diff options
Diffstat (limited to 'arch/um')
33 files changed, 318 insertions, 710 deletions
diff --git a/arch/um/Kconfig b/arch/um/Kconfig index 6b9938919f0b..ec9711d068b7 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -12,6 +12,7 @@ config UML select HAVE_UID16 select HAVE_FUTEX_CMPXCHG if FUTEX select HAVE_DEBUG_KMEMLEAK + select HAVE_DEBUG_BUGVERBOSE select GENERIC_IRQ_SHOW select GENERIC_CPU_DEVICES select GENERIC_CLOCKEVENTS @@ -31,12 +32,6 @@ config ISA config SBUS bool -config PCI - bool - -config PCMCIA - bool - config TRACE_IRQFLAGS_SUPPORT bool default y diff --git a/arch/um/Kconfig.debug b/arch/um/Kconfig.debug index 2014597605ea..85726eeec345 100644 --- a/arch/um/Kconfig.debug +++ b/arch/um/Kconfig.debug @@ -16,6 +16,7 @@ config GPROF config GCOV bool "Enable gcov support" depends on DEBUG_INFO + depends on !KCOV help This option allows developers to retrieve coverage data from a UML session. diff --git a/arch/um/Makefile b/arch/um/Makefile index ab1066c38944..273130cf91d1 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -23,8 +23,6 @@ OS := $(shell uname -s) # features. SHELL := /bin/bash -filechk_gen_header = $< - core-y += $(ARCH_DIR)/kernel/ \ $(ARCH_DIR)/drivers/ \ $(ARCH_DIR)/os-$(OS)/ @@ -116,7 +114,8 @@ endef archheaders: $(Q)$(MAKE) -f $(srctree)/Makefile ARCH=$(HEADER_ARCH) asm-generic archheaders -archprepare: include/generated/user_constants.h +archprepare: + $(Q)$(MAKE) $(build)=$(HOST_DIR)/um include/generated/user_constants.h LINK-$(CONFIG_LD_SCRIPT_STATIC) += -static LINK-$(CONFIG_LD_SCRIPT_DYN) += -Wl,-rpath,/lib $(call cc-option, -no-pie) @@ -146,25 +145,4 @@ archclean: @find . \( -name '*.bb' -o -name '*.bbg' -o -name '*.da' \ -o -name '*.gcov' \) -type f -print | xargs rm -f -# Generated files - -$(HOST_DIR)/um/user-offsets.s: __headers FORCE - $(Q)$(MAKE) $(build)=$(HOST_DIR)/um $@ - -define filechk_gen-asm-offsets - (set -e; \ - echo "/*"; \ - echo " * DO NOT MODIFY."; \ - echo " *"; \ - echo " * This file was generated by arch/$(ARCH)/Makefile"; \ - echo " *"; \ - echo " */"; \ - echo ""; \ - sed -ne "/^->/{s:^->\([^ ]*\) [\$$#]*\([^ ]*\) \(.*\):#define \1 \2 /* \3 */:; s:->::; p;}" < $<; \ - echo ""; ) -endef - -include/generated/user_constants.h: $(HOST_DIR)/um/user-offsets.s - $(call filechk,gen-asm-offsets) - export HEADER_ARCH SUBARCH USER_CFLAGS CFLAGS_NO_HARDENING OS DEV_NULL_PATH diff --git a/arch/um/drivers/chan_kern.c b/arch/um/drivers/chan_kern.c index 05588f9466c7..a4e64edb8f38 100644 --- a/arch/um/drivers/chan_kern.c +++ b/arch/um/drivers/chan_kern.c @@ -211,12 +211,6 @@ void deactivate_chan(struct chan *chan, int irq) deactivate_fd(chan->fd, irq); } -void reactivate_chan(struct chan *chan, int irq) -{ - if (chan && chan->enabled) - reactivate_fd(chan->fd, irq); -} - int write_chan(struct chan *chan, const char *buf, int len, int write_irq) { @@ -228,8 +222,6 @@ int write_chan(struct chan *chan, const char *buf, int len, n = chan->ops->write(chan->fd, buf, len, chan->data); if (chan->primary) { ret = n; - if ((ret == -EAGAIN) || ((ret >= 0) && (ret < len))) - reactivate_fd(chan->fd, write_irq); } return ret; } @@ -527,8 +519,6 @@ void chan_interrupt(struct line *line, int irq) tty_insert_flip_char(port, c, TTY_NORMAL); } while (err > 0); - if (err == 0) - reactivate_fd(chan->fd, irq); if (err == -EIO) { if (chan->primary) { tty_port_tty_hangup(&line->port, false); diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index 8d80b27502e6..e0e63931fb2b 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -235,14 +235,6 @@ void line_unthrottle(struct tty_struct *tty) line->throttled = 0; chan_interrupt(line, line->driver->read_irq); - - /* - * Maybe there is enough stuff pending that calling the interrupt - * throttles us again. In this case, line->throttled will be 1 - * again and we shouldn't turn the interrupt back on. - */ - if (!line->throttled) - reactivate_chan(line->chan_in, line->driver->read_irq); } static irqreturn_t line_write_interrupt(int irq, void *data) @@ -261,7 +253,7 @@ static irqreturn_t line_write_interrupt(int irq, void *data) if (err == 0) { spin_unlock(&line->lock); return IRQ_NONE; - } else if (err < 0) { + } else if ((err < 0) && (err != -EAGAIN)) { line->head = line->buffer; line->tail = line->buffer; } @@ -284,7 +276,7 @@ int line_setup_irq(int fd, int input, int output, struct line *line, void *data) if (err) return err; if (output) - err = um_request_irq(driver->write_irq, fd, IRQ_NONE, + err = um_request_irq(driver->write_irq, fd, IRQ_WRITE, line_write_interrupt, IRQF_SHARED, driver->write_irq_name, data); return err; @@ -667,8 +659,6 @@ static irqreturn_t winch_interrupt(int irq, void *data) tty_kref_put(tty); } out: - if (winch->fd != -1) - reactivate_fd(winch->fd, WINCH_IRQ); return IRQ_HANDLED; } diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index d5f9a2d1da1b..ff3ab72fd90f 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -96,7 +96,6 @@ static irqreturn_t mconsole_interrupt(int irq, void *dev_id) } if (!list_empty(&mc_requests)) schedule_work(&mconsole_work); - reactivate_fd(fd, MCONSOLE_IRQ); return IRQ_HANDLED; } @@ -240,7 +239,6 @@ void mconsole_stop(struct mc_request *req) (*req->cmd->handler)(req); } os_set_fd_block(req->originating_fd, 0); - reactivate_fd(req->originating_fd, MCONSOLE_IRQ); mconsole_reply(req, "", 0, 0); } diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c index 3ef1b48e064a..d80cfb1d9430 100644 --- a/arch/um/drivers/net_kern.c +++ b/arch/um/drivers/net_kern.c @@ -6,7 +6,7 @@ * Licensed under the GPL. */ -#include <linux/bootmem.h> +#include <linux/memblock.h> #include <linux/etherdevice.h> #include <linux/ethtool.h> #include <linux/inetdevice.h> @@ -137,8 +137,6 @@ static irqreturn_t uml_net_interrupt(int irq, void *dev_id) schedule_work(&lp->work); goto out; } - reactivate_fd(lp->fd, UM_ETH_IRQ); - out: spin_unlock(&lp->lock); return IRQ_HANDLED; @@ -650,7 +648,7 @@ static int __init eth_setup(char *str) return 1; } - new = alloc_bootmem(sizeof(*new)); + new = memblock_alloc(sizeof(*new), SMP_CACHE_BYTES); INIT_LIST_HEAD(&new->list); new->index = n; diff --git a/arch/um/drivers/port_kern.c b/arch/um/drivers/port_kern.c index 40ca5cc275e9..b0e9ff35daee 100644 --- a/arch/um/drivers/port_kern.c +++ b/arch/um/drivers/port_kern.c @@ -137,7 +137,6 @@ static void port_work_proc(struct work_struct *unused) if (!port->has_connection) continue; - reactivate_fd(port->fd, ACCEPT_IRQ); while (port_accept(port)) ; port->has_connection = 0; diff --git a/arch/um/drivers/port_user.c b/arch/um/drivers/port_user.c index 9a8e1b64c22e..5f56d11b886f 100644 --- a/arch/um/drivers/port_user.c +++ b/arch/um/drivers/port_user.c @@ -168,7 +168,7 @@ int port_connection(int fd, int *socket, int *pid_out) { int new, err; char *argv[] = { "/usr/sbin/in.telnetd", "-L", - "/usr/lib/uml/port-helper", NULL }; + OS_LIB_PATH "/uml/port-helper", NULL }; struct port_pre_exec_data data; new = accept(fd, NULL, 0); diff --git a/arch/um/drivers/random.c b/arch/um/drivers/random.c index 778a0e52d5a5..1d5d3057e6f1 100644 --- a/arch/um/drivers/random.c +++ b/arch/um/drivers/random.c @@ -73,7 +73,6 @@ static ssize_t rng_dev_read (struct file *filp, char __user *buf, size_t size, return ret ? : -EAGAIN; atomic_inc(&host_sleep_count); - reactivate_fd(random_fd, RANDOM_IRQ); add_sigio_fd(random_fd); add_wait_queue(&host_read_wait, &wait); diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 74c002ddc0ce..a4a41421c5e2 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -1,4 +1,5 @@ /* + * Copyright (C) 2018 Cambridge Greys Ltd * Copyright (C) 2015-2016 Anton Ivanov ([email protected]) * Copyright (C) 2000 Jeff Dike ([email protected]) * Licensed under the GPL @@ -43,11 +44,11 @@ #include <os.h> #include "cow.h" -enum ubd_req { UBD_READ, UBD_WRITE, UBD_FLUSH }; +/* Max request size is determined by sector mask - 32K */ +#define UBD_MAX_REQUEST (8 * sizeof(long)) struct io_thread_req { struct request *req; - enum ubd_req op; int fds[2]; unsigned long offsets[2]; unsigned long long offset; @@ -153,6 +154,7 @@ struct ubd { struct openflags openflags; unsigned shared:1; unsigned no_cow:1; + unsigned no_trim:1; struct cow cow; struct platform_device pdev; struct request_queue *queue; @@ -176,6 +178,7 @@ struct ubd { .boot_openflags = OPEN_FLAGS, \ .openflags = OPEN_FLAGS, \ .no_cow = 0, \ + .no_trim = 0, \ .shared = 0, \ .cow = DEFAULT_COW, \ .lock = __SPIN_LOCK_UNLOCKED(ubd_devs.lock), \ @@ -322,7 +325,7 @@ static int ubd_setup_common(char *str, int *index_out, char **error_out) *index_out = n; err = -EINVAL; - for (i = 0; i < sizeof("rscd="); i++) { + for (i = 0; i < sizeof("rscdt="); i++) { switch (*str) { case 'r': flags.w = 0; @@ -336,12 +339,15 @@ static int ubd_setup_common(char *str, int *index_out, char **error_out) case 'c': ubd_dev->shared = 1; break; + case 't': + ubd_dev->no_trim = 1; + break; case '=': str++; goto break_loop; default: *error_out = "Expected '=' or flag letter " - "(r, s, c, or d)"; + "(r, s, c, t or d)"; goto out; } str++; @@ -414,6 +420,7 @@ __uml_help(ubd_setup, " 'c' will cause the device to be treated as being shared between multiple\n" " UMLs and file locking will be turned off - this is appropriate for a\n" " cluster filesystem and inappropriate at almost all other times.\n\n" +" 't' will disable trim/discard support on the device (enabled by default).\n\n" ); static int udb_setup(char *str) @@ -511,16 +518,21 @@ static void ubd_handler(void) } for (count = 0; count < n/sizeof(struct io_thread_req *); count++) { struct io_thread_req *io_req = (*irq_req_buffer)[count]; - int err = io_req->error ? BLK_STS_IOERR : BLK_STS_OK; - - if (!blk_update_request(io_req->req, err, io_req->length)) - __blk_mq_end_request(io_req->req, err); + if ((io_req->error == BLK_STS_NOTSUPP) && (req_op(io_req->req) == REQ_OP_DISCARD)) { + blk_queue_max_discard_sectors(io_req->req->q, 0); + blk_queue_max_write_zeroes_sectors(io_req->req->q, 0); + blk_queue_flag_clear(QUEUE_FLAG_DISCARD, io_req->req->q); + } + if ((io_req->error) || (io_req->buffer == NULL)) + blk_mq_end_request(io_req->req, io_req->error); + else { + if (!blk_update_request(io_req->req, io_req->error, io_req->length)) + __blk_mq_end_request(io_req->req, io_req->error); + } kfree(io_req); } } - - reactivate_fd(thread_fd, UBD_IRQ); } static irqreturn_t ubd_intr(int irq, void *dev) @@ -789,7 +801,7 @@ static int ubd_open_dev(struct ubd *ubd_dev) if((fd == -ENOENT) && create_cow){ fd = create_cow_file(ubd_dev->file, ubd_dev->cow.file, - ubd_dev->openflags, 1 << 9, PAGE_SIZE, + ubd_dev->openflags, SECTOR_SIZE, PAGE_SIZE, &ubd_dev->cow.bitmap_offset, &ubd_dev->cow.bitmap_len, &ubd_dev->cow.data_offset); @@ -830,6 +842,14 @@ static int ubd_open_dev(struct ubd *ubd_dev) if(err < 0) goto error; ubd_dev->cow.fd = err; } + if (ubd_dev->no_trim == 0) { + ubd_dev->queue->limits.discard_granularity = SECTOR_SIZE; + ubd_dev->queue->limits.discard_alignment = SECTOR_SIZE; + blk_queue_max_discard_sectors(ubd_dev->queue, UBD_MAX_REQUEST); + blk_queue_max_write_zeroes_sectors(ubd_dev->queue, UBD_MAX_REQUEST); + blk_queue_flag_set(QUEUE_FLAG_DISCARD, ubd_dev->queue); + } + blk_queue_flag_set(QUEUE_FLAG_NONROT, ubd_dev->queue); return 0; error: os_close_file(ubd_dev->fd); @@ -882,7 +902,7 @@ static int ubd_disk_register(int major, u64 size, int unit, return 0; } -#define ROUND_BLOCK(n) ((n + ((1 << 9) - 1)) & (-1 << 9)) +#define ROUND_BLOCK(n) ((n + (SECTOR_SIZE - 1)) & (-SECTOR_SIZE)) static const struct blk_mq_ops ubd_mq_ops = { .queue_rq = ubd_queue_rq, @@ -1234,10 +1254,10 @@ static void cowify_bitmap(__u64 io_offset, int length, unsigned long *cow_mask, __u64 bitmap_offset, unsigned long *bitmap_words, __u64 bitmap_len) { - __u64 sector = io_offset >> 9; + __u64 sector = io_offset >> SECTOR_SHIFT; int i, update_bitmap = 0; - for(i = 0; i < length >> 9; i++){ + for (i = 0; i < length >> SECTOR_SHIFT; i++) { if(cow_mask != NULL) ubd_set_bit(i, (unsigned char *) cow_mask); if(ubd_test_bit(sector + i, (unsigned char *) bitmap)) @@ -1271,14 +1291,14 @@ static void cowify_bitmap(__u64 io_offset, int length, unsigned long *cow_mask, static void cowify_req(struct io_thread_req *req, unsigned long *bitmap, __u64 bitmap_offset, __u64 bitmap_len) { - __u64 sector = req->offset >> 9; + __u64 sector = req->offset >> SECTOR_SHIFT; int i; - if(req->length > (sizeof(req->sector_mask) * 8) << 9) + if (req->length > (sizeof(req->sector_mask) * 8) << SECTOR_SHIFT) panic("Operation too long"); - if(req->op == UBD_READ) { - for(i = 0; i < req->length >> 9; i++){ + if (req_op(req->req) == REQ_OP_READ) { + for (i = 0; i < req->length >> SECTOR_SHIFT; i++) { if(ubd_test_bit(sector + i, (unsigned char *) bitmap)) ubd_set_bit(i, (unsigned char *) &req->sector_mask); @@ -1305,28 +1325,27 @@ static int ubd_queue_one_vec(struct blk_mq_hw_ctx *hctx, struct request *req, io_req->fds[0] = dev->cow.fd; else io_req->fds[0] = dev->fd; + io_req->error = 0; - if (req_op(req) == REQ_OP_FLUSH) { - io_req->op = UBD_FLUSH; - } else { - io_req->fds[1] = dev->fd; - io_req->cow_offset = -1; - io_req->offset = off; + if (bvec != NULL) { + io_req->buffer = page_address(bvec->bv_page) + bvec->bv_offset; io_req->length = bvec->bv_len; - io_req->error = 0; - io_req->sector_mask = 0; + } else { + io_req->buffer = NULL; + io_req->length = blk_rq_bytes(req); + } - io_req->op = rq_data_dir(req) == READ ? UBD_READ : UBD_WRITE; - io_req->offsets[0] = 0; - io_req->offsets[1] = dev->cow.data_offset; - io_req->buffer = page_address(bvec->bv_page) + bvec->bv_offset; - io_req->sectorsize = 1 << 9; + io_req->sectorsize = SECTOR_SIZE; + io_req->fds[1] = dev->fd; + io_req->cow_offset = -1; + io_req->offset = off; + io_req->sector_mask = 0; + io_req->offsets[0] = 0; + io_req->offsets[1] = dev->cow.data_offset; - if (dev->cow.file) { - cowify_req(io_req, dev->cow.bitmap, - dev->cow.bitmap_offset, dev->cow.bitmap_len); - } - } + if (dev->cow.file) + cowify_req(io_req, dev->cow.bitmap, + dev->cow.bitmap_offset, dev->cow.bitmap_len); ret = os_write_file(thread_fd, &io_req, sizeof(io_req)); if (ret != sizeof(io_req)) { @@ -1334,37 +1353,60 @@ static int ubd_queue_one_vec(struct blk_mq_hw_ctx *hctx, struct request *req, pr_err("write to io thread failed: %d\n", -ret); kfree(io_req); } - return ret; } +static int queue_rw_req(struct blk_mq_hw_ctx *hctx, struct request *req) +{ + struct req_iterator iter; + struct bio_vec bvec; + int ret; + u64 off = (u64)blk_rq_pos(req) << SECTOR_SHIFT; + + rq_for_each_segment(bvec, req, iter) { + ret = ubd_queue_one_vec(hctx, req, off, &bvec); + if (ret < 0) + return ret; + off += bvec.bv_len; + } + return 0; +} + static blk_status_t ubd_queue_rq(struct blk_mq_hw_ctx *hctx, const struct blk_mq_queue_data *bd) { + struct ubd *ubd_dev = hctx->queue->queuedata; struct request *req = bd->rq; - int ret = 0; + int ret = 0, res = BLK_STS_OK; blk_mq_start_request(req); - if (req_op(req) == REQ_OP_FLUSH) { + spin_lock_irq(&ubd_dev->lock); + + switch (req_op(req)) { + /* operations with no lentgth/offset arguments */ + case REQ_OP_FLUSH: ret = ubd_queue_one_vec(hctx, req, 0, NULL); - } else { - struct req_iterator iter; - struct bio_vec bvec; - u64 off = (u64)blk_rq_pos(req) << 9; - - rq_for_each_segment(bvec, req, iter) { - ret = ubd_queue_one_vec(hctx, req, off, &bvec); - if (ret < 0) - goto out; - off += bvec.bv_len; - } + break; + case REQ_OP_READ: + case REQ_OP_WRITE: + ret = queue_rw_req(hctx, req); + break; + case REQ_OP_DISCARD: + case REQ_OP_WRITE_ZEROES: + ret = ubd_queue_one_vec(hctx, req, (u64)blk_rq_pos(req) << 9, NULL); + break; + default: + WARN_ON_ONCE(1); + res = BLK_STS_NOTSUPP; } -out: - if (ret < 0) { + + spin_unlock_irq(&ubd_dev->lock); + + if (ret < 0) blk_mq_requeue_request(req, true); - } - return BLK_STS_OK; + + return res; } static int ubd_getgeo(struct block_device *bdev, struct hd_geometry *geo) @@ -1409,39 +1451,60 @@ static int ubd_ioctl(struct block_device *bdev, fmode_t mode, return -EINVAL; } +static int map_error(int error_code) +{ + switch (error_code) { + case 0: + return BLK_STS_OK; + case ENOSYS: + case EOPNOTSUPP: + return BLK_STS_NOTSUPP; + case ENOSPC: + return BLK_STS_NOSPC; + } + return BLK_STS_IOERR; +} + +/* + * Everything from here onwards *IS NOT PART OF THE KERNEL* + * + * The following functions are part of UML hypervisor code. + * All functions from here onwards are executed as a helper + * thread and are not allowed to execute any kernel functions. + * + * Any communication must occur strictly via shared memory and IPC. + * + * Do not add printks, locks, kernel memory operations, etc - it + * will result in unpredictable behaviour and/or crashes. + */ + static int update_bitmap(struct io_thread_req *req) { int n; if(req->cow_offset == -1) - return 0; + return map_error(0); n = os_pwrite_file(req->fds[1], &req->bitmap_words, sizeof(req->bitmap_words), req->cow_offset); - if(n != sizeof(req->bitmap_words)){ - printk("do_io - bitmap update failed, err = %d fd = %d\n", -n, - req->fds[1]); - return 1; - } + if (n != sizeof(req->bitmap_words)) + return map_error(-n); - return 0; + return map_error(0); } static void do_io(struct io_thread_req *req) { - char *buf; + char *buf = NULL; unsigned long len; int n, nsectors, start, end, bit; __u64 off; - if (req->op == UBD_FLUSH) { + /* FLUSH is really a special case, we cannot "case" it with others */ + + if (req_op(req->req) == REQ_OP_FLUSH) { /* fds[0] is always either the rw image or our cow file */ - n = os_sync_file(req->fds[0]); - if (n != 0) { - printk("do_io - sync failed err = %d " - "fd = %d\n", -n, req->fds[0]); - req->error = 1; - } + req->error = map_error(-os_sync_file(req->fds[0])); return; } @@ -1458,30 +1521,42 @@ static void do_io(struct io_thread_req *req) off = req->offset + req->offsets[bit] + start * req->sectorsize; len = (end - start) * req->sectorsize; - buf = &req->buffer[start * req->sectorsize]; + if (req->buffer != NULL) + buf = &req->buffer[start * req->sectorsize]; - if(req->op == UBD_READ){ + switch (req_op(req->req)) { + case REQ_OP_READ: n = 0; do { buf = &buf[n]; len -= n; n = os_pread_file(req->fds[bit], buf, len, off); if (n < 0) { - printk("do_io - read failed, err = %d " - "fd = %d\n", -n, req->fds[bit]); - req->error = 1; + req->error = map_error(-n); return; } } while((n < len) && (n != 0)); if (n < len) memset(&buf[n], 0, len - n); - } else { + break; + case REQ_OP_WRITE: n = os_pwrite_file(req->fds[bit], buf, len, off); if(n != len){ - printk("do_io - write failed err = %d " - "fd = %d\n", -n, req->fds[bit]); - req->error = 1; + req->error = map_error(-n); + return; + } + break; + case REQ_OP_DISCARD: + case REQ_OP_WRITE_ZEROES: + n = os_falloc_punch(req->fds[bit], off, len); + if (n) { + req->error = map_error(-n); return; } + break; + default: + WARN_ON_ONCE(1); + req->error = BLK_STS_NOTSUPP; + return; } start = end; @@ -1516,11 +1591,6 @@ int io_thread(void *arg) if (n == -EAGAIN) { ubd_read_poll(-1); continue; - } else { - printk("io_thread - read failed, fd = %d, " - "err = %d," - "reminder = %d\n", - kernel_fd, -n, io_remainder_size); } } @@ -1535,11 +1605,6 @@ int io_thread(void *arg) res = os_write_file(kernel_fd, ((char *) io_req_buffer) + written, n); if (res >= 0) { written += res; - } else { - if (res != -EAGAIN) { - printk("io_thread - write failed, fd = %d, " - "err = %d\n", kernel_fd, -n); - } } if (written < n) { ubd_write_poll(-1); diff --git a/arch/um/drivers/vector_kern.c b/arch/um/drivers/vector_kern.c index 50ee3bb5a63a..046fa9ea0ccc 100644 --- a/arch/um/drivers/vector_kern.c +++ b/arch/um/drivers/vector_kern.c @@ -9,7 +9,7 @@ */ #include <linux/version.h> -#include <linux/bootmem.h> +#include <linux/memblock.h> #include <linux/etherdevice.h> #include <linux/ethtool.h> #include <linux/inetdevice.h> @@ -1118,16 +1118,11 @@ static int vector_net_close(struct net_device *dev) os_close_file(vp->fds->tx_fd); vp->fds->tx_fd = -1; } - if (vp->bpf != NULL) - kfree(vp->bpf); - if (vp->fds->remote_addr != NULL) - kfree(vp->fds->remote_addr); - if (vp->transport_data != NULL) - kfree(vp->transport_data); - if (vp->header_rxbuffer != NULL) - kfree(vp->header_rxbuffer); - if (vp->header_txbuffer != NULL) - kfree(vp->header_txbuffer); + kfree(vp->bpf); + kfree(vp->fds->remote_addr); + kfree(vp->transport_data); + kfree(vp->header_rxbuffer); + kfree(vp->header_txbuffer); if (vp->rx_queue != NULL) destroy_queue(vp->rx_queue); if (vp->tx_queue != NULL) @@ -1580,7 +1575,7 @@ static int __init vector_setup(char *str) str, error); return 1; } - new = alloc_bootmem(sizeof(*new)); + new = memblock_alloc(sizeof(*new), SMP_CACHE_BYTES); INIT_LIST_HEAD(&new->list); new->unit = n; new->arguments = str; diff --git a/arch/um/drivers/vector_user.c b/arch/um/drivers/vector_user.c index 4d6a78e31089..d2c17dd74620 100644 --- a/arch/um/drivers/vector_user.c +++ b/arch/um/drivers/vector_user.c @@ -25,11 +25,13 @@ #include <linux/if_packet.h> #include <sys/socket.h> #include <sys/wait.h> +#include <sys/uio.h> #include <linux/virtio_net.h> #include <netdb.h> #include <stdlib.h> #include <os.h> #include <um_malloc.h> +#include <sys/uio.h> #include "vector_user.h" #define ID_GRE 0 @@ -267,8 +269,7 @@ cleanup: os_close_file(rxfd); if (txfd >= 0) os_close_file(txfd); - if (result != NULL) - kfree(result); + kfree(result); return NULL; } @@ -434,8 +435,7 @@ cleanup: if (fd >= 0) os_close_file(fd); if (result != NULL) { - if (result->remote_addr != NULL) - kfree(result->remote_addr); + kfree(result->remote_addr); kfree(result); } return NULL; diff --git a/arch/um/include/asm/Kbuild b/arch/um/include/asm/Kbuild index b10dde6cb793..00bcbe2326d9 100644 --- a/arch/um/include/asm/Kbuild +++ b/arch/um/include/asm/Kbuild @@ -10,9 +10,7 @@ generic-y += exec.h generic-y += extable.h generic-y += ftrace.h generic-y += futex.h -generic-y += hardirq.h generic-y += hw_irq.h -generic-y += io.h generic-y += irq_regs.h generic-y += irq_work.h generic-y += kdebug.h diff --git a/arch/um/include/asm/a.out-core.h b/arch/um/include/asm/a.out-core.h deleted file mode 100644 index 995643b18309..000000000000 --- a/arch/um/include/asm/a.out-core.h +++ /dev/null @@ -1,27 +0,0 @@ -/* a.out coredump register dumper - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells ([email protected]) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. - */ - -#ifndef __UM_A_OUT_CORE_H -#define __UM_A_OUT_CORE_H - -#ifdef __KERNEL__ - -#include <linux/user.h> - -/* - * fill in the user structure for an a.out core dump - */ -static inline void aout_dump_thread(struct pt_regs *regs, struct user *u) -{ -} - -#endif /* __KERNEL__ */ -#endif /* __UM_A_OUT_CORE_H */ diff --git a/arch/um/include/asm/pgalloc.h b/arch/um/include/asm/pgalloc.h index bf90b2aa2002..99eb5682792a 100644 --- a/arch/um/include/asm/pgalloc.h +++ b/arch/um/include/asm/pgalloc.h @@ -25,8 +25,8 @@ extern pgd_t *pgd_alloc(struct mm_struct *); extern void pgd_free(struct mm_struct *mm, pgd_t *pgd); -extern pte_t *pte_alloc_one_kernel(struct mm_struct *, unsigned long); -extern pgtable_t pte_alloc_one(struct mm_struct *, unsigned long); +extern pte_t *pte_alloc_one_kernel(struct mm_struct *); +extern pgtable_t pte_alloc_one(struct mm_struct *); static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte) { diff --git a/arch/um/include/asm/pgtable.h b/arch/um/include/asm/pgtable.h index 7485398d0737..9c04562310b3 100644 --- a/arch/um/include/asm/pgtable.h +++ b/arch/um/include/asm/pgtable.h @@ -197,12 +197,17 @@ static inline pte_t pte_mkold(pte_t pte) static inline pte_t pte_wrprotect(pte_t pte) { - pte_clear_bits(pte, _PAGE_RW); + if (likely(pte_get_bits(pte, _PAGE_RW))) + pte_clear_bits(pte, _PAGE_RW); + else + return pte; return(pte_mknewprot(pte)); } static inline pte_t pte_mkread(pte_t pte) { + if (unlikely(pte_get_bits(pte, _PAGE_USER))) + return pte; pte_set_bits(pte, _PAGE_USER); return(pte_mknewprot(pte)); } @@ -221,6 +226,8 @@ static inline pte_t pte_mkyoung(pte_t pte) static inline pte_t pte_mkwrite(pte_t pte) { + if (unlikely(pte_get_bits(pte, _PAGE_RW))) + return pte; pte_set_bits(pte, _PAGE_RW); return(pte_mknewprot(pte)); } diff --git a/arch/um/include/shared/aio.h b/arch/um/include/shared/aio.h deleted file mode 100644 index 423bae9153f8..000000000000 --- a/arch/um/include/shared/aio.h +++ /dev/null @@ -1,28 +0,0 @@ -/* - * Copyright (C) 2004 Jeff Dike ([email protected]) - * Licensed under the GPL - */ - -#ifndef AIO_H__ -#define AIO_H__ - -enum aio_type { AIO_READ, AIO_WRITE, AIO_MMAP }; - -struct aio_thread_reply { - void *data; - int err; -}; - -struct aio_context { - int reply_fd; - struct aio_context *next; -}; - -#define INIT_AIO_CONTEXT { .reply_fd = -1, \ - .next = NULL } - -extern int submit_aio(enum aio_type type, int fd, char *buf, int len, - unsigned long long offset, int reply_fd, - struct aio_context *aio); - -#endif diff --git a/arch/um/include/shared/irq_user.h b/arch/um/include/shared/irq_user.h index a7a6120f19d5..e7242a0ae489 100644 --- a/arch/um/include/shared/irq_user.h +++ b/arch/um/include/shared/irq_user.h @@ -31,7 +31,6 @@ struct irq_fd { struct siginfo; extern void sigio_handler(int sig, struct siginfo *unused_si, struct uml_pt_regs *regs); extern void free_irq_by_fd(int fd); -extern void reactivate_fd(int fd, int irqnum); extern void deactivate_fd(int fd, int irqnum); extern int deactivate_all_fds(void); extern int activate_ipi(int fd, int pid); diff --git a/arch/um/include/shared/os.h b/arch/um/include/shared/os.h index 048ae37eb5aa..ebf23012a59b 100644 --- a/arch/um/include/shared/os.h +++ b/arch/um/include/shared/os.h @@ -175,6 +175,7 @@ extern int os_fchange_dir(int fd); extern unsigned os_major(unsigned long long dev); extern unsigned os_minor(unsigned long long dev); extern unsigned long long os_makedev(unsigned major, unsigned minor); +extern int os_falloc_punch(int fd, unsigned long long offset, int count); /* start_up.c */ extern void os_early_checks(void); diff --git a/arch/um/kernel/initrd.c b/arch/um/kernel/initrd.c index 6f6e7896e53f..ce169ea87e61 100644 --- a/arch/um/kernel/initrd.c +++ b/arch/um/kernel/initrd.c @@ -4,7 +4,7 @@ */ #include <linux/init.h> -#include <linux/bootmem.h> +#include <linux/memblock.h> #include <linux/initrd.h> #include <asm/types.h> #include <init.h> @@ -36,7 +36,7 @@ int __init read_initrd(void) return 0; } - area = alloc_bootmem(size); + area = memblock_alloc(size, SMP_CACHE_BYTES); if (load_initrd(initrd, area, size) == -1) return 0; diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c index 6b7f3827d6e4..f4874b7ec503 100644 --- a/arch/um/kernel/irq.c +++ b/arch/um/kernel/irq.c @@ -244,8 +244,7 @@ static void garbage_collect_irq_entries(void) to_free = NULL; } walk = walk->next; - if (to_free != NULL) - kfree(to_free); + kfree(to_free); } } @@ -351,11 +350,6 @@ static void free_irq_by_irq_and_dev(unsigned int irq, void *dev) } -void reactivate_fd(int fd, int irqnum) -{ - /** NOP - we do auto-EOI now **/ -} - void deactivate_fd(int fd, int irqnum) { struct irq_entry *to_free; @@ -450,7 +444,6 @@ int um_request_irq(unsigned int irq, int fd, int type, } EXPORT_SYMBOL(um_request_irq); -EXPORT_SYMBOL(reactivate_fd); /* * irq_chip must define at least enable/disable and ack when diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c index 3c0e470ea646..799b571a8f88 100644 --- a/arch/um/kernel/mem.c +++ b/arch/um/kernel/mem.c @@ -5,7 +5,7 @@ #include <linux/stddef.h> #include <linux/module.h> -#include <linux/bootmem.h> +#include <linux/memblock.h> #include <linux/highmem.h> #include <linux/mm.h> #include <linux/swap.h> @@ -46,13 +46,13 @@ void __init mem_init(void) */ brk_end = (unsigned long) UML_ROUND_UP(sbrk(0)); map_memory(brk_end, __pa(brk_end), uml_reserved - brk_end, 1, 1, 0); - free_bootmem(__pa(brk_end), uml_reserved - brk_end); + memblock_free(__pa(brk_end), uml_reserved - brk_end); uml_reserved = brk_end; /* this will put all low memory onto the freelists */ - free_all_bootmem(); - max_low_pfn = totalram_pages; - max_pfn = totalram_pages; + memblock_free_all(); + max_low_pfn = totalram_pages(); + max_pfn = max_low_pfn; mem_init_print_info(NULL); kmalloc_ok = 1; } @@ -64,7 +64,8 @@ void __init mem_init(void) static void __init one_page_table_init(pmd_t *pmd) { if (pmd_none(*pmd)) { - pte_t *pte = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE); + pte_t *pte = (pte_t *) memblock_alloc_low(PAGE_SIZE, + PAGE_SIZE); set_pmd(pmd, __pmd(_KERNPG_TABLE + (unsigned long) __pa(pte))); if (pte != pte_offset_kernel(pmd, 0)) @@ -75,7 +76,7 @@ static void __init one_page_table_init(pmd_t *pmd) static void __init one_md_table_init(pud_t *pud) { #ifdef CONFIG_3_LEVEL_PGTABLES - pmd_t *pmd_table = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE); + pmd_t *pmd_table = (pmd_t *) memblock_alloc_low(PAGE_SIZE, PAGE_SIZE); set_pud(pud, __pud(_KERNPG_TABLE + (unsigned long) __pa(pmd_table))); if (pmd_table != pmd_offset(pud, 0)) BUG(); @@ -124,7 +125,7 @@ static void __init fixaddr_user_init( void) return; fixrange_init( FIXADDR_USER_START, FIXADDR_USER_END, swapper_pg_dir); - v = (unsigned long) alloc_bootmem_low_pages(size); + v = (unsigned long) memblock_alloc_low(size, PAGE_SIZE); memcpy((void *) v , (void *) FIXADDR_USER_START, size); p = __pa(v); for ( ; size > 0; size -= PAGE_SIZE, vaddr += PAGE_SIZE, @@ -143,7 +144,8 @@ void __init paging_init(void) unsigned long zones_size[MAX_NR_ZONES], vaddr; int i; - empty_zero_page = (unsigned long *) alloc_bootmem_low_pages(PAGE_SIZE); + empty_zero_page = (unsigned long *) memblock_alloc_low(PAGE_SIZE, + PAGE_SIZE); for (i = 0; i < ARRAY_SIZE(zones_size); i++) zones_size[i] = 0; @@ -197,7 +199,7 @@ void pgd_free(struct mm_struct *mm, pgd_t *pgd) free_page((unsigned long) pgd); } -pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) +pte_t *pte_alloc_one_kernel(struct mm_struct *mm) { pte_t *pte; @@ -205,7 +207,7 @@ pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) return pte; } -pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address) +pgtable_t pte_alloc_one(struct mm_struct *mm) { struct page *pte; diff --git a/arch/um/kernel/physmem.c b/arch/um/kernel/physmem.c index f02596e9931d..5bf56af4d5b9 100644 --- a/arch/um/kernel/physmem.c +++ b/arch/um/kernel/physmem.c @@ -4,7 +4,7 @@ */ #include <linux/module.h> -#include <linux/bootmem.h> +#include <linux/memblock.h> #include <linux/mm.h> #include <linux/pfn.h> #include <asm/page.h> @@ -80,28 +80,23 @@ void __init setup_physmem(unsigned long start, unsigned long reserve_end, unsigned long len, unsigned long long highmem) { unsigned long reserve = reserve_end - start; - unsigned long pfn = PFN_UP(__pa(reserve_end)); - unsigned long delta = (len - reserve) >> PAGE_SHIFT; - unsigned long offset, bootmap_size; - long map_size; + long map_size = len - reserve; int err; - offset = uml_reserved - uml_physmem; - map_size = len - offset; if(map_size <= 0) { os_warn("Too few physical memory! Needed=%lu, given=%lu\n", - offset, len); + reserve, len); exit(1); } physmem_fd = create_mem_file(len + highmem); - err = os_map_memory((void *) uml_reserved, physmem_fd, offset, + err = os_map_memory((void *) reserve_end, physmem_fd, reserve, map_size, 1, 1, 1); if (err < 0) { os_warn("setup_physmem - mapping %ld bytes of memory at 0x%p " "failed - errno = %d\n", map_size, - (void *) uml_reserved, err); + (void *) reserve_end, err); exit(1); } @@ -113,9 +108,11 @@ void __init setup_physmem(unsigned long start, unsigned long reserve_end, os_write_file(physmem_fd, __syscall_stub_start, PAGE_SIZE); os_fsync_file(physmem_fd); - bootmap_size = init_bootmem(pfn, pfn + delta); - free_bootmem(__pa(reserve_end) + bootmap_size, - len - bootmap_size - reserve); + memblock_add(__pa(start), len + highmem); + memblock_reserve(__pa(start), reserve); + + min_low_pfn = PFN_UP(__pa(reserve_end)); + max_low_pfn = min_low_pfn + (map_size >> PAGE_SHIFT); } int phys_mapping(unsigned long phys, unsigned long long *offset_out) diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c index 1a1d88a4d940..5f47422401e1 100644 --- a/arch/um/kernel/ptrace.c +++ b/arch/um/kernel/ptrace.c @@ -66,7 +66,7 @@ long arch_ptrace(struct task_struct *child, long request, #ifdef PTRACE_GETREGS case PTRACE_GETREGS: { /* Get all gp regs from the child. */ - if (!access_ok(VERIFY_WRITE, p, MAX_REG_OFFSET)) { + if (!access_ok(p, MAX_REG_OFFSET)) { ret = -EIO; break; } @@ -81,7 +81,7 @@ long arch_ptrace(struct task_struct *child, long request, #ifdef PTRACE_SETREGS case PTRACE_SETREGS: { /* Set all gp regs in the child. */ unsigned long tmp = 0; - if (!access_ok(VERIFY_READ, p, MAX_REG_OFFSET)) { + if (!access_ok(p, MAX_REG_OFFSET)) { ret = -EIO; break; } diff --git a/arch/um/kernel/sigio.c b/arch/um/kernel/sigio.c index b5e0cbb34382..3fb6a4041ed6 100644 --- a/arch/um/kernel/sigio.c +++ b/arch/um/kernel/sigio.c @@ -16,7 +16,6 @@ static irqreturn_t sigio_interrupt(int irq, void *data) char c; os_read_file(sigio_irq_fd, &c, sizeof(c)); - reactivate_fd(sigio_irq_fd, SIGIO_WRITE_IRQ); return IRQ_HANDLED; } diff --git a/arch/um/kernel/skas/uaccess.c b/arch/um/kernel/skas/uaccess.c index d450797a3a7c..7f06fdbc7ee1 100644 --- a/arch/um/kernel/skas/uaccess.c +++ b/arch/um/kernel/skas/uaccess.c @@ -62,27 +62,28 @@ static int do_op_one_page(unsigned long addr, int len, int is_write, jmp_buf buf; struct page *page; pte_t *pte; - int n, faulted; + int n; pte = maybe_map(addr, is_write); if (pte == NULL) return -1; page = pte_page(*pte); +#ifdef CONFIG_64BIT + pagefault_disable(); + addr = (unsigned long) page_address(page) + + (addr & ~PAGE_MASK); +#else addr = (unsigned long) kmap_atomic(page) + (addr & ~PAGE_MASK); +#endif + n = (*op)(addr, len, arg); - current->thread.fault_catcher = &buf; - - faulted = UML_SETJMP(&buf); - if (faulted == 0) - n = (*op)(addr, len, arg); - else - n = -1; - - current->thread.fault_catcher = NULL; - +#ifdef CONFIG_64BIT + pagefault_enable(); +#else kunmap_atomic((void *)addr); +#endif return n; } diff --git a/arch/um/kernel/tlb.c b/arch/um/kernel/tlb.c index 37508b190106..8347161c2ae0 100644 --- a/arch/um/kernel/tlb.c +++ b/arch/um/kernel/tlb.c @@ -37,17 +37,19 @@ struct host_vm_change { } mprotect; } u; } ops[1]; + int userspace; int index; - struct mm_id *id; + struct mm_struct *mm; void *data; int force; }; -#define INIT_HVC(mm, force) \ +#define INIT_HVC(mm, force, userspace) \ ((struct host_vm_change) \ { .ops = { { .type = NONE } }, \ - .id = &mm->context.id, \ + .mm = mm, \ .data = NULL, \ + .userspace = userspace, \ .index = 0, \ .force = force }) @@ -68,18 +70,40 @@ static int do_ops(struct host_vm_change *hvc, int end, op = &hvc->ops[i]; switch (op->type) { case MMAP: - ret = map(hvc->id, op->u.mmap.addr, op->u.mmap.len, - op->u.mmap.prot, op->u.mmap.fd, - op->u.mmap.offset, finished, &hvc->data); + if (hvc->userspace) + ret = map(&hvc->mm->context.id, op->u.mmap.addr, + op->u.mmap.len, op->u.mmap.prot, + op->u.mmap.fd, + op->u.mmap.offset, finished, + &hvc->data); + else + map_memory(op->u.mmap.addr, op->u.mmap.offset, + op->u.mmap.len, 1, 1, 1); break; case MUNMAP: - ret = unmap(hvc->id, op->u.munmap.addr, - op->u.munmap.len, finished, &hvc->data); + if (hvc->userspace) + ret = unmap(&hvc->mm->context.id, + op->u.munmap.addr, + op->u.munmap.len, finished, + &hvc->data); + else + ret = os_unmap_memory( + (void *) op->u.munmap.addr, + op->u.munmap.len); + break; case MPROTECT: - ret = protect(hvc->id, op->u.mprotect.addr, - op->u.mprotect.len, op->u.mprotect.prot, - finished, &hvc->data); + if (hvc->userspace) + ret = protect(&hvc->mm->context.id, + op->u.mprotect.addr, + op->u.mprotect.len, + op->u.mprotect.prot, + finished, &hvc->data); + else + ret = os_protect_memory( + (void *) op->u.mprotect.addr, + op->u.mprotect.len, + 1, 1, 1); break; default: printk(KERN_ERR "Unknown op type %d in do_ops\n", @@ -100,9 +124,12 @@ static int add_mmap(unsigned long virt, unsigned long phys, unsigned long len, { __u64 offset; struct host_vm_op *last; - int fd, ret = 0; + int fd = -1, ret = 0; - fd = phys_mapping(phys, &offset); + if (hvc->userspace) + fd = phys_mapping(phys, &offset); + else + offset = phys; if (hvc->index != 0) { last = &hvc->ops[hvc->index - 1]; if ((last->type == MMAP) && @@ -215,10 +242,11 @@ static inline int update_pte_range(pmd_t *pmd, unsigned long addr, prot = ((r ? UM_PROT_READ : 0) | (w ? UM_PROT_WRITE : 0) | (x ? UM_PROT_EXEC : 0)); if (hvc->force || pte_newpage(*pte)) { - if (pte_present(*pte)) - ret = add_mmap(addr, pte_val(*pte) & PAGE_MASK, - PAGE_SIZE, prot, hvc); - else + if (pte_present(*pte)) { + if (pte_newpage(*pte)) + ret = add_mmap(addr, pte_val(*pte) & PAGE_MASK, + PAGE_SIZE, prot, hvc); + } else ret = add_munmap(addr, PAGE_SIZE, hvc); } else if (pte_newprot(*pte)) ret = add_mprotect(addr, PAGE_SIZE, prot, hvc); @@ -277,9 +305,9 @@ void fix_range_common(struct mm_struct *mm, unsigned long start_addr, pgd_t *pgd; struct host_vm_change hvc; unsigned long addr = start_addr, next; - int ret = 0; + int ret = 0, userspace = 1; - hvc = INIT_HVC(mm, force); + hvc = INIT_HVC(mm, force, userspace); pgd = pgd_offset(mm, addr); do { next = pgd_addr_end(addr, end_addr); @@ -314,9 +342,11 @@ static int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) pmd_t *pmd; pte_t *pte; unsigned long addr, last; - int updated = 0, err; + int updated = 0, err = 0, force = 0, userspace = 0; + struct host_vm_change hvc; mm = &init_mm; + hvc = INIT_HVC(mm, force, userspace); for (addr = start; addr < end;) { pgd = pgd_offset(mm, addr); if (!pgd_present(*pgd)) { @@ -325,8 +355,7 @@ static int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) last = end; if (pgd_newpage(*pgd)) { updated = 1; - err = os_unmap_memory((void *) addr, - last - addr); + err = add_munmap(addr, last - addr, &hvc); if (err < 0) panic("munmap failed, errno = %d\n", -err); @@ -342,8 +371,7 @@ static int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) last = end; if (pud_newpage(*pud)) { updated = 1; - err = os_unmap_memory((void *) addr, - last - addr); + err = add_munmap(addr, last - addr, &hvc); if (err < 0) panic("munmap failed, errno = %d\n", -err); @@ -359,8 +387,7 @@ static int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) last = end; if (pmd_newpage(*pmd)) { updated = 1; - err = os_unmap_memory((void *) addr, - last - addr); + err = add_munmap(addr, last - addr, &hvc); if (err < 0) panic("munmap failed, errno = %d\n", -err); @@ -372,22 +399,25 @@ static int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) pte = pte_offset_kernel(pmd, addr); if (!pte_present(*pte) || pte_newpage(*pte)) { updated = 1; - err = os_unmap_memory((void *) addr, - PAGE_SIZE); + err = add_munmap(addr, PAGE_SIZE, &hvc); if (err < 0) panic("munmap failed, errno = %d\n", -err); if (pte_present(*pte)) - map_memory(addr, - pte_val(*pte) & PAGE_MASK, - PAGE_SIZE, 1, 1, 1); + err = add_mmap(addr, pte_val(*pte) & PAGE_MASK, + PAGE_SIZE, 0, &hvc); } else if (pte_newprot(*pte)) { updated = 1; - os_protect_memory((void *) addr, PAGE_SIZE, 1, 1, 1); + err = add_mprotect(addr, PAGE_SIZE, 0, &hvc); } addr += PAGE_SIZE; } + if (!err) + err = do_ops(&hvc, hvc.index, 1); + + if (err < 0) + panic("flush_tlb_kernel failed, errno = %d\n", err); return updated; } @@ -491,6 +521,13 @@ pte_t *addr_pte(struct task_struct *task, unsigned long addr) void flush_tlb_all(void) { + /* + * Don't bother flushing if this address space is about to be + * destroyed. + */ + if (atomic_read(¤t->mm->mm_users) == 0) + return; + flush_tlb_mm(current->mm); } @@ -512,6 +549,13 @@ void __flush_tlb_one(unsigned long addr) static void fix_range(struct mm_struct *mm, unsigned long start_addr, unsigned long end_addr, int force) { + /* + * Don't bother flushing if this address space is about to be + * destroyed. + */ + if (atomic_read(&mm->mm_users) == 0) + return; + fix_range_common(mm, start_addr, end_addr, force); } @@ -527,13 +571,6 @@ EXPORT_SYMBOL(flush_tlb_range); void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, unsigned long end) { - /* - * Don't bother flushing if this address space is about to be - * destroyed. - */ - if (atomic_read(&mm->mm_users) == 0) - return; - fix_range(mm, start, end, 0); } diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c index cced82946042..0e8b6158f224 100644 --- a/arch/um/kernel/trap.c +++ b/arch/um/kernel/trap.c @@ -19,7 +19,7 @@ #include <skas.h> /* - * Note this is constrained to return 0, -EFAULT, -EACCESS, -ENOMEM by + * Note this is constrained to return 0, -EFAULT, -EACCES, -ENOMEM by * segv(). */ int handle_page_fault(unsigned long address, unsigned long ip, diff --git a/arch/um/os-Linux/Makefile b/arch/um/os-Linux/Makefile index ada473bf6f46..455b500afe97 100644 --- a/arch/um/os-Linux/Makefile +++ b/arch/um/os-Linux/Makefile @@ -6,18 +6,14 @@ # Don't instrument UML-specific code KCOV_INSTRUMENT := n -obj-y = aio.o execvp.o file.o helper.o irq.o main.o mem.o process.o \ +obj-y = execvp.o file.o helper.o irq.o main.o mem.o process.o \ registers.o sigio.o signal.o start_up.o time.o tty.o \ umid.o user_syms.o util.o drivers/ skas/ obj-$(CONFIG_ARCH_REUSE_HOST_VSYSCALL_AREA) += elf_aux.o -USER_OBJS := $(user-objs-y) aio.o elf_aux.o execvp.o file.o helper.o irq.o \ +USER_OBJS := $(user-objs-y) elf_aux.o execvp.o file.o helper.o irq.o \ main.o mem.o process.o registers.o sigio.o signal.o start_up.o time.o \ tty.o umid.o util.o -HAVE_AIO_ABI := $(shell [ -r /usr/include/linux/aio_abi.h ] && \ - echo -DHAVE_AIO_ABI ) -CFLAGS_aio.o += $(HAVE_AIO_ABI) - include arch/um/scripts/Makefile.rules diff --git a/arch/um/os-Linux/aio.c b/arch/um/os-Linux/aio.c deleted file mode 100644 index 014eb35fd13b..000000000000 --- a/arch/um/os-Linux/aio.c +++ /dev/null @@ -1,390 +0,0 @@ -/* - * Copyright (C) 2004 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) - * Licensed under the GPL - */ - -#include <unistd.h> -#include <sched.h> -#include <signal.h> -#include <errno.h> -#include <sys/time.h> -#include <asm/unistd.h> -#include <aio.h> -#include <init.h> -#include <kern_util.h> -#include <os.h> - -struct aio_thread_req { - enum aio_type type; - int io_fd; - unsigned long long offset; - char *buf; - int len; - struct aio_context *aio; -}; - -#if defined(HAVE_AIO_ABI) -#include <linux/aio_abi.h> - -/* - * If we have the headers, we are going to build with AIO enabled. - * If we don't have aio in libc, we define the necessary stubs here. - */ - -#if !defined(HAVE_AIO_LIBC) - -static long io_setup(int n, aio_context_t *ctxp) -{ - return syscall(__NR_io_setup, n, ctxp); -} - -static long io_submit(aio_context_t ctx, long nr, struct iocb **iocbpp) -{ - return syscall(__NR_io_submit, ctx, nr, iocbpp); -} - -static long io_getevents(aio_context_t ctx_id, long min_nr, long nr, - struct io_event *events, struct timespec *timeout) -{ - return syscall(__NR_io_getevents, ctx_id, min_nr, nr, events, timeout); -} - -#endif - -/* - * The AIO_MMAP cases force the mmapped page into memory here - * rather than in whatever place first touches the data. I used - * to do this by touching the page, but that's delicate because - * gcc is prone to optimizing that away. So, what's done here - * is we read from the descriptor from which the page was - * mapped. The caller is required to pass an offset which is - * inside the page that was mapped. Thus, when the read - * returns, we know that the page is in the page cache, and - * that it now backs the mmapped area. - */ - -static int do_aio(aio_context_t ctx, enum aio_type type, int fd, char *buf, - int len, unsigned long long offset, struct aio_context *aio) -{ - struct iocb *iocbp = & ((struct iocb) { - .aio_data = (unsigned long) aio, - .aio_fildes = fd, - .aio_buf = (unsigned long) buf, - .aio_nbytes = len, - .aio_offset = offset - }); - char c; - - switch (type) { - case AIO_READ: - iocbp->aio_lio_opcode = IOCB_CMD_PREAD; - break; - case AIO_WRITE: - iocbp->aio_lio_opcode = IOCB_CMD_PWRITE; - break; - case AIO_MMAP: - iocbp->aio_lio_opcode = IOCB_CMD_PREAD; - iocbp->aio_buf = (unsigned long) &c; - iocbp->aio_nbytes = sizeof(c); - break; - default: - printk(UM_KERN_ERR "Bogus op in do_aio - %d\n", type); - return -EINVAL; - } - - return (io_submit(ctx, 1, &iocbp) > 0) ? 0 : -errno; -} - -/* Initialized in an initcall and unchanged thereafter */ -static aio_context_t ctx = 0; - -static int aio_thread(void *arg) -{ - struct aio_thread_reply reply; - struct io_event event; - int err, n, reply_fd; - - os_fix_helper_signals(); - while (1) { - n = io_getevents(ctx, 1, 1, &event, NULL); - if (n < 0) { - if (errno == EINTR) - continue; - printk(UM_KERN_ERR "aio_thread - io_getevents failed, " - "errno = %d\n", errno); - } - else { - reply = ((struct aio_thread_reply) - { .data = (void *) (long) event.data, - .err = event.res }); - reply_fd = ((struct aio_context *) reply.data)->reply_fd; - err = write(reply_fd, &reply, sizeof(reply)); - if (err != sizeof(reply)) - printk(UM_KERN_ERR "aio_thread - write failed, " - "fd = %d, err = %d\n", reply_fd, errno); - } - } - return 0; -} - -#endif - -static int do_not_aio(struct aio_thread_req *req) -{ - char c; - unsigned long long actual; - int n; - - actual = lseek64(req->io_fd, req->offset, SEEK_SET); - if (actual != req->offset) - return -errno; - - switch (req->type) { - case AIO_READ: - n = read(req->io_fd, req->buf, req->len); - break; - case AIO_WRITE: - n = write(req->io_fd, req->buf, req->len); - break; - case AIO_MMAP: - n = read(req->io_fd, &c, sizeof(c)); - break; - default: - printk(UM_KERN_ERR "do_not_aio - bad request type : %d\n", - req->type); - return -EINVAL; - } - - if (n < 0) - return -errno; - return 0; -} - -/* These are initialized in initcalls and not changed */ -static int aio_req_fd_r = -1; -static int aio_req_fd_w = -1; -static int aio_pid = -1; -static unsigned long aio_stack; - -static int not_aio_thread(void *arg) -{ - struct aio_thread_req req; - struct aio_thread_reply reply; - int err; - - os_fix_helper_signals(); - while (1) { - err = read(aio_req_fd_r, &req, sizeof(req)); - if (err != sizeof(req)) { - if (err < 0) - printk(UM_KERN_ERR "not_aio_thread - " - "read failed, fd = %d, err = %d\n", - aio_req_fd_r, - errno); - else { - printk(UM_KERN_ERR "not_aio_thread - short " - "read, fd = %d, length = %d\n", - aio_req_fd_r, err); - } - continue; - } - err = do_not_aio(&req); - reply = ((struct aio_thread_reply) { .data = req.aio, - .err = err }); - err = write(req.aio->reply_fd, &reply, sizeof(reply)); - if (err != sizeof(reply)) - printk(UM_KERN_ERR "not_aio_thread - write failed, " - "fd = %d, err = %d\n", req.aio->reply_fd, errno); - } - - return 0; -} - -static int init_aio_24(void) -{ - int fds[2], err; - - err = os_pipe(fds, 1, 1); - if (err) - goto out; - - aio_req_fd_w = fds[0]; - aio_req_fd_r = fds[1]; - - err = os_set_fd_block(aio_req_fd_w, 0); - if (err) - goto out_close_pipe; - - err = run_helper_thread(not_aio_thread, NULL, - CLONE_FILES | CLONE_VM, &aio_stack); - if (err < 0) - goto out_close_pipe; - - aio_pid = err; - goto out; - -out_close_pipe: - close(fds[0]); - close(fds[1]); - aio_req_fd_w = -1; - aio_req_fd_r = -1; -out: -#ifndef HAVE_AIO_ABI - printk(UM_KERN_INFO "/usr/include/linux/aio_abi.h not present during " - "build\n"); -#endif - printk(UM_KERN_INFO "2.6 host AIO support not used - falling back to " - "I/O thread\n"); - return 0; -} - -#ifdef HAVE_AIO_ABI -#define DEFAULT_24_AIO 0 -static int init_aio_26(void) -{ - int err; - - if (io_setup(256, &ctx)) { - err = -errno; - printk(UM_KERN_ERR "aio_thread failed to initialize context, " - "err = %d\n", errno); - return err; - } - - err = run_helper_thread(aio_thread, NULL, - CLONE_FILES | CLONE_VM, &aio_stack); - if (err < 0) - return err; - - aio_pid = err; - - printk(UM_KERN_INFO "Using 2.6 host AIO\n"); - return 0; -} - -static int submit_aio_26(enum aio_type type, int io_fd, char *buf, int len, - unsigned long long offset, struct aio_context *aio) -{ - struct aio_thread_reply reply; - int err; - - err = do_aio(ctx, type, io_fd, buf, len, offset, aio); - if (err) { - reply = ((struct aio_thread_reply) { .data = aio, - .err = err }); - err = write(aio->reply_fd, &reply, sizeof(reply)); - if (err != sizeof(reply)) { - err = -errno; - printk(UM_KERN_ERR "submit_aio_26 - write failed, " - "fd = %d, err = %d\n", aio->reply_fd, -err); - } - else err = 0; - } - - return err; -} - -#else -#define DEFAULT_24_AIO 1 -static int init_aio_26(void) -{ - return -ENOSYS; -} - -static int submit_aio_26(enum aio_type type, int io_fd, char *buf, int len, - unsigned long long offset, struct aio_context *aio) -{ - return -ENOSYS; -} -#endif - -/* Initialized in an initcall and unchanged thereafter */ -static int aio_24 = DEFAULT_24_AIO; - -static int __init set_aio_24(char *name, int *add) -{ - aio_24 = 1; - return 0; -} - -__uml_setup("aio=2.4", set_aio_24, -"aio=2.4\n" -" This is used to force UML to use 2.4-style AIO even when 2.6 AIO is\n" -" available. 2.4 AIO is a single thread that handles one request at a\n" -" time, synchronously. 2.6 AIO is a thread which uses the 2.6 AIO \n" -" interface to handle an arbitrary number of pending requests. 2.6 AIO \n" -" is not available in tt mode, on 2.4 hosts, or when UML is built with\n" -" /usr/include/linux/aio_abi.h not available. Many distributions don't\n" -" include aio_abi.h, so you will need to copy it from a kernel tree to\n" -" your /usr/include/linux in order to build an AIO-capable UML\n\n" -); - -static int init_aio(void) -{ - int err; - - if (!aio_24) { - err = init_aio_26(); - if (err && (errno == ENOSYS)) { - printk(UM_KERN_INFO "2.6 AIO not supported on the " - "host - reverting to 2.4 AIO\n"); - aio_24 = 1; - } - else return err; - } - - if (aio_24) - return init_aio_24(); - - return 0; -} - -/* - * The reason for the __initcall/__uml_exitcall asymmetry is that init_aio - * needs to be called when the kernel is running because it calls run_helper, - * which needs get_free_page. exit_aio is a __uml_exitcall because the generic - * kernel does not run __exitcalls on shutdown, and can't because many of them - * break when called outside of module unloading. - */ -__initcall(init_aio); - -static void exit_aio(void) -{ - if (aio_pid != -1) { - os_kill_process(aio_pid, 1); - free_stack(aio_stack, 0); - } -} - -__uml_exitcall(exit_aio); - -static int submit_aio_24(enum aio_type type, int io_fd, char *buf, int len, - unsigned long long offset, struct aio_context *aio) -{ - struct aio_thread_req req = { .type = type, - .io_fd = io_fd, - .offset = offset, - .buf = buf, - .len = len, - .aio = aio, - }; - int err; - - err = write(aio_req_fd_w, &req, sizeof(req)); - if (err == sizeof(req)) - err = 0; - else err = -errno; - - return err; -} - -int submit_aio(enum aio_type type, int io_fd, char *buf, int len, - unsigned long long offset, int reply_fd, - struct aio_context *aio) -{ - aio->reply_fd = reply_fd; - if (aio_24) - return submit_aio_24(type, io_fd, buf, len, offset, aio); - else - return submit_aio_26(type, io_fd, buf, len, offset, aio); -} diff --git a/arch/um/os-Linux/file.c b/arch/um/os-Linux/file.c index c0197097c86e..f25b110d4e70 100644 --- a/arch/um/os-Linux/file.c +++ b/arch/um/os-Linux/file.c @@ -610,3 +610,13 @@ unsigned long long os_makedev(unsigned major, unsigned minor) { return makedev(major, minor); } + +int os_falloc_punch(int fd, unsigned long long offset, int len) +{ + int n = fallocate(fd, FALLOC_FL_PUNCH_HOLE|FALLOC_FL_KEEP_SIZE, offset, len); + + if (n < 0) + return -errno; + return n; +} + diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c index c94c3bd70ccd..df4a985716eb 100644 --- a/arch/um/os-Linux/skas/process.c +++ b/arch/um/os-Linux/skas/process.c @@ -610,6 +610,11 @@ int start_idle_thread(void *stack, jmp_buf *switch_buf) fatal_sigsegv(); } longjmp(*switch_buf, 1); + + /* unreachable */ + printk(UM_KERN_ERR "impossible long jump!"); + fatal_sigsegv(); + return 0; } void initial_thread_cb_skas(void (*proc)(void *), void *arg) |