Daniel Veillard wrote:

On Mon, Feb 01, 2010 at 10:18:27PM +0100, Jim Meyering wrote:

...

In src/qemu/qemu_driver.c, coverity reports this:

Event negative_return_fn: Called negative-returning function "lseek(logfile, 0L, 2)" Event var_assign: NEGATIVE return value of "lseek" assigned to signed variable "pos" At conditional (1): "(pos = lseek(logfile, 0L, 2)) < 0" taking true path 2877 if ((pos = lseek(logfile, 0, SEEK_END)) < 0) 2878 VIR_WARN(_("Unable to seek to end of logfile: %s"), 2879 virStrerror(errno, ebuf, sizeof ebuf));

since later in that same function, a negative "pos" may be used like this:

Event negative_returns: Tracked variable "pos" was passed to a negative sink. [details] 2930 if (qemudWaitForMonitor(conn, driver, vm, pos) < 0) 2931 goto abort; 2932

which is a legitimate problem, since qemudWaitForMonitor calls qemudLogReadFD, which calls lseek with that same "pos" value:

Event neg_sink_parm_call: Parameter "pos" passed to negative sink "lseek" 560 if (lseek(fd, pos, SEEK_SET) < 0) { 561 virReportSystemError(conn, errno, 562 _("Unable to seek to %lld in %s"), 563 (long long) pos, logfile); 564 close(fd); 565 }

One approach is to detect the negative offset in that final bit of code and skip the lseek:

...

From 0ef617935462c314ed0b44bcaa3dd5bf58ccbc1b Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering@redhat.com> Date: Mon, 1 Feb 2010 22:17:44 +0100 Subject: [PATCH] avoid a probable EINVAL from lseek

* src/qemu/qemu_driver.c (qemudLogReadFD): Don't pass a negative offset (from a preceding failed attempt to seek to EOF) to this use of lseek. --- src/qemu/qemu_driver.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 22593bf..676a27b 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -558,8 +558,8 @@ qemudLogReadFD(virConnectPtr conn, const char* logDir, const char* name, off_t p close(fd); return -1; } - if (lseek(fd, pos, SEEK_SET) < 0) { - virReportSystemError(conn, errno, + if (pos < 0 || lseek(fd, pos, SEEK_SET) < 0) { + virReportSystemError(conn, pos < 0 ? 0 : errno, _("Unable to seek to %lld in %s"), (long long) pos, logfile); close(fd);

I was wondering if it wasn't simpler to abort earlier on when pos < 0 was returned from lseek, but after rereading the code I agree with your patch,

Returning early (failing) after the initial log-lseek failure did not seem justified, considering other log-related failures merely get a warning there. Pushed.

Daniel P. Berrange wrote:

On Mon, Feb 01, 2010 at 10:18:27PM +0100, Jim Meyering wrote:

...

In src/qemu/qemu_driver.c, coverity reports this:

Event negative_return_fn: Called negative-returning function "lseek(logfile, 0L, 2)" Event var_assign: NEGATIVE return value of "lseek" assigned to signed variable "pos" At conditional (1): "(pos = lseek(logfile, 0L, 2)) < 0" taking true path 2877 if ((pos = lseek(logfile, 0, SEEK_END)) < 0) 2878 VIR_WARN(_("Unable to seek to end of logfile: %s"), 2879 virStrerror(errno, ebuf, sizeof ebuf));

I think it'd be less surprising to just set 'pos = 0' inside the if branch here, so later code doesn't have to worry about unexpected negative values.

Oh. I pushed after DV's ACK. That would let the later code continue, but using (lseek'ing to) an invalid position. Sounds like it could result in a cascade of additional errors, or worse, silent malfunction. But this is largely hypothetical, since failing to lseek-to-EOF on a valid file descriptor is not likely to happen.

...

diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 22593bf..676a27b 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -558,8 +558,8 @@ qemudLogReadFD(virConnectPtr conn, const char* logDir, const char* name, off_t p close(fd); return -1; } - if (lseek(fd, pos, SEEK_SET) < 0) { - virReportSystemError(conn, errno, + if (pos < 0 || lseek(fd, pos, SEEK_SET) < 0) { + virReportSystemError(conn, pos < 0 ? 0 : errno, _("Unable to seek to %lld in %s"), (long long) pos, logfile); close(fd);