Clean up use of error_printf()
v2 cover Clean up use of error_printf()
This series cleans up two kinds of error_printf() misuse:
* Errors and warnings should be reported with error_report() and warn_report().
* Help output should be go to stdout, not stderr.
This is obviously for 4.1. If nobody objects, I'll take the whole series through my tree.
v2:
* PATCH 02: Commit message tweaked, indentation fixed [Eric]
* PATCH 10: Indentation fixed [Marcel]
* PATCH 11: Use-after-free fixed [Patchew]
* PATCH 12: test-util-sockets.c updated along with stubs/monitor.c
* PATCH 14: Commit message typo [Eric]
Markus Armbruster (15):
qemu-img: Use error_vreport() in error_exit() block/ssh: Do not report read/write/flush errors to the user
char-pty: Drop "char device redirected to" message
loader-fit: Wean off error_printf()
mips/boston: Report errors with error_report(), not error_printf()
pci: Report fatal errors with error_report(), not error_printf()
hpet: Report warnings with warn_report(), not error_printf()
vfio: Report warnings with warn_report(), not error_printf() s390x/kvm: Report warnings with warn_report(), not error_printf()
vl: Make -machine $TYPE,help and -accel help print to stdout monitor error: Make printf()-like functions return a value
qemu-print: New qemu_printf(), qemu_vprintf() etc.
blockdev: Make -drive format=help print to stdout
char: Make -chardev help print to stdout
monitor: Simplify how -device/device_add print help
MAINTAINERS | 2 ++ block/ssh.c | 38 +++++++------------- block/trace-events | 3 ++ blockdev.c | 9 ++--- chardev/char-pty.c | 2 -- chardev/char.c | 3 +- hw/core/loader-fit.c | 62 +++++++++++++++++++-------------- hw/mips/boston.c | 6 ++-- hw/pci/pci.c | 2 +- hw/timer/hpet.c | 2 +- hw/vfio/pci.c | 19 ++++++---- include/monitor/monitor.h | 7 ++-- include/qemu/error-report.h | 8 ++--- include/qemu/qemu-print.h | 19 ++++++++++ monitor.c | 69 ++++++++++++++++++------------------- qdev-monitor.c | 36 ++++++++----------- qemu-img.c | 6 ++-- stubs/error-printf.c | 13 ++++--- stubs/monitor.c | 5 +++ target/s390x/kvm.c | 2 +- tests/test-util-sockets.c | 1 + util/Makefile.objs | 1 + util/qemu-error.c | 12 ++++--- util/qemu-print.c | 42 ++++++++++++++++++++++ vl.c | 10 +++--- 25 files changed, 224 insertions(+), 155 deletions(-) create mode 100644 include/qemu/qemu-print.h create mode 100644 util/qemu-print.c
Signature · 1 line
2.17.2
PATCH v2 1/15 qemu-img: Use error_vreport() in error_exit()
error_exit() uses low-level error_printf() to report errors. Modernize it to use error_vreport().
--- qemu-img.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
qemu-img.c +2 -4
@@ -85,13 +85,11 @@ static void QEMU_NORETURN GCC_FMT_ATTR(1, 2) error_exit(const char *fmt, ...) {va_list ap; -error_printf ("qemu-img: " ); -va_start (ap, fmt); -error_vprintf (fmt, ap); +error_vreport (fmt, ap);va_end (ap); -error_printf ("\nTry 'qemu-img --help' for more information\n" ); +error_printf ("Try 'qemu-img --help' for more information\n" );exit (EXIT_FAILURE ); }
Signature · 1 line
2.17.2
PATCH v2 2/15 block/ssh: Do not report read/write/flush errors to the user
In reply to Markus ArmbrusterCallbacks ssh_co_readv(), ssh_co_writev(), ssh_co_flush() report errors to the user with error_printf(). They shouldn't, it's their caller's job. Replace by a suitable trace point. While there, drop the unreachable !s->sftp case.
Perhaps we should convert this part of the block driver interface to Error, so block drivers can pass more detail to their callers. Not today.
--- block/ssh.c | 38 +++++++++++++------------------------- block/trace-events | 3 +++ 2 files changed, 16 insertions(+), 25 deletions(-)
block/ssh.c +13 -25
@@ -159,31 +159,19 @@ sftp_error_setg(Error **errp, BDRVSSHState *s, const char *fs, ...)g_free (msg); } -static void GCC_FMT_ATTR(2, 3) -sftp_error_report(BDRVSSHState *s, const char *fs, ...) +static void sftp_error_trace (BDRVSSHState *s,const char *op) { -va_list args; +char *ssh_err; +int ssh_err_code; +unsigned long sftp_err_code; -va_start (args, fs); -error_vprintf (fs, args); +/* This is not an errno. See <libssh2.h>. */ + ssh_err_code =libssh2_session_last_error (s->session, + &ssh_err,NULL ,0 ); +/* See <libssh2_sftp.h>. */ + sftp_err_code =libssh2_sftp_last_error ((s)->sftp); -if ((s)->sftp) { -char *ssh_err; -int ssh_err_code; -unsigned long sftp_err_code; - -/* This is not an errno. See <libssh2.h>. */ - ssh_err_code =libssh2_session_last_error (s->session, - &ssh_err,NULL ,0 ); -/* See <libssh2_sftp.h>. */ - sftp_err_code =libssh2_sftp_last_error ((s)->sftp); - -error_printf (": %s (libssh2 error code: %d, sftp error code: %lu)" , - ssh_err, ssh_err_code, sftp_err_code); - } - -va_end (args); -error_printf ("\n" ); +trace_sftp_error (op, ssh_err, ssh_err_code, sftp_err_code); }static int parse_uri (const char *filename,QDict *options,Error **errp)
@@ -1035,7 +1023,7 @@ static coroutine_fn int ssh_read(BDRVSSHState *s, BlockDriverState *bs, goto again; }if (r <0 ) { -sftp_error_report (s,"read failed" ); +sftp_error_trace (s,"read" ); s->offset =-1 ;return -EIO ; }
@@ -1105,7 +1093,7 @@ static int ssh_write(BDRVSSHState *s, BlockDriverState *bs, goto again; }if (r <0 ) { -sftp_error_report (s,"write failed" ); +sftp_error_trace (s,"write" ); s->offset =-1 ;return -EIO ; }
@@ -1188,7 +1176,7 @@ static coroutine_fn int ssh_flush(BDRVSSHState *s, BlockDriverState *bs)return 0 ; }if (r <0 ) { -sftp_error_report (s,"fsync failed" ); +sftp_error_trace (s,"fsync" );return -EIO ; }
block/trace-events +3 -0
@@ -208,3 +208,6 @@ sheepdog_co_rw_vector_new(uint64_t oid) "new oid 0x%" PRIx64 sheepdog_snapshot_create_info(const char *sn_name, const char *id, const char *name, int64_t size, int is_snapshot) "sn_info: name %s id_str %s s: name %s vm_state_size %" PRId64 " " "is_snapshot %d" sheepdog_snapshot_create(const char *sn_name, const char *id) "%s %s" sheepdog_snapshot_create_inode(const char *name, uint32_t snap, uint32_t vdi) "s->inode: name %s snap_id 0x%" PRIx32 " vdi 0x%" PRIx32 + +# ssh.c +sftp_error(const char *op, const char *ssh_err, int ssh_err_code, unsigned long sftp_err_code) "%s failed: %s (libssh2 error code: %d, sftp error code: %lu)"
Signature · 1 line
2.17.2
PATCH v2 3/15 char-pty: Drop "char device redirected to" message
In reply to Markus Armbrusterchar_pty_open() prints a "char device redirected to PTY_NAME (label LABEL)" message to the current monitor or else to stderr. No other ChardevClass::open() prints anything on success. Drop the message.
--- chardev/char-pty.c | 2 -- 1 file changed, 2 deletions(-)
chardev/char-pty.c +0 -2
@@ -211,8 +211,6 @@ static void char_pty_open(Chardev *chr,qemu_set_nonblock (master_fd); chr->filename =g_strdup_printf ("pty:%s" , pty_name); -error_printf ("char device redirected to %s (label %s)\n" , - pty_name, chr->label); s =PTY_CHARDEV (chr); s->ioc =QIO_CHANNEL (qio_channel_file_new_fd (master_fd));
Signature · 1 line
2.17.2
Re: [Qemu-devel] [PATCH v2 03/15] char-pty: Drop "char device redirected to" message
Quoted 17 lines · Markus Armbruster
char_pty_open()prints a "char device redirected toPTY_NAME(label LABEL)" message to the current monitor or else to stderr. No otherChardevClass::open()prints anything on success. Drop the message.Cc "Marc-André Lureau" <[email protected]>Cc Paolo Bonzini <[email protected]>Signed-off-by Markus Armbruster <[email protected]>Reviewed-by Marc-André Lureau <[email protected]>--- chardev/char-pty.c | 2 -- 1 file changed, 2 deletions(-)chardev/char-pty.c +0
-0
chardev/char-pty.c static void char_pty_open(Chardev *chr, · line 216
Quoted 8 lines
chardev/char-pty.c +0
-2@@ -211,8 +211,6 @@ static void char_pty_open(Chardev *chr,qemu_set_nonblock (master_fd); chr->filename =g_strdup_printf ("pty:%s" , pty_name); -error_printf ("char device redirected to %s (label %s)\n" , - pty_name, chr->label); s =PTY_CHARDEV (chr); s->ioc =QIO_CHANNEL (qio_channel_file_new_fd (master_fd));
The reason for the message is that the char device is completely useless until the user knows the /dev/pts/N path[1]. You can get it with "info chardev" (aka query-chardev for QMP) but there's an interesting chicken and egg problem if the pty is for your monitor...
Paolo
[1] once you know it, you can use the monitor's readline interface with e.g. "socat STDIO,cfmakeraw FILE:/dev/pts/1"
Re: [Qemu-devel] [PATCH v2 03/15] char-pty: Drop "char device redirected to" message
Quoted 32 lines · Paolo Bonzini
Quoted 17 lines · Markus Armbruster
char_pty_open()prints a "char device redirected toPTY_NAME(label LABEL)" message to the current monitor or else to stderr. No otherChardevClass::open()prints anything on success. Drop the message.Cc "Marc-André Lureau" <[email protected]>Cc Paolo Bonzini <[email protected]>Signed-off-by Markus Armbruster <[email protected]>Reviewed-by Marc-André Lureau <[email protected]>--- chardev/char-pty.c | 2 -- 1 file changed, 2 deletions(-)chardev/char-pty.c +0
-0chardev/char-pty.c static void char_pty_open(Chardev *chr, · line 216
Quoted 8 lines
chardev/char-pty.c +0
-2@@ -211,8 +211,6 @@ static void char_pty_open(Chardev *chr,qemu_set_nonblock (master_fd); chr->filename =g_strdup_printf ("pty:%s" , pty_name); -error_printf ("char device redirected to %s (label %s)\n" , - pty_name, chr->label); s =PTY_CHARDEV (chr); s->ioc =QIO_CHANNEL (qio_channel_file_new_fd (master_fd));The reason for the message is that the char device is completely useless until the user knows the
/dev/pts/Npath[1]. You can get it with "info chardev" (aka query-chardev for QMP) but there's an interesting chicken and egg problem if the pty is for your monitor...
Quoted 1 line
Paolo
During review of v1, I wrote:
If we should decide the message is still useful enough to be worth keeping, I could direct it to stdout instead of dropping it.
No clear conclusion emerged, so I did nothing for v2. If we conclude to keep the message now, I'll gladly do that.
Quoted 2 lines
[1] once you know it, you can use the monitor's readline interface with e.g. "socat STDIO,cfmakeraw FILE:/dev/pts/1"
There's also
$ socat UNIX:/path/to/socket READLINE,history=$HOME/.hmp_history,prompt='(qemu) '
Lacks completion. But then our very own reimplementation of readline lacks any number of other features.
Re: [Qemu-devel] [PATCH v2 03/15] char-pty: Drop "char device redirected to" message
Quoted 36 lines · Markus Armbruster
Quoted 32 lines · Paolo Bonzini
Quoted 17 lines · Markus Armbruster
char_pty_open()prints a "char device redirected toPTY_NAME(label LABEL)" message to the current monitor or else to stderr. No otherChardevClass::open()prints anything on success. Drop the message.Cc "Marc-André Lureau" <[email protected]>Cc Paolo Bonzini <[email protected]>Signed-off-by Markus Armbruster <[email protected]>Reviewed-by Marc-André Lureau <[email protected]>--- chardev/char-pty.c | 2 -- 1 file changed, 2 deletions(-)chardev/char-pty.c +0
-0chardev/char-pty.c static void char_pty_open(Chardev *chr, · line 216
Quoted 8 lines
chardev/char-pty.c +0
-2@@ -211,8 +211,6 @@ static void char_pty_open(Chardev *chr,qemu_set_nonblock (master_fd); chr->filename =g_strdup_printf ("pty:%s" , pty_name); -error_printf ("char device redirected to %s (label %s)\n" , - pty_name, chr->label); s =PTY_CHARDEV (chr); s->ioc =QIO_CHANNEL (qio_channel_file_new_fd (master_fd));The reason for the message is that the char device is completely useless until the user knows the
/dev/pts/Npath[1]. You can get it with "info chardev" (aka query-chardev for QMP) but there's an interesting chicken and egg problem if the pty is for your monitor...Quoted 1 line
Paolo
Quoted 7 lines
During review of v1, I wrote:
If we should decide the message is still useful enough to be worth keeping, I could direct it to stdout instead of dropping it.
No clear conclusion emerged, so I did nothing for v2. If we conclude to keep the message now, I'll gladly do that.
No further comments, no clear conclusion, so let's stick to the status quo. I'll drop this patch and add one to print the message to stdout.
[...]
Re: [Qemu-devel] [PATCH v2 03/15] char-pty: Drop "char device redirected to" message
In reply to Markus ArmbrusterQuoted 17 lines · Markus Armbruster
char_pty_open()prints a "char device redirected toPTY_NAME(label LABEL)" message to the current monitor or else to stderr. No otherChardevClass::open()prints anything on success. Drop the message.Cc "Marc-André Lureau" <[email protected]>Cc Paolo Bonzini <[email protected]>Signed-off-by Markus Armbruster <[email protected]>Reviewed-by Marc-André Lureau <[email protected]>--- chardev/char-pty.c | 2 -- 1 file changed, 2 deletions(-)chardev/char-pty.c +0
-0
chardev/char-pty.c static void char_pty_open(Chardev *chr, · line 215
Quoted 5 lines
chardev/char-pty.c +0
-2@@ -211,8 +211,6 @@ static void char_pty_open(Chardev *chr,qemu_set_nonblock (master_fd); chr->filename =g_strdup_printf ("pty:%s" , pty_name); -error_printf ("char device redirected to %s (label %s)\n" , - pty_name, chr->label);
Don't be scared by the one occurence of this string in a 'strstr' call in libvirt. We thankfully deleted the code that was using it to extract the PTY path a long time ago. The only leftover part is to skip it when we are attempting to extract an useful error from qemu which would be reported to stderr prior to initiating QMP so getting rid of the message actually has no effect.
https://libvirt.org/git/?p=libvirt.git;a=commitdiff;h=932534e85f34a479c7eac174e997bfd9c85bd22d
PATCH v2 4/15 loader-fit: Wean off error_printf()
In reply to Markus Armbrusterload_fit() reports errors with error_printf() instead of error_report(). Worse, it even reports errors it actually recovers from, in fit_cfg_compatible() and fit_load_fdt(). Messed up in initial commit 51b58561c1d.
Convert the helper functions for load_fit() to Error. Make sure each failure path sets an error.
Fix fit_cfg_compatible() and fit_load_fdt() not to report errors they actually recover from.
Convert load_fit() to error_report().
--- hw/core/loader-fit.c | 62 +++++++++++++++++++++++++------------------- 1 file changed, 36 insertions(+), 26 deletions(-)
hw/core/loader-fit.c +36 -26
@@ -18,6 +18,7 @@*/ #include "qemu/osdep.h" +#include "qapi/error.h" #include "qemu/units.h" #include "exec/memory.h" #include "hw/loader.h"
@@ -33,7 +34,7 @@#define FIT_LOADER_MAX_PATH (128)static const void *fit_load_image_alloc (const void *itb,const char *name, -int *poff,size_t *psz) +int *poff,size_t *psz,Error **errp) {const void *data;const char *comp;
@@ -46,6 +47,7 @@ static const void *fit_load_image_alloc(const void *itb, const char *name, off =fdt_path_offset (itb, path);if (off <0 ) { +error_setg (errp,"can't find node %s" , path);return NULL ; }if (poff) {
@@ -54,6 +56,7 @@ static const void *fit_load_image_alloc(const void *itb, const char *name, data =fdt_getprop (itb, off,"data" , &sz);if (!data) { +error_setg (errp,"can't get %s/data" , path);return NULL ; }
@@ -73,7 +76,7 @@ static const void *fit_load_image_alloc(const void *itb, const char *name, uncomp_len =gunzip (uncomp_data, uncomp_len, (void *) data, sz);if (uncomp_len <0 ) { -error_printf ("unable to decompress %s image\n" , name); +error_setg (errp,"unable to decompress %s image" , name);g_free (uncomp_data);return NULL ; }
@@ -85,18 +88,19 @@ static const void *fit_load_image_alloc(const void *itb, const char *name,return data; } -error_printf ("unknown compression '%s'\n" , comp); +error_setg (errp,"unknown compression '%s'" , comp);return NULL ; }static int fit_image_addr (const void *itb,int img,const char *name, -hwaddr *addr) +hwaddr *addr,Error **errp) {const void *prop;int len; prop =fdt_getprop (itb, img, name, &len);if (!prop) { +error_setg (errp,"can't find %s address" , name);return -ENOENT ; }
@@ -108,13 +112,14 @@ static int fit_image_addr(const void *itb, int img, const char *name, *addr =fdt64_to_cpu (*(fdt64_t *)prop);return 0 ;default : -error_printf ("invalid %s address length %d\n" , name, len); +error_setg (errp,"invalid %s address length %d" , name, len);return -EINVAL ; } }static int fit_load_kernel (const struct fit_loader *ldr,const void *itb, -int cfg,void *opaque,hwaddr *pend) +int cfg,void *opaque,hwaddr *pend, +Error **errp) {const char *name;const void *data;
@@ -126,26 +131,26 @@ static int fit_load_kernel(const struct fit_loader *ldr, const void *itb, name =fdt_getprop (itb, cfg,"kernel" ,NULL );if (!name) { -error_printf ("no kernel specified by FIT configuration\n" ); +error_setg (errp,"no kernel specified by FIT configuration" );return -EINVAL ; } - load_data = data =fit_load_image_alloc (itb, name, &img_off, &sz); + load_data = data =fit_load_image_alloc (itb, name, &img_off, &sz, errp);if (!data) { -error_printf ("unable to load kernel image from FIT\n" ); +error_prepend (errp,"unable to load kernel image from FIT: " );return -EINVAL ; } - err =fit_image_addr (itb, img_off,"load" , &load_addr); + err =fit_image_addr (itb, img_off,"load" , &load_addr, errp);if (err) { -error_printf ("unable to read kernel load address from FIT\n" ); +error_prepend (errp,"unable to read kernel load address from FIT: " ); ret = err; goto out; } - err =fit_image_addr (itb, img_off,"entry" , &entry_addr); + err =fit_image_addr (itb, img_off,"entry" , &entry_addr, errp);if (err) { -error_printf ("unable to read kernel entry address from FIT\n" ); +error_prepend (errp,"unable to read kernel entry address from FIT: " ); ret = err; goto out; }
@@ -172,7 +177,7 @@ out:static int fit_load_fdt (const struct fit_loader *ldr,const void *itb,int cfg,void *opaque,const void *match_data, -hwaddr kernel_end) +hwaddr kernel_end,Error **errp) {const char *name;const void *data;
@@ -187,16 +192,18 @@ static int fit_load_fdt(const struct fit_loader *ldr, const void *itb,return 0 ; } - load_data = data =fit_load_image_alloc (itb, name, &img_off, &sz); + load_data = data =fit_load_image_alloc (itb, name, &img_off, &sz, errp);if (!data) { -error_printf ("unable to load FDT image from FIT\n" ); +error_prepend (errp,"unable to load FDT image from FIT: " );return -EINVAL ; } - err =fit_image_addr (itb, img_off,"load" , &load_addr); + err =fit_image_addr (itb, img_off,"load" , &load_addr, errp);if (err == -ENOENT ) { load_addr =ROUND_UP (kernel_end,64 * KiB) + (10 * MiB); +error_free (*errp); }else if (err) { +error_prepend (errp,"unable to read FDT load address from FIT: " ); ret = err; goto out; }
@@ -229,7 +236,7 @@ static bool fit_cfg_compatible(const void *itb, int cfg, const char *compat)return false; } - fdt =fit_load_image_alloc (itb, fdt_name,NULL ,NULL ); + fdt =fit_load_image_alloc (itb, fdt_name,NULL ,NULL ,NULL );if (!fdt) {return false; }
@@ -252,11 +259,12 @@ out:int load_fit (const struct fit_loader *ldr,const char *filename,void *opaque) { +Error *err =NULL ;const struct fit_loader_match *match;const void *itb, *match_data =NULL ;const char *def_cfg_name;char path[FIT_LOADER_MAX_PATH ]; -int itb_size, configs, cfg_off, off, err; +int itb_size, configs, cfg_off, off;hwaddr kernel_end;int ret;
@@ -267,6 +275,7 @@ int load_fit(const struct fit_loader *ldr, const char *filename, void *opaque) configs =fdt_path_offset (itb,"/configurations" );if (configs <0 ) { +error_report ("can't find node /configurations" ); ret = configs; goto out; }
@@ -301,20 +310,21 @@ int load_fit(const struct fit_loader *ldr, const char *filename, void *opaque) }if (cfg_off <0 ) { -/* couldn't find a configuration to use */ +error_report ("can't find configuration" ); ret = cfg_off; goto out; } - err =fit_load_kernel (ldr, itb, cfg_off, opaque, &kernel_end); -if (err) { - ret = err; + ret =fit_load_kernel (ldr, itb, cfg_off, opaque, &kernel_end, &err); +if (ret) { +error_report_err (err); goto out; } - err =fit_load_fdt (ldr, itb, cfg_off, opaque, match_data, kernel_end); -if (err) { - ret = err; + ret =fit_load_fdt (ldr, itb, cfg_off, opaque, match_data, kernel_end, + &err); +if (ret) { +error_report_err (err); goto out; }
Signature · 1 line
2.17.2
PATCH v2 5/15 mips/boston: Report errors with error_report(), not error_printf()
In reply to Markus Armbruster--- hw/mips/boston.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
hw/mips/boston.c +3 -3
@@ -528,21 +528,21 @@ static void boston_mach_init(MachineState *machine) fw_size =load_image_targphys (machine->firmware,0x1fc00000 ,4 * MiB);if (fw_size ==-1 ) { -error_printf ("unable to load firmware image '%s'\n" , +error_report ("unable to load firmware image '%s'" , machine->firmware);exit (1 ); } }else if (machine->kernel_filename) { fit_err =load_fit (&boston_fit_loader, machine->kernel_filename, s);if (fit_err) { -error_printf ("unable to load FIT image\n" ); +error_report ("unable to load FIT image" );exit (1 ); }gen_firmware (memory_region_get_ram_ptr (flash) +0x7c00000 , s->kernel_entry, s->fdt_base, is_64b); }else if (!qtest_enabled ()) { -error_printf ("Please provide either a -kernel or -bios argument\n" ); +error_report ("Please provide either a -kernel or -bios argument" );exit (1 ); } }
Signature · 1 line
2.17.2
PATCH v2 6/15 pci: Report fatal errors with error_report(), not error_printf()
In reply to Markus Armbruster--- hw/pci/pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
hw/pci/pci.c +1 -1
@@ -951,7 +951,7 @@ static uint16_t pci_req_id_cache_extract(PCIReqIDCache *cache) result =PCI_BUILD_BDF (bus_n,0 );break ;default : -error_printf ("Invalid PCI requester ID cache type: %d\n" , +error_report ("Invalid PCI requester ID cache type: %d" , cache->type);exit (1 );break ;
Signature · 1 line
2.17.2
PATCH v2 7/15 hpet: Report warnings with warn_report(), not error_printf()
In reply to Markus Armbruster--- hw/timer/hpet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
hw/timer/hpet.c +1 -1
@@ -744,7 +744,7 @@ static void hpet_realize(DeviceState *dev, Error **errp)HPETTimer *timer;if (!s->intcap) { -error_printf ("Hpet's intcap not initialized.\n" ); +warn_report ("Hpet's intcap not initialized" ); }if (hpet_cfg.count ==UINT8_MAX ) {/* first instance */
Signature · 1 line
2.17.2
PATCH v2 8/15 vfio: Report warnings with warn_report(), not error_printf()
In reply to Markus Armbruster--- hw/vfio/pci.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-)
hw/vfio/pci.c +13 -6
@@ -947,8 +947,10 @@ static void vfio_pci_size_rom(VFIOPCIDevice *vdev)if (vdev->pdev.romfile || !vdev->pdev.rom_bar) {/* Since pci handles romfile, just print a message and return */ if (vfio_blacklist_opt_rom (vdev) && vdev->pdev.romfile) { -error_printf ("Warning : Device at %s is known to cause system instability issues during option rom execution. Proceeding anyway since user specified romfile\n" , - vdev->vbasedev.name); +warn_report ("Device at %s is known to cause system instability" +" issues during option rom execution" , + vdev->vbasedev.name); +error_printf ("Proceeding anyway since user specified romfile\n" ); }return ; }
@@ -973,11 +975,16 @@ static void vfio_pci_size_rom(VFIOPCIDevice *vdev)if (vfio_blacklist_opt_rom (vdev)) {if (dev->opts &&qemu_opt_get (dev->opts,"rombar" )) { -error_printf ("Warning : Device at %s is known to cause system instability issues during option rom execution. Proceeding anyway since user specified non zero value for rombar\n" , - vdev->vbasedev.name); +warn_report ("Device at %s is known to cause system instability" +" issues during option rom execution" , + vdev->vbasedev.name); +error_printf ("Proceeding anyway since user specified" +" non zero value for rombar\n" ); }else { -error_printf ("Warning : Rom loading for device at %s has been disabled due to system instability issues. Specify rombar=1 or romfile to force\n" , - vdev->vbasedev.name); +warn_report ("Rom loading for device at %s has been disabled" +" due to system instability issues" , + vdev->vbasedev.name); +error_printf ("Specify rombar=1 or romfile to force\n" );return ; } }
Signature · 1 line
2.17.2
PATCH v2 9/15 s390x/kvm: Report warnings with warn_report(), not error_printf()
In reply to Markus Armbrusterkvm_s390_mem_op() can fail in two ways: when !cap_mem_op, it returns -ENOSYS, and when kvm_vcpu_ioctl() fails, it returns -errno set by ioctl(). Its caller s390_cpu_virt_mem_rw() recovers from both failures.
kvm_s390_mem_op() prints "KVM_S390_MEM_OP failed" with error_printf() in the latter failure mode. Since this is obviously a warning, use warn_report().
Perhaps the reporting should be left to the caller. It could warn on failure other than -ENOSYS.
--- target/s390x/kvm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
target/s390x/kvm.c +1 -1
@@ -782,7 +782,7 @@ int kvm_s390_mem_op(S390CPU *cpu, vaddr addr, uint8_t ar, void *hostbuf, ret =kvm_vcpu_ioctl (CPU (cpu),KVM_S390_MEM_OP , &mem_op);if (ret <0 ) { -error_printf ("KVM_S390_MEM_OP failed: %s\n" ,strerror (-ret)); +warn_report ("KVM_S390_MEM_OP failed: %s" ,strerror (-ret)); }return ret; }
Signature · 1 line
2.17.2
PATCH v2 10/15 vl: Make -machine $TYPE, help and -accel help print to stdout
In reply to Markus ArmbrusterCommand line help help explicitly requested by the user should be printed to stdout, not stderr. We do elsewhere. Adjust -machine $TYPE,help and -accel help to match: use printf() instead of error_printf().
--- vl.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
vl.c +5 -5
@@ -1556,12 +1556,12 @@ static int machine_help_func(QemuOpts *opts, MachineState *machine)continue ; } -error_printf ("%s.%s=%s" ,MACHINE_GET_CLASS (machine)->name, - prop->name, prop->type); +printf ("%s.%s=%s" ,MACHINE_GET_CLASS (machine)->name, + prop->name, prop->type);if (prop->description) { -error_printf (" (%s)\n" , prop->description); +printf (" (%s)\n" , prop->description); }else { -error_printf ("\n" ); +printf ("\n" ); } }
@@ -3643,7 +3643,7 @@ int main(int argc, char **argv, char **envp) optarg, true); optarg =qemu_opt_get (accel_opts,"accel" );if (!optarg ||is_help_option (optarg)) { -error_printf ("Possible accelerators: kvm, xen, hax, tcg\n" ); +printf ("Possible accelerators: kvm, xen, hax, tcg\n" );exit (0 ); } opts = qemu_opts_create(qemu_find_opts("machine"), NULL,
Signature · 1 line
2.17.2
PATCH v2 11/15 monitor error: Make printf()-like functions return a value
In reply to Markus Armbrusterprintf() & friends return the number of characters written on success, negative value on error.
monitor_printf(), monitor_vfprintf(), monitor_vprintf(), error_printf(), error_printf_unless_qmp(), error_vprintf(), and error_vprintf_unless_qmp() return void. Some of them carry a TODO comment asking for int instead.
Improve them to return int like printf() does.
This makes our use of monitor_printf() as fprintf_function slightly less dirty: the function cast no longer adds a return value that isn't there. It still changes a parameter's pointer type. That will be addressed in a future commit.
monitor_vfprintf() always returns zero. Improve it to return the proper value.
--- include/monitor/monitor.h | 8 ++--- include/qemu/error-report.h | 8 ++--- monitor.c | 61 ++++++++++++++++++++----------------- stubs/error-printf.c | 13 +++++--- util/qemu-error.c | 12 +++++--- 5 files changed, 57 insertions(+), 45 deletions(-)
include/monitor/monitor.h +4 -4
@@ -28,9 +28,9 @@ void monitor_resume(Monitor *mon);int monitor_get_fd (Monitor *mon,const char *fdname,Error **errp);int monitor_fd_param (Monitor *mon,const char *fdname,Error **errp); -void monitor_vprintf(Monitor *mon, const char *fmt, va_list ap) +int monitor_vprintf(Monitor *mon, const char *fmt, va_list ap) GCC_FMT_ATTR(2, 0); -void monitor_printf(Monitor *mon, const char *fmt, ...) GCC_FMT_ATTR(2, 3); +int monitor_printf(Monitor *mon, const char *fmt, ...) GCC_FMT_ATTR(2, 3); int monitor_fprintf(FILE *stream, const char *fmt, ...) GCC_FMT_ATTR(2, 3);void monitor_flush (Monitor *mon);int monitor_set_cpu (int cpu_index);
@@ -48,7 +48,7 @@ int monitor_fdset_dup_fd_add(int64_t fdset_id, int dup_fd);void monitor_fdset_dup_fd_remove (int dup_fd);int monitor_fdset_dup_fd_find (int dup_fd); -void monitor_vfprintf(FILE *stream, - const char *fmt, va_list ap) GCC_FMT_ATTR(2, 0); +int monitor_vfprintf(FILE *stream, + const char *fmt, va_list ap) GCC_FMT_ATTR(2, 0);#endif /* MONITOR_H */
include/qemu/error-report.h +4 -4
@@ -30,10 +30,10 @@ void loc_set_none(void);void loc_set_cmdline (char **argv,int idx,int cnt);void loc_set_file (const char *fname,int lno); -void error_vprintf(const char *fmt, va_list ap) GCC_FMT_ATTR(1, 0); -void error_printf(const char *fmt, ...) GCC_FMT_ATTR(1, 2); -void error_vprintf_unless_qmp(const char *fmt, va_list ap) GCC_FMT_ATTR(1, 0); -void error_printf_unless_qmp(const char *fmt, ...) GCC_FMT_ATTR(1, 2); +int error_vprintf(const char *fmt, va_list ap) GCC_FMT_ATTR(1, 0); +int error_printf(const char *fmt, ...) GCC_FMT_ATTR(1, 2); +int error_vprintf_unless_qmp(const char *fmt, va_list ap) GCC_FMT_ATTR(1, 0); +int error_printf_unless_qmp(const char *fmt, ...) GCC_FMT_ATTR(1, 2);void error_set_progname (const char *argv0); void error_vreport(const char *fmt, va_list ap) GCC_FMT_ATTR(1, 0);
monitor.c +33 -28
122 lines not on this page · raw message
stubs/error-printf.c +8 -5
@@ -2,19 +2,22 @@#include "qemu-common.h" #include "qemu/error-report.h" -void error_vprintf (const char *fmt,va_list ap) +int error_vprintf (const char *fmt,va_list ap) { +int ret; +if (g_test_initialized () && !g_test_subprocess () &&getenv ("QTEST_SILENT_ERRORS" )) {char *msg =g_strdup_vprintf (fmt, ap);g_test_message ("%s" , msg); + ret =strlen (msg);g_free (msg); - }else { -vfprintf (stderr, fmt, ap); +return ret; } +return vfprintf (stderr, fmt, ap); } -void error_vprintf_unless_qmp (const char *fmt,va_list ap) +int error_vprintf_unless_qmp (const char *fmt,va_list ap) { -error_vprintf (fmt, ap); +return error_vprintf (fmt, ap); }
util/qemu-error.c +8 -4
@@ -24,22 +24,26 @@ typedef enum {REPORT_TYPE_INFO , } report_type; -void error_printf (const char *fmt, ...) +int error_printf (const char *fmt, ...) {va_list ap; +int ret;va_start (ap, fmt); -error_vprintf (fmt, ap); + ret =error_vprintf (fmt, ap);va_end (ap); +return ret; } -void error_printf_unless_qmp (const char *fmt, ...) +int error_printf_unless_qmp (const char *fmt, ...) {va_list ap; +int ret;va_start (ap, fmt); -error_vprintf_unless_qmp (fmt, ap); + ret =error_vprintf_unless_qmp (fmt, ap);va_end (ap); +return ret; } static Location std_loc = {
Signature · 1 line
2.17.2
PATCH v2 12/15 qemu-print: New qemu_printf(), qemu_vprintf() etc.
In reply to Markus ArmbrusterWe commonly want to print to the current monitor if we have one, else to stdout/stderr. For stderr, have error_printf(). For stdout, all we have is monitor_vfprintf(), which is rather unwieldy. We often print to stderr just because error_printf() is easier.
New qemu_printf() and qemu_vprintf() do exactly what's needed. The next commits will put them to use.
--- MAINTAINERS | 2 ++ include/qemu/qemu-print.h | 19 ++++++++++++++++++ stubs/monitor.c | 5 +++++ tests/test-util-sockets.c | 1 + util/Makefile.objs | 1 + util/qemu-print.c | 42 +++++++++++++++++++++++++++++++++++++++ 6 files changed, 70 insertions(+) create mode 100644 include/qemu/qemu-print.h create mode 100644 util/qemu-print.c
MAINTAINERS +2 -0
@@ -1895,6 +1895,8 @@ F: hmp.[ch] F: hmp-commands*.hx F: include/monitor/hmp-target.h F: tests/test-hmp.c +F: include/qemu/qemu-print.h +F: util/qemu-print.c Network device backends M: Jason Wang <[email protected]>
include/qemu/qemu-print.h new file +19 -0
@@ -0,0 +1,19 @@ +/* +* Print to stream or current monitor +* +* Copyright (C) 2019 Red Hat Inc. +* +* Authors: +* Markus Armbruster <[email protected]>, +* +* This work is licensed under the terms of the GNU GPL, version 2 or later. +* See the COPYING file in the top-level directory. +*/ + +#ifndef QEMU_PRINT_H +#define QEMU_PRINT_H + +int qemu_vprintf(const char *fmt, va_list ap) GCC_FMT_ATTR(1, 0); +int qemu_printf(const char *fmt, ...) GCC_FMT_ATTR(1, 2); + +#endif
stubs/monitor.c +5 -0
@@ -6,6 +6,11 @@ __threadMonitor *cur_mon; +int monitor_vprintf (Monitor *mon,const char *fmt,va_list ap) +{ +abort (); +} +int monitor_get_fd (Monitor *mon,const char *name,Error **errp) {error_setg (errp,"only QEMU supports file descriptor passing" );
tests/test-util-sockets.c +1 -0
@@ -70,6 +70,7 @@ int monitor_get_fd(Monitor *mon, const char *fdname, Error **errp)* otherwise we get duplicate syms at link time. */ __threadMonitor *cur_mon; +int monitor_vprintf (Monitor *mon,const char *fmt,va_list ap) {abort (); }void monitor_init (Chardev *chr,int flags) {}
util/Makefile.objs +1 -0
@@ -23,6 +23,7 @@ util-obj-y += bitmap.o bitops.o hbitmap.o util-obj-y += fifo8.o util-obj-y += cacheinfo.o util-obj-y += error.o qemu-error.o +util-obj-y += qemu-print.o util-obj-y += id.o util-obj-y += iov.o qemu-config.o qemu-sockets.o uri.o notify.o util-obj-y += qemu-option.o qemu-progress.o
util/qemu-print.c new file +42 -0
42 lines not on this page · raw message
Signature · 1 line
2.17.2
PATCH v2 13/15 blockdev: Make -drive format=help print to stdout
In reply to Markus ArmbrusterCommand line help explicitly requested by the user should be printed to stdout, not stderr. We do elsewhere. Adjust -drive to match: use qemu_printf() instead of error_printf(). Plain printf() would be wrong because we need to print to the current monitor for "drive_add ... format=help".
--- blockdev.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
blockdev.c +5 -4
@@ -40,6 +40,7 @@#include "monitor/monitor.h" #include "qemu/error-report.h" #include "qemu/option.h" +#include "qemu/qemu-print.h" #include "qemu/config-file.h" #include "qapi/qapi-commands-block.h" #include "qapi/qapi-commands-transaction.h"
@@ -301,7 +302,7 @@ DriveInfo *drive_get_next(BlockInterfaceType type)static void bdrv_format_print (void *opaque,const char *name) { -error_printf (" %s" , name); +qemu_printf (" %s" , name); } typedef struct {
@@ -530,11 +531,11 @@ static BlockBackend *blockdev_init(const char *file, QDict *bs_opts,if ((buf =qemu_opt_get (opts,"format" )) !=NULL ) {if (is_help_option (buf)) { -error_printf ("Supported formats:" ); +qemu_printf ("Supported formats:" );bdrv_iterate_format (bdrv_format_print,NULL , false); -error_printf ("\nSupported formats (read-only):" ); +qemu_printf ("\nSupported formats (read-only):" );bdrv_iterate_format (bdrv_format_print,NULL , true); -error_printf ("\n" ); +qemu_printf ("\n" ); goto early_err; }
Signature · 1 line
2.17.2
PATCH v2 14/15 char: Make -chardev help print to stdout
In reply to Markus ArmbrusterCommand line help explicitly requested by the user should be printed to stdout, not stderr. We do elsewhere. Adjust -chardev to match:
use qemu_printf() instead of error_printf(). Plain printf() would be wrong because we need to print to the current monitor for "chardev-add help".
--- chardev/char.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
chardev/char.c +2 -1
@@ -28,6 +28,7 @@#include "sysemu/sysemu.h" #include "qemu/config-file.h" #include "qemu/error-report.h" +#include "qemu/qemu-print.h" #include "chardev/char.h" #include "qapi/error.h" #include "qapi/qapi-commands-char.h"
@@ -651,7 +652,7 @@ Chardev *qemu_chr_new_from_opts(QemuOpts *opts, GMainContext *context,chardev_name_foreach (help_string_append, str); -error_printf ("Available chardev backend types: %s\n" , str->str); +qemu_printf ("Available chardev backend types: %s\n" , str->str);g_string_free (str, true);return NULL ; }
Signature · 1 line
2.17.2
PATCH v2 15/15 monitor: Simplify how -device/device_add print help
In reply to Markus ArmbrusterCommit a95db58f210 added monitor_vfprintf() as an error_printf() generalized from stderr to arbitrary streams, then used it wrapped in helper out_printf() to print -device/device_add help to stdout. Use qemu_printf() instead, and delete monitor_vfprintf() and out_printf().
--- include/monitor/monitor.h | 3 --- monitor.c | 16 ++++------------ qdev-monitor.c | 36 ++++++++++++++---------------------- 3 files changed, 18 insertions(+), 37 deletions(-)
include/monitor/monitor.h +0 -3
@@ -48,7 +48,4 @@ int monitor_fdset_dup_fd_add(int64_t fdset_id, int dup_fd);void monitor_fdset_dup_fd_remove (int dup_fd);int monitor_fdset_dup_fd_find (int dup_fd); -int monitor_vfprintf(FILE *stream, - const char *fmt, va_list ap) GCC_FMT_ATTR(2, 0); -#endif /* MONITOR_H */
monitor.c +4 -12
@@ -4541,23 +4541,15 @@ static void monitor_readline_flush(void *opaque)monitor_flush (opaque); } -/* -* Print to current monitor if we have one, else to stream. -*/ -int monitor_vfprintf (FILE *stream,const char *fmt,va_list ap) -{ -if (cur_mon && !monitor_cur_is_qmp ()) { -return monitor_vprintf (cur_mon, fmt, ap); - } -return vfprintf (stream, fmt, ap); -} -/* * Print to current monitor if we have one, else to stderr. */ int error_vprintf (const char *fmt,va_list ap) { -return monitor_vfprintf (stderr, fmt, ap); +if (cur_mon && !monitor_cur_is_qmp ()) { +return monitor_vprintf (cur_mon, fmt, ap); + } +return vfprintf (stderr, fmt, ap); }int error_vprintf_unless_qmp (const char *fmt,va_list ap)
qdev-monitor.c +14 -22
@@ -31,6 +31,7 @@#include "qemu/error-report.h" #include "qemu/help_option.h" #include "qemu/option.h" +#include "qemu/qemu-print.h" #include "sysemu/block-backend.h" #include "migration/misc.h"
@@ -104,31 +105,22 @@ static bool qdev_class_has_alias(DeviceClass *dc)return (qdev_class_get_alias (dc) !=NULL ); } -static void out_printf (const char *fmt, ...) -{ -va_list ap; - -va_start (ap, fmt); -monitor_vfprintf (stdout, fmt, ap); -va_end (ap); -} -static void qdev_print_devinfo (DeviceClass *dc) { -out_printf ("name \"%s\"" ,object_class_get_name (OBJECT_CLASS (dc))); +qemu_printf ("name \"%s\"" ,object_class_get_name (OBJECT_CLASS (dc)));if (dc->bus_type) { -out_printf (", bus %s" , dc->bus_type); +qemu_printf (", bus %s" , dc->bus_type); }if (qdev_class_has_alias (dc)) { -out_printf (", alias \"%s\"" ,qdev_class_get_alias (dc)); +qemu_printf (", alias \"%s\"" ,qdev_class_get_alias (dc)); }if (dc->desc) { -out_printf (", desc \"%s\"" , dc->desc); +qemu_printf (", desc \"%s\"" , dc->desc); }if (!dc->user_creatable) { -out_printf (", no-user" ); +qemu_printf (", no-user" ); } -out_printf ("\n" ); +qemu_printf ("\n" ); }static void qdev_print_devinfos (bool show_no_user)
@@ -164,7 +156,7 @@ static void qdev_print_devinfos(bool show_no_user)continue ; }if (!cat_printed) { -out_printf ("%s%s devices:\n" , i ?"\n" :"" , cat_name[i]); +qemu_printf ("%s%s devices:\n" , i ?"\n" :"" , cat_name[i]); cat_printed = true; }qdev_print_devinfo (dc);
@@ -286,20 +278,20 @@ int qdev_device_help(QemuOpts *opts) }if (prop_list) { -out_printf ("%s options:\n" , driver); +qemu_printf ("%s options:\n" , driver); }else { -out_printf ("There are no options for %s.\n" , driver); +qemu_printf ("There are no options for %s.\n" , driver); }for (prop = prop_list; prop; prop = prop->next) {int len; -out_printf (" %s=<%s>%n" , prop->value->name, prop->value->type, &len); +qemu_printf (" %s=<%s>%n" , prop->value->name, prop->value->type, &len);if (prop->value->has_description) {if (len <24 ) { -out_printf ("%*s" ,24 - len,"" ); +qemu_printf ("%*s" ,24 - len,"" ); } -out_printf (" - %s\n" , prop->value->description); +qemu_printf (" - %s\n" , prop->value->description); }else { -out_printf ("\n" ); +qemu_printf ("\n" ); } }
Signature · 1 line
2.17.2
PATCH v2 1/1 char-pty: Print "char device redirected" message to stdout
In reply to Markus Armbrusterchar_pty_open() prints a "char device redirected to PTY_NAME (label LABEL)" message to the current monitor or else to stderr. This is not an error, so it shouldn't go to stderr. Print it to stdout instead.
Why is it even printed? No other ChardevClass::open() prints anything on success. It's because you need to know PTY_NAME to actually use this char device, e.g. like e.g. "socat STDIO,cfmakeraw FILE:PTY_NAME" to use the monitor's readline interface. You can get PTY_NAME with "info chardev" (a.k.a. query-chardev for QMP), but only if you already have a monitor.
--- chardev/char-pty.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
chardev/char-pty.c +2 -2
@@ -211,8 +211,8 @@ static void char_pty_open(Chardev *chr,qemu_set_nonblock (master_fd); chr->filename =g_strdup_printf ("pty:%s" , pty_name); -error_printf ("char device redirected to %s (label %s)\n" , - pty_name, chr->label); +qemu_printf ("char device redirected to %s (label %s)\n" , + pty_name, chr->label); s =PTY_CHARDEV (chr); s->ioc =QIO_CHANNEL (qio_channel_file_new_fd (master_fd));
Signature · 1 line
2.17.2
Re: [Qemu-devel] [PATCH v2 14.5/15] char-pty: Print "char device redirected" message to stdout
Patchew URL: https://patchew.org/QEMU/[email protected]/
Patchew URL: https://patchew.org/QEMU/[email protected]/
Hi,
This series failed the asan build test. Please find the testing commands and their output below. If you have Docker installed, you can probably reproduce it locally.
=== TEST SCRIPT BEGIN ===
#!/bin/bash
time make docker-test-debug@fedora TARGET_LIST=x86_64-softmmu J=14 NETWORK=1 === TEST SCRIPT END ===
AR libqemuutil.a LINK elf2dmp AS optionrom/multiboot.o
/tmp/qemu-test/src/chardev/char-pty.c:214:5: error: implicit declaration of function 'qemu_printf' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
qemu_printf("char device redirected to %s (label %s)\n",
^/tmp/qemu-test/src/chardev/char-pty.c:214:5: error: this function declaration is not a prototype [-Werror,-Wstrict-prototypes]
2 errors generated.
LINK ivshmem-client
LINK ivshmem-server
The full log is available at
http://patchew.org/logs/[email protected]/testing.asan/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to [email protected]