) + check_expected "${out}" "" rc=$? fi log_test $rc 0 "Prefix route removed on link down"
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 28/81 drm/amd/display: If one stream full updates, full update all planes
In reply to Sasha LevinFrom: David Francis <[email protected]>
[ Upstream commit c238bfe0be9ef7420f7669a69e27c8c8f4d8a568 ]
[Why]
On some compositors, with two monitors attached, VT terminal switch can cause a graphical issue by the following means:
There are two streams, one for each monitor. Each stream has one plane
current state:
M1:S1->P1 M2:S2->P2
The user calls for a terminal switch and a commit is made to change both planes to linear swizzle mode. In atomic check, a new dc_state is constructed with new planes on each stream
new state:
M1:S1->P3 M2:S2->P4
In commit tail, each stream is committed, one at a time. The first stream (S1) updates properly, triggerring a full update and replacing the state
current state:
M1:S1->P3 M2:S2->P4
The update for S2 comes in, but dc detects that there is no difference between the stream and plane in the new and current states, and so triggers a fast update. The fast update does not program swizzle, so the second monitor is corrupted
[How]
Add a flag to dc_plane_state that forces full updates
When a stream undergoes a full update, set this flag on all changed planes, then clear it on the current stream
Subsequent streams will get full updates as a result
--- drivers/gpu/drm/amd/display/dc/core/dc.c | 19 +++++++++++++++++++ drivers/gpu/drm/amd/display/dc/dc.h | 3 +++ 2 files changed, 22 insertions(+)
drivers/gpu/drm/amd/display/dc/core/dc.c +19 -0
@@ -1213,6 +1213,11 @@ static enum surface_update_type det_surface_update(const struct dc *dc,return UPDATE_TYPE_FULL ; } +if (u->surface->force_full_update) { + update_flags->bits.full_update =1 ; +return UPDATE_TYPE_FULL ; + } + type =get_plane_info_update_type (u);elevate_update_type (&overall_type, type);
@@ -1467,6 +1472,14 @@ void dc_commit_updates_for_stream(struct dc *dc, }dc_resource_state_copy_construct (state, context); + +for (i =0 ; i < dc->res_pool->pipe_count; i++) { +struct pipe_ctx *new_pipe = &context->res_ctx.pipe_ctx[i]; +struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i]; + +if (new_pipe->plane_state && new_pipe->plane_state != old_pipe->plane_state) + new_pipe->plane_state->force_full_update = true; + } }
@@ -1510,6 +1523,12 @@ void dc_commit_updates_for_stream(struct dc *dc, dc->current_state = context;dc_release_state (old); +for (i =0 ; i < dc->res_pool->pipe_count; i++) { +struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + +if (pipe_ctx->plane_state && pipe_ctx->stream == stream) + pipe_ctx->plane_state->force_full_update = false; + } }/*let's use current_state to update watermark etc*/ if (update_type >=UPDATE_TYPE_FULL )
drivers/gpu/drm/amd/display/dc/dc.h +3 -0
@@ -505,6 +505,9 @@ struct dc_plane_state {struct dc_plane_status status;struct dc_context *ctx; +/* HACK: Workaround for forcing full reprogramming under some conditions */ +bool force_full_update; +/* private to dc_surface.c */ enum dc_irq_source irq_source;struct kref refcount;
Signature · 6 lines
2.20.1 _______________________________________________ amd-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/amd-gfx
Sasha Levin
maintainerPATCH 29/81 s390/pkey: add one more argument space for debug feature entry
In reply to Sasha LevinFrom: Harald Freudenberger <[email protected]>
[ Upstream commit 6b1f16ba730d4c0cda1247568c3a1bf4fa3a2f2f ]
The debug feature entries have been used with up to 5 arguents (including the pointer to the format string) but there was only space reserved for 4 arguemnts. So now the registration does reserve space for 5 times a long value.
This fixes a sometime appearing weired value as the last value of an debug feature entry like this:
... pkey_sec2protkey zcrypt_send_cprb (cardnr=10 domain=12)
failed with errno -2143346254
--- drivers/s390/crypto/pkey_api.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/s390/crypto/pkey_api.c +2 -1
@@ -45,7 +45,8 @@ static debug_info_t *debug_info;static void __init pkey_debug_init (void ) { - debug_info =debug_register ("pkey" ,1 ,1 ,4 *sizeof (long )); +/* 5 arguments per dbf entry (including the format string ptr) */ + debug_info =debug_register ("pkey" ,1 ,1 ,5 *sizeof (long ));debug_register_view (debug_info, &debug_sprintf_view);debug_set_level (debug_info,3 ); }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 31/81 x86/reboot, efi: Use EFI reboot for Acer TravelMate X514-51T
In reply to Sasha LevinFrom: Jian-Hong Pan <[email protected]>
[ Upstream commit 0082517fa4bce073e7cf542633439f26538a14cc ]
Upon reboot, the Acer TravelMate X514-51T laptop appears to complete the shutdown process, but then it hangs in BIOS POST with a black screen.
The problem is intermittent - at some points it has appeared related to Secure Boot settings or different kernel builds, but ultimately we have not been able to identify the exact conditions that trigger the issue to come and go.
Besides, the EFI mode cannot be disabled in the BIOS of this model.
However, after extensive testing, we observe that using the EFI reboot method reliably avoids the issue in all cases.
So add a boot time quirk to use EFI reboot on such systems.
--- arch/x86/kernel/reboot.c | 21 +++++++++++++++++++++ include/linux/efi.h | 7 ++++++- 2 files changed, 27 insertions(+), 1 deletion(-)
arch/x86/kernel/reboot.c +21 -0
@@ -81,6 +81,19 @@ static int __init set_bios_reboot(const struct dmi_system_id *d)return 0 ; } +/* +* Some machines don't handle the default ACPI reboot method and +* require the EFI reboot method: +*/ +static int __init set_efi_reboot (const struct dmi_system_id *d) +{ +if (reboot_type !=BOOT_EFI && !efi_runtime_disabled ()) { + reboot_type =BOOT_EFI ; +pr_info ("%s series board detected. Selecting EFI-method for reboot.\n" , d->ident); + } +return 0 ; +} +void __noreturn machine_real_restart (unsigned int type) {local_irq_disable ();
@@ -166,6 +179,14 @@ static const struct dmi_system_id reboot_dmi_table[] __initconst = { DMI_MATCH(DMI_PRODUCT_NAME, "AOA110"), }, }, + { /* Handle reboot issue on Acer TravelMate X514-51T */ + .callback = set_efi_reboot, + .ident = "Acer TravelMate X514-51T", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Acer"), + DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate X514-51T"), + }, + },/* Apple */ {/* Handle problems with rebooting on Apple MacBook5 */
include/linux/efi.h +6 -1
@@ -1564,7 +1564,12 @@ efi_status_t efi_setup_gop(efi_system_table_t *sys_table_arg, struct screen_info *si, efi_guid_t *proto, unsigned long size); -bool efi_runtime_disabled (void ); +#ifdef CONFIG_EFI +extern bool efi_runtime_disabled (void ); +#else +static inline bool efi_runtime_disabled (void ) {return true; } +#endif +extern void efi_call_virt_check_flags (unsigned long flags,const char *call);enum efi_secureboot_mode {
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 40/81 scsi: aic7xxx: fix EISA support
In reply to Sasha LevinFrom: Christoph Hellwig <[email protected]>
[ Upstream commit 144ec97493af34efdb77c5aba146e9c7de8d0a06 ]
Instead of relying on the now removed NULL argument to pci_alloc_consistent, switch to the generic DMA API, and store the struct device so that we can pass it.
--- drivers/scsi/aic7xxx/aic7770_osm.c | 1 + drivers/scsi/aic7xxx/aic7xxx.h | 1 + drivers/scsi/aic7xxx/aic7xxx_osm.c | 10 ++++------ drivers/scsi/aic7xxx/aic7xxx_osm_pci.c | 1 + 4 files changed, 7 insertions(+), 6 deletions(-)
drivers/scsi/aic7xxx/aic7770_osm.c +1 -0
@@ -91,6 +91,7 @@ aic7770_probe(struct device *dev) ahc =ahc_alloc (&aic7xxx_driver_template, name);if (ahc ==NULL )return (ENOMEM ); + ahc->dev = dev; error =aic7770_config (ahc, aic7770_ident_table + edev->id.driver_data, eisaBase);if (error !=0 ) {
drivers/scsi/aic7xxx/aic7xxx.h +1 -0
@@ -949,6 +949,7 @@ struct ahc_softc {* Platform specific device information. */ ahc_dev_softc_t dev_softc; +struct device *dev;/* * Bus specific device information.
drivers/scsi/aic7xxx/aic7xxx_osm.c +4 -6
@@ -861,8 +861,8 @@ int ahc_dmamem_alloc(struct ahc_softc *ahc, bus_dma_tag_t dmat, void** vaddr, int flags, bus_dmamap_t *mapp) { - *vaddr =pci_alloc_consistent (ahc->dev_softc, - dmat->maxsize, mapp); +/* XXX: check if we really need the GFP_ATOMIC and unwind this mess! */ + *vaddr =dma_alloc_coherent (ahc->dev, dmat->maxsize, mapp,GFP_ATOMIC );if (*vaddr ==NULL )return ENOMEM ;return 0 ;
@@ -872,8 +872,7 @@ void ahc_dmamem_free(struct ahc_softc *ahc, bus_dma_tag_t dmat, void* vaddr, bus_dmamap_t map) { - pci_free_consistent(ahc->dev_softc, dmat->maxsize, - vaddr, map); + dma_free_coherent(ahc->dev, dmat->maxsize, vaddr, map); }int
@@ -1124,8 +1123,7 @@ ahc_linux_register_host(struct ahc_softc *ahc, struct scsi_host_template *templa host->transportt = ahc_linux_transport_template; - retval =scsi_add_host (host, - (ahc->dev_softc ? &ahc->dev_softc->dev :NULL )); + retval =scsi_add_host (host, ahc->dev);if (retval) {printk (KERN_WARNING "aic7xxx: scsi_add_host failed\n" );scsi_host_put (host);
drivers/scsi/aic7xxx/aic7xxx_osm_pci.c +1 -0
@@ -250,6 +250,7 @@ ahc_linux_pci_dev_probe(struct pci_dev *pdev, const struct pci_device_id *ent) } } ahc->dev_softc = pci; + ahc->dev = &pci->dev; error =ahc_pci_config (ahc, entry);if (error !=0 ) {ahc_free (ahc);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 43/81 selftests: netfilter: check icmp pkttoobig errors are set as related
In reply to Sasha LevinFrom: Florian Westphal <fw at strlen.de>
[ Upstream commit becf2319f320cae43e20cf179cc51a355a0deb5f ]
When an icmp error such as pkttoobig is received, conntrack checks if the "inner" header (header of packet that did not fit link mtu) is matches an existing connection, and, if so, sets that packet as being related to the conntrack entry it found.
It was recently reported that this "related" setting also works if the inner header is from another, different connection (i.e., artificial/forged icmp error).
Add a test, followup patch will add additional "inner dst matches outer dst in reverse direction" check before setting related state.
--- tools/testing/selftests/netfilter/Makefile | 2 +- .../netfilter/conntrack_icmp_related.sh | 283 ++++++++++++++++++ 2 files changed, 284 insertions(+), 1 deletion(-) create mode 100755 tools/testing/selftests/netfilter/conntrack_icmp_related.sh
tools/testing/selftests/netfilter/Makefile +1 -1
@@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 # Makefile for netfilter selftests -TEST_PROGS := nft_trans_stress.sh nft_nat.sh +TEST_PROGS := nft_trans_stress.sh nft_nat.sh conntrack_icmp_related.sh include ../lib.mk
tools/testing/selftests/netfilter/conntrack_icmp_related.sh new file +283 -0
283 lines not on this page · raw message
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 44/81 ipvs: do not schedule icmp errors from tunnels
In reply to Sasha LevinFrom: Julian Anastasov <[email protected]>
[ Upstream commit 0261ea1bd1eb0da5c0792a9119b8655cf33c80a3 ]
We can receive ICMP errors from client or from
tunneling real server. While the former can be
scheduled to real server, the latter should
not be scheduled, they are decapsulated only when
existing connection is found.
--- net/netfilter/ipvs/ip_vs_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/netfilter/ipvs/ip_vs_core.c +1 -1
@@ -1647,7 +1647,7 @@ ip_vs_in_icmp(struct netns_ipvs *ipvs, struct sk_buff *skb, int *related,if (!cp) {int v; -if (!sysctl_schedule_icmp (ipvs)) +if (ipip || !sysctl_schedule_icmp (ipvs))return NF_ACCEPT ;if (!ip_vs_try_to_schedule (ipvs,AF_INET , skb, pd, &v, &cp, &ciph))
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 48/81 s390: ctcm: fix ctcm_new_device error return code
In reply to Sasha LevinFrom: Arnd Bergmann <[email protected]>
[ Upstream commit 27b141fc234a3670d21bd742c35d7205d03cbb3a ]
clang points out that the return code from this function is undefined for one of the error paths:
../drivers/s390/net/ctcm_main.c:1595:7: warning: variable 'result' is used uninitialized whenever 'if' condition is true
[-Wsometimes-uninitialized]
if (priv->channel[direction] == NULL) {
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../drivers/s390/net/ctcm_main.c:1638:9: note: uninitialized use occurs here
return result;
^~~~~~
../drivers/s390/net/ctcm_main.c:1595:3: note: remove the 'if' if its condition is always false
if (priv->channel[direction] == NULL) {
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../drivers/s390/net/ctcm_main.c:1539:12: note: initialize the variable 'result' to silence this warning
int result;
^Make it return -ENODEV here, as in the related failure cases. gcc has a known bug in underreporting some of these warnings when it has already eliminated the assignment of the return code based on some earlier optimization step.
--- drivers/s390/net/ctcm_main.c | 1 + 1 file changed, 1 insertion(+)
drivers/s390/net/ctcm_main.c +1 -0
@@ -1595,6 +1595,7 @@ static int ctcm_new_device(struct ccwgroup_device *cgdev)if (priv->channel[direction] ==NULL ) {if (direction ==CTCM_WRITE )channel_free (priv->channel[CTCM_READ ]); + result = -ENODEV ; goto out_dev; } priv->channel[direction]->netdev = dev;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 51/81 selftests/net: correct the return value for run_netsocktests
In reply to Sasha LevinFrom: Po-Hsu Lin <po-hsu.lin at canonical.com>
[ Upstream commit 30c04d796b693e22405c38e9b78e9a364e4c77e6 ]
The run_netsocktests will be marked as passed regardless the actual test result from the ./socket:
selftests: net: run_netsocktests
========================================
--------------------
running socket test
--------------------
[FAIL]
ok 1..6 selftests: net: run_netsocktests [PASS]This is because the test script itself has been successfully executed. Fix this by exit 1 when the test failed.
--- tools/testing/selftests/net/run_netsocktests | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/testing/selftests/net/run_netsocktests +1 -1
@@ -7,7 +7,7 @@ echo "--------------------" ./socket if [ $? -ne 0 ]; then echo "[FAIL]" + exit 1 else echo "[PASS]" fi -
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 60/81 Input: synaptics-rmi4 - fix possible double free
In reply to Sasha LevinFrom: Pan Bian <[email protected]>
[ Upstream commit bce1a78423961fce676ac65540a31b6ffd179e6d ]
The RMI4 function structure has been released in rmi_register_function if error occurs. However, it will be released again in the function rmi_create_function, which may result in a double-free bug.
--- drivers/input/rmi4/rmi_driver.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-)
drivers/input/rmi4/rmi_driver.c +1 -5
@@ -860,7 +860,7 @@ static int rmi_create_function(struct rmi_device *rmi_dev, error =rmi_register_function (fn);if (error) - goto err_put_fn; +return error;if (pdt->function_number ==0x01 ) data->f01_container = fn;
@@ -870,10 +870,6 @@ static int rmi_create_function(struct rmi_device *rmi_dev,list_add_tail (&fn->node, &data->function_list);return RMI_SCAN_CONTINUE ; - -err_put_fn: -put_device (&fn->dev); -return error; }void rmi_enable_irq (struct rmi_device *rmi_dev,bool clear_wake)
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 62/81 RDMA/hns: Bugfix for mapping user db
In reply to Sasha LevinFrom: Lijun Ou <[email protected]>
[ Upstream commit 2557fabd6e29f349bfa0ac13f38ac98aa5eafc74 ]
When the maximum send wr delivered by the user is zero, the qp does not have a sq.
When allocating the sq db buffer to store the user sq pi pointer and map it to the kernel mode, max_send_wr is used as the trigger condition, while the kernel does not consider the max_send_wr trigger condition when mapmping db. It will cause sq record doorbell map fail and create qp fail.
The failed print information as follows:
hns3 0000:7d:00.1: Send cmd: tail - 418, opcode - 0x8504, flag - 0x0011, retval - 0x0000 hns3 0000:7d:00.1: Send cmd: 0xe59dc000 0x00000000 0x00000000 0x00000000 0x00000116 0x0000ffff hns3 0000:7d:00.1: sq record doorbell map failed!
hns3 0000:7d:00.1: Create RC QP failed
--- drivers/infiniband/hw/hns/hns_roce_qp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/hw/hns/hns_roce_qp.c +1 -1
@@ -494,7 +494,7 @@ static int hns_roce_set_kernel_sq_size(struct hns_roce_dev *hr_dev,static int hns_roce_qp_has_sq (struct ib_qp_init_attr *attr) { -if (attr->qp_type ==IB_QPT_XRC_TGT ) +if (attr->qp_type ==IB_QPT_XRC_TGT || !attr->cap.max_send_wr)return 0 ;return 1 ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 1/81 iio: adc: xilinx: fix potential use-after-free on remove
From: Sven Van Asbroeck <[email protected]>
[ Upstream commit 62039b6aef63380ba7a37c113bbaeee8a55c5342 ]
When cancel_delayed_work() returns, the delayed work may still be running. This means that the core could potentially free the private structure (struct xadc) while the delayed work is still using it. This is a potential use-after-free.
Fix by calling cancel_delayed_work_sync(), which waits for any residual work to finish before returning.
--- drivers/iio/adc/xilinx-xadc-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/adc/xilinx-xadc-core.c +1 -1
@@ -1320,7 +1320,7 @@ static int xadc_remove(struct platform_device *pdev) }free_irq (xadc->irq, indio_dev);clk_disable_unprepare (xadc->clk); -cancel_delayed_work (&xadc->zynq_unmask_work); +cancel_delayed_work_sync (&xadc->zynq_unmask_work);kfree (xadc->data);kfree (indio_dev->channels);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 5/81 libnvdimm/namespace: Fix a potential NULL pointer dereference
From: Kangjie Lu <[email protected]>
[ Upstream commit 55c1fc0af29a6c1b92f217b7eb7581a882e0c07c ]
In case kmemdup fails, the fix goes to blk_err to avoid NULL pointer dereference.
--- drivers/nvdimm/namespace_devs.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/nvdimm/namespace_devs.c +4 -1
@@ -2251,9 +2251,12 @@ static struct device *create_namespace_blk(struct nd_region *nd_region,if (!nsblk->uuid) goto blk_err;memcpy (name, nd_label->name,NSLABEL_NAME_LEN ); -if (name[0 ]) +if (name[0 ]) { nsblk->alt_name =kmemdup (name,NSLABEL_NAME_LEN ,GFP_KERNEL ); +if (!nsblk->alt_name) + goto blk_err; + } res =nsblk_add_resource (nd_region, ndd, nsblk,__le64_to_cpu (nd_label->dpa));if (!res)
Signature · 6 lines
2.20.1 _______________________________________________ Linux-nvdimm mailing list [email protected] https://lists.01.org/mailman/listinfo/linux-nvdimm
Sasha Levin
maintainerPATCH 3/81 iio: adc: xilinx: prevent touching unclocked h/w on remove
In reply to Sasha LevinFrom: Sven Van Asbroeck <[email protected]>
[ Upstream commit 2e4b88f73966adead360e47621df0183586fac32 ]
In remove, the clock is disabled before canceling the delayed work. This means that the delayed work may be touching unclocked hardware.
Fix by disabling the clock after the delayed work is fully canceled. This is consistent with the probe error path order.
--- drivers/iio/adc/xilinx-xadc-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/adc/xilinx-xadc-core.c +1 -1
@@ -1320,8 +1320,8 @@ static int xadc_remove(struct platform_device *pdev)iio_triggered_buffer_cleanup (indio_dev); }free_irq (xadc->irq, indio_dev); -clk_disable_unprepare (xadc->clk);cancel_delayed_work_sync (&xadc->zynq_unmask_work); +clk_disable_unprepare (xadc->clk);kfree (xadc->data);kfree (indio_dev->channels);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 9/81 libnvdimm/btt: Fix a kmemdup failure check
In reply to Sasha LevinFrom: Aditya Pakki <[email protected]>
[ Upstream commit 486fa92df4707b5df58d6508728bdb9321a59766 ]
In case kmemdup fails, the fix releases resources and returns to avoid the NULL pointer dereference.
--- drivers/nvdimm/btt_devs.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-)
drivers/nvdimm/btt_devs.c +13 -5
@@ -190,14 +190,15 @@ static struct device *__nd_btt_create(struct nd_region *nd_region,return NULL ; nd_btt->id =ida_simple_get (&nd_region->btt_ida,0 ,0 ,GFP_KERNEL ); -if (nd_btt->id <0 ) { -kfree (nd_btt); -return NULL ; - } +if (nd_btt->id <0 ) + goto out_nd_btt; nd_btt->lbasize = lbasize; -if (uuid) +if (uuid) { uuid =kmemdup (uuid,16 ,GFP_KERNEL ); +if (!uuid) + goto out_put_id; + } nd_btt->uuid = uuid; dev = &nd_btt->dev;dev_set_name (dev,"btt%d.%d" , nd_region->id, nd_btt->id);
@@ -212,6 +213,13 @@ static struct device *__nd_btt_create(struct nd_region *nd_region,return NULL ; }return dev; + +out_put_id: +ida_simple_remove (&nd_region->btt_ida, nd_btt->id); + +out_nd_btt: +kfree (nd_btt); +return NULL ; }struct device *nd_btt_create (struct nd_region *nd_region)
Signature · 6 lines
2.20.1 _______________________________________________ Linux-nvdimm mailing list [email protected] https://lists.01.org/mailman/listinfo/linux-nvdimm
Sasha Levin
maintainerPATCH 13/81 cfg80211: Handle WMM rules in regulatory domain intersection
In reply to Sasha LevinFrom: Ilan Peer <[email protected]>
[ Upstream commit 08a75a887ee46828b54600f4bb7068d872a5edd5 ]
The support added for regulatory WMM rules did not handle the case of regulatory domain intersections. Fix it.
--- net/wireless/reg.c | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+)
net/wireless/reg.c +39 -0
@@ -1287,6 +1287,16 @@ reg_intersect_dfs_region(const enum nl80211_dfs_regions dfs_region1,return dfs_region1; } +static void reg_wmm_rules_intersect (const struct ieee80211_wmm_ac *wmm_ac1, +const struct ieee80211_wmm_ac *wmm_ac2, +struct ieee80211_wmm_ac *intersect) +{ + intersect->cw_min =max_t (u16, wmm_ac1->cw_min, wmm_ac2->cw_min); + intersect->cw_max =max_t (u16, wmm_ac1->cw_max, wmm_ac2->cw_max); + intersect->cot =min_t (u16, wmm_ac1->cot, wmm_ac2->cot); + intersect->aifsn =max_t (u8, wmm_ac1->aifsn, wmm_ac2->aifsn); +} +/* * Helper for regdom_intersect(), this does the real * mathematical intersection fun
@@ -1301,6 +1311,8 @@ static int reg_rules_intersect(const struct ieee80211_regdomain *rd1,struct ieee80211_freq_range *freq_range;const struct ieee80211_power_rule *power_rule1, *power_rule2;struct ieee80211_power_rule *power_rule; +const struct ieee80211_wmm_rule *wmm_rule1, *wmm_rule2; +struct ieee80211_wmm_rule *wmm_rule;u32 freq_diff, max_bandwidth1, max_bandwidth2; freq_range1 = &rule1->freq_range;
@@ -1311,6 +1323,10 @@ static int reg_rules_intersect(const struct ieee80211_regdomain *rd1, power_rule2 = &rule2->power_rule; power_rule = &intersected_rule->power_rule; + wmm_rule1 = &rule1->wmm_rule; + wmm_rule2 = &rule2->wmm_rule; + wmm_rule = &intersected_rule->wmm_rule; + freq_range->start_freq_khz =max (freq_range1->start_freq_khz, freq_range2->start_freq_khz); freq_range->end_freq_khz = min(freq_range1->end_freq_khz,
@@ -1354,6 +1370,29 @@ static int reg_rules_intersect(const struct ieee80211_regdomain *rd1, intersected_rule->dfs_cac_ms =max (rule1->dfs_cac_ms, rule2->dfs_cac_ms); +if (rule1->has_wmm && rule2->has_wmm) { +u8 ac; + +for (ac =0 ; ac <IEEE80211_NUM_ACS ; ac++) { +reg_wmm_rules_intersect (&wmm_rule1->client[ac], + &wmm_rule2->client[ac], + &wmm_rule->client[ac]); +reg_wmm_rules_intersect (&wmm_rule1->ap[ac], + &wmm_rule2->ap[ac], + &wmm_rule->ap[ac]); + } + + intersected_rule->has_wmm = true; + }else if (rule1->has_wmm) { + *wmm_rule = *wmm_rule1; + intersected_rule->has_wmm = true; + }else if (rule2->has_wmm) { + *wmm_rule = *wmm_rule2; + intersected_rule->has_wmm = true; + }else { + intersected_rule->has_wmm = false; + } +if (!is_valid_reg_rule (intersected_rule))return -EINVAL ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 17/81 libnvdimm/pmem: fix a possible OOB access when read and write pmem
In reply to Sasha LevinFrom: Li RongQing <[email protected]>
[ Upstream commit 9dc6488e84b0f64df17672271664752488cd6a25 ]
If offset is not zero and length is bigger than PAGE_SIZE, this will cause to out of boundary access to a page memory
--- drivers/nvdimm/pmem.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/nvdimm/pmem.c +4 -4
@@ -113,13 +113,13 @@ static void write_pmem(void *pmem_addr, struct page *page,while (len) { mem =kmap_atomic (page); - chunk = min_t(unsigned int, len, PAGE_SIZE); + chunk = min_t(unsigned int, len, PAGE_SIZE - off);memcpy_flushcache (pmem_addr, mem + off, chunk);kunmap_atomic (mem); len -= chunk; off =0 ; page++; - pmem_addr +=PAGE_SIZE ; + pmem_addr += chunk; } }
@@ -132,7 +132,7 @@ static blk_status_t read_pmem(struct page *page, unsigned int off,while (len) { mem =kmap_atomic (page); - chunk = min_t(unsigned int, len, PAGE_SIZE); + chunk = min_t(unsigned int, len, PAGE_SIZE - off); rem =memcpy_mcsafe (mem + off, pmem_addr, chunk);kunmap_atomic (mem);if (rem)
@@ -140,7 +140,7 @@ static blk_status_t read_pmem(struct page *page, unsigned int off, len -= chunk; off =0 ; page++; - pmem_addr +=PAGE_SIZE ; + pmem_addr += chunk; }return BLK_STS_OK ; }
Signature · 6 lines
2.20.1 _______________________________________________ Linux-nvdimm mailing list [email protected] https://lists.01.org/mailman/listinfo/linux-nvdimm
Sasha Levin
maintainerPATCH 15/81 nl80211: Add NL80211_FLAG_CLEAR_SKB flag for other NL commands
In reply to Sasha LevinFrom: Sunil Dutt <[email protected]>
[ Upstream commit d6db02a88a4aaa1cd7105137c67ddec7f3bdbc05 ]
This commit adds NL80211_FLAG_CLEAR_SKB flag to other NL commands that carry key data to ensure they do not stick around on heap after the SKB is freed.
Also introduced this flag for NL80211_CMD_VENDOR as there are sub commands which configure the keys.
--- net/wireless/nl80211.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-)
net/wireless/nl80211.c +12 -6
@@ -13392,7 +13392,8 @@ static const struct genl_ops nl80211_ops[] = { .policy = nl80211_policy, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + NL80211_FLAG_NEED_RTNL | + NL80211_FLAG_CLEAR_SKB, }, { .cmd = NL80211_CMD_DEAUTHENTICATE,
@@ -13443,7 +13444,8 @@ static const struct genl_ops nl80211_ops[] = { .policy = nl80211_policy, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + NL80211_FLAG_NEED_RTNL | + NL80211_FLAG_CLEAR_SKB, }, { .cmd = NL80211_CMD_UPDATE_CONNECT_PARAMS,
@@ -13451,7 +13453,8 @@ static const struct genl_ops nl80211_ops[] = { .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + NL80211_FLAG_NEED_RTNL | + NL80211_FLAG_CLEAR_SKB, }, { .cmd = NL80211_CMD_DISCONNECT,
@@ -13480,7 +13483,8 @@ static const struct genl_ops nl80211_ops[] = { .policy = nl80211_policy, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + NL80211_FLAG_NEED_RTNL | + NL80211_FLAG_CLEAR_SKB, }, { .cmd = NL80211_CMD_DEL_PMKSA,
@@ -13832,7 +13836,8 @@ static const struct genl_ops nl80211_ops[] = { .policy = nl80211_policy, .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WIPHY | - NL80211_FLAG_NEED_RTNL, + NL80211_FLAG_NEED_RTNL | + NL80211_FLAG_CLEAR_SKB, }, { .cmd = NL80211_CMD_SET_QOS_MAP,
@@ -13887,7 +13892,8 @@ static const struct genl_ops nl80211_ops[] = { .doit = nl80211_set_pmk, .policy = nl80211_policy, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | - NL80211_FLAG_NEED_RTNL, + NL80211_FLAG_NEED_RTNL | + NL80211_FLAG_CLEAR_SKB, }, { .cmd = NL80211_CMD_DEL_PMK,
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 18/81 mac80211: Honor SW_CRYPTO_CONTROL for unicast keys in AP VLAN mode
In reply to Sasha LevinFrom: Alexander Wetzel <[email protected]>
[ Upstream commit 78ad2341521d5ea96cb936244ed4c4c4ef9ec13b ]
Restore SW_CRYPTO_CONTROL operation on AP_VLAN interfaces for unicast keys, the original override was intended to be done for group keys as those are treated specially by mac80211 and would always have been rejected.
Now the situation is that AP_VLAN support must be enabled by the driver if it can support it (meaning it can support software crypto GTK TX).
Thus, also simplify the code - if we get here with AP_VLAN and non- pairwise key, software crypto must be used (driver doesn't know about the interface) and can be used (driver must've advertised AP_VLAN if it also uses SW_CRYPTO_CONTROL).
--- net/mac80211/key.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-)
net/mac80211/key.c +4 -5
@@ -167,8 +167,10 @@ static int ieee80211_key_enable_hw_accel(struct ieee80211_key *key)* The driver doesn't know anything about VLAN interfaces. * Hence, don't send GTKs for VLAN interfaces to the driver. */ -if (!(key->conf.flags &IEEE80211_KEY_FLAG_PAIRWISE )) +if (!(key->conf.flags &IEEE80211_KEY_FLAG_PAIRWISE )) { + ret =1 ; goto out_unsupported; + } } ret = drv_set_key(key->local, SET_KEY, sdata,
@@ -213,11 +215,8 @@ static int ieee80211_key_enable_hw_accel(struct ieee80211_key *key)/* all of these we can do in software - if driver can */ if (ret ==1 )return 0 ; -if (ieee80211_hw_check (&key->local->hw,SW_CRYPTO_CONTROL )) { -if (sdata->vif.type ==NL80211_IFTYPE_AP_VLAN ) -return 0 ; +if (ieee80211_hw_check (&key->local->hw,SW_CRYPTO_CONTROL ))return -EINVAL ; - }return 0 ;default :return -EINVAL ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 22/81 clocksource/drivers/oxnas: Fix OX820 compatible
In reply to Sasha LevinFrom: Neil Armstrong <[email protected]>
[ Upstream commit fbc87aa0f7c429999dc31f1bac3b2615008cac32 ]
The OX820 compatible is wrong is the driver, fix it.
--- drivers/clocksource/timer-oxnas-rps.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/clocksource/timer-oxnas-rps.c +1 -1
@@ -296,4 +296,4 @@ static int __init oxnas_rps_timer_init(struct device_node *np)TIMER_OF_DECLARE (ox810se_rps,"oxsemi,ox810se-rps-timer" , oxnas_rps_timer_init);TIMER_OF_DECLARE (ox820_rps, -"oxsemi,ox820se-rps-timer" , oxnas_rps_timer_init); +"oxsemi,ox820-rps-timer" , oxnas_rps_timer_init);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 24/81 mISDN: Check address length before reading address family
In reply to Sasha LevinFrom: Tetsuo Handa <[email protected]>
[ Upstream commit 238ffdc49ef98b15819cfd5e3fb23194e3ea3d39 ]
KMSAN will complain if valid address length passed to bind() is shorter than sizeof("struct sockaddr_mISDN"->family) bytes.
--- drivers/isdn/mISDN/socket.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/isdn/mISDN/socket.c +2 -2
@@ -711,10 +711,10 @@ base_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_len)struct sock *sk = sock->sk;int err =0 ; -if (!maddr || maddr->family !=AF_ISDN ) +if (addr_len <sizeof (struct sockaddr_mISDN ))return -EINVAL ; -if (addr_len <sizeof (struct sockaddr_mISDN )) +if (!maddr || maddr->family !=AF_ISDN )return -EINVAL ;lock_sock (sk);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 25/81 vxge: fix return of a free'd memblock on a failed dma mapping
In reply to Sasha LevinFrom: Colin Ian King <[email protected]>
[ Upstream commit 0a2c34f18c94b596562bf3d019fceab998b8b584 ]
Currently if a pci dma mapping failure is detected a free'd memblock address is returned rather than a NULL (that indicates an error). Fix this by ensuring NULL is returned on this error case.
Addresses-Coverity: ("Use after free")
--- drivers/net/ethernet/neterion/vxge/vxge-config.c | 1 + 1 file changed, 1 insertion(+)
drivers/net/ethernet/neterion/vxge/vxge-config.c +1 -0
@@ -2365,6 +2365,7 @@ static void *__vxge_hw_blockpool_malloc(struct __vxge_hw_device *devh, u32 size, dma_object->addr))) {vxge_os_dma_free (devh->pdev, memblock, &dma_object->acc_handle); + memblock =NULL ; goto exit; }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 33/81 KVM: x86: Raise #GP when guest vCPU do not support PMU
In reply to Sasha LevinFrom: Liran Alon <[email protected]>
[ Upstream commit 672ff6cff80ca43bf3258410d2b887036969df5f ]
Before this change, reading a VMware pseduo PMC will succeed even when PMU is not supported by guest. This can easily be seen by running kvm-unit-test vmware_backdoors with "-cpu host,-pmu" option.
--- arch/x86/kvm/pmu.c | 4 ++++ 1 file changed, 4 insertions(+)
arch/x86/kvm/pmu.c +4 -0
@@ -281,9 +281,13 @@ static int kvm_pmu_rdpmc_vmware(struct kvm_vcpu *vcpu, unsigned idx, u64 *data)int kvm_pmu_rdpmc (struct kvm_vcpu *vcpu,unsigned idx,u64 *data) {bool fast_mode = idx & (1u <<31 ); +struct kvm_pmu *pmu =vcpu_to_pmu (vcpu);struct kvm_pmc *pmc;u64 ctr_val; +if (!pmu->version) +return 1 ; +if (is_vmware_backdoor_pmc (idx))return kvm_pmu_rdpmc_vmware (vcpu, idx, data);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 34/81 KVM: fix spectrev1 gadgets
In reply to Sasha LevinFrom: Paolo Bonzini <[email protected]>
[ Upstream commit 1d487e9bf8ba66a7174c56a0029c54b1eca8f99c ]
These were found with smatch, and then generalized when applicable.
--- arch/x86/kvm/lapic.c | 4 +++- include/linux/kvm_host.h | 10 ++++++---- virt/kvm/irqchip.c | 5 +++-- virt/kvm/kvm_main.c | 6 ++++-- 4 files changed, 16 insertions(+), 9 deletions(-)
arch/x86/kvm/lapic.c +3 -1
@@ -133,6 +133,7 @@ static inline bool kvm_apic_map_get_logical_dest(struct kvm_apic_map *map,if (offset <= max_apic_id) {u8 cluster_size =min (max_apic_id - offset +1 ,16U ); + offset =array_index_nospec (offset, map->max_apic_id +1 ); *cluster = &map->phys_map[offset]; *mask = dest_id & (0xffff >> (16 - cluster_size)); }else {
@@ -896,7 +897,8 @@ static inline bool kvm_apic_map_get_dest_lapic(struct kvm *kvm,if (irq->dest_id > map->max_apic_id) { *bitmap =0 ; }else { - *dst = &map->phys_map[irq->dest_id]; +u32 dest_id =array_index_nospec (irq->dest_id, map->max_apic_id +1 ); + *dst = &map->phys_map[dest_id]; *bitmap =1 ; }return true;
include/linux/kvm_host.h +6 -4
@@ -28,6 +28,7 @@#include <linux/irqbypass.h> #include <linux/swait.h> #include <linux/refcount.h> +#include <linux/nospec.h> #include <asm/signal.h> #include <linux/kvm.h>
@@ -491,10 +492,10 @@ static inline struct kvm_io_bus *kvm_get_bus(struct kvm *kvm, enum kvm_bus idx)static inline struct kvm_vcpu *kvm_get_vcpu (struct kvm *kvm,int i) { -/* Pairs with smp_wmb() in kvm_vm_ioctl_create_vcpu, in case -* the caller has read kvm->online_vcpus before (as is the case -* for kvm_for_each_vcpu, for example). -*/ +int num_vcpus =atomic_read (&kvm->online_vcpus); + i =array_index_nospec (i, num_vcpus); + +/* Pairs with smp_wmb() in kvm_vm_ioctl_create_vcpu. */ smp_rmb ();return kvm->vcpus[i]; }
@@ -578,6 +579,7 @@ void kvm_put_kvm(struct kvm *kvm);static inline struct kvm_memslots *__kvm_memslots (struct kvm *kvm,int as_id) { + as_id =array_index_nospec (as_id,KVM_ADDRESS_SPACE_NUM );return srcu_dereference_check (kvm->memslots[as_id], &kvm->srcu,lockdep_is_held (&kvm->slots_lock) || !refcount_read (&kvm->users_count));
virt/kvm/irqchip.c +3 -2
@@ -144,18 +144,19 @@ static int setup_routing_entry(struct kvm *kvm, {struct kvm_kernel_irq_routing_entry *ei;int r; +u32 gsi =array_index_nospec (ue->gsi,KVM_MAX_IRQ_ROUTES );/* * Do not allow GSI to be mapped to the same irqchip more than once. * Allow only one to one mapping between GSI and non-irqchip routing. */ -hlist_for_each_entry (ei, &rt->map[ue->gsi], link) +hlist_for_each_entry (ei, &rt->map[gsi], link)if (ei->type !=KVM_IRQ_ROUTING_IRQCHIP || ue->type !=KVM_IRQ_ROUTING_IRQCHIP || ue->u.irqchip.irqchip == ei->irqchip.irqchip)return -EINVAL ; - e->gsi = ue->gsi; + e->gsi = gsi; e->type = ue->type; r =kvm_set_routing_entry (kvm, e, ue);if (r)
virt/kvm/kvm_main.c +4 -2
@@ -2887,12 +2887,14 @@ static int kvm_ioctl_create_device(struct kvm *kvm,struct kvm_device_ops *ops =NULL ;struct kvm_device *dev;bool test = cd->flags &KVM_CREATE_DEVICE_TEST ; +int type;int ret;if (cd->type >=ARRAY_SIZE (kvm_device_ops_table))return -ENODEV ; - ops = kvm_device_ops_table[cd->type]; + type =array_index_nospec (cd->type,ARRAY_SIZE (kvm_device_ops_table)); + ops = kvm_device_ops_table[type];if (ops ==NULL )return -ENODEV ;
@@ -2907,7 +2909,7 @@ static int kvm_ioctl_create_device(struct kvm *kvm, dev->kvm = kvm;mutex_lock (&kvm->lock); - ret = ops->create (dev, cd->type); + ret = ops->create (dev, type);if (ret <0 ) {mutex_unlock (&kvm->lock);kfree (dev);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 36/81 tools lib traceevent: Fix missing equality check for strcmp
In reply to Sasha LevinFrom: Rikard Falkeborn <[email protected]>
[ Upstream commit f32c2877bcb068a718bb70094cd59ccc29d4d082 ]
There was a missing comparison with 0 when checking if type is "s64" or "u64". Therefore, the body of the if-statement was entered if "type" was "u64" or not "s64", which made the first strcmp() redundant since if type is "u64", it's not "s64".
If type is "s64", the body of the if-statement is not entered but since the remainder of the function consists of if-statements which will not be entered if type is "s64", we will just return "val", which is correct, albeit at the cost of a few more calls to strcmp(), i.e., it will behave just as if the if-statement was entered.
If type is neither "s64" or "u64", the body of the if-statement will be entered incorrectly and "val" returned. This means that any type that is checked after "s64" and "u64" is handled the same way as "s64" and "u64", i.e., the limiting of "val" to fit in for example "s8" is never reached.
This was introduced in the kernel tree when the sources were copied from trace-cmd in commit f7d82350e597 ("tools/events: Add files to create libtraceevent.a"), and in the trace-cmd repo in 1cdbae6035cei ("Implement typecasting in parser") when the function was introduced, i.e., it has always behaved the wrong way.
Detected by cppcheck.
--- tools/lib/traceevent/event-parse.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/lib/traceevent/event-parse.c +1 -1
@@ -2192,7 +2192,7 @@ eval_type_str(unsigned long long val, const char *type, int pointer)return val &0xffffffff ;if (strcmp (type,"u64" ) ==0 || -strcmp (type,"s64" )) +strcmp (type,"s64" ) ==0 )return val;if (strcmp (type,"s8" ) ==0 )
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 38/81 ocelot: Don't sleep in atomic context (irqs_disabled())
In reply to Sasha LevinFrom: Claudiu Manoil <[email protected]>
[ Upstream commit a8fd48b50deaa20808bbf0f6685f6f1acba6a64c ]
Preemption disabled at:
[<ffff000008cabd54>] dev_set_rx_mode+0x1c/0x38 Call trace: [<ffff00000808a5c0>] dump_backtrace+0x0/0x3d0 [<ffff00000808a9a4>] show_stack+0x14/0x20 [<ffff000008e6c0c0>] dump_stack+0xac/0xe4 [<ffff0000080fe76c>] ___might_sleep+0x164/0x238 [<ffff0000080fe890>] __might_sleep+0x50/0x88 [<ffff0000082261e4>] kmem_cache_alloc+0x17c/0x1d0 [<ffff000000ea0ae8>] ocelot_set_rx_mode+0x108/0x188 [mscc_ocelot_common] [<ffff000008cabcf0>] __dev_set_rx_mode+0x58/0xa0 [<ffff000008cabd5c>] dev_set_rx_mode+0x24/0x38
--- drivers/net/ethernet/mscc/ocelot.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mscc/ocelot.c +1 -1
@@ -605,7 +605,7 @@ static int ocelot_mact_mc_add(struct ocelot_port *port, struct netdev_hw_addr *hw_addr) {struct ocelot *ocelot = port->ocelot; -struct netdev_hw_addr *ha =kzalloc (sizeof (*ha),GFP_KERNEL ); +struct netdev_hw_addr *ha =kzalloc (sizeof (*ha),GFP_ATOMIC );if (!ha)return -ENOMEM ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 41/81 mm: fix inactive list balancing between NUMA nodes and cgroups
In reply to Sasha LevinFrom: Johannes Weiner <[email protected]>
[ Upstream commit 3b991208b897f52507168374033771a984b947b1 ]
During !CONFIG_CGROUP reclaim, we expand the inactive list size if it's thrashing on the node that is about to be reclaimed. But when cgroups are enabled, we suddenly ignore the node scope and use the cgroup scope only. The result is that pressure bleeds between NUMA nodes depending on whether cgroups are merely compiled into Linux. This behavioral difference is unexpected and undesirable.
When the refault adaptivity of the inactive list was first introduced, there were no statistics at the lruvec level - the intersection of node and memcg - so it was better than nothing.
But now that we have that infrastructure, use lruvec_page_state() to make the list balancing decision always NUMA aware.
--- mm/vmscan.c | 29 +++++++++-------------------- 1 file changed, 9 insertions(+), 20 deletions(-)
mm/vmscan.c +9 -20
@@ -2190,7 +2190,6 @@ static void shrink_active_list(unsigned long nr_to_scan,* 10TB 320 32GB */ static bool inactive_list_is_low (struct lruvec *lruvec,bool file, -struct mem_cgroup *memcg,struct scan_control *sc,bool actual_reclaim) {enum lru_list active_lru = file *LRU_FILE +LRU_ACTIVE ;
@@ -2211,16 +2210,12 @@ static bool inactive_list_is_low(struct lruvec *lruvec, bool file, inactive =lruvec_lru_size (lruvec, inactive_lru, sc->reclaim_idx); active =lruvec_lru_size (lruvec, active_lru, sc->reclaim_idx); -if (memcg) - refaults =memcg_page_state (memcg,WORKINGSET_ACTIVATE ); -else - refaults =node_page_state (pgdat,WORKINGSET_ACTIVATE ); -/* * When refaults are being observed, it means a new workingset * is being established. Disable active list protection to get * rid of the stale workingset quickly. */ + refaults =lruvec_page_state (lruvec,WORKINGSET_ACTIVATE );if (file && actual_reclaim && lruvec->refaults != refaults) { inactive_ratio =0 ; }else {
@@ -2241,12 +2236,10 @@ static bool inactive_list_is_low(struct lruvec *lruvec, bool file, }static unsigned long shrink_list (enum lru_list lru,unsigned long nr_to_scan, -struct lruvec *lruvec,struct mem_cgroup *memcg, -struct scan_control *sc) +struct lruvec *lruvec,struct scan_control *sc) {if (is_active_lru (lru)) { -if (inactive_list_is_low (lruvec,is_file_lru (lru), - memcg, sc, true)) +if (inactive_list_is_low (lruvec,is_file_lru (lru), sc, true))shrink_active_list (nr_to_scan, lruvec, sc, lru);return 0 ; }
@@ -2346,7 +2339,7 @@ static void get_scan_count(struct lruvec *lruvec, struct mem_cgroup *memcg,* anonymous pages on the LRU in eligible zones. * Otherwise, the small LRU gets thrashed. */ -if (!inactive_list_is_low (lruvec, false, memcg, sc, false) && +if (!inactive_list_is_low (lruvec, false, sc, false) &&lruvec_lru_size (lruvec,LRU_INACTIVE_ANON , sc->reclaim_idx) >> sc->priority) { scan_balance =SCAN_ANON ;
@@ -2364,7 +2357,7 @@ static void get_scan_count(struct lruvec *lruvec, struct mem_cgroup *memcg,* lruvec even if it has plenty of old anonymous pages unless the * system is under heavy pressure. */ -if (!inactive_list_is_low (lruvec, true, memcg, sc, false) && +if (!inactive_list_is_low (lruvec, true, sc, false) &&lruvec_lru_size (lruvec,LRU_INACTIVE_FILE , sc->reclaim_idx) >> sc->priority) { scan_balance =SCAN_FILE ; goto out;
@@ -2517,7 +2510,7 @@ static void shrink_node_memcg(struct pglist_data *pgdat, struct mem_cgroup *memc nr[lru] -= nr_to_scan; nr_reclaimed +=shrink_list (lru, nr_to_scan, - lruvec, memcg, sc); + lruvec, sc); } }
@@ -2584,7 +2577,7 @@ static void shrink_node_memcg(struct pglist_data *pgdat, struct mem_cgroup *memc* Even if we did not try to evict anon pages at all, we want to * rebalance the anon lru active/inactive ratio. */ -if (inactive_list_is_low (lruvec, false, memcg, sc, true)) +if (inactive_list_is_low (lruvec, false, sc, true))shrink_active_list (SWAP_CLUSTER_MAX , lruvec, sc,LRU_ACTIVE_ANON ); }
@@ -2982,12 +2975,8 @@ static void snapshot_refaults(struct mem_cgroup *root_memcg, pg_data_t *pgdat)unsigned long refaults;struct lruvec *lruvec; -if (memcg) - refaults =memcg_page_state (memcg,WORKINGSET_ACTIVATE ); -else - refaults =node_page_state (pgdat,WORKINGSET_ACTIVATE ); - lruvec =mem_cgroup_lruvec (pgdat, memcg); + refaults =lruvec_page_state (lruvec,WORKINGSET_ACTIVATE ); lruvec->refaults = refaults; }while ((memcg =mem_cgroup_iter (root_memcg, memcg,NULL ))); }
@@ -3344,7 +3333,7 @@ static void age_active_anon(struct pglist_data *pgdat, do { struct lruvec *lruvec = mem_cgroup_lruvec(pgdat, memcg); - if (inactive_list_is_low(lruvec, false, memcg, sc, true)) + if (inactive_list_is_low(lruvec, false, sc, true)) shrink_active_list(SWAP_CLUSTER_MAX, lruvec, sc, LRU_ACTIVE_ANON);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 42/81 init: initialize jump labels before command line option parsing
In reply to Sasha LevinFrom: Dan Williams <[email protected]>
[ Upstream commit 6041186a32585fc7a1d0f6cfe2f138b05fdc3c82 ]
When a module option, or core kernel argument, toggles a static-key it requires jump labels to be initialized early. While x86, PowerPC, and ARM64 arrange for jump_label_init() to be called before parse_args(), ARM does not.
Kernel command line: rdinit=/sbin/init page_alloc.shuffle=1 panic=-1 console=ttyAMA0,115200 page_alloc.shuffle=1
------------[ cut here ]------------ WARNING: CPU: 0 PID: 0 at ./include/linux/jump_label.h:303 page_alloc_shuffle+0x12c/0x1ac static_key_enable(): static key 'page_alloc_shuffle_key+0x0/0x4' used before call to jump_label_init() Modules linked in: CPU: 0 PID: 0 Comm: swapper Not tainted 5.1.0-rc4-next-20190410-00003-g3367c36ce744 #1 Hardware name: ARM Integrator/CP (Device Tree) [<c0011c68>] (unwind_backtrace) from [<c000ec48>] (show_stack+0x10/0x18) [<c000ec48>] (show_stack) from [<c07e9710>] (dump_stack+0x18/0x24) [<c07e9710>] (dump_stack) from [<c001bb1c>] (__warn+0xe0/0x108) [<c001bb1c>] (__warn) from [<c001bb88>] (warn_slowpath_fmt+0x44/0x6c) [<c001bb88>] (warn_slowpath_fmt) from [<c0b0c4a8>] (page_alloc_shuffle+0x12c/0x1ac) [<c0b0c4a8>] (page_alloc_shuffle) from [<c0b0c550>] (shuffle_store+0x28/0x48) [<c0b0c550>] (shuffle_store) from [<c003e6a0>] (parse_args+0x1f4/0x350) [<c003e6a0>] (parse_args) from [<c0ac3c00>] (start_kernel+0x1c0/0x488)
Move the fallback call to jump_label_init() to occur before parse_args().
The redundant calls to jump_label_init() in other archs are left intact in case they have static key toggling use cases that are even earlier than option parsing.
--- init/main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
init/main.c +2 -2
@@ -568,6 +568,8 @@ asmlinkage __visible void __init start_kernel(void)page_alloc_init ();pr_notice ("Kernel command line: %s\n" , boot_command_line); +/* parameters may set static keys */ +jump_label_init ();parse_early_param (); after_dashes = parse_args("Booting kernel", static_command_line, __start___param,
@@ -577,8 +579,6 @@ asmlinkage __visible void __init start_kernel(void)parse_args ("Setting init args" , after_dashes,NULL ,0 ,-1 ,-1 ,NULL , set_init_arg); -jump_label_init (); -/* * These use large bootmem allocations and must precede * kmem_cache_init()
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 45/81 netfilter: ctnetlink: don't use conntrack/expect object addresses as id
In reply to Sasha LevinFrom: Florian Westphal <[email protected]>
[ Upstream commit 3c79107631db1f7fd32cf3f7368e4672004a3010 ]
else, we leak the addresses to userspace via ctnetlink events and dumps.
Compute an ID on demand based on the immutable parts of nf_conn struct.
Another advantage compared to using an address is that there is no immediate re-use of the same ID in case the conntrack entry is freed and reallocated again immediately.
--- include/net/netfilter/nf_conntrack.h | 2 ++ net/netfilter/nf_conntrack_core.c | 35 ++++++++++++++++++++++++++++ net/netfilter/nf_conntrack_netlink.c | 34 +++++++++++++++++++++++---- 3 files changed, 66 insertions(+), 5 deletions(-)
include/net/netfilter/nf_conntrack.h +2 -0
@@ -313,6 +313,8 @@ struct nf_conn *nf_ct_tmpl_alloc(struct net *net, gfp_t flags);void nf_ct_tmpl_free (struct nf_conn *tmpl); +u32 nf_ct_get_id (const struct nf_conn *ct); +static inline void nf_ct_set (struct sk_buff *skb,struct nf_conn *ct,enum ip_conntrack_info info) {
net/netfilter/nf_conntrack_core.c +35 -0
@@ -25,6 +25,7 @@#include <linux/slab.h> #include <linux/random.h> #include <linux/jhash.h> +#include <linux/siphash.h> #include <linux/err.h> #include <linux/percpu.h> #include <linux/moduleparam.h>
@@ -424,6 +425,40 @@ nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, }EXPORT_SYMBOL_GPL (nf_ct_invert_tuple); +/* Generate a almost-unique pseudo-id for a given conntrack. +* +* intentionally doesn't re-use any of the seeds used for hash +* table location, we assume id gets exposed to userspace. +* +* Following nf_conn items do not change throughout lifetime +* of the nf_conn after it has been committed to main hash table: +* +* 1. nf_conn address +* 2. nf_conn->ext address +* 3. nf_conn->master address (normally NULL) +* 4. tuple +* 5. the associated net namespace +*/ +u32 nf_ct_get_id (const struct nf_conn *ct) +{ +static __read_mostly siphash_key_t ct_id_seed; +unsigned long a, b, c, d; + +net_get_random_once (&ct_id_seed,sizeof (ct_id_seed)); + + a = (unsigned long )ct; + b = (unsigned long )ct->master ^net_hash_mix (nf_ct_net (ct)); + c = (unsigned long )ct->ext; + d = (unsigned long )siphash (&ct->tuplehash,sizeof (ct->tuplehash), + &ct_id_seed); +#ifdef CONFIG_64BIT +return siphash_4u64 ((u64 )a, (u64 )b, (u64 )c, (u64 )d, &ct_id_seed); +#else +return siphash_4u32 ((u32 )a, (u32 )b, (u32 )c, (u32 )d, &ct_id_seed); +#endif +} +EXPORT_SYMBOL_GPL (nf_ct_get_id); +static void clean_from_lists (struct nf_conn *ct) {
net/netfilter/nf_conntrack_netlink.c +29 -5
@@ -29,6 +29,7 @@#include <linux/spinlock.h> #include <linux/interrupt.h> #include <linux/slab.h> +#include <linux/siphash.h> #include <linux/netfilter.h> #include <net/netlink.h>
@@ -487,7 +488,9 @@ static int ctnetlink_dump_ct_synproxy(struct sk_buff *skb, struct nf_conn *ct)static int ctnetlink_dump_id (struct sk_buff *skb,const struct nf_conn *ct) { -if (nla_put_be32 (skb,CTA_ID ,htonl ((unsigned long )ct))) +__be32 id = (__force __be32 )nf_ct_get_id (ct); + +if (nla_put_be32 (skb,CTA_ID , id)) goto nla_put_failure;return 0 ;
@@ -1275,8 +1278,9 @@ static int ctnetlink_del_conntrack(struct net *net, struct sock *ctnl, }if (cda[CTA_ID ]) { -u_int32_t id =ntohl (nla_get_be32 (cda[CTA_ID ])); -if (id != (u32 )(unsigned long )ct) { +__be32 id =nla_get_be32 (cda[CTA_ID ]); + +if (id != (__force __be32 )nf_ct_get_id (ct)) {nf_ct_put (ct);return -ENOENT ; }
@@ -2675,6 +2679,25 @@ static int ctnetlink_exp_dump_mask(struct sk_buff *skb,static const union nf_inet_addr any_addr; +static __be32 nf_expect_get_id (const struct nf_conntrack_expect *exp) +{ +static __read_mostly siphash_key_t exp_id_seed; +unsigned long a, b, c, d; + +net_get_random_once (&exp_id_seed,sizeof (exp_id_seed)); + + a = (unsigned long )exp; + b = (unsigned long )exp->helper; + c = (unsigned long )exp->master; + d = (unsigned long )siphash (&exp->tuple,sizeof (exp->tuple), &exp_id_seed); + +#ifdef CONFIG_64BIT +return (__force __be32 )siphash_4u64 ((u64 )a, (u64 )b, (u64 )c, (u64 )d, &exp_id_seed); +#else +return (__force __be32 )siphash_4u32 ((u32 )a, (u32 )b, (u32 )c, (u32 )d, &exp_id_seed); +#endif +} +static int ctnetlink_exp_dump_expect (struct sk_buff *skb,const struct nf_conntrack_expect *exp)
@@ -2722,7 +2745,7 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb, }#endif if (nla_put_be32 (skb,CTA_EXPECT_TIMEOUT ,htonl (timeout)) || -nla_put_be32 (skb,CTA_EXPECT_ID ,htonl ((unsigned long )exp)) || +nla_put_be32 (skb,CTA_EXPECT_ID ,nf_expect_get_id (exp)) ||nla_put_be32 (skb,CTA_EXPECT_FLAGS ,htonl (exp->flags)) ||nla_put_be32 (skb,CTA_EXPECT_CLASS ,htonl (exp->class))) goto nla_put_failure;
@@ -3027,7 +3050,8 @@ static int ctnetlink_get_expect(struct net *net, struct sock *ctnl,if (cda[CTA_EXPECT_ID ]) {__be32 id =nla_get_be32 (cda[CTA_EXPECT_ID ]); -if (ntohl (id) != (u32 )(unsigned long )exp) { + +if (id !=nf_expect_get_id (exp)) {nf_ct_expect_put (exp);return -ENOENT ; }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 49/81 drm/sun4i: Set device driver data at bind time for use in unbind
In reply to Sasha LevinFrom: Paul Kocialkowski <[email protected]>
[ Upstream commit 02b92adbe33e6dbd15dc6e32540b22f47c4ff0a2 ]
Our sun4i_drv_unbind gets the drm device using dev_get_drvdata. However, that driver data is never set in sun4i_drv_bind.
Set it there to avoid getting a NULL pointer at unbind time.
--- drivers/gpu/drm/sun4i/sun4i_drv.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/gpu/drm/sun4i/sun4i_drv.c +2 -0
@@ -92,6 +92,8 @@ static int sun4i_drv_bind(struct device *dev) ret = -ENOMEM ; goto free_drm; } + +dev_set_drvdata (dev, drm); drm->dev_private = drv;INIT_LIST_HEAD (&drv->frontend_list);INIT_LIST_HEAD (&drv->engine_list);
Signature · 6 lines
2.20.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 50/81 drm/sun4i: Fix component unbinding and component master deletion
In reply to Sasha LevinFrom: Paul Kocialkowski <[email protected]>
[ Upstream commit f5a9ed867c83875546c9aadd4ed8e785e9adcc3c ]
For our component-backed driver to be properly removed, we need to delete the component master in sun4i_drv_remove and make sure to call component_unbind_all in the master's unbind so that all components are unbound when the master is.
--- drivers/gpu/drm/sun4i/sun4i_drv.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/gpu/drm/sun4i/sun4i_drv.c +4 -0
@@ -160,6 +160,8 @@ static void sun4i_drv_unbind(struct device *dev)drm_mode_config_cleanup (drm);of_reserved_mem_device_release (dev);drm_dev_put (drm); + +component_unbind_all (dev,NULL ); }static const struct component_master_ops sun4i_drv_master_ops = {
@@ -407,6 +409,8 @@ static int sun4i_drv_probe(struct platform_device *pdev)static int sun4i_drv_remove (struct platform_device *pdev) { +component_master_del (&pdev->dev, &sun4i_drv_master_ops); +return 0 ; }
Signature · 6 lines
2.20.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 47/81 MIPS: perf: ath79: Fix perfcount IRQ assignment
In reply to Sasha LevinFrom: Petr Štetiar <[email protected]>
[ Upstream commit a1e8783db8e0d58891681bc1e6d9ada66eae8e20 ]
Currently it's not possible to use perf on ath79 due to genirq flags mismatch happening on static virtual IRQ 13 which is used for performance counters hardware IRQ 5.
On TP-Link Archer C7v5:
CPU0
2: 0 MIPS 2 ath9k 4: 318 MIPS 4 19000000.eth 7: 55034 MIPS 7 timer 8: 1236 MISC 3 ttyS0 12: 0 INTC 1 ehci_hcd:usb1 13: 0 gpio-ath79 2 keys 14: 0 gpio-ath79 5 keys 15: 31 AR724X PCI 1 ath10k_pci
$ perf top genirq: Flags mismatch irq 13. 00014c83 (mips_perf_pmu) vs. 00002003 (keys)
On TP-Link Archer C7v4:
CPU0
4: 0 MIPS 4 19000000.eth 5: 7135 MIPS 5 1a000000.eth 7: 98379 MIPS 7 timer 8: 30 MISC 3 ttyS0 12: 90028 INTC 0 ath9k 13: 5520 INTC 1 ehci_hcd:usb1 14: 4623 INTC 2 ehci_hcd:usb2 15: 32844 AR724X PCI 1 ath10k_pci 16: 0 gpio-ath79 16 keys 23: 0 gpio-ath79 23 keys
$ perf top genirq: Flags mismatch irq 13. 00014c80 (mips_perf_pmu) vs. 00000080 (ehci_hcd:usb1)
This problem is happening, because currently statically assigned virtual IRQ 13 for performance counters is not claimed during the initialization of MIPS PMU during the bootup, so the IRQ subsystem doesn't know, that this interrupt isn't available for further use.
So this patch fixes the issue by simply booking hardware IRQ 5 for MIPS PMU.
--- arch/mips/ath79/setup.c | 6 ------ drivers/irqchip/irq-ath79-misc.c | 11 +++++++++++ 2 files changed, 11 insertions(+), 6 deletions(-)
arch/mips/ath79/setup.c +0 -6
@@ -211,12 +211,6 @@ const char *get_system_type(void)return ath79_sys_type; } -int get_c0_perfcount_int (void ) -{ -return ATH79_MISC_IRQ (5 ); -} -EXPORT_SYMBOL_GPL (get_c0_perfcount_int); -unsigned int get_c0_compare_int (void ) {return CP0_LEGACY_COMPARE_IRQ ;
drivers/irqchip/irq-ath79-misc.c +11 -0
@@ -22,6 +22,15 @@#define AR71XX_RESET_REG_MISC_INT_ENABLE 4#define ATH79_MISC_IRQ_COUNT 32 +#define ATH79_MISC_PERF_IRQ 5 + +static int ath79_perfcount_irq; + +int get_c0_perfcount_int (void ) +{ +return ath79_perfcount_irq; +} +EXPORT_SYMBOL_GPL (get_c0_perfcount_int);static void ath79_misc_irq_handler (struct irq_desc *desc) {
@@ -113,6 +122,8 @@ static void __init ath79_misc_intc_domain_init( {void __iomem *base = domain->host_data; + ath79_perfcount_irq =irq_create_mapping (domain,ATH79_MISC_PERF_IRQ ); +/* Disable and clear all interrupts */ __raw_writel (0 , base +AR71XX_RESET_REG_MISC_INT_ENABLE );__raw_writel (0 , base +AR71XX_RESET_REG_MISC_INT_STATUS );
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 53/81 gpu: ipu-v3: dp: fix CSC handling
In reply to Sasha LevinFrom: Lucas Stach <[email protected]>
[ Upstream commit d4fad0a426c6e26f48c9a7cdd21a7fe9c198d645 ]
Initialize the flow input colorspaces to unknown and reset to that value when the channel gets disabled. This avoids the state getting mixed up with a previous mode.
Also keep the CSC settings for the background flow intact when disabling the foreground flow.
--- drivers/gpu/ipu-v3/ipu-dp.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)
drivers/gpu/ipu-v3/ipu-dp.c +9 -3
@@ -195,7 +195,8 @@ int ipu_dp_setup_channel(struct ipu_dp *dp,ipu_dp_csc_init (flow, flow->foreground.in_cs, flow->out_cs,DP_COM_CONF_CSC_DEF_BOTH ); }else { -if (flow->foreground.in_cs == flow->out_cs) +if (flow->foreground.in_cs ==IPUV3_COLORSPACE_UNKNOWN || + flow->foreground.in_cs == flow->out_cs)/* * foreground identical to output, apply color * conversion on background
@@ -261,6 +262,8 @@ void ipu_dp_disable_channel(struct ipu_dp *dp, bool sync)struct ipu_dp_priv *priv = flow->priv;u32 reg, csc; + dp->in_cs =IPUV3_COLORSPACE_UNKNOWN ; +if (!dp->foreground)return ;
@@ -268,8 +271,9 @@ void ipu_dp_disable_channel(struct ipu_dp *dp, bool sync) reg =readl (flow->base +DP_COM_CONF ); csc = reg &DP_COM_CONF_CSC_DEF_MASK ; -if (csc ==DP_COM_CONF_CSC_DEF_FG ) - reg &= ~DP_COM_CONF_CSC_DEF_MASK ; + reg &= ~DP_COM_CONF_CSC_DEF_MASK ; +if (csc ==DP_COM_CONF_CSC_DEF_BOTH || csc ==DP_COM_CONF_CSC_DEF_BG ) + reg |=DP_COM_CONF_CSC_DEF_BG ; reg &= ~DP_COM_CONF_FG_EN ;writel (reg, flow->base +DP_COM_CONF );
@@ -347,6 +351,8 @@ int ipu_dp_init(struct ipu_soc *ipu, struct device *dev, unsigned long base)mutex_init (&priv->mutex);for (i =0 ; i <IPUV3_NUM_FLOWS ; i++) { + priv->flow[i].background.in_cs =IPUV3_COLORSPACE_UNKNOWN ; + priv->flow[i].foreground.in_cs =IPUV3_COLORSPACE_UNKNOWN ; priv->flow[i].foreground.foreground = true; priv->flow[i].base = priv->base + ipu_dp_flow_base[i]; priv->flow[i].priv = priv;
Signature · 6 lines
2.20.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 54/81 drm/imx: don't skip DP channel disable for background plane
In reply to Sasha LevinFrom: Lucas Stach <[email protected]>
[ Upstream commit 7bcde275eb1d0ac8793c77c7e666a886eb16633d ]
In order to make sure that the plane color space gets reset correctly.
--- drivers/gpu/drm/imx/ipuv3-crtc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/imx/ipuv3-crtc.c +1 -1
@@ -78,7 +78,7 @@ static void ipu_crtc_disable_planes(struct ipu_crtc *ipu_crtc,if (disable_partial)ipu_plane_disable (ipu_crtc->plane[1 ], true);if (disable_full) -ipu_plane_disable (ipu_crtc->plane[0 ], false); +ipu_plane_disable (ipu_crtc->plane[0 ], true); } static void ipu_crtc_atomic_disable(struct drm_crtc *crtc,
Signature · 6 lines
2.20.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 55/81 ARM: 8856/1: NOMMU: Fix CCR register faulty initialization when MPU is disabled
In reply to Sasha LevinFrom: Tigran Tadevosyan <[email protected]>
[ Upstream commit c3143967807adb1357c36b68a7563fc0c4e1f615 ]
When CONFIG_ARM_MPU is not defined, the base address of v7M SCB register is not initialized with correct value. This prevents enabling I/D caches when the L1 cache poilcy is applied in kernel.
--- arch/arm/kernel/head-nommu.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/kernel/head-nommu.S +1 -1
@@ -133,9 +133,9 @@ __secondary_data: */ .text __after_proc_init: -#ifdef CONFIG_ARM_MPU M_CLASS(movw r12, #:lower16:BASEADDR_V7M_SCB) M_CLASS(movt r12, #:upper16:BASEADDR_V7M_SCB) +#ifdef CONFIG_ARM_MPU M_CLASS(ldr r3, [r12, 0x50]) AR_CLASS(mrc p15, 0, r3, c0, c1, 4) @ Read ID_MMFR0 and r3, r3, #(MMFR0_PMSA) @ PMSA field
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 59/81 drm/sun4i: Unbind components before releasing DRM and memory
In reply to Sasha LevinFrom: Paul Kocialkowski <[email protected]>
[ Upstream commit e02bc29b2cfa7806830d6da8b2322cddd67e8dfe ]
Our components may still be using the DRM device driver (if only to access our driver's private data), so make sure to unbind them before the final drm_dev_put.
Also release our reserved memory after component unbind instead of before to match reverse creation order.
--- drivers/gpu/drm/sun4i/sun4i_drv.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/gpu/drm/sun4i/sun4i_drv.c +3 -2
@@ -158,10 +158,11 @@ static void sun4i_drv_unbind(struct device *dev)drm_kms_helper_poll_fini (drm);sun4i_framebuffer_free (drm);drm_mode_config_cleanup (drm); -of_reserved_mem_device_release (dev); -drm_dev_put (drm);component_unbind_all (dev,NULL ); +of_reserved_mem_device_release (dev); + +drm_dev_put (drm); }static const struct component_master_ops sun4i_drv_master_ops = {
Signature · 6 lines
2.20.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 57/81 spi: ST ST95HF NFC: declare missing of table
In reply to Sasha LevinFrom: Daniel Gomez <[email protected]>
[ Upstream commit d04830531d0c4a99c897a44038e5da3d23331d2f ]
Add missing <of_device_id> table for SPI driver relying on SPI device match since compatible is in a DT binding or in a DTS.
Before this patch:
modinfo drivers/nfc/st95hf/st95hf.ko | grep alias
alias: spi:st95hf
After this patch:
modinfo drivers/nfc/st95hf/st95hf.ko | grep alias
alias: spi:st95hf alias: of:N*T*Cst,st95hfC* alias: of:N*T*Cst,st95hf
--- drivers/nfc/st95hf/core.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/nfc/st95hf/core.c +7 -0
@@ -1074,6 +1074,12 @@ static const struct spi_device_id st95hf_id[] = { };MODULE_DEVICE_TABLE (spi, st95hf_id); +static const struct of_device_id st95hf_spi_of_match[] = { + { .compatible ="st,st95hf" }, + { }, +}; +MODULE_DEVICE_TABLE (of, st95hf_spi_of_match); +static int st95hf_probe (struct spi_device *nfc_spi_dev) {int ret;
@@ -1260,6 +1266,7 @@ static struct spi_driver st95hf_driver = { .driver = { .name = "st95hf", .owner = THIS_MODULE, + .of_match_table = of_match_ptr(st95hf_spi_of_match), }, .id_table = st95hf_id, .probe = st95hf_probe,
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 61/81 btrfs: Switch memory allocations in async csum calculation path to kvmalloc
In reply to Sasha LevinFrom: Nikolay Borisov <[email protected]>
[ Upstream commit a3d46aea46f99d134b4e0726e4826b824c3e5980 ]
Recent multi-page biovec rework allowed creation of bios that can span large regions - up to 128 megabytes in the case of btrfs. OTOH btrfs' submission path currently allocates a contiguous array to store the checksums for every bio submitted. This means we can request up to (128mb / BTRFS_SECTOR_SIZE) * 4 bytes + 32bytes of memory from kmalloc. On busy systems with possibly fragmented memory said kmalloc can fail which will trigger BUG_ON due to improper error handling IO submission context in btrfs.
Until error handling is improved or bios in btrfs limited to a more manageable size (e.g. 1m) let's use kvmalloc to fallback to vmalloc for such large allocations. There is no hard requirement that the memory allocated for checksums during IO submission has to be contiguous, but this is a simple fix that does not require several non-contiguous allocations.
For small writes this is unlikely to have any visible effect since kmalloc will still satisfy allocation requests as usual. For larger requests the code will just fallback to vmalloc.
We've performed evaluation on several workload types and there was no significant difference kmalloc vs kvmalloc.
--- fs/btrfs/file-item.c | 15 +++++++++++---- fs/btrfs/ordered-data.c | 3 ++- 2 files changed, 13 insertions(+), 5 deletions(-)
fs/btrfs/file-item.c +11 -4
@@ -7,6 +7,7 @@#include <linux/slab.h> #include <linux/pagemap.h> #include <linux/highmem.h> +#include <linux/sched/mm.h> #include "ctree.h" #include "disk-io.h" #include "transaction.h"
@@ -434,9 +435,13 @@ blk_status_t btrfs_csum_one_bio(struct inode *inode, struct bio *bio,unsigned long this_sum_bytes =0 ;int i;u64 offset; +unsigned nofs_flag; + + nofs_flag =memalloc_nofs_save (); + sums =kvzalloc (btrfs_ordered_sum_size (fs_info, bio->bi_iter.bi_size), +GFP_KERNEL ); +memalloc_nofs_restore (nofs_flag); - sums =kzalloc (btrfs_ordered_sum_size (fs_info, bio->bi_iter.bi_size), -GFP_NOFS );if (!sums)return BLK_STS_RESOURCE ;
@@ -479,8 +484,10 @@ blk_status_t btrfs_csum_one_bio(struct inode *inode, struct bio *bio, bytes_left = bio->bi_iter.bi_size - total_bytes; - sums =kzalloc (btrfs_ordered_sum_size (fs_info, bytes_left), -GFP_NOFS ); + nofs_flag =memalloc_nofs_save (); + sums =kvzalloc (btrfs_ordered_sum_size (fs_info, + bytes_left),GFP_KERNEL ); +memalloc_nofs_restore (nofs_flag);BUG_ON (!sums);/* -ENOMEM */ sums->len = bytes_left; ordered = btrfs_lookup_ordered_extent(inode,
fs/btrfs/ordered-data.c +2 -1
@@ -6,6 +6,7 @@#include <linux/slab.h> #include <linux/blkdev.h> #include <linux/writeback.h> +#include <linux/sched/mm.h> #include "ctree.h" #include "transaction.h" #include "btrfs_inode.h"
@@ -442,7 +443,7 @@ void btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry) cur = entry->list.next; sum = list_entry(cur, struct btrfs_ordered_sum, list);list_del (&sum->list); -kfree (sum); +kvfree (sum); }kmem_cache_free (btrfs_ordered_extent_cache, entry); }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 63/81 mm/memory_hotplug.c: drop memory device reference after find_memory_block()
In reply to Sasha LevinFrom: David Hildenbrand <[email protected]>
[ Upstream commit 89c02e69fc5245f8a2f34b58b42d43a737af1a5e ]
Right now we are using find_memory_block() to get the node id for the pfn range to online. We are missing to drop a reference to the memory block device. While the device still gets unregistered via device_unregister(), resulting in no user visible problem, the device is never released via device_release(), resulting in a memory leak. Fix that by properly using a put_device().
--- mm/memory_hotplug.c | 1 + 1 file changed, 1 insertion(+)
mm/memory_hotplug.c +1 -0
@@ -901,6 +901,7 @@ int __ref online_pages(unsigned long pfn, unsigned long nr_pages, int online_typ*/ mem =find_memory_block (__pfn_to_section (pfn)); nid = mem->nid; +put_device (&mem->dev);/* associate pfn range with the zone */ zone =move_pfn_range (online_type, nid, pfn, nr_pages);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 67/81 mm/memory.c: fix modifying of page protection by insert_pfn()
In reply to Sasha LevinFrom: Jan Kara <[email protected]>
[ Upstream commit cae85cb8add35f678cf487139d05e083ce2f570a ]
Aneesh has reported that PPC triggers the following warning when excercising DAX code:
IP set_pte_at+0x3c/0x190
LR insert_pfn+0x208/0x280
Call Trace:
insert_pfn+0x68/0x280
dax_iomap_pte_fault.isra.7+0x734/0xa40
__xfs_filemap_fault+0x280/0x2d0
do_wp_page+0x48c/0xa40
__handle_mm_fault+0x8d0/0x1fd0
handle_mm_fault+0x140/0x250
__do_page_fault+0x300/0xd60
handle_page_fault+0x18Now that is WARN_ON in set_pte_at which is
VM_WARN_ON(pte_hw_valid(*ptep) && !pte_protnone(*ptep));
The problem is that on some architectures set_pte_at() cannot cope with a situation where there is already some (different) valid entry present.
Use ptep_set_access_flags() instead to modify the pfn which is built to deal with modifying existing PTE.
--- mm/memory.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
mm/memory.c +6 -5
@@ -1796,10 +1796,12 @@ static int insert_pfn(struct vm_area_struct *vma, unsigned long addr,WARN_ON_ONCE (!is_zero_pfn (pte_pfn (*pte))); goto out_unlock; } - entry = *pte; - goto out_mkwrite; - }else - goto out_unlock; + entry =pte_mkyoung (*pte); + entry =maybe_mkwrite (pte_mkdirty (entry), vma); +if (ptep_set_access_flags (vma, addr, pte, entry,1 )) +update_mmu_cache (vma, addr, pte); + } + goto out_unlock; }/* Ok, finally just insert the thing.. */
@@ -1808,7 +1810,6 @@ static int insert_pfn(struct vm_area_struct *vma, unsigned long addr,else entry =pte_mkspecial (pfn_t_pte (pfn, prot)); -out_mkwrite:if (mkwrite) { entry =pte_mkyoung (entry); entry =maybe_mkwrite (pte_mkdirty (entry), vma);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 66/81 net: dsa: mv88e6xxx: fix few issues in mv88e6390x_port_set_cmode
In reply to Sasha LevinFrom: Heiner Kallweit <[email protected]>
[ Upstream commit 5ceaeb99ffb4dc002d20f6ac243c19a85e2c7a76 ]
This patches fixes few issues in mv88e6390x_port_set_cmode().
1. When entering the function the old cmode may be 0, in this case mv88e6390x_serdes_get_lane() returns -ENODEV. As result we bail out and have no chance to set a new mode. Therefore deal properly with -ENODEV.
2. Once we have disabled power and irq, let's set the cached cmode to 0. This reflects the actual status and is cleaner if we bail out with an error in the following function calls.
3. The cached cmode is used by mv88e6390x_serdes_get_lane(), mv88e6390_serdes_power_lane() and mv88e6390_serdes_irq_enable(). Currently we set the cached mode to the new one at the very end of the function only, means until then we use the old one what may be wrong.
4. When calling mv88e6390_serdes_irq_enable() we use the lane value belonging to the old cmode. Get the lane belonging to the new cmode before calling this function.
It's hard to provide a good "Fixes" tag because quite a few smaller changes have been done to the code in question recently.
--- drivers/net/dsa/mv88e6xxx/port.c | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-)
drivers/net/dsa/mv88e6xxx/port.c +16 -8
@@ -379,18 +379,22 @@ int mv88e6390x_port_set_cmode(struct mv88e6xxx_chip *chip, int port,return 0 ; lane =mv88e6390x_serdes_get_lane (chip, port); -if (lane <0 ) +if (lane <0 && lane != -ENODEV )return lane; -if (chip->ports[port].serdes_irq) { - err =mv88e6390_serdes_irq_disable (chip, port, lane); +if (lane >=0 ) { +if (chip->ports[port].serdes_irq) { + err =mv88e6390_serdes_irq_disable (chip, port, lane); +if (err) +return err; + } + + err =mv88e6390x_serdes_power (chip, port, false);if (err)return err; } - err =mv88e6390x_serdes_power (chip, port, false); -if (err) -return err; + chip->ports[port].cmode =0 ;if (cmode) { err =mv88e6xxx_port_read (chip, port,MV88E6XXX_PORT_STS , ®);
@@ -404,6 +408,12 @@ int mv88e6390x_port_set_cmode(struct mv88e6xxx_chip *chip, int port,if (err)return err; + chip->ports[port].cmode = cmode; + + lane =mv88e6390x_serdes_get_lane (chip, port); +if (lane <0 ) +return lane; + err =mv88e6390x_serdes_power (chip, port, true);if (err)return err;
@@ -415,8 +425,6 @@ int mv88e6390x_port_set_cmode(struct mv88e6xxx_chip *chip, int port, } } - chip->ports[port].cmode = cmode; -return 0 ; }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 68/81 usb: typec: Fix unchecked return value
In reply to Sasha LevinFrom: "Gustavo A. R. Silva" <[email protected]>
[ Upstream commit e82adc1074a7356f1158233551df9e86b7ebfb82 ]
Currently there is no check on platform_get_irq() return value in case it fails, hence never actually reporting any errors and causing unexpected behavior when using such value as argument for function regmap_irq_get_virq().
Fix this by adding a proper check, a message error and return *irq* in case platform_get_irq() fails.
Addresses-Coverity-ID: 1443899 ("Improper use of negative value")
--- drivers/usb/typec/typec_wcove.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/usb/typec/typec_wcove.c +7 -2
@@ -615,8 +615,13 @@ static int wcove_typec_probe(struct platform_device *pdev) wcove->dev = &pdev->dev; wcove->regmap = pmic->regmap; - irq =regmap_irq_get_virq (pmic->irq_chip_data_chgr, -platform_get_irq (pdev,0 )); + irq =platform_get_irq (pdev,0 ); +if (irq <0 ) { +dev_err (&pdev->dev,"Failed to get IRQ: %d\n" , irq); +return irq; + } + + irq =regmap_irq_get_virq (pmic->irq_chip_data_chgr, irq);if (irq <0 )return irq;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 69/81 f2fs: fix to data block override node segment by mistake
In reply to Sasha LevinFrom: zhengliang <[email protected]>
[ Upstream commit a0770e13c8da83bdb64738c0209ab02dd3cfff8b ]
v4: Rearrange the previous three versions.
The following scenario could lead to data block override by mistake.
TASK A | TASK kworker | TASK B | TASK C
| | |
open | | |
write | | |
close | | |
| f2fs_write_data_pages | |
| f2fs_write_cache_pages | |
| f2fs_outplace_write_data | |
| f2fs_allocate_data_block (get block in seg S, | |
| S is full, and only | |
| have this valid data | |
| block) | |
| allocate_segment | |
| locate_dirty_segment (mark S as PRE) | |
| f2fs_submit_page_write (submit but is not | |
| written on dev) | |
unlink | | |
iput_final | | |
f2fs_drop_inode | | |
f2fs_truncate | | |
(not evict) | | |
| | write_checkpoint |
| | flush merged bio but not wait file data writeback |
| | set_prefree_as_free (mark S as FREE) |
| | | update NODE/DATA
| | | allocate_segment (select S)
| writeback done | |So we need to guarantee io complete before truncate inode in f2fs_drop_inode.
--- fs/f2fs/super.c | 4 ++++ 1 file changed, 4 insertions(+)
fs/f2fs/super.c +4 -0
@@ -890,6 +890,10 @@ static int f2fs_drop_inode(struct inode *inode)sb_start_intwrite (inode->i_sb);f2fs_i_size_write (inode,0 ); +f2fs_submit_merged_write_cond (F2FS_I_SB (inode), + inode,NULL ,0 ,DATA ); +truncate_inode_pages_final (inode->i_mapping); +if (F2FS_HAS_BLOCKS (inode))f2fs_truncate (inode);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 74/81 mlxsw: core: Do not use WQ_MEM_RECLAIM for EMAD workqueue
In reply to Sasha LevinFrom: Ido Schimmel <[email protected]>
[ Upstream commit a8c133b06183c529c51cd0d54eb57d6b7078370c ]
The EMAD workqueue is used to handle retransmission of EMAD packets that contain configuration data for the device's firmware.
Given the workers need to allocate these packets and that the code is not called as part of memory reclaim path, remove the WQ_MEM_RECLAIM flag.
--- drivers/net/ethernet/mellanox/mlxsw/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlxsw/core.c +1 -1
@@ -568,7 +568,7 @@ static int mlxsw_emad_init(struct mlxsw_core *mlxsw_core)if (!(mlxsw_core->bus->features &MLXSW_BUS_F_TXRX ))return 0 ; - emad_wq =alloc_workqueue ("mlxsw_core_emad" ,WQ_MEM_RECLAIM ,0 ); + emad_wq =alloc_workqueue ("mlxsw_core_emad" ,0 ,0 );if (!emad_wq)return -ENOMEM ; mlxsw_core->emad_wq = emad_wq;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 76/81 mlxsw: core: Do not use WQ_MEM_RECLAIM for mlxsw workqueue
In reply to Sasha LevinFrom: Ido Schimmel <[email protected]>
[ Upstream commit b442fed1b724af0de087912a5718ddde1b87acbb ]
The workqueue is used to periodically update the networking stack about activity / statistics of various objects such as neighbours and TC actions.
It should not be called as part of memory reclaim path, so remove the WQ_MEM_RECLAIM flag.
--- drivers/net/ethernet/mellanox/mlxsw/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlxsw/core.c +1 -1
@@ -1875,7 +1875,7 @@ static int __init mlxsw_core_module_init(void) {int err; - mlxsw_wq =alloc_workqueue (mlxsw_core_driver_name,WQ_MEM_RECLAIM ,0 ); + mlxsw_wq =alloc_workqueue (mlxsw_core_driver_name,0 ,0 );if (!mlxsw_wq)return -ENOMEM ; mlxsw_owq = alloc_ordered_workqueue("%s_ordered", 0,
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 75/81 mlxsw: core: Do not use WQ_MEM_RECLAIM for mlxsw ordered workqueue
In reply to Sasha LevinFrom: Ido Schimmel <[email protected]>
[ Upstream commit 4af0699782e2cc7d0d89db9eb6f8844dd3df82dc ]
The ordered workqueue is used to offload various objects such as routes and neighbours in the order they are notified.
It should not be called as part of memory reclaim path, so remove the WQ_MEM_RECLAIM flag. This can also result in a warning [1], if a worker tries to flush a non-WQ_MEM_RECLAIM workqueue.
[1]
[97703.542861] workqueue: WQ_MEM_RECLAIM mlxsw_core_ordered:mlxsw_sp_router_fib6_event_work [mlxsw_spectrum] is flushing !WQ_MEM_RECLAIM events:rht_deferred_worker [97703.542884] WARNING: CPU: 1 PID: 32492 at kernel/workqueue.c:2605 check_flush_dependency+0xb5/0x130 ... [97703.542988] Hardware name: Mellanox Technologies Ltd. MSN3700C/VMOD0008, BIOS 5.11 10/10/2018 [97703.543049] Workqueue: mlxsw_core_ordered mlxsw_sp_router_fib6_event_work [mlxsw_spectrum] [97703.543061] RIP: 0010:check_flush_dependency+0xb5/0x130 ... [97703.543071] RSP: 0018:ffffb3f08137bc00 EFLAGS: 00010086 [97703.543076] RAX: 0000000000000000 RBX: ffff96e07740ae00 RCX: 0000000000000000 [97703.543080] RDX: 0000000000000094 RSI: ffffffff82dc1934 RDI: 0000000000000046 [97703.543084] RBP: ffffb3f08137bc20 R08: ffffffff82dc18a0 R09: 00000000000225c0 [97703.543087] R10: 0000000000000000 R11: 0000000000007eec R12: ffffffff816e4ee0 [97703.543091] R13: ffff96e06f6a5c00 R14: ffff96e077ba7700 R15: ffffffff812ab0c0 [97703.543097] FS: 0000000000000000(0000) GS:ffff96e077a80000(0000) knlGS:0000000000000000 [97703.543101] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [97703.543104] CR2: 00007f8cd135b280 CR3: 00000001e860e003 CR4: 00000000003606e0 [97703.543109] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [97703.543112] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [97703.543115] Call Trace: [97703.543129] __flush_work+0xbd/0x1e0 [97703.543137] ? __cancel_work_timer+0x136/0x1b0 [97703.543145] ? pwq_dec_nr_in_flight+0x49/0xa0 [97703.543154] __cancel_work_timer+0x136/0x1b0 [97703.543175] ? mlxsw_reg_trans_bulk_wait+0x145/0x400 [mlxsw_core] [97703.543184] cancel_work_sync+0x10/0x20 [97703.543191] rhashtable_free_and_destroy+0x23/0x140 [97703.543198] rhashtable_destroy+0xd/0x10 [97703.543254] mlxsw_sp_fib_destroy+0xb1/0xf0 [mlxsw_spectrum] [97703.543310] mlxsw_sp_vr_put+0xa8/0xc0 [mlxsw_spectrum] [97703.543364] mlxsw_sp_fib_node_put+0xbf/0x140 [mlxsw_spectrum] [97703.543418] ? mlxsw_sp_fib6_entry_destroy+0xe8/0x110 [mlxsw_spectrum] [97703.543475] mlxsw_sp_router_fib6_event_work+0x6cd/0x7f0 [mlxsw_spectrum] [97703.543484] process_one_work+0x1fd/0x400 [97703.543493] worker_thread+0x34/0x410 [97703.543500] kthread+0x121/0x140 [97703.543507] ? process_one_work+0x400/0x400 [97703.543512] ? kthread_park+0x90/0x90 [97703.543523] ret_from_fork+0x35/0x40
--- drivers/net/ethernet/mellanox/mlxsw/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlxsw/core.c +1 -1
@@ -1878,7 +1878,7 @@ static int __init mlxsw_core_module_init(void) mlxsw_wq =alloc_workqueue (mlxsw_core_driver_name,WQ_MEM_RECLAIM ,0 );if (!mlxsw_wq)return -ENOMEM ; - mlxsw_owq =alloc_ordered_workqueue ("%s_ordered" ,WQ_MEM_RECLAIM , + mlxsw_owq =alloc_ordered_workqueue ("%s_ordered" ,0 , mlxsw_core_driver_name);if (!mlxsw_owq) { err = -ENOMEM ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 78/81 net: strparser: partially revert "strparser: Call skb_unclone conditionally"
In reply to Sasha LevinFrom: Jakub Kicinski <[email protected]>
[ Upstream commit 4a9c2e3746e6151fd5d077259d79ce9ca86d47d7 ]
This reverts the first part of commit 4e485d06bb8c ("strparser: Call skb_unclone conditionally"). To build a message with multiple fragments we need our own root of frag_list. We can't simply use the frag_list of orig_skb, because it will lead to linking all orig_skbs together creating very long frag chains, and causing stack overflow on kfree_skb() (which is called recursively on the frag_lists).
BUG: stack guard page was hit at 00000000d40fad41 (stack is 0000000029dde9f4..000000008cce03d5) kernel stack overflow (double-fault): 0000 [#1] PREEMPT SMP RIP: 0010:free_one_page+0x2b/0x490 Call Trace: __free_pages_ok+0x143/0x2c0 skb_release_data+0x8e/0x140 ? skb_release_data+0xad/0x140 kfree_skb+0x32/0xb0
[...]
skb_release_data+0xad/0x140 ? skb_release_data+0xad/0x140 kfree_skb+0x32/0xb0 skb_release_data+0xad/0x140 ? skb_release_data+0xad/0x140 kfree_skb+0x32/0xb0 skb_release_data+0xad/0x140 ? skb_release_data+0xad/0x140 kfree_skb+0x32/0xb0 skb_release_data+0xad/0x140 ? skb_release_data+0xad/0x140 kfree_skb+0x32/0xb0 skb_release_data+0xad/0x140 __kfree_skb+0xe/0x20 tcp_disconnect+0xd6/0x4d0 tcp_close+0xf4/0x430 ? tcp_check_oom+0xf0/0xf0 tls_sk_proto_close+0xe4/0x1e0 [tls] inet_release+0x36/0x60 __sock_release+0x37/0xa0 sock_close+0x11/0x20 __fput+0xa2/0x1d0 task_work_run+0x89/0xb0 exit_to_usermode_loop+0x9a/0xa0 do_syscall_64+0xc0/0xf0 entry_SYSCALL_64_after_hwframe+0x44/0xa9
Let's leave the second unclone conditional, as I'm not entirely sure what is its purpose :)
--- net/strparser/strparser.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-)
net/strparser/strparser.c +5 -7
@@ -140,13 +140,11 @@ static int __strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,/* We are going to append to the frags_list of head. * Need to unshare the frag_list. */ -if (skb_has_frag_list (head)) { - err =skb_unclone (head,GFP_ATOMIC ); -if (err) { -STRP_STATS_INCR (strp->stats.mem_fail); - desc->error = err; -return 0 ; - } + err =skb_unclone (head,GFP_ATOMIC ); +if (err) { +STRP_STATS_INCR (strp->stats.mem_fail); + desc->error = err; +return 0 ; }if (unlikely (skb_shinfo (head)->frag_list)) {
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 79/81 NFC: nci: Add some bounds checking in nci_hci_cmd_received()
In reply to Sasha LevinFrom: Dan Carpenter <[email protected]>
[ Upstream commit d7ee81ad09f072eab1681877fc71ec05f9c1ae92 ]
This is similar to commit 674d9de02aa7 ("NFC: Fix possible memory corruption when handling SHDLC I-Frame commands").
I'm not totally sure, but I think that commit description may have overstated the danger. I was under the impression that this data came from the firmware? If you can't trust your networking firmware, then you're already in trouble.
Anyway, these days we add bounds checking where ever we can and we call it kernel hardening. Better safe than sorry.
--- net/nfc/nci/hci.c | 8 ++++++++ 1 file changed, 8 insertions(+)
net/nfc/nci/hci.c +8 -0
@@ -312,6 +312,10 @@ static void nci_hci_cmd_received(struct nci_dev *ndev, u8 pipe, create_info = (struct nci_hci_create_pipe_resp *)skb->data; dest_gate = create_info->dest_gate; new_pipe = create_info->pipe; +if (new_pipe >=NCI_HCI_MAX_PIPES ) { + status =NCI_HCI_ANY_E_NOK ; + goto exit; + }/* Save the new created pipe and bind with local gate, * the description for skb->data[3] is destination gate id
@@ -336,6 +340,10 @@ static void nci_hci_cmd_received(struct nci_dev *ndev, u8 pipe, goto exit; } delete_info = (struct nci_hci_delete_pipe_noti *)skb->data; +if (delete_info->pipe >=NCI_HCI_MAX_PIPES ) { + status =NCI_HCI_ANY_E_NOK ; + goto exit; + } ndev->hci_dev->pipes[delete_info->pipe].gate =NCI_HCI_INVALID_GATE ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 80/81 nfc: nci: Potential off by one in ->pipes[] array
In reply to Sasha LevinFrom: Dan Carpenter <[email protected]>
[ Upstream commit 6491d698396fd5da4941980a35ca7c162a672016 ]
This is similar to commit e285d5bfb7e9 ("NFC: Fix the number of pipes") where we changed NFC_HCI_MAX_PIPES from 127 to 128.
As the comment next to the define explains, the pipe identifier is 7 bits long. The highest possible pipe is 127, but the number of possible pipes is 128. As the code is now, then there is potential for an out of bounds array access:
net/nfc/nci/hci.c:297 nci_hci_cmd_received() warn: array off by one? 'ndev->hci_dev->pipes[pipe]' '0-127 == 127'
--- include/net/nfc/nci_core.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/net/nfc/nci_core.h +1 -1
@@ -166,7 +166,7 @@ struct nci_conn_info {* According to specification 102 622 chapter 4.4 Pipes, * the pipe identifier is 7 bits long. */ -#define NCI_HCI_MAX_PIPES 127 +#define NCI_HCI_MAX_PIPES 128struct nci_hci_gate {u8 gate;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 64/81 powerpc/smp: Fix NMI IPI timeout
In reply to Sasha LevinFrom: Nicholas Piggin <[email protected]>
[ Upstream commit 1b5fc84aba170bdfe3533396ca9662ceea1609b7 ]
The NMI IPI timeout logic is broken, if __smp_send_nmi_ipi() times out on the first condition, delay_us will be zero which will send it into the second spin loop with no timeout so it will spin forever.
--- arch/powerpc/kernel/smp.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
arch/powerpc/kernel/smp.c +3 -2
@@ -499,7 +499,7 @@ int __smp_send_nmi_ipi(int cpu, void (*fn)(struct pt_regs *), u64 delay_us, boolif (delay_us) { delay_us--;if (!delay_us) -break ; + goto timeout; } }
@@ -510,10 +510,11 @@ int __smp_send_nmi_ipi(int cpu, void (*fn)(struct pt_regs *), u64 delay_us, boolif (delay_us) { delay_us--;if (!delay_us) -break ; + goto timeout; } } +timeout:if (!cpumask_empty (&nmi_ipi_pending_mask)) {/* Timeout waiting for CPUs to call smp_handle_nmi_ipi */ ret =0 ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 65/81 powerpc/smp: Fix NMI IPI xmon timeout
In reply to Sasha LevinFrom: Nicholas Piggin <[email protected]>
[ Upstream commit 88b9a3d1425a436e95c41f09986fdae2daee437a ]
The xmon debugger IPI handler waits in the callback function while xmon is still active. This means they don't complete the IPI, and the initiator always times out waiting for them.
Things manage to work after the timeout because there is some fallback logic to keep NMI IPI state sane in case of the timeout, but this is a bit ugly.
This patch changes NMI IPI back to half-asynchronous (i.e., wait for everyone to call in, do not wait for IPI function to complete), but the complexity is avoided by going one step further and allowing new IPIs to be issued before the IPI functions to all complete.
If synchronization against that is required, it is left up to the caller, but current callers don't require that. In fact with the timeout handling, callers must be able to cope with this already.
--- arch/powerpc/kernel/smp.c | 93 ++++++++++++--------------------------- 1 file changed, 29 insertions(+), 64 deletions(-)
arch/powerpc/kernel/smp.c +29 -64
@@ -338,13 +338,12 @@ void arch_send_call_function_ipi_mask(const struct cpumask *mask)* NMI IPIs may not be recoverable, so should not be used as ongoing part of * a running system. They can be used for crash, debug, halt/reboot, etc. * -* NMI IPIs are globally single threaded. No more than one in progress at -* any time. -* * The IPI call waits with interrupts disabled until all targets enter the -* NMI handler, then the call returns. +* NMI handler, then returns. Subsequent IPIs can be issued before targets +* have returned from their handlers, so there is no guarantee about +* concurrency or re-entrancy. * -* No new NMI can be initiated until targets exit the handler. +* A new NMI can be issued before all targets exit the handler. * * The IPI call may time out without all targets entering the NMI handler. * In that case, there is some logic to recover (and ignore subsequent
@@ -355,7 +354,7 @@ void arch_send_call_function_ipi_mask(const struct cpumask *mask)static atomic_t __nmi_ipi_lock =ATOMIC_INIT (0 );static struct cpumask nmi_ipi_pending_mask; -static int nmi_ipi_busy_count =0 ; +static bool nmi_ipi_busy = false;static void (*nmi_ipi_function)(struct pt_regs *) =NULL ;static void nmi_ipi_lock_start (unsigned long *flags)
@@ -394,7 +393,7 @@ static void nmi_ipi_unlock_end(unsigned long *flags)*/ int smp_handle_nmi_ipi (struct pt_regs *regs) { -void (*fn)(struct pt_regs *); +void (*fn)(struct pt_regs *) =NULL ;unsigned long flags;int me =raw_smp_processor_id ();int ret =0 ;
@@ -405,29 +404,17 @@ int smp_handle_nmi_ipi(struct pt_regs *regs)* because the caller may have timed out. */ nmi_ipi_lock_start (&flags); -if (!nmi_ipi_busy_count) - goto out; -if (!cpumask_test_cpu (me, &nmi_ipi_pending_mask)) - goto out; - - fn = nmi_ipi_function; -if (!fn) - goto out; - -cpumask_clear_cpu (me, &nmi_ipi_pending_mask); - nmi_ipi_busy_count++; -nmi_ipi_unlock (); - - ret =1 ; - -fn (regs); - -nmi_ipi_lock (); -if (nmi_ipi_busy_count >1 )/* Can race with caller time-out */ - nmi_ipi_busy_count--; -out: +if (cpumask_test_cpu (me, &nmi_ipi_pending_mask)) { +cpumask_clear_cpu (me, &nmi_ipi_pending_mask); + fn =READ_ONCE (nmi_ipi_function); +WARN_ON_ONCE (!fn); + ret =1 ; + }nmi_ipi_unlock_end (&flags); +if (fn) +fn (regs); +return ret; }
@@ -453,7 +440,7 @@ static void do_smp_send_nmi_ipi(int cpu, bool safe)* - cpu is the target CPU (must not be this CPU), or NMI_IPI_ALL_OTHERS. * - fn is the target callback function. * - delay_us > 0 is the delay before giving up waiting for targets to -* complete executing the handler, == 0 specifies indefinite delay. +* begin executing the handler, == 0 specifies indefinite delay. */ int __smp_send_nmi_ipi (int cpu,void (*fn)(struct pt_regs *),u64 delay_us,bool safe) {
@@ -467,31 +454,33 @@ int __smp_send_nmi_ipi(int cpu, void (*fn)(struct pt_regs *), u64 delay_us, boolif (unlikely (!smp_ops))return 0 ; -/* Take the nmi_ipi_busy count/lock with interrupts hard disabled */ nmi_ipi_lock_start (&flags); -while (nmi_ipi_busy_count) { +while (nmi_ipi_busy) {nmi_ipi_unlock_end (&flags); -spin_until_cond (nmi_ipi_busy_count ==0 ); +spin_until_cond (!nmi_ipi_busy);nmi_ipi_lock_start (&flags); } - + nmi_ipi_busy = true; nmi_ipi_function = fn; +WARN_ON_ONCE (!cpumask_empty (&nmi_ipi_pending_mask)); +if (cpu <0 ) {/* ALL_OTHERS */ cpumask_copy (&nmi_ipi_pending_mask, cpu_online_mask);cpumask_clear_cpu (me, &nmi_ipi_pending_mask); }else { -/* cpumask starts clear */ cpumask_set_cpu (cpu, &nmi_ipi_pending_mask); } - nmi_ipi_busy_count++; +nmi_ipi_unlock (); +/* Interrupts remain hard disabled */ +do_smp_send_nmi_ipi (cpu, safe);nmi_ipi_lock (); -/* nmi_ipi_busy_count is held here, so unlock/lock is okay */ +/* nmi_ipi_busy is set here, so unlock/lock is okay */ while (!cpumask_empty (&nmi_ipi_pending_mask)) {nmi_ipi_unlock ();udelay (1 );
@@ -499,34 +488,19 @@ int __smp_send_nmi_ipi(int cpu, void (*fn)(struct pt_regs *), u64 delay_us, boolif (delay_us) { delay_us--;if (!delay_us) - goto timeout; +break ; } } -while (nmi_ipi_busy_count >1 ) { -nmi_ipi_unlock (); -udelay (1 ); -nmi_ipi_lock (); -if (delay_us) { - delay_us--; -if (!delay_us) - goto timeout; - } - } - -timeout:if (!cpumask_empty (&nmi_ipi_pending_mask)) {/* Timeout waiting for CPUs to call smp_handle_nmi_ipi */ ret =0 ;cpumask_clear (&nmi_ipi_pending_mask); } -if (nmi_ipi_busy_count >1 ) { -/* Timeout waiting for CPUs to execute fn */ - ret =0 ; - nmi_ipi_busy_count =1 ; - } - nmi_ipi_busy_count--; + nmi_ipi_function =NULL ; + nmi_ipi_busy = false; +nmi_ipi_unlock_end (&flags);return ret;
@@ -594,17 +568,8 @@ void crash_send_ipi(void (*crash_ipi_callback)(struct pt_regs *))static void nmi_stop_this_cpu (struct pt_regs *regs) {/* -* This is a special case because it never returns, so the NMI IPI -* handling would never mark it as done, which makes any later -* smp_send_nmi_ipi() call spin forever. Mark it done now. -* * IRQs are already hard disabled by the smp_handle_nmi_ipi. */ -nmi_ipi_lock (); -if (nmi_ipi_busy_count >1 ) - nmi_ipi_busy_count--; -nmi_ipi_unlock (); -spin_begin ();while (1 )spin_cpu_relax ();
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 81/81 x86/kprobes: Avoid kretprobe recursion bug
In reply to Sasha LevinFrom: Masami Hiramatsu <[email protected]>
[ Upstream commit b191fa96ea6dc00d331dcc28c1f7db5e075693a0 ]
Avoid kretprobe recursion loop bg by setting a dummy kprobes to current_kprobe per-CPU variable.
This bug has been introduced with the asm-coded trampoline code, since previously it used another kprobe for hooking the function return placeholder (which only has a nop) and trampoline handler was called from that kprobe.
This revives the old lost kprobe again.
With this fix, we don't see deadlock anymore.
And you can see that all inner-called kretprobe are skipped.
event_1 235 0
event_2 19375 19612
The 1st column is recorded count and the 2nd is missed count. Above shows (event_1 rec) + (event_2 rec) ~= (event_2 missed) (some difference are here because the counter is racy)
--- arch/x86/kernel/kprobes/core.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-)
arch/x86/kernel/kprobes/core.c +20 -2
@@ -749,11 +749,16 @@ asm(NOKPROBE_SYMBOL (kretprobe_trampoline);STACK_FRAME_NON_STANDARD (kretprobe_trampoline); +static struct kprobe kretprobe_kprobe = { + .addr = (void *)kretprobe_trampoline, +}; +/* * Called from kretprobe_trampoline */ __visible __used void *trampoline_handler(struct pt_regs *regs) { +struct kprobe_ctlblk *kcb;struct kretprobe_instance *ri =NULL ;struct hlist_head *head, empty_rp;struct hlist_node *tmp;
@@ -763,6 +768,17 @@ __visible __used void *trampoline_handler(struct pt_regs *regs)void *frame_pointer;bool skipped = false; +preempt_disable (); + +/* +* Set a dummy kprobe for avoiding kretprobe recursion. +* Since kretprobe never run in kprobe handler, kprobe must not +* be running at this point. +*/ + kcb =get_kprobe_ctlblk (); +__this_cpu_write (current_kprobe, &kretprobe_kprobe); + kcb->kprobe_status =KPROBE_HIT_ACTIVE ; +INIT_HLIST_HEAD (&empty_rp);kretprobe_hash_lock (current, &head, &flags);/* fixup registers */
@@ -838,10 +854,9 @@ __visible __used void *trampoline_handler(struct pt_regs *regs) orig_ret_address = (unsigned long )ri->ret_addr;if (ri->rp && ri->rp->handler) {__this_cpu_write (current_kprobe, &ri->rp->kp); -get_kprobe_ctlblk ()->kprobe_status =KPROBE_HIT_ACTIVE ; ri->ret_addr = correct_ret_addr; ri->rp->handler (ri, regs); -__this_cpu_write (current_kprobe,NULL ); +__this_cpu_write (current_kprobe, &kretprobe_kprobe); }recycle_rp_inst (ri, &empty_rp);
@@ -857,6 +872,9 @@ __visible __used void *trampoline_handler(struct pt_regs *regs)kretprobe_hash_unlock (current, &flags); +__this_cpu_write (current_kprobe,NULL ); +preempt_enable (); +hlist_for_each_entry_safe (ri, tmp, &empty_rp, hlist) {hlist_del (&ri->hlist);kfree (ri);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 77/81 net/tls: fix the IV leaks
In reply to Sasha LevinFrom: Jakub Kicinski <[email protected]>
[ Upstream commit 5a03bc73abed6ae196c15e9950afde19d48be12c ]
Commit f66de3ee2c16 ("net/tls: Split conf to rx + tx") made freeing of IV and record sequence number conditional to SW path only, but commit e8f69799810c ("net/tls: Add generic NIC offload infrastructure") also allocates that state for the device offload configuration. Remember to free it.
--- net/tls/tls_device.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
net/tls/tls_device.c +4 -1
@@ -52,8 +52,11 @@ static DEFINE_SPINLOCK(tls_device_lock);static void tls_device_free_ctx (struct tls_context *ctx) { -if (ctx->tx_conf ==TLS_HW ) +if (ctx->tx_conf ==TLS_HW ) {kfree (tls_offload_ctx_tx (ctx)); +kfree (ctx->tx.rec_seq); +kfree (ctx->tx.iv); + }if (ctx->rx_conf ==TLS_HW )kfree (tls_offload_ctx_rx (ctx));
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 70/81 netfilter: nf_tables: use-after-free in dynamic operations
In reply to Sasha LevinFrom: Pablo Neira Ayuso <[email protected]>
[ Upstream commit 3f3a390dbd59d236f62cff8e8b20355ef7069e3d ]
Smatch reports:
net/netfilter/nf_tables_api.c:2167 nf_tables_expr_destroy()
error: dereferencing freed memory 'expr->ops'net/netfilter/nf_tables_api.c
2162 static void nf_tables_expr_destroy(const struct nft_ctx *ctx,
2163 struct nft_expr *expr)
2164 {
2165 if (expr->ops->destroy)
2166 expr->ops->destroy(ctx, expr);
^^^^
--> 2167 module_put(expr->ops->type->owner);
^^^^^^^^^
2168 }Smatch says there are three functions which free expr->ops.
--- net/netfilter/nf_tables_api.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
net/netfilter/nf_tables_api.c +3 -1
@@ -2113,9 +2113,11 @@ static int nf_tables_newexpr(const struct nft_ctx *ctx,static void nf_tables_expr_destroy (const struct nft_ctx *ctx,struct nft_expr *expr) { +const struct nft_expr_type *type = expr->ops->type; +if (expr->ops->destroy) expr->ops->destroy (ctx, expr); -module_put (expr->ops->type->owner); +module_put (type->owner); } struct nft_expr *nft_expr_init(const struct nft_ctx *ctx,
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 73/81 mlxsw: spectrum_switchdev: Add MDB entries in prepare phase
In reply to Sasha LevinFrom: Ido Schimmel <[email protected]>
[ Upstream commit d4d0e40977ac450f32f2db5e4d8e23c9d2578899 ]
The driver cannot guarantee in the prepare phase that it will be able to write an MDB entry to the device. In case the driver returned success during the prepare phase, but then failed to add the entry in the commit phase, a WARNING [1] will be generated by the switchdev core.
Fix this by doing the work in the prepare phase instead.
[1]
[ 358.544486] swp12s0: Commit of object (id=2) failed. [ 358.550061] WARNING: CPU: 0 PID: 30 at net/switchdev/switchdev.c:281 switchdev_port_obj_add_now+0x9b/0xe0 [ 358.560754] CPU: 0 PID: 30 Comm: kworker/0:1 Not tainted 5.0.0-custom-13382-gf2449babf221 #1350 [ 358.570472] Hardware name: Mellanox Technologies Ltd. MSN2100-CB2FO/SA001017, BIOS 5.6.5 06/07/2016 [ 358.580582] Workqueue: events switchdev_deferred_process_work [ 358.587001] RIP: 0010:switchdev_port_obj_add_now+0x9b/0xe0 ... [ 358.614109] RSP: 0018:ffffa6b900d6fe18 EFLAGS: 00010286 [ 358.619943] RAX: 0000000000000000 RBX: ffff8b00797ff000 RCX: 0000000000000000 [ 358.627912] RDX: ffff8b00b7a1d4c0 RSI: ffff8b00b7a152e8 RDI: ffff8b00b7a152e8 [ 358.635881] RBP: ffff8b005c3f5bc0 R08: 000000000000022b R09: 0000000000000000 [ 358.643850] R10: 0000000000000000 R11: ffffa6b900d6fcc8 R12: 0000000000000000 [ 358.651819] R13: dead000000000100 R14: ffff8b00b65a23c0 R15: 0ffff8b00b7a2200 [ 358.659790] FS: 0000000000000000(0000) GS:ffff8b00b7a00000(0000) knlGS:0000000000000000 [ 358.668820] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 358.675228] CR2: 00007f00aad90de0 CR3: 00000001ca80d000 CR4: 00000000001006f0 [ 358.683188] Call Trace: [ 358.685918] switchdev_port_obj_add_deferred+0x13/0x60 [ 358.691655] switchdev_deferred_process+0x6b/0xf0 [ 358.696907] switchdev_deferred_process_work+0xa/0x10 [ 358.702548] process_one_work+0x1f5/0x3f0 [ 358.707022] worker_thread+0x28/0x3c0 [ 358.711099] ? process_one_work+0x3f0/0x3f0 [ 358.715768] kthread+0x10d/0x130 [ 358.719369] ? __kthread_create_on_node+0x180/0x180 [ 358.724815] ret_from_fork+0x35/0x40
--- drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +1 -1
@@ -1585,7 +1585,7 @@ static int mlxsw_sp_port_mdb_add(struct mlxsw_sp_port *mlxsw_sp_port,u16 fid_index;int err =0 ; -if (switchdev_trans_ph_prepare (trans)) +if (switchdev_trans_ph_commit (trans))return 0 ; bridge_port =mlxsw_sp_bridge_port_find (mlxsw_sp->bridge, orig_dev);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 72/81 net: fec: manage ahb clock in runtime pm
In reply to Sasha LevinFrom: Andy Duan <[email protected]>
[ Upstream commit d7c3a206e6338e4ccdf030719dec028e26a521d5 ]
Some SOC like i.MX6SX clock have some limits:
- ahb clock should be disabled before ipg.
- ahb and ipg clocks are required for MAC MII bus. So, move the ahb clock to runtime management together with ipg clock.
--- drivers/net/ethernet/freescale/fec_main.c | 30 ++++++++++++++++------- 1 file changed, 21 insertions(+), 9 deletions(-)
drivers/net/ethernet/freescale/fec_main.c +21 -9
@@ -1850,13 +1850,9 @@ static int fec_enet_clk_enable(struct net_device *ndev, bool enable)int ret;if (enable) { - ret =clk_prepare_enable (fep->clk_ahb); -if (ret) -return ret; - ret =clk_prepare_enable (fep->clk_enet_out);if (ret) - goto failed_clk_enet_out; +return ret;if (fep->clk_ptp) {mutex_lock (&fep->ptp_clk_mutex);
@@ -1876,7 +1872,6 @@ static int fec_enet_clk_enable(struct net_device *ndev, bool enable)phy_reset_after_clk_enable (ndev->phydev); }else { -clk_disable_unprepare (fep->clk_ahb);clk_disable_unprepare (fep->clk_enet_out);if (fep->clk_ptp) {mutex_lock (&fep->ptp_clk_mutex);
@@ -1895,8 +1890,6 @@ static int fec_enet_clk_enable(struct net_device *ndev, bool enable) failed_clk_ptp:if (fep->clk_enet_out)clk_disable_unprepare (fep->clk_enet_out); -failed_clk_enet_out: -clk_disable_unprepare (fep->clk_ahb);return ret; }
@@ -3485,6 +3478,9 @@ fec_probe(struct platform_device *pdev) ret =clk_prepare_enable (fep->clk_ipg);if (ret) goto failed_clk_ipg; + ret =clk_prepare_enable (fep->clk_ahb); +if (ret) + goto failed_clk_ahb; fep->reg_phy =devm_regulator_get (&pdev->dev,"phy" );if (!IS_ERR (fep->reg_phy)) {
@@ -3578,6 +3574,9 @@ fec_probe(struct platform_device *pdev)pm_runtime_put (&pdev->dev);pm_runtime_disable (&pdev->dev); failed_regulator: +clk_disable_unprepare (fep->clk_ahb); +failed_clk_ahb: +clk_disable_unprepare (fep->clk_ipg); failed_clk_ipg:fec_enet_clk_enable (ndev, false); failed_clk:
@@ -3701,6 +3700,7 @@ static int __maybe_unused fec_runtime_suspend(struct device *dev)struct net_device *ndev =dev_get_drvdata (dev);struct fec_enet_private *fep =netdev_priv (ndev); +clk_disable_unprepare (fep->clk_ahb);clk_disable_unprepare (fep->clk_ipg);return 0 ;
@@ -3710,8 +3710,20 @@ static int __maybe_unused fec_runtime_resume(struct device *dev) {struct net_device *ndev =dev_get_drvdata (dev);struct fec_enet_private *fep =netdev_priv (ndev); +int ret; -return clk_prepare_enable (fep->clk_ipg); + ret =clk_prepare_enable (fep->clk_ahb); +if (ret) +return ret; + ret =clk_prepare_enable (fep->clk_ipg); +if (ret) + goto failed_clk_ipg; + +return 0 ; + +failed_clk_ipg: +clk_disable_unprepare (fep->clk_ahb); +return ret; }static const struct dev_pm_ops fec_pm_ops = {
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 71/81 netfilter: nf_tables: add missing ->release_ops() in error path of newrule()
In reply to Sasha LevinFrom: Taehee Yoo <[email protected]>
[ Upstream commit b25a31bf0ca091aa8bdb9ab329b0226257568bbe ]
->release_ops() callback releases resources and this is used in error path. If nf_tables_newrule() fails after ->select_ops(), it should release resources. but it can not call ->destroy() because that should be called after ->init().
At this point, ->release_ops() should be used for releasing resources.
Test commands:
modprobe -rv xt_tcpudp
iptables-nft -I INPUT -m tcp <-- error command
lsmod
Result:
Module Size Used by
xt_tcpudp 20480 2 <-- it should be 0
--- net/netfilter/nf_tables_api.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
net/netfilter/nf_tables_api.c +4 -1
@@ -2719,8 +2719,11 @@ static int nf_tables_newrule(struct net *net, struct sock *nlsk,nf_tables_rule_release (&ctx, rule); err1:for (i =0 ; i < n; i++) { -if (info[i].ops !=NULL ) +if (info[i].ops) {module_put (info[i].ops->type->owner); +if (info[i].ops->type->release_ops) + info[i].ops->type->release_ops (info[i].ops); + } }kvfree (info);return err;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 58/81 x86/mm: Fix a crash with kmemleak_scan()
In reply to Sasha LevinFrom: Qian Cai <[email protected]>
[ Upstream commit 0d02113b31b2017dd349ec9df2314e798a90fa6e ]
The first kmemleak_scan() call after boot would trigger the crash below because this callpath:
kernel_init
free_initmem
mem_encrypt_free_decrypted_mem
free_init_pages
unmaps memory inside the .bss when DEBUG_PAGEALLOC=y.
kmemleak_init() will register the .data/.bss sections and then kmemleak_scan() will scan those addresses and dereference them looking for pointer references. If free_init_pages() frees and unmaps pages in those sections, kmemleak_scan() will crash if referencing one of those addresses:
BUG: unable to handle kernel paging request at ffffffffbd402000 CPU: 12 PID: 325 Comm: kmemleak Not tainted 5.1.0-rc4+ #4 RIP: 0010:scan_block Call Trace: scan_gray_list kmemleak_scan kmemleak_scan_thread kthread ret_from_fork
Since kmemleak_free_part() is tolerant to unknown objects (not tracked by kmemleak), it is fine to call it from free_init_pages() even if not all address ranges passed to this function are known to kmemleak.
[ bp: Massage. ]
--- arch/x86/mm/init.c | 6 ++++++ 1 file changed, 6 insertions(+)
arch/x86/mm/init.c +6 -0
@@ -6,6 +6,7 @@#include <linux/bootmem.h> /* for max_low_pfn */ #include <linux/swapfile.h> #include <linux/swapops.h> +#include <linux/kmemleak.h> #include <asm/set_memory.h> #include <asm/e820/api.h>
@@ -767,6 +768,11 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)if (debug_pagealloc_enabled ()) {pr_info ("debug: unmapping init [mem %#010lx-%#010lx]\n" , begin, end -1 ); +/* +* Inform kmemleak about the hole in the memory since the +* corresponding pages will be unmapped. +*/ +kmemleak_free_part ((void *)begin, end - begin);set_memory_np (begin, (end - begin) >>PAGE_SHIFT ); }else {/*
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 56/81 spi: Micrel eth switch: declare missing of table
In reply to Sasha LevinFrom: Daniel Gomez <[email protected]>
[ Upstream commit 2f23a2a768bee7ad2ff1e9527c3f7e279e794a46 ]
Add missing <of_device_id> table for SPI driver relying on SPI device match since compatible is in a DT binding or in a DTS.
Before this patch:
modinfo drivers/net/phy/spi_ks8995.ko | grep alias
alias: spi:ksz8795 alias: spi:ksz8864 alias: spi:ks8995
After this patch:
modinfo drivers/net/phy/spi_ks8995.ko | grep alias
alias: spi:ksz8795 alias: spi:ksz8864 alias: spi:ks8995 alias: of:N*T*Cmicrel,ksz8795C* alias: of:N*T*Cmicrel,ksz8795 alias: of:N*T*Cmicrel,ksz8864C* alias: of:N*T*Cmicrel,ksz8864 alias: of:N*T*Cmicrel,ks8995C* alias: of:N*T*Cmicrel,ks8995
--- drivers/net/phy/spi_ks8995.c | 9 +++++++++ 1 file changed, 9 insertions(+)
drivers/net/phy/spi_ks8995.c +9 -0
@@ -162,6 +162,14 @@ static const struct spi_device_id ks8995_id[] = { };MODULE_DEVICE_TABLE (spi, ks8995_id); +static const struct of_device_id ks8895_spi_of_match[] = { + { .compatible ="micrel,ks8995" }, + { .compatible ="micrel,ksz8864" }, + { .compatible ="micrel,ksz8795" }, + { }, + }; +MODULE_DEVICE_TABLE (of, ks8895_spi_of_match); +static inline u8 get_chip_id (u8 val) {return (val >>ID1_CHIPID_S ) &ID1_CHIPID_M ;
@@ -529,6 +537,7 @@ static int ks8995_remove(struct spi_device *spi)static struct spi_driver ks8995_driver = { .driver = { .name ="spi-ks8995" , + .of_match_table =of_match_ptr (ks8895_spi_of_match), }, .probe = ks8995_probe, .remove = ks8995_remove,
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 52/81 netfilter: fix nf_l4proto_log_invalid to log invalid packets
In reply to Sasha LevinFrom: Andrei Vagin <[email protected]>
[ Upstream commit d48668052b2603b6262459625c86108c493588dd ]
It doesn't log a packet if sysctl_log_invalid isn't equal to protonum OR sysctl_log_invalid isn't equal to IPPROTO_RAW. This sentence is always true. I believe we need to replace OR to AND.
--- net/netfilter/nf_conntrack_proto.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/netfilter/nf_conntrack_proto.c +1 -1
@@ -86,7 +86,7 @@ void nf_l4proto_log_invalid(const struct sk_buff *skb,struct va_format vaf;va_list args; -if (net->ct.sysctl_log_invalid != protonum || +if (net->ct.sysctl_log_invalid != protonum && net->ct.sysctl_log_invalid !=IPPROTO_RAW )return ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 46/81 netfilter: nf_tables: prevent shift wrap in nft_chain_parse_hook()
In reply to Sasha LevinFrom: Dan Carpenter <[email protected]>
[ Upstream commit 33d1c018179d0a30c39cc5f1682b77867282694b ]
I believe that "hook->num" can be up to UINT_MAX. Shifting more than 31 bits would is undefined in C but in practice it would lead to shift wrapping. That would lead to an array overflow in nf_tables_addchain():
ops->hook = hook.type->hooks[ops->hooknum];
--- net/netfilter/nf_tables_api.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/netfilter/nf_tables_api.c +1 -1
@@ -1496,7 +1496,7 @@ static int nft_chain_parse_hook(struct net *net,if (IS_ERR (type))return PTR_ERR (type); } -if (!(type->hook_mask & (1 << hook->num))) +if (hook->num >NF_MAX_HOOKS || !(type->hook_mask & (1 << hook->num)))return -EOPNOTSUPP ; if (type->type == NFT_CHAIN_T_NAT &&
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 37/81 ipmi: ipmi_si_hardcode.c: init si_type array to fix a crash
In reply to Sasha LevinFrom: Tony Camuso <[email protected]>
[ Upstream commit a885bcfd152f97b25005298ab2d6b741aed9b49c ]
The intended behavior of function ipmi_hardcode_init_one() is to default to kcs interface when no type argument is presented when initializing ipmi with hard coded addresses.
However, the array of char pointers allocated on the stack by function ipmi_hardcode_init() was not inited to zeroes, so it contained stack debris.
Consequently, passing the cruft stored in this array to function ipmi_hardcode_init_one() caused a crash when it was unable to detect that the char * being passed was nonsense and tried to access the address specified by the bogus pointer.
The fix is simply to initialize the si_type array to zeroes, so if there were no type argument given to at the command line, function ipmi_hardcode_init_one() could properly default to the kcs interface.
Message-Id: <[email protected]>
--- drivers/char/ipmi/ipmi_si_hardcode.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/char/ipmi/ipmi_si_hardcode.c +2 -0
@@ -200,6 +200,8 @@ void __init ipmi_hardcode_init(void)char *str;char *si_type[SI_MAX_PARMS ]; +memset (si_type,0 ,sizeof (si_type)); +/* Parse out the si_type string into its components. */ str = si_type_str;if (*str !='\0' ) {
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 39/81 x86/mm/KASLR: Fix the size of the direct mapping section
In reply to Sasha LevinFrom: Baoquan He <[email protected]>
[ Upstream commit ec3937107ab43f3e8b2bc9dad95710043c462ff7 ]
kernel_randomize_memory() uses __PHYSICAL_MASK_SHIFT to calculate the maximum amount of system RAM supported. The size of the direct mapping section is obtained from the smaller one of the below two values:
(actual system RAM size + padding size) vs (max system RAM size supported)
This calculation is wrong since commit
b83ce5ee9147 ("x86/mm/64: Make __PHYSICAL_MASK_SHIFT always 52").
In it, __PHYSICAL_MASK_SHIFT was changed to be 52, regardless of whether the kernel is using 4-level or 5-level page tables. Thus, it will always use 4 PB as the maximum amount of system RAM, even in 4-level paging mode where it should actually be 64 TB.
Thus, the size of the direct mapping section will always be the sum of the actual system RAM size plus the padding size.
Even when the amount of system RAM is 64 TB, the following layout will still be used. Obviously KALSR will be weakened significantly.
|____|_______actual RAM_______|_padding_|______the rest_______| 0 64TB ~120TB
Instead, it should be like this:
|____|_______actual RAM_______|_________the rest______________| 0 64TB ~120TB
The size of padding region is controlled by CONFIG_RANDOMIZE_MEMORY_PHYSICAL_PADDING, which is 10 TB by default.
The above issue only exists when CONFIG_RANDOMIZE_MEMORY_PHYSICAL_PADDING is set to a non-zero value, which is the case when CONFIG_MEMORY_HOTPLUG is enabled. Otherwise, using __PHYSICAL_MASK_SHIFT doesn't affect KASLR.
Fix it by replacing __PHYSICAL_MASK_SHIFT with MAX_PHYSMEM_BITS.
[ bp: Massage commit message. ]
--- arch/x86/mm/kaslr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/mm/kaslr.c +1 -1
@@ -93,7 +93,7 @@ void __init kernel_randomize_memory(void)if (!kaslr_memory_enabled ())return ; - kaslr_regions[0 ].size_tb =1 << (__PHYSICAL_MASK_SHIFT -TB_SHIFT ); + kaslr_regions[0 ].size_tb =1 << (MAX_PHYSMEM_BITS -TB_SHIFT ); kaslr_regions[1 ].size_tb =VMALLOC_SIZE_TB ;/*
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 35/81 KVM: x86: avoid misreporting level-triggered irqs as edge-triggered in tracing
In reply to Sasha LevinFrom: Vitaly Kuznetsov <[email protected]>
[ Upstream commit 7a223e06b1a411cef6c4cd7a9b9a33c8d225b10e ]
In __apic_accept_irq() interface trig_mode is int and actually on some code paths it is set above u8:
kvm_apic_set_irq() extracts it from 'struct kvm_lapic_irq' where trig_mode is u16. This is done on purpose as e.g. kvm_set_msi_irq() sets it to (1 << 15) & e->msi.data
kvm_apic_local_deliver sets it to reg & (1 << 15).
Fix the immediate issue by making 'tm' into u16. We may also want to adjust __apic_accept_irq() interface and use proper sizes for vector, level, trig_mode but this is not urgent.
--- arch/x86/kvm/trace.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/x86/kvm/trace.h +2 -2
@@ -438,13 +438,13 @@ TRACE_EVENT(kvm_apic_ipi, ); TRACE_EVENT(kvm_apic_accept_irq, - TP_PROTO(__u32 apicid, __u16 dm, __u8 tm, __u8 vec), + TP_PROTO(__u32 apicid, __u16 dm, __u16 tm, __u8 vec), TP_ARGS(apicid, dm, tm, vec), TP_STRUCT__entry( __field( __u32, apicid ) __field( __u16, dm ) - __field( __u8, tm ) + __field( __u16, tm ) __field( __u8, vec ) ),
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 32/81 x86/mm/tlb: Revert "x86/mm: Align TLB invalidation info"
In reply to Sasha LevinFrom: Peter Zijlstra <[email protected]>
[ Upstream commit 780e0106d468a2962b16b52fdf42898f2639e0a0 ]
Revert the following commit:
515ab7c41306: ("x86/mm: Align TLB invalidation info")
I found out (the hard way) that under some .config options (notably L1_CACHE_SHIFT=7) and compiler combinations this on-stack alignment leads to a 320 byte stack usage, which then triggers a KASAN stack warning elsewhere.
Using 320 bytes of stack space for a 40 byte structure is ludicrous and clearly not right.
--- arch/x86/mm/tlb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/mm/tlb.c +1 -1
@@ -694,7 +694,7 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, {int cpu; -struct flush_tlb_info info__aligned(SMP_CACHE_BYTES) = { +struct flush_tlb_info info = { .mm = mm, };
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 30/81 x86/build/lto: Fix truncated .bss with -fdata-sections
In reply to Sasha LevinFrom: Sami Tolvanen <[email protected]>
[ Upstream commit 6a03469a1edc94da52b65478f1e00837add869a3 ]
With CONFIG_LD_DEAD_CODE_DATA_ELIMINATION=y, we compile the kernel with -fdata-sections, which also splits the .bss section.
The new section, with a new .bss.* name, which pattern gets missed by the main x86 linker script which only expects the '.bss' name. This results in the discarding of the second part and a too small, truncated .bss section and an unhappy, non-working kernel.
Use the common BSS_MAIN macro in the linker script to properly capture and merge all the generated BSS sections.
--- arch/x86/kernel/vmlinux.lds.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kernel/vmlinux.lds.S +1 -1
@@ -372,7 +372,7 @@ SECTIONS .bss : AT(ADDR(.bss) - LOAD_OFFSET) { __bss_start = .; *(.bss..page_aligned) - *(.bss) + *(BSS_MAIN) BSS_DECRYPTED . = ALIGN(PAGE_SIZE); __bss_stop = .;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 27/81 afs: Unlock pages for __pagevec_release()
In reply to Sasha LevinFrom: Marc Dionne <[email protected]>
[ Upstream commit 21bd68f196ca91fc0f3d9bd1b32f6e530e8c1c88 ]
__pagevec_release() complains loudly if any page in the vector is still locked. The pages need to be locked for generic_error_remove_page(), but that function doesn't actually unlock them.
Unlock the pages afterwards.
--- fs/afs/write.c | 1 + 1 file changed, 1 insertion(+)
fs/afs/write.c +1 -0
@@ -253,6 +253,7 @@ static void afs_kill_pages(struct address_space *mapping, first = page->index +1 ;lock_page (page);generic_error_remove_page (mapping, page); +unlock_page (page); }__pagevec_release (&pv);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 26/81 qede: fix write to free'd pointer error and double free of ptp
In reply to Sasha LevinFrom: Colin Ian King <[email protected]>
[ Upstream commit 1dc2b3d65523780ed1972d446c76e62e13f3e8f5 ]
The err2 error return path calls qede_ptp_disable that cleans up on an error and frees ptp. After this, the free'd ptp is dereferenced when ptp->clock is set to NULL and the code falls-through to error path err1 that frees ptp again.
Fix this by calling qede_ptp_disable and exiting via an error return path that does not set ptp->clock or kfree ptp.
Addresses-Coverity: ("Write to pointer after free")
--- drivers/net/ethernet/qlogic/qede/qede_ptp.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-)
drivers/net/ethernet/qlogic/qede/qede_ptp.c +3 -4
@@ -490,18 +490,17 @@ int qede_ptp_enable(struct qede_dev *edev, bool init_tc) ptp->clock =ptp_clock_register (&ptp->clock_info, &edev->pdev->dev);if (IS_ERR (ptp->clock)) { - rc = -EINVAL ;DP_ERR (edev,"PTP clock registration failed\n" ); +qede_ptp_disable (edev); + rc = -EINVAL ; goto err2; }return 0 ; -err2: -qede_ptp_disable (edev); - ptp->clock =NULL ; err1:kfree (ptp); +err2: edev->ptp =NULL ;return rc;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 21/81 clocksource/drivers/npcm: select TIMER_OF
In reply to Sasha LevinFrom: Arnd Bergmann <[email protected]>
[ Upstream commit 99834eead2a04e93a120abb112542b87c42ff5e1 ]
When this is disabled, we get a link failure:
drivers/clocksource/timer-npcm7xx.o: In function `npcm7xx_timer_init':
timer-npcm7xx.c:(.init.text+0xf): undefined reference to `timer_of_init'
--- drivers/clocksource/Kconfig | 1 + 1 file changed, 1 insertion(+)
drivers/clocksource/Kconfig +1 -0
@@ -136,6 +136,7 @@ config VT8500_TIMER config NPCM7XX_TIMER bool "NPCM7xx timer driver" if COMPILE_TEST depends on HAS_IOMEM + select TIMER_OF select CLKSRC_MMIO help Enable 24-bit TIMER0 and TIMER1 counters in the NPCM7xx architecture,
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 12/81 mac80211: Increase MAX_MSG_LEN
In reply to Sasha LevinFrom: Andrei Otcheretianski <[email protected]>
[ Upstream commit 78be2d21cc1cd3069c6138dcfecec62583130171 ]
Looks that 100 chars isn't enough for messages, as we keep getting warnings popping from different places due to message shortening. Instead of trying to shorten the prints, just increase the buffer size.
--- net/mac80211/trace_msg.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
net/mac80211/trace_msg.h +6 -1
@@ -1,4 +1,9 @@/* SPDX-License-Identifier: GPL-2.0 */ +/* +* Portions of this file +* Copyright (C) 2019 Intel Corporation +*/ +#ifdef CONFIG_MAC80211_MESSAGE_TRACING #if !defined(__MAC80211_MSG_DRIVER_TRACE) || defined(TRACE_HEADER_MULTI_READ )
@@ -11,7 +16,7 @@#undef TRACE_SYSTEM#define TRACE_SYSTEM mac80211_msg -#define MAX_MSG_LEN 100 +#define MAX_MSG_LEN 120 DECLARE_EVENT_CLASS(mac80211_msg_event, TP_PROTO(struct va_format *vaf),
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 14/81 mac80211: fix memory accounting with A-MSDU aggregation
In reply to Sasha LevinFrom: Felix Fietkau <[email protected]>
[ Upstream commit eb9b64e3a9f8483e6e54f4e03b2ae14ae5db2690 ]
skb->truesize can change due to memory reallocation or when adding extra fragments. Adjust fq->memory_usage accordingly
--- net/mac80211/tx.c | 3 +++ 1 file changed, 3 insertions(+)
net/mac80211/tx.c +3 -0
@@ -3185,6 +3185,7 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata,u8 max_subframes = sta->sta.max_amsdu_subframes;int max_frags = local->hw.max_tx_fragments;int max_amsdu_len = sta->sta.max_amsdu_len; +int orig_truesize;__be16 len;void *data;bool ret = false;
@@ -3218,6 +3219,7 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata,if (!head) goto out; + orig_truesize = head->truesize; orig_len = head->len;if (skb->len + head->len > max_amsdu_len)
@@ -3272,6 +3274,7 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata, *frag_tail = skb; out_recalc: + fq->memory_usage += head->truesize - orig_truesize;if (head->len != orig_len) { flow->backlog += head->len - orig_len; tin->backlog_bytes += head->len - orig_len;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 11/81 mac80211: fix unaligned access in mesh table hash function
In reply to Sasha LevinFrom: Felix Fietkau <[email protected]>
[ Upstream commit 40586e3fc400c00c11151804dcdc93f8c831c808 ]
The pointer to the last four bytes of the address is not guaranteed to be aligned, so we need to use __get_unaligned_cpu32 here
--- net/mac80211/mesh_pathtbl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/mac80211/mesh_pathtbl.c +1 -1
@@ -23,7 +23,7 @@ static void mesh_path_free_rcu(struct mesh_table *tbl, struct mesh_path *mpath);static u32 mesh_table_hash (const void *addr,u32 len,u32 seed) {/* Use last four bytes of hw addr as hash index */ -return jhash_1word (*(u32 *)(addr+2 ), seed); +return jhash_1word (__get_unaligned_cpu32 ((u8 *)addr +2 ), seed); }static const struct rhashtable_params mesh_rht_params = {
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 2/81 iio: adc: xilinx: fix potential use-after-free on probe
In reply to Sasha LevinFrom: Sven Van Asbroeck <[email protected]>
[ Upstream commit 862e4644fd2d7df8998edc65e0963ea2f567bde9 ]
If probe errors out after request_irq(), its error path does not explicitly cancel the delayed work, which may have been scheduled by the interrupt handler.
This means the delayed work may still be running when the core frees the private structure (struct xadc). This is a potential use-after-free.
Fix by inserting cancel_delayed_work_sync() in the probe error path.
--- drivers/iio/adc/xilinx-xadc-core.c | 1 + 1 file changed, 1 insertion(+)
drivers/iio/adc/xilinx-xadc-core.c +1 -0
@@ -1290,6 +1290,7 @@ static int xadc_probe(struct platform_device *pdev) err_free_irq:free_irq (xadc->irq, indio_dev); +cancel_delayed_work_sync (&xadc->zynq_unmask_work); err_clk_disable_unprepare:clk_disable_unprepare (xadc->clk); err_free_samplerate_trigger:
Signature · 1 line
2.20.1