diff options
Diffstat (limited to 'arch/um')
29 files changed, 401 insertions, 952 deletions
diff --git a/arch/um/Kconfig.um b/arch/um/Kconfig index 3e7f228b22e1..6b9938919f0b 100644 --- a/arch/um/Kconfig.um +++ b/arch/um/Kconfig @@ -1,4 +1,70 @@ # SPDX-License-Identifier: GPL-2.0 + +menu "UML-specific options" + +config UML + bool + default y + select ARCH_HAS_KCOV + select ARCH_NO_PREEMPT + select HAVE_ARCH_AUDITSYSCALL + select HAVE_ARCH_SECCOMP_FILTER + select HAVE_UID16 + select HAVE_FUTEX_CMPXCHG if FUTEX + select HAVE_DEBUG_KMEMLEAK + select GENERIC_IRQ_SHOW + select GENERIC_CPU_DEVICES + select GENERIC_CLOCKEVENTS + select HAVE_GCC_PLUGINS + select TTY # Needed for line.c + +config MMU + bool + default y + +config NO_IOMEM + def_bool y + +config ISA + bool + +config SBUS + bool + +config PCI + bool + +config PCMCIA + bool + +config TRACE_IRQFLAGS_SUPPORT + bool + default y + +config LOCKDEP_SUPPORT + bool + default y + +config STACKTRACE_SUPPORT + bool + default y + select STACKTRACE + +config GENERIC_CALIBRATE_DELAY + bool + default y + +config HZ + int + default 100 + +config NR_CPUS + int + range 1 1 + default 1 + +source "arch/$(HEADER_ARCH)/um/Kconfig" + config STATIC_LINK bool "Force a static link" default n @@ -10,8 +76,6 @@ config STATIC_LINK Additionally, this option enables using higher memory spaces (up to 2.75G) for UML. -source "mm/Kconfig" - config LD_SCRIPT_STATIC bool default y @@ -23,8 +87,6 @@ config LD_SCRIPT_DYN depends on !LD_SCRIPT_STATIC select MODULE_REL_CRCS if MODVERSIONS -source "fs/Kconfig.binfmt" - config HOSTFS tristate "Host filesystem" help @@ -80,7 +142,7 @@ config MAGIC_SYSRQ On UML, this is accomplished by sending a "sysrq" command with mconsole, followed by the letter for the requested command. - The keys are documented in <file:Documentation/sysrq.txt>. Don't say Y + The keys are documented in <file:Documentation/admin-guide/sysrq.rst>. Don't say Y unless you really know what this hack does. config KERNEL_STACK_ORDER @@ -122,3 +184,7 @@ config SECCOMP defined by each seccomp mode. If unsure, say Y. + +endmenu + +source "arch/um/drivers/Kconfig" diff --git a/arch/um/Kconfig.char b/arch/um/Kconfig.char deleted file mode 100644 index f184bde7030e..000000000000 --- a/arch/um/Kconfig.char +++ /dev/null @@ -1,124 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0 -menu "UML Character Devices" - -config STDERR_CONSOLE - bool "stderr console" - default y - help - console driver which dumps all printk messages to stderr. - -config SSL - bool "Virtual serial line" - help - The User-Mode Linux environment allows you to create virtual serial - lines on the UML that are usually made to show up on the host as - ttys or ptys. - - See <http://user-mode-linux.sourceforge.net/old/input.html> for more - information and command line examples of how to use this facility. - - Unless you have a specific reason for disabling this, say Y. - -config NULL_CHAN - bool "null channel support" - help - This option enables support for attaching UML consoles and serial - lines to a device similar to /dev/null. Data written to it disappears - and there is never any data to be read. - -config PORT_CHAN - bool "port channel support" - help - This option enables support for attaching UML consoles and serial - lines to host portals. They may be accessed with 'telnet <host> - <port number>'. Any number of consoles and serial lines may be - attached to a single portal, although what UML device you get when - you telnet to that portal will be unpredictable. - It is safe to say 'Y' here. - -config PTY_CHAN - bool "pty channel support" - help - This option enables support for attaching UML consoles and serial - lines to host pseudo-terminals. Access to both traditional - pseudo-terminals (/dev/pty*) and pts pseudo-terminals are controlled - with this option. The assignment of UML devices to host devices - will be announced in the kernel message log. - It is safe to say 'Y' here. - -config TTY_CHAN - bool "tty channel support" - help - This option enables support for attaching UML consoles and serial - lines to host terminals. Access to both virtual consoles - (/dev/tty*) and the slave side of pseudo-terminals (/dev/ttyp* and - /dev/pts/*) are controlled by this option. - It is safe to say 'Y' here. - -config XTERM_CHAN - bool "xterm channel support" - help - This option enables support for attaching UML consoles and serial - lines to xterms. Each UML device so assigned will be brought up in - its own xterm. - It is safe to say 'Y' here. - -config NOCONFIG_CHAN - bool - default !(XTERM_CHAN && TTY_CHAN && PTY_CHAN && PORT_CHAN && NULL_CHAN) - -config CON_ZERO_CHAN - string "Default main console channel initialization" - default "fd:0,fd:1" - help - This is the string describing the channel to which the main console - will be attached by default. This value can be overridden from the - command line. The default value is "fd:0,fd:1", which attaches the - main console to stdin and stdout. - It is safe to leave this unchanged. - -config CON_CHAN - string "Default console channel initialization" - default "xterm" - help - This is the string describing the channel to which all consoles - except the main console will be attached by default. This value can - be overridden from the command line. The default value is "xterm", - which brings them up in xterms. - It is safe to leave this unchanged, although you may wish to change - this if you expect the UML that you build to be run in environments - which don't have X or xterm available. - -config SSL_CHAN - string "Default serial line channel initialization" - default "pty" - help - This is the string describing the channel to which the serial lines - will be attached by default. This value can be overridden from the - command line. The default value is "pty", which attaches them to - traditional pseudo-terminals. - It is safe to leave this unchanged, although you may wish to change - this if you expect the UML that you build to be run in environments - which don't have a set of /dev/pty* devices. - -config UML_SOUND - tristate "Sound support" - help - This option enables UML sound support. If enabled, it will pull in - soundcore and the UML hostaudio relay, which acts as a intermediary - between the host's dsp and mixer devices and the UML sound system. - It is safe to say 'Y' here. - -config SOUND - tristate - default UML_SOUND - -config SOUND_OSS_CORE - bool - default UML_SOUND - -config HOSTAUDIO - tristate - default UML_SOUND - -endmenu diff --git a/arch/um/Kconfig.common b/arch/um/Kconfig.common deleted file mode 100644 index c68add8df3ae..000000000000 --- a/arch/um/Kconfig.common +++ /dev/null @@ -1,64 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0 -config UML - bool - default y - select ARCH_HAS_KCOV - select HAVE_ARCH_AUDITSYSCALL - select HAVE_ARCH_SECCOMP_FILTER - select HAVE_UID16 - select HAVE_FUTEX_CMPXCHG if FUTEX - select HAVE_DEBUG_KMEMLEAK - select GENERIC_IRQ_SHOW - select GENERIC_CPU_DEVICES - select GENERIC_CLOCKEVENTS - select HAVE_GCC_PLUGINS - select TTY # Needed for line.c - -config MMU - bool - default y - -config NO_IOMEM - def_bool y - -config ISA - bool - -config SBUS - bool - -config PCI - bool - -config PCMCIA - bool - -config TRACE_IRQFLAGS_SUPPORT - bool - default y - -config LOCKDEP_SUPPORT - bool - default y - -config STACKTRACE_SUPPORT - bool - default y - select STACKTRACE - -config GENERIC_CALIBRATE_DELAY - bool - default y - -config HZ - int - default 100 - -config SUBARCH - string - option env="SUBARCH" - -config NR_CPUS - int - range 1 1 - default 1 diff --git a/arch/um/Kconfig.debug b/arch/um/Kconfig.debug index 967d3109689f..2014597605ea 100644 --- a/arch/um/Kconfig.debug +++ b/arch/um/Kconfig.debug @@ -1,7 +1,4 @@ # SPDX-License-Identifier: GPL-2.0 -menu "Kernel hacking" - -source "lib/Kconfig.debug" config GPROF bool "Enable gprof support" @@ -37,5 +34,3 @@ config EARLY_PRINTK This is useful for kernel debugging when your machine crashes very early before the console code is initialized. - -endmenu diff --git a/arch/um/Kconfig.rest b/arch/um/Kconfig.rest deleted file mode 100644 index 08327b9c0cbe..000000000000 --- a/arch/um/Kconfig.rest +++ /dev/null @@ -1,22 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0 -source "init/Kconfig" - -source "kernel/Kconfig.freezer" - -source "arch/um/Kconfig.char" - -source "drivers/Kconfig" - -source "net/Kconfig" - -source "arch/um/Kconfig.net" - -source "fs/Kconfig" - -source "security/Kconfig" - -source "crypto/Kconfig" - -source "lib/Kconfig" - -source "arch/um/Kconfig.debug" diff --git a/arch/um/Makefile b/arch/um/Makefile index e54dda8a0363..ab1066c38944 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -113,17 +113,8 @@ define archhelp echo ' find in the kernel root.' endef -KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig - archheaders: - $(Q)$(MAKE) -f $(srctree)/scripts/Makefile.asm-generic \ - kbuild-file=$(HOST_DIR)/include/asm/Kbuild \ - obj=$(HOST_DIR)/include/generated/asm - $(Q)$(MAKE) -f $(srctree)/scripts/Makefile.asm-generic \ - kbuild-file=$(HOST_DIR)/include/uapi/asm/Kbuild \ - obj=$(HOST_DIR)/include/generated/uapi/asm - $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders - + $(Q)$(MAKE) -f $(srctree)/Makefile ARCH=$(HEADER_ARCH) asm-generic archheaders archprepare: include/generated/user_constants.h @@ -142,7 +133,7 @@ export LDS_ELF_FORMAT := $(ELF_FORMAT) # The wrappers will select whether using "malloc" or the kernel allocator. LINK_WRAPS = -Wl,--wrap,malloc -Wl,--wrap,free -Wl,--wrap,calloc -LD_FLAGS_CMDLINE = $(foreach opt,$(LDFLAGS),-Wl,$(opt)) +LD_FLAGS_CMDLINE = $(foreach opt,$(KBUILD_LDFLAGS),-Wl,$(opt)) # Used by link-vmlinux.sh which has special support for um link export CFLAGS_vmlinux := $(LINK-y) $(LINK_WRAPS) $(LD_FLAGS_CMDLINE) @@ -169,11 +160,11 @@ define filechk_gen-asm-offsets echo " *"; \ echo " */"; \ echo ""; \ - sed -ne "/^->/{s:^->\([^ ]*\) [\$$#]*\([^ ]*\) \(.*\):#define \1 \2 /* \3 */:; s:->::; p;}"; \ + 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 SUBARCH USER_CFLAGS CFLAGS_NO_HARDENING OS DEV_NULL_PATH +export HEADER_ARCH SUBARCH USER_CFLAGS CFLAGS_NO_HARDENING OS DEV_NULL_PATH diff --git a/arch/um/Kconfig.net b/arch/um/drivers/Kconfig index c390f3deb0dc..2b1aaf7755aa 100644 --- a/arch/um/Kconfig.net +++ b/arch/um/drivers/Kconfig @@ -1,5 +1,129 @@ # SPDX-License-Identifier: GPL-2.0 +menu "UML Character Devices" + +config STDERR_CONSOLE + bool "stderr console" + default y + help + console driver which dumps all printk messages to stderr. + +config SSL + bool "Virtual serial line" + help + The User-Mode Linux environment allows you to create virtual serial + lines on the UML that are usually made to show up on the host as + ttys or ptys. + + See <http://user-mode-linux.sourceforge.net/old/input.html> for more + information and command line examples of how to use this facility. + + Unless you have a specific reason for disabling this, say Y. + +config NULL_CHAN + bool "null channel support" + help + This option enables support for attaching UML consoles and serial + lines to a device similar to /dev/null. Data written to it disappears + and there is never any data to be read. + +config PORT_CHAN + bool "port channel support" + help + This option enables support for attaching UML consoles and serial + lines to host portals. They may be accessed with 'telnet <host> + <port number>'. Any number of consoles and serial lines may be + attached to a single portal, although what UML device you get when + you telnet to that portal will be unpredictable. + It is safe to say 'Y' here. + +config PTY_CHAN + bool "pty channel support" + help + This option enables support for attaching UML consoles and serial + lines to host pseudo-terminals. Access to both traditional + pseudo-terminals (/dev/pty*) and pts pseudo-terminals are controlled + with this option. The assignment of UML devices to host devices + will be announced in the kernel message log. + It is safe to say 'Y' here. + +config TTY_CHAN + bool "tty channel support" + help + This option enables support for attaching UML consoles and serial + lines to host terminals. Access to both virtual consoles + (/dev/tty*) and the slave side of pseudo-terminals (/dev/ttyp* and + /dev/pts/*) are controlled by this option. + It is safe to say 'Y' here. + +config XTERM_CHAN + bool "xterm channel support" + help + This option enables support for attaching UML consoles and serial + lines to xterms. Each UML device so assigned will be brought up in + its own xterm. + It is safe to say 'Y' here. + +config NOCONFIG_CHAN + bool + default !(XTERM_CHAN && TTY_CHAN && PTY_CHAN && PORT_CHAN && NULL_CHAN) + +config CON_ZERO_CHAN + string "Default main console channel initialization" + default "fd:0,fd:1" + help + This is the string describing the channel to which the main console + will be attached by default. This value can be overridden from the + command line. The default value is "fd:0,fd:1", which attaches the + main console to stdin and stdout. + It is safe to leave this unchanged. + +config CON_CHAN + string "Default console channel initialization" + default "xterm" + help + This is the string describing the channel to which all consoles + except the main console will be attached by default. This value can + be overridden from the command line. The default value is "xterm", + which brings them up in xterms. + It is safe to leave this unchanged, although you may wish to change + this if you expect the UML that you build to be run in environments + which don't have X or xterm available. + +config SSL_CHAN + string "Default serial line channel initialization" + default "pty" + help + This is the string describing the channel to which the serial lines + will be attached by default. This value can be overridden from the + command line. The default value is "pty", which attaches them to + traditional pseudo-terminals. + It is safe to leave this unchanged, although you may wish to change + this if you expect the UML that you build to be run in environments + which don't have a set of /dev/pty* devices. + +config UML_SOUND + tristate "Sound support" + help + This option enables UML sound support. If enabled, it will pull in + soundcore and the UML hostaudio relay, which acts as a intermediary + between the host's dsp and mixer devices and the UML sound system. + It is safe to say 'Y' here. + +config SOUND + tristate + default UML_SOUND + +config SOUND_OSS_CORE + bool + default UML_SOUND + +config HOSTAUDIO + tristate + default UML_SOUND + +endmenu + menu "UML Network Devices" depends on NET @@ -211,4 +335,3 @@ config UML_NET_SLIRP Startup example: "eth0=slirp,FE:FD:01:02:03:04,/usr/local/bin/slirp" endmenu - diff --git a/arch/um/drivers/Makefile b/arch/um/drivers/Makefile index 16b3cebddafb..693319839f69 100644 --- a/arch/um/drivers/Makefile +++ b/arch/um/drivers/Makefile @@ -25,10 +25,10 @@ LDFLAGS_vde.o := -r $(shell $(CC) $(CFLAGS) -print-file-name=libvdeplug.a) targets := pcap_kern.o pcap_user.o vde_kern.o vde_user.o $(obj)/pcap.o: $(obj)/pcap_kern.o $(obj)/pcap_user.o - $(LD) -r -dp -o $@ $^ $(LDFLAGS) $(LDFLAGS_pcap.o) + $(LD) -r -dp -o $@ $^ $(ld_flags) $(obj)/vde.o: $(obj)/vde_kern.o $(obj)/vde_user.o - $(LD) -r -dp -o $@ $^ $(LDFLAGS) $(LDFLAGS_vde.o) + $(LD) -r -dp -o $@ $^ $(ld_flags) #XXX: The call below does not work because the flags are added before the # object name, so nothing from the library gets linked. diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index 8d80b27502e6..7e524efed584 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -261,7 +261,7 @@ 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 +284,7 @@ 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; diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c index 3ef1b48e064a..624cb47cc9cd 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> @@ -650,7 +650,7 @@ 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_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/ubd_kern.c b/arch/um/drivers/ubd_kern.c index d4e8c497ae86..28c40624bcb6 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -23,6 +23,7 @@ #include <linux/module.h> #include <linux/init.h> #include <linux/blkdev.h> +#include <linux/blk-mq.h> #include <linux/ata.h> #include <linux/hdreg.h> #include <linux/cdrom.h> @@ -142,7 +143,6 @@ struct cow { #define MAX_SG 64 struct ubd { - struct list_head restart; /* name (and fd, below) of the file opened for writing, either the * backing or the cow file. */ char *file; @@ -156,11 +156,8 @@ struct ubd { struct cow cow; struct platform_device pdev; struct request_queue *queue; + struct blk_mq_tag_set tag_set; spinlock_t lock; - struct scatterlist sg[MAX_SG]; - struct request *request; - int start_sg, end_sg; - sector_t rq_pos; }; #define DEFAULT_COW { \ @@ -182,10 +179,6 @@ struct ubd { .shared = 0, \ .cow = DEFAULT_COW, \ .lock = __SPIN_LOCK_UNLOCKED(ubd_devs.lock), \ - .request = NULL, \ - .start_sg = 0, \ - .end_sg = 0, \ - .rq_pos = 0, \ } /* Protected by ubd_lock */ @@ -196,6 +189,9 @@ static int fake_ide = 0; static struct proc_dir_entry *proc_ide_root = NULL; static struct proc_dir_entry *proc_ide = NULL; +static blk_status_t ubd_queue_rq(struct blk_mq_hw_ctx *hctx, + const struct blk_mq_queue_data *bd); + static void make_proc_ide(void) { proc_ide_root = proc_mkdir("ide", NULL); @@ -208,19 +204,6 @@ static int fake_ide_media_proc_show(struct seq_file *m, void *v) return 0; } -static int fake_ide_media_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, fake_ide_media_proc_show, NULL); -} - -static const struct file_operations fake_ide_media_proc_fops = { - .owner = THIS_MODULE, - .open = fake_ide_media_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - static void make_ide_entries(const char *dev_name) { struct proc_dir_entry *dir, *ent; @@ -231,7 +214,8 @@ static void make_ide_entries(const char *dev_name) dir = proc_mkdir(dev_name, proc_ide); if(!dir) return; - ent = proc_create("media", S_IRUGO, dir, &fake_ide_media_proc_fops); + ent = proc_create_single("media", S_IRUGO, dir, + fake_ide_media_proc_show); if(!ent) return; snprintf(name, sizeof(name), "ide0/%s", dev_name); proc_symlink(dev_name, proc_ide_root, name); @@ -448,11 +432,8 @@ __uml_help(udb_setup, " in the boot output.\n\n" ); -static void do_ubd_request(struct request_queue * q); - /* Only changed by ubd_init, which is an initcall. */ static int thread_fd = -1; -static LIST_HEAD(restart); /* Function to read several request pointers at a time * handling fractional reads if (and as) needed @@ -510,9 +491,6 @@ static int bulk_req_safe_read( /* Called without dev->lock held, and only in interrupt context. */ static void ubd_handler(void) { - struct ubd *ubd; - struct list_head *list, *next_ele; - unsigned long flags; int n; int count; @@ -532,23 +510,17 @@ static void ubd_handler(void) return; } for (count = 0; count < n/sizeof(struct io_thread_req *); count++) { - blk_end_request( - (*irq_req_buffer)[count]->req, - BLK_STS_OK, - (*irq_req_buffer)[count]->length - ); - kfree((*irq_req_buffer)[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); + + kfree(io_req); } } - reactivate_fd(thread_fd, UBD_IRQ); - list_for_each_safe(list, next_ele, &restart){ - ubd = container_of(list, struct ubd, restart); - list_del_init(&ubd->restart); - spin_lock_irqsave(&ubd->lock, flags); - do_ubd_request(ubd->queue); - spin_unlock_irqrestore(&ubd->lock, flags); - } + reactivate_fd(thread_fd, UBD_IRQ); } static irqreturn_t ubd_intr(int irq, void *dev) @@ -869,6 +841,7 @@ static void ubd_device_release(struct device *dev) struct ubd *ubd_dev = dev_get_drvdata(dev); blk_cleanup_queue(ubd_dev->queue); + blk_mq_free_tag_set(&ubd_dev->tag_set); *ubd_dev = ((struct ubd) DEFAULT_UBD); } @@ -903,7 +876,7 @@ static int ubd_disk_register(int major, u64 size, int unit, disk->private_data = &ubd_devs[unit]; disk->queue = ubd_devs[unit].queue; - device_add_disk(parent, disk); + device_add_disk(parent, disk, NULL); *disk_out = disk; return 0; @@ -911,6 +884,10 @@ static int ubd_disk_register(int major, u64 size, int unit, #define ROUND_BLOCK(n) ((n + ((1 << 9) - 1)) & (-1 << 9)) +static const struct blk_mq_ops ubd_mq_ops = { + .queue_rq = ubd_queue_rq, +}; + static int ubd_add(int n, char **error_out) { struct ubd *ubd_dev = &ubd_devs[n]; @@ -927,15 +904,23 @@ static int ubd_add(int n, char **error_out) ubd_dev->size = ROUND_BLOCK(ubd_dev->size); - INIT_LIST_HEAD(&ubd_dev->restart); - sg_init_table(ubd_dev->sg, MAX_SG); + ubd_dev->tag_set.ops = &ubd_mq_ops; + ubd_dev->tag_set.queue_depth = 64; + ubd_dev->tag_set.numa_node = NUMA_NO_NODE; + ubd_dev->tag_set.flags = BLK_MQ_F_SHOULD_MERGE; + ubd_dev->tag_set.driver_data = ubd_dev; + ubd_dev->tag_set.nr_hw_queues = 1; - err = -ENOMEM; - ubd_dev->queue = blk_init_queue(do_ubd_request, &ubd_dev->lock); - if (ubd_dev->queue == NULL) { - *error_out = "Failed to initialize device queue"; + err = blk_mq_alloc_tag_set(&ubd_dev->tag_set); + if (err) goto out; + + ubd_dev->queue = blk_mq_init_queue(&ubd_dev->tag_set); + if (IS_ERR(ubd_dev->queue)) { + err = PTR_ERR(ubd_dev->queue); + goto out_cleanup; } + ubd_dev->queue->queuedata = ubd_dev; blk_queue_write_cache(ubd_dev->queue, true, false); @@ -943,7 +928,7 @@ static int ubd_add(int n, char **error_out) err = ubd_disk_register(UBD_MAJOR, ubd_dev->size, n, &ubd_gendisk[n]); if(err){ *error_out = "Failed to register device"; - goto out_cleanup; + goto out_cleanup_tags; } if (fake_major != UBD_MAJOR) @@ -961,6 +946,8 @@ static int ubd_add(int n, char **error_out) out: return err; +out_cleanup_tags: + blk_mq_free_tag_set(&ubd_dev->tag_set); out_cleanup: blk_cleanup_queue(ubd_dev->queue); goto out; @@ -1139,9 +1126,9 @@ static int __init ubd_init(void) return -1; } - irq_req_buffer = kmalloc( - sizeof(struct io_thread_req *) * UBD_REQ_BUFFER_SIZE, - GFP_KERNEL + irq_req_buffer = kmalloc_array(UBD_REQ_BUFFER_SIZE, + sizeof(struct io_thread_req *), + GFP_KERNEL ); irq_remainder = 0; @@ -1149,9 +1136,9 @@ static int __init ubd_init(void) printk(KERN_ERR "Failed to initialize ubd buffering\n"); return -1; } - io_req_buffer = kmalloc( - sizeof(struct io_thread_req *) * UBD_REQ_BUFFER_SIZE, - GFP_KERNEL + io_req_buffer = kmalloc_array(UBD_REQ_BUFFER_SIZE, + sizeof(struct io_thread_req *), + GFP_KERNEL ); io_remainder = 0; @@ -1302,123 +1289,86 @@ static void cowify_req(struct io_thread_req *req, unsigned long *bitmap, req->bitmap_words, bitmap_len); } -/* Called with dev->lock held */ -static void prepare_request(struct request *req, struct io_thread_req *io_req, - unsigned long long offset, int page_offset, - int len, struct page *page) +static int ubd_queue_one_vec(struct blk_mq_hw_ctx *hctx, struct request *req, + u64 off, struct bio_vec *bvec) { - struct gendisk *disk = req->rq_disk; - struct ubd *ubd_dev = disk->private_data; - - io_req->req = req; - io_req->fds[0] = (ubd_dev->cow.file != NULL) ? ubd_dev->cow.fd : - ubd_dev->fd; - io_req->fds[1] = ubd_dev->fd; - io_req->cow_offset = -1; - io_req->offset = offset; - io_req->length = len; - io_req->error = 0; - io_req->sector_mask = 0; - - io_req->op = (rq_data_dir(req) == READ) ? UBD_READ : UBD_WRITE; - io_req->offsets[0] = 0; - io_req->offsets[1] = ubd_dev->cow.data_offset; - io_req->buffer = page_address(page) + page_offset; - io_req->sectorsize = 1 << 9; - - if(ubd_dev->cow.file != NULL) - cowify_req(io_req, ubd_dev->cow.bitmap, - ubd_dev->cow.bitmap_offset, ubd_dev->cow.bitmap_len); - -} + struct ubd *dev = hctx->queue->queuedata; + struct io_thread_req *io_req; + int ret; -/* Called with dev->lock held */ -static void prepare_flush_request(struct request *req, - struct io_thread_req *io_req) -{ - struct gendisk *disk = req->rq_disk; - struct ubd *ubd_dev = disk->private_data; + io_req = kmalloc(sizeof(struct io_thread_req), GFP_ATOMIC); + if (!io_req) + return -ENOMEM; io_req->req = req; - io_req->fds[0] = (ubd_dev->cow.file != NULL) ? ubd_dev->cow.fd : - ubd_dev->fd; - io_req->op = UBD_FLUSH; -} + if (dev->cow.file) + io_req->fds[0] = dev->cow.fd; + else + io_req->fds[0] = dev->fd; + io_req->error = 0; -static bool submit_request(struct io_thread_req *io_req, struct ubd *dev) -{ - int n = os_write_file(thread_fd, &io_req, - sizeof(io_req)); - if (n != sizeof(io_req)) { - if (n != -EAGAIN) - printk("write to io thread failed, " - "errno = %d\n", -n); - else if (list_empty(&dev->restart)) - list_add(&dev->restart, &restart); + 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; + io_req->length = bvec->bv_len; + io_req->sector_mask = 0; + 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; + + 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)) { + if (ret != -EAGAIN) + pr_err("write to io thread failed: %d\n", -ret); kfree(io_req); - return false; } - return true; + + return ret; } -/* Called with dev->lock held */ -static void do_ubd_request(struct request_queue *q) +static blk_status_t ubd_queue_rq(struct blk_mq_hw_ctx *hctx, + const struct blk_mq_queue_data *bd) { - struct io_thread_req *io_req; - struct request *req; - - while(1){ - struct ubd *dev = q->queuedata; - if(dev->request == NULL){ - struct request *req = blk_fetch_request(q); - if(req == NULL) - return; + struct ubd *ubd_dev = hctx->queue->queuedata; + struct request *req = bd->rq; + int ret = 0; - dev->request = req; - dev->rq_pos = blk_rq_pos(req); - dev->start_sg = 0; - dev->end_sg = blk_rq_map_sg(q, req, dev->sg); - } + blk_mq_start_request(req); - req = dev->request; + spin_lock_irq(&ubd_dev->lock); - if (req_op(req) == REQ_OP_FLUSH) { - io_req = kmalloc(sizeof(struct io_thread_req), - GFP_ATOMIC); - if (io_req == NULL) { - if (list_empty(&dev->restart)) - list_add(&dev->restart, &restart); - return; - } - prepare_flush_request(req, io_req); - if (submit_request(io_req, dev) == false) - return; + if (req_op(req) == 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; } + } +out: + spin_unlock_irq(&ubd_dev->lock); - while(dev->start_sg < dev->end_sg){ - struct scatterlist *sg = &dev->sg[dev->start_sg]; - - io_req = kmalloc(sizeof(struct io_thread_req), - GFP_ATOMIC); - if(io_req == NULL){ - if(list_empty(&dev->restart)) - list_add(&dev->restart, &restart); - return; - } - prepare_request(req, io_req, - (unsigned long long)dev->rq_pos << 9, - sg->offset, sg->length, sg_page(sg)); - - if (submit_request(io_req, dev) == false) - return; + if (ret < 0) + blk_mq_requeue_request(req, true); - dev->rq_pos += sg->length >> 9; - dev->start_sg++; - } - dev->end_sg = 0; - dev->request = NULL; - } + return BLK_STS_OK; } static int ubd_getgeo(struct block_device *bdev, struct hd_geometry *geo) diff --git a/arch/um/drivers/vector_kern.c b/arch/um/drivers/vector_kern.c index 02168fe25105..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> @@ -188,7 +188,7 @@ static int get_transport_options(struct arglist *def) if (strncmp(transport, TRANS_TAP, TRANS_TAP_LEN) == 0) return (vec_rx | VECTOR_BPF); if (strncmp(transport, TRANS_RAW, TRANS_RAW_LEN) == 0) - return (vec_rx | vec_tx); + return (vec_rx | vec_tx | VECTOR_QDISC_BYPASS); return (vec_rx | vec_tx); } @@ -504,15 +504,19 @@ static struct vector_queue *create_queue( result = kmalloc(sizeof(struct vector_queue), GFP_KERNEL); if (result == NULL) - goto out_fail; + return NULL; result->max_depth = max_size; result->dev = vp->dev; result->mmsg_vector = kmalloc( (sizeof(struct mmsghdr) * max_size), GFP_KERNEL); + if (result->mmsg_vector == NULL) + goto out_mmsg_fail; result->skbuff_vector = kmalloc( (sizeof(void *) * max_size), GFP_KERNEL); - if (result->mmsg_vector == NULL || result->skbuff_vector == NULL) - goto out_fail; + if (result->skbuff_vector == NULL) + goto out_skb_fail; + + /* further failures can be handled safely by destroy_queue*/ mmsg_vector = result->mmsg_vector; for (i = 0; i < max_size; i++) { @@ -527,14 +531,14 @@ static struct vector_queue *create_queue( result->max_iov_frags = num_extra_frags; for (i = 0; i < max_size; i++) { if (vp->header_size > 0) - iov = kmalloc( - sizeof(struct iovec) * (3 + num_extra_frags), - GFP_KERNEL + iov = kmalloc_array(3 + num_extra_frags, + sizeof(struct iovec), + GFP_KERNEL ); else - iov = kmalloc( - sizeof(struct iovec) * (2 + num_extra_frags), - GFP_KERNEL + iov = kmalloc_array(2 + num_extra_frags, + sizeof(struct iovec), + GFP_KERNEL ); if (iov == NULL) goto out_fail; @@ -563,6 +567,11 @@ static struct vector_queue *create_queue( result->head = 0; result->tail = 0; return result; +out_skb_fail: + kfree(result->mmsg_vector); +out_mmsg_fail: + kfree(result); + return NULL; out_fail: destroy_queue(result); return NULL; @@ -1109,16 +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) @@ -1232,9 +1236,8 @@ static int vector_net_open(struct net_device *dev) if ((vp->options & VECTOR_QDISC_BYPASS) != 0) { if (!uml_raw_enable_qdisc_bypass(vp->fds->rx_fd)) - vp->options = vp->options | VECTOR_BPF; + vp->options |= VECTOR_BPF; } - if ((vp->options & VECTOR_BPF) != 0) vp->bpf = uml_vector_default_bpf(vp->fds->rx_fd, dev->dev_addr); @@ -1572,7 +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_transports.c b/arch/um/drivers/vector_transports.c index 9065047f844b..77e4ebc206ae 100644 --- a/arch/um/drivers/vector_transports.c +++ b/arch/um/drivers/vector_transports.c @@ -120,7 +120,8 @@ static int raw_form_header(uint8_t *header, skb, vheader, virtio_legacy_is_little_endian(), - false + false, + 0 ); return 0; diff --git a/arch/um/drivers/vector_user.c b/arch/um/drivers/vector_user.c index 4d6a78e31089..3d8cdbdb4e66 100644 --- a/arch/um/drivers/vector_user.c +++ b/arch/um/drivers/vector_user.c @@ -267,8 +267,7 @@ cleanup: os_close_file(rxfd); if (txfd >= 0) os_close_file(txfd); - if (result != NULL) - kfree(result); + kfree(result); return NULL; } @@ -434,8 +433,7 @@ 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 bb5a196c3061..b10dde6cb793 100644 --- a/arch/um/include/asm/Kbuild +++ b/arch/um/include/asm/Kbuild @@ -1,6 +1,7 @@ generic-y += barrier.h generic-y += bpf_perf_event.h generic-y += bug.h +generic-y += compat.h generic-y += current.h generic-y += delay.h generic-y += device.h diff --git a/arch/um/include/asm/common.lds.S b/arch/um/include/asm/common.lds.S index b30d73ca29d0..4049f2c46387 100644 --- a/arch/um/include/asm/common.lds.S +++ b/arch/um/include/asm/common.lds.S @@ -53,14 +53,6 @@ CON_INITCALL } - .uml.initcall.init : { - __uml_initcall_start = .; - *(.uml.initcall.init) - __uml_initcall_end = .; - } - - SECURITY_INIT - .exitcall : { __exitcall_begin = .; *(.exitcall.exit) 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 (jdike@karaya.com) - * 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/init.h b/arch/um/include/shared/init.h index b3f5865a92c9..c66de434a983 100644 --- a/arch/um/include/shared/init.h +++ b/arch/um/include/shared/init.h @@ -64,14 +64,10 @@ struct uml_param { int (*setup_func)(char *, int *); }; -extern initcall_t __uml_initcall_start, __uml_initcall_end; extern initcall_t __uml_postsetup_start, __uml_postsetup_end; extern const char *__uml_help_start, *__uml_help_end; #endif -#define __uml_initcall(fn) \ - static initcall_t __uml_initcall_##fn __uml_init_call = fn - #define __uml_exitcall(fn) \ static exitcall_t __uml_exitcall_##fn __uml_exit_call = fn @@ -108,7 +104,6 @@ extern struct uml_param __uml_setup_start, __uml_setup_end; */ #define __uml_init_setup __used __section(.uml.setup.init) #define __uml_setup_help __used __section(.uml.help.init) -#define __uml_init_call __used __section(.uml.initcall.init) #define __uml_postsetup_call __used __section(.uml.postsetup.init) #define __uml_exit_call __used __section(.uml.exitcall.exit) 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..8360fa3f676d 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); } } diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c index 3c0e470ea646..1067469ba2ea 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> @@ -46,11 +46,11 @@ 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(); + memblock_free_all(); max_low_pfn = totalram_pages; max_pfn = totalram_pages; mem_init_print_info(NULL); @@ -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; 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 bc2a516c190f..1a1d88a4d940 100644 --- a/arch/um/kernel/ptrace.c +++ b/arch/um/kernel/ptrace.c @@ -115,17 +115,10 @@ long arch_ptrace(struct task_struct *child, long request, static void send_sigtrap(struct task_struct *tsk, struct uml_pt_regs *regs, int error_code) { - struct siginfo info; - - memset(&info, 0, sizeof(info)); - info.si_signo = SIGTRAP; - info.si_code = TRAP_BRKPT; - - /* User-mode eip? */ - info.si_addr = UPT_IS_USER(regs) ? (void __user *) UPT_IP(regs) : NULL; - /* Send us the fake SIGTRAP */ - force_sig_info(SIGTRAP, &info, tsk); + force_sig_fault(SIGTRAP, TRAP_BRKPT, + /* User-mode eip? */ + UPT_IS_USER(regs) ? (void __user *) UPT_IP(regs) : NULL, tsk); } /* diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c index b2b02df9896e..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, @@ -72,7 +72,7 @@ good_area: } do { - int fault; + vm_fault_t fault; fault = handle_mm_fault(vma, address, flags); @@ -162,13 +162,9 @@ static void show_segv_info(struct uml_pt_regs *regs) static void bad_segv(struct faultinfo fi, unsigned long ip) { - struct siginfo si; - - si.si_signo = SIGSEGV; - si.si_code = SEGV_ACCERR; - si.si_addr = (void __user *) FAULT_ADDRESS(fi); current->thread.arch.faultinfo = fi; - force_sig_info(SIGSEGV, &si, current); + force_sig_fault(SIGSEGV, SEGV_ACCERR, (void __user *) FAULT_ADDRESS(fi), + current); } void fatal_sigsegv(void) @@ -214,8 +210,8 @@ void segv_handler(int sig, struct siginfo *unused_si, struct uml_pt_regs *regs) unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, struct uml_pt_regs *regs) { - struct siginfo si; jmp_buf *catcher; + int si_code; int err; int is_write = FAULT_WRITE(fi); unsigned long address = FAULT_ADDRESS(fi); @@ -239,7 +235,7 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, if (SEGV_IS_FIXABLE(&fi)) err = handle_page_fault(address, ip, is_write, is_user, - &si.si_code); + &si_code); else { err = -EFAULT; /* @@ -271,18 +267,14 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, show_segv_info(regs); if (err == -EACCES) { - si.si_signo = SIGBUS; - si.si_errno = 0; - si.si_code = BUS_ADRERR; - si.si_addr = (void __user *)address; current->thread.arch.faultinfo = fi; - force_sig_info(SIGBUS, &si, current); + force_sig_fault(SIGBUS, BUS_ADRERR, (void __user *)address, + current); } else { BUG_ON(err != -EFAULT); - si.si_signo = SIGSEGV; - si.si_addr = (void __user *) address; current->thread.arch.faultinfo = fi; - force_sig_info(SIGSEGV, &si, current); + force_sig_fault(SIGSEGV, si_code, (void __user *) address, + current); } out: @@ -294,9 +286,7 @@ out: void relay_signal(int sig, struct siginfo *si, struct uml_pt_regs *regs) { - struct faultinfo *fi; - struct siginfo clean_si; - + int code, err; if (!UPT_IS_USER(regs)) { if (sig == SIGBUS) printk(KERN_ERR "Bus error - the host /dev/shm or /tmp " @@ -306,29 +296,21 @@ void relay_signal(int sig, struct siginfo *si, struct uml_pt_regs *regs) arch_examine_signal(sig, regs); - clear_siginfo(&clean_si); - clean_si.si_signo = si->si_signo; - clean_si.si_errno = si->si_errno; - clean_si.si_code = si->si_code; - switch (sig) { - case SIGILL: - case SIGFPE: - case SIGSEGV: - case SIGBUS: - case SIGTRAP: - fi = UPT_FAULTINFO(regs); - clean_si.si_addr = (void __user *) FAULT_ADDRESS(*fi); + /* Is the signal layout for the signal known? + * Signal data must be scrubbed to prevent information leaks. + */ + code = si->si_code; + err = si->si_errno; + if ((err == 0) && (siginfo_layout(sig, code) == SIL_FAULT)) { + struct faultinfo *fi = UPT_FAULTINFO(regs); current->thread.arch.faultinfo = *fi; -#ifdef __ARCH_SI_TRAPNO - clean_si.si_trapno = si->si_trapno; -#endif - break; - default: - printk(KERN_ERR "Attempted to relay unknown signal %d (si_code = %d)\n", - sig, si->si_code); + force_sig_fault(sig, code, (void __user *)FAULT_ADDRESS(*fi), + current); + } else { + printk(KERN_ERR "Attempted to relay unknown signal %d (si_code = %d) with errno %d\n", + sig, code, err); + force_sig(sig, current); } - - force_sig_info(sig, &clean_si, current); } void bus_handler(int sig, struct siginfo *si, struct uml_pt_regs *regs) 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/main.c b/arch/um/os-Linux/main.c index 5f970ece5ac3..f1fee2b91239 100644 --- a/arch/um/os-Linux/main.c +++ b/arch/um/os-Linux/main.c @@ -40,17 +40,6 @@ static void set_stklim(void) } } -static __init void do_uml_initcalls(void) -{ - initcall_t *call; - - call = &__uml_initcall_start; - while (call < &__uml_initcall_end) { - (*call)(); - call++; - } -} - static void last_ditch_exit(int sig) { uml_cleanup(); @@ -151,7 +140,6 @@ int __init main(int argc, char **argv, char **envp) scan_elf_aux(envp); #endif - do_uml_initcalls(); change_sig(SIGPIPE, 0); ret = linux_main(argc, argv); 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) |
